Marina Paola Banchetti-Robino - The Chemical Philosophy of Robert Boyle - Mechanicism, Chymical Atoms, and Emergence-Oxford University Press (2020)

The Chemical Philosophy of Robert Boyle
The Chemical
Philosophy
of Robert Boyle
Mechanicism, Chymical Atoms,
and Emergence
M A R I NA PAO L A BA N C H E T T I - R
OBINO
1
3
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2020
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.
You must not circulate this work in any other form

and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data

Names: Banchetti-Robino, Marina Paola, author.
Title: The chemical philosophy of Robert Boyle : mechanicism, chymical atoms, and emergence /
Marina Paola Banchetti-Robino.
Description: New York, NY : Oxford University Press, [2020] |
Includes bibliographical references and index.
Identifiers: LCCN 2020008358 (print) | LCCN 2020008359 (ebook) |
ISBN 9780197502501 (hardback) | ISBN 9780197502525 (epub)
Subjects: LCSH: Boyle, Robert, 1627–1691. | Chemistry—History—17th century. |
Chemistry—Philosophy—History—17th century.
Classification: LCC QD22.B 76 B36 2020 (print) |
LCC QD22.B 76 (ebook) | DDC 540.1—dc23
LC record available at https://lccn.loc.gov/2020008358
LC ebook record available at https://lccn.loc.gov/2020008359
1 3 5 7 9 8 6 4 2
Printed by Integrated Books International, United States of America
For my dear Ariel
and
my beloved parents
Contents
Preface ix
Introduction 1
1. Chemical Philosophy in the 16th and 17th Centuries: Vitalism,
Paracelsian Alchemy, and Aristotelian Hylomorphism 8
1.1 The Vitalistic Character of Renaissance Alchemy 11
1.2 The Scholastic Theory of Substantial Form 13
1.3 Paracelsian Spagyria and the Tria Prima 16
1.4 Semina Rerum, Minima Naturalia, and Vitalistic Corpuscularianism 20
1.5 Daniel Sennert’s Structural Hylomorphism and Atomicity as a
Negative-Empirical Concept 26
1.6 Jan Baptista van Helmont and the Chemical Interpretation of Spirit
and Ferment 34
2. Chemical Philosophy vs. Rationalistic Mechanicism: The Heuristic
Limits of Cartesianism for Chemistry 45
2.1 The Cartesian Rejection of Substantial Forms 45
2.2 Pierre Gassendi and the Reformation of Epicurean Atomism 47
2.3 The Limitations of the Cartesian Project for Chemistry and
Chemical Philosophy 51
2.4 Mechanistic Corpuscularianism and Experimental Natural
Philosophy 60
2.5 Boyle’s Relation to the Cartesian Project in Natural Philosophy 69
2.6 The Negative and Positive Heuristic Functions of the Mechanical
Philosophy in Boyle’s Scientific Research Programme 73
3. The Ontological Complexity of Boyle’s Corpuscularian
Theory: Microstructure, Natural Kinds, and Essential Form 79
3.1 The Sceptical Chymist: Against Scholastics and Spagyrists 79
3.2 Boyle’s Corpuscularian Theory of Matter 84
3.3 Composition vs. Microstructure 90
3.4 Taxonomical Classification, Natural Kinds, and Essential Form 99
3.5 The Empirical Nature of Essential Form: The Reduction to the
Pristine State 104
viii Contents
4. Boyle’s View of Chemical Properties as Dispositional, Relational,

and Emergent Properties 109
4.1 The Hierarchy of Properties in Boyle’s Chemical Ontology 110
4.2 Sensible Properties as Dispositional and Relational 113
4.3 Chemical Properties as Dispositional and Relational 121
4.4 Chemical Properties as Emergent and Supervenient 130
4.5 Supervenience, Non-Summative Difference, and
Underdetermination 138
4.6 Cosmical Qualities as Dispositional and Relational Properties 139
5. The Relation between Parts and Wholes: The Complex Mereology
of Chymical Atoms 146
5.1 Boylean Chemistry as Mereological 147
5.2 Continuous vs. Contiguous Integral Wholes 151
5.3 Integral Parts and Essential Parts 153
5.4 Aquinas, Abelard, and Boyle on Substantial Unity 154
5.5 The Mereology of Boyle’s Chymical Atoms as Chemically
Elementary Entities 156
5.6 A Brief Excursion into the Mereology of Epicurean Semantics 163
Conclusion 168
Bibliography 173
Index 191
Preface
Writing this book was both more challenging than I ever anticipated and more
rewarding than I ever imagined. Yet, although only one author’s name appears
on its cover, this book is a product not only of my own work but is indirectly also
a product of the people who have stood by me and my choices throughout the
years. First and foremost, this project would not have come to fruition without
the unconditional love and support of my husband and philosophical soul mate,
Clevis Headley, whose constant encouragement was invaluable to the progress
of my work. I am also grateful for the warm friendship and generous guidance of
Hasok Chang, Jean-Pierre Llored, Eric Scerri, Brigitte van Tiggelen, and the late
Rom Harré. Their pioneering work in the history and philosophy of chemistry
has been truly inspiring and has served as the standard of excellence that I have
sought to emulate in my own work.
I am indebted to my institution, Florida Atlantic University, for generously
supporting the writing of this book with a sabbatical semester in spring 2015,
and I also want to express my appreciation to my former student Ms. Cameron
Black for her detailed and meticulous proofreading of the final draft of the man-
uscript. I want to thank Jeremy Lewis at Oxford University Press and his editorial
assistant, Bronwyn Geyer, for their unwavering support of this project. They are
both truly a pleasure to work with. Finally, I am also heavily indebted to the two
anonymous reviewers of my original book proposal for their close and careful
reading, their constructive criticisms, and their many thoughtful comments and
suggestions, which helped to greatly improve the final version of this work.
I benefited considerably from presenting papers on the various topics of
this book at numerous conferences, which included those of the International
Society for the Philosophy of Chemistry, the Philosophy of Science Association,
the History of Science Society, the Science History Institute (formerly the
Chemical Heritage Foundation), and the Seminar on the History and Philosophy
of Chemistry at the University of Paris.
Some sections of this book draw upon and revise material previously published
in the following articles and chapters: “Ontological Tensions in Sixteenth
and Seventeenth Century Chemistry: Between Mechanism and Vitalism,”
Foundations of Chemistry 13 (2011), 173– 186; “The Ontological Function
of First-Order and Second- Order Corpuscles in the Chemical Philosophy
of Robert Boyle: The Redintegration of Potassium Nitrate,” Foundations of
Chemistry 14 (2012), 221–234; “The Relevance of Boyle’s Chemical Philosophy
x Preface
for Contemporary Philosophy of Chemistry,” in Jean- Pierre Llored (ed.),

Philosophy of Chemistry: Practices, Methodologies, and Concepts (Newcastle upon
Tyne: Cambridge Scholars Publishing, 2013), 240–265; “From Corpuscles to
Elements: Chemical Ontologies from Van Helmont to Lavoisier,” in Lee McIntyre
and Eric Scerri (eds.), Philosophy of Chemistry: Growth of a New Discipline
(Dordrecht: Springer, 2014), 141–154; “Van Helmont’s Hybrid Ontology and Its
Influence on the Chemical Interpretation of Spirit and Ferment,” Foundations
of Chemistry 18 (2015), 103–112; “Reality without Reification: Philosophy of
Chemistry’s Contribution to the Philosophy of Mind,” in Grant Fisher and Eric
Scerri (eds.), Essays in the Philosophy of Chemistry (Oxford: Oxford University
Press, 2016), 83–110; and “The Function of Microstructure in Boyle’s Chemical
Philosophy: ‘Chymical Atoms’ and Structural Explanation,” Foundations of
Chemistry 21 (2019), 51–59.
Although the writing of this book would not have been possible without the
patience, advice, and constant support of my husband Clevis, I am also eternally
grateful to my lifelong friend, Rebekah Guerra, for standing by me through all
my struggles and successes and to my aunts, Vittoria and Giuseppina Robino, for
their continued love, generosity, and support.
I wish to dedicate this book to the memory of my beloved parents, Paris
Edoardo Banchetti and Fortunata Robino di Banchetti. My mother, Fortunata,
passed away in 1996 as I was beginning my teaching and scholarly career at
Florida Atlantic University. My father, Paris, passed away in March 2019, only
one week prior to my receiving final word from Oxford University Press that
my book had been accepted for publication. I believe that they would have been
proud. Throughout their lives, my parents served as models of dedication, per-
severance, and integrity and their influence made me who I am today. Their un-
conditional love and support will continue to sustain me for the rest of my life.
Finally, I also dedicate this book to one of my sweet and beautiful cats, Ariel
Mahogany, who was diagnosed with cancer in January 2015 as I began my sab-
batical semester. The months following her diagnosis were spent taking regular
breaks from the writing of this book to feed and medicate Ariel, in a desperate
attempt to keep her cancer in remission. Each and every day, I wrote incessantly
as she held on to life, sleeping next to me as I worked and at first even showing
signs of improvement. After six months, the time that it took for me to finish the
manuscript, her little body succumbed to the disease and she passed away one
week after I wrote the final words of the conclusion. Ariel lived just long enough
to see me finish this book and her soul lies deep within its pages. I will always feel
grateful and honored that this beautiful creature chose us as her family. Although
she was here only for a moment, her unconditional love will live in my heart
forever.
Introduction
In the introduction to his book The Aspiring Adept: Robert Boyle and His
Alchemical Quest, Lawrence Principe states that “The work I have done on
Robert Boyle since 1998 has actually been an extended digression.”1 Following
this seemingly surprising statement, Principe goes on to recount how, in the
process of researching the work of the mysterious 15th century alchemist Basil
Valentine, he came across a 19th-century text referencing Boyle’s experiments
on the volatility of gold. Surprised by the similarity between Boyle’s experiments
and those described by Valentine, Principe decided to investigate Boyle’s
writings further, particularly regarding what these revealed about Boyle’s interest
in chrysopoeia. The discovery of Boyle’s significant involvement with alchemy
thus pulled Principe away from his researches on Valentine and resulted in the
groundbreaking book cited above.
I open this introduction with Principe’s account precisely because, as I read
his words, they struck a familiar chord with me. My own work on Boyle for the
past 10 years, and its culmination in this book, are also the result of an extended
digression. In 2008, while working on a different project, I researched early
modern theories of matter and, in particular, the transition between the vital-
istic theories of the alchemical tradition and the mechanistic theories that came
to define early modern science. To this end, I came across two books that altered
my perspective on early modern chemistry and, in the process, also altered the
path of my research. These books were Antonio Clericuzio’s Elements, Principles,
and Corpuscles: A Study of Atomism and Chemistry in the Seventeenth Century
and Hiro Hirai’s incredibly erudite Le concept de semence dans les theories de la
matière à la Renaissance: de Marsile Ficin à Pierre Gassendi. Since I was only be-
ginning to research this topic at the time, I was surprised to discover by reading
these fascinating texts that vitalism and alchemical concepts had continued to
influence theories of matter well into the 17th century. I thus decided to further
research these ideas and was inevitably led to the work of Robert Boyle, whose
1 Principe, The Aspiring Adept, 5.
The Chemical Philosophy of Robert Boyle. Marina Paola Banchetti-Robino, Oxford University Press (2020). © Oxford
University Press.
DOI: 10.1093/oso/9780197502501.001.0001
2 Introduction
relationship to alchemy and to the mechanistic theory of matter have been the
subject of much discussion and revaluation for the past 25 years.
This revaluation has certainly been justified since, as Michael Hunter points
out in his important collection Robert Boyle Reconsidered, Boyle is “a figure
of undeniable significance.”2 This significance is not limited to his scientific
contributions but is also philosophical, particularly with regard to Boyle’s rec-
onciliation of the mechanistic conception of fundamental particles with his
complex and nuanced ontology of chemical properties and substances. In fact,
Hunter identifies a need to “abandon simplistic preconceptions about the me-
chanical philosophy in the seventeenth century”3 as they relate to Boyle’s chem-
ical work, pointing out that Boyle was a sophisticated thinker who understood
that “the principles of Cartesian mechanism were too limited, and that ulti-
mately, a true new philosophy would emerge that would do justice to consider-
ations of this kind.”4
Boyle is a much more important philosophical figure than had been pre-
viously realized and his philosophical contributions are crucial to our un-
derstanding of the creation of modern chemistry and physics. This is the case
because they help to define the transition between the physics of the mechan-
ical philosophy and the modern conception of chemistry. Boyle’s writings reveal
that the mechanical philosophy was just one prominent feature of a very distinc-
tive and well-developed chemical philosophy. As Peter Anstey’s important book
The Philosophy of Robert Boyle shows, Boyle made important and significant
contributions to early modern philosophical discussions regarding the theory of
qualities, of matter in motion, and of mind/body interaction.
I have taken much inspiration from Anstey’s work, especially from his excel-
lent argument that Boyle regarded sensible qualities as dispositional and rela-
tional properties. However, although Anstey suggests that Boyle also regarded
chemical properties as dispositional and relational, he never pursues this argu-
ment and restricts himself to the discussion of sensible qualities. I have there-
fore taken it upon myself to extend Anstey’s arguments to Boyle’s conception
of chemical properties, while going a step further and also arguing that Boyle
regarded such properties as emergent and supervenient. Furthermore, since the
philosophy of chemistry recognizes the intimate connection between the emer-
gence and supervenience of chemical properties and the relation between chem-
ical wholes and their parts, I considered it imperative also to examine the kind
of mereology of chymical atoms that is most consistent with Boyle’s emergentist
conception of chemical qualities. Therefore, the two main goals of this book are,
2 Hunter, Robert Boyle Reconsidered, 5.

3 Ibid., 8.
4 Ibid.
Introduction 3
first, to demonstrate that Boyle regarded chemical properties as properties that

emerge from and supervene upon the mechanical and structural properties of
chymical atoms and, second, to demonstrate that such an emergentist position
entails a non-extensional and non-summative mereology of chymical atoms.
The emergentist reading of Boyle and the mereological theory of chymical atoms
that I am proposing are both novel contributions to the scholarship on Boyle,
and I am convinced that a sustained analysis of these ideas can significantly en-
rich our understanding of Boyle’s chemical philosophy.
To explain the structure of this book, I must first say that I agree with
Michael Hunter’s view that, although contextual considerations must always be
addressed, Boyle studies should also remain sensitive to “the power and com-
plexity of intellectual traditions in their own right.”5 Consequently, I follow
Hunter’s admonition to balance the contextual historiographical approach with
a critical analysis of “the content of his ideas and the extent to which the intel-
lectual developments with which he was associated had an internal momentum
of their own.”6 Therefore, I not only situate Boyle’s ideas within the intellectual
context of the early modern period and in relation to the intellectual traditions
that helped shape that context, but I also thoroughly examine the internal con-
tent and logic of Boyle’s ideas regarding chymical atoms and chemical properties.
Additionally, I consider the textual evidence that supports the emergentist posi-
tion and non-extensional mereology that I am proposing.
To this end, the book begins by setting the historical context within which to
situate Boyle’s ideas. I pursue this task by examining the dominant paradigms
of 16th-and 17th-century chemical philosophy. More particularly, Chapter 1
focuses on three of these important paradigms, the doctrine of vitalism to which
Cartesian mechanicism was responding, the theory of substantial form that had
dominated Scholastic science, and the Paracelsian tria prima that functioned as
the theory of principles for Renaissance and early modern alchemists. This first
chapter also discusses other important concepts that are relevant throughout
the book, such as the notions of semina rerum and minima naturalia and the
idea that corpuscularianism was compatible with both vitalistic and mechanistic
theories of matter. The chapter concludes with detailed discussions of the vital-
istic corpuscularian chemical ontologies of Daniel Sennert and Jan Baptista van
Helmont. These respective ontologies both served as important antecedents to
the chemical philosophy of Robert Boyle and also influenced some of his more
important chemical experiments.
Chapter 2 contextualizes Boyle’s distinctive mechanistic corpuscularianism
within the larger framework of the mechanical philosophy by, first, providing
5 Ibid., 5.
6 Ibid.
4 Introduction
a detailed account of the relationship between Cartesian mechanicism and

early modern chemistry. This discussion establishes that an intrinsic ten-
sion existed between Cartesian mechanicism and chemical philosophy, par-
ticularly regarding the heuristic value and scientific autonomy of chemistry.
The chapter then discusses the revival of Epicurean atomism in early modern
Europe and contrasts Gassendi’s mechanistic atomism with Descartes’ mech-
anistic corpuscularianism to show why the former view was more compatible
with chemical experimental practice. The chapter also examines the competing
explanations of Boyle and Spinoza regarding the redintegration of potassium ni-
trate to establish why Boyle rejected Spinoza’s attempt to explain experimental
results strictly in terms of the positions and motions of fundamental particles.
The chapter concludes with a discussion of the role played by the mechanistic
philosophy in Boyle’s overall experimental program. Here, Boyle’s experi-
mental work in chemistry, pneumatics, and hydrostatics is considered as a type
of Lakatosian research programme within which the mechanical philosophy
functions as the “metaphysical hard core.” As such, the mechanical philosophy
serves as a negative heuristic, that is, as a limiting principle that forbids explana-
tory appeal to principles that are incompatible with it. As the “metaphysical hard
core” of the Boylean research programme, the mechanical philosophy also serves
as a positive heuristic by providing an explanatory model based on empirical
clockwork mechanisms and simple machines.
Following the historical and philosophical context provided by the first two
chapters, Chapter 3 engages in a thorough and detailed account of Boyle’s chem-
ical ontology. This chapter begins by discussing Boyle’s rejection of two key
principles of Scholastic and Renaissance chemical ontology, that is, the theory
of substantial forms and the Paracelsian theory of the tria prima. The chapter
then examines Boyle’s mechanistic corpuscularian theory of matter, focusing
particularly on the role played by microstructure in determining chemical prop-
erties. Boyle’s corpuscularian chemistry presupposes stable concretions of fun-
damental particles and I follow William Newman in referring to these stable
concretions as “chymical atoms.” Boyle’s chymical atoms are structures resulting
from combinations of fundamental particles possessing only shape, size, and
motion. However, Boyle accounts for the chemical properties of substances in
terms of the microstructure of such chymical atoms, rather than by appealing
to the shapes, sizes, and motions of their constituent fundamental particles.
Boyle refers to the microstructure of chymical atoms as their texture or essential
form, and he regards essential form as a mechanical property of chymical atoms,
thereby accounting for stability in mechanistic terms rather than by appealing
to mysterious entities such as substantial forms. Because microstructure is what
accounts for a substance’s essential chemical properties, it is that which makes
a particular substance the kind of substance that it is and that which allows for
Introduction 5
the taxonomical classification of material species. Here, the book compares and
contrasts Boyle’s views on taxonomical classification with those of Locke, with
whom Boyle engaged in numerous debates. Finally, the chapter concludes with
a close examination of the experiment of reduction to the pristine state, an ex-
periment that Boyle inherits from Sennert but which he uses to argue against
the concept of substantial form and in favor of microstructure as the source of
the stability of chymical atoms, even when these are combined, separated, and
transposed through chemical reactions. Boyle purports to demonstrate that,
since chemical substances are composed of such stable concretions, they pre-
serve their identity through various chemical processes. The ability to recover
such substances from heterogeneous mixts thus lends confirmation to the con-
cept of essential form or structure as the source of chemical stability.
Following upon this discussion of Boyle’s chemical ontology, Chapter 4 argues
that Boyle holds an emergentist conception of chemical qualities, also regarding
them as dispositional and relational properties. This discussion is relevant to
more recent debates in the philosophy of chemistry on the emergence of chem-
ical properties and on the reducibility of chemistry to physics. In this chapter,
I build upon Anstey’s very thorough arguments for the dispositionality and
relationality of sensible qualities and I extend his analysis to chemical proper-
ties. I then argue that, in addition to considering such properties to be disposi-
tional and relational, Boyle also considers them to be emergent and supervenient
properties. I then conclude the chapter by discussing Boyle’s conception of cos-
mical qualities such as gravity and magnetism to demonstrate that, although he
regards them as properties of corpuscular effluvia, he also regards these qualities
as dispositional and relational properties.
To the extent that mereological questions regarding chemical wholes and
their parts are intimately connected to emergentist accounts of chemical prop-
erties, the final chapter focuses on the mereology that is implied by Boyle’s con-
ception of chymical atoms. A lot of recent work in the philosophy of chemistry,
particularly the collaborative work of Jean-Pierre Llored and Rom Harré, has
focused on the need to develop a non-extensional, non-summative mereology
for quantum chemical wholes to account for the emergence of chemical prop-
erties from such wholes. What I propose, in Chapter 5, is that the mereology
of Boylean chymical atoms must also be a non-extensional mereology for the
following two reasons. First, chymical atoms are more than simply the summa-
tion of their mechanistic fundamental parts and, second, novel properties are
generated by the structural arrangement of their parts. In order to develop this
argument, the chapter first examines in detail the leading mereological theories
of Boyle’s time, particularly those theories that had been developed by medi-
eval logicians and metaphysicians. The purpose of this discussion is to deter-
mine which of these theories is most consistent with the Boylean conception of
6 Introduction
chymical atoms, that is, with a conception of chemical wholes from which emerge
novel properties that are not properties of their parts. Based on this mereological
analysis, the chapter then proposes that Boyle regarded chymical atoms as chem-
ically elementary entities, despite their ontological reducibility to fundamental
mechanistic particles, thereby anticipating the conception of elementarity later
proposed by John Dalton. The chapter ends by discussing Boyle’s explicit analogy
between Lucretius’ conception of the relation between individual letters and lin-
guistic expressions and his own conception of the relation between fundamental
particles and chymical atoms. Once this analogy is unpacked, it reveals a great
deal regarding the manner in which Boyle conceptualizes the mereological re-
lation between chemical wholes and their parts. In fact, the chapter extends
Boyle’s analogy to help support the claim that his conception of chymical atoms,
as wholes from whose microstructure emerge chemical properties, is most
consistent with a non-extensional and non-summative mereology of wholes
and parts.
The conclusion recapitulates many of the book’s arguments and establishes
the relevance of Boyle’s chemical philosophy for contemporary debates in the
philosophy of chemistry. More specifically, the book reviews Boyle’s position re-
garding the microstructure of chymical atoms, the relational and dispositional
nature of chemical properties, as well as their emergence and supervenience, and
the autonomy of chemical explanations, and it compares these views to similar
positions in contemporary philosophy of chemistry. The conclusion establishes
that Boyle’s position on these matters was prescient and, in many regards, also
ahead of its time, thereby justifying the current interest in Boyle’s significant
philosophical contributions.
It is my hope that this book will not simply contribute to our understanding
of Boyle’s nuanced and sophisticated chemical philosophy but, more important,
that it will situate the relevance of Boyle’s ideas in relation to the contemporary
philosophy of chemistry. The philosophy of chemistry is the youngest subfield
within the philosophy of science, and it is still in the process of defining its re-
lation to the history of chemistry. Among other things, this book contributes to
this process by highlighting the relevance of historical considerations for con-
temporary issues in the philosophy of chemistry. Some of the more important
of these issues pertain to reduction, emergence, and supervenience, and there is
great historical and philosophical significance in establishing that these concerns
also informed the theoretical and experimental work of Boyle. This book also
establishes that Boyle was deeply engaged with the philosophical implications of
the practice of chemistry, particularly as it related to the mechanistic theory of
matter. Furthermore, Boyle’s theories were not only associated with the mecha-
nistic corpuscularianism of his time but were also intimately linked to themes in
medieval and Renaissance philosophy.
Introduction 7
Although the study of such an important historical figure as Boyle is valu-

able in its own right, this study also has great significance for our under-
standing of early modern science to the extent that Boyle made important
contributions to discussions on scientific taxonomies, the nature of chem-
ical elements and chemical properties, natural kinds, and the debates over at-
omism vs. corpuscularianism. His important philosophical contributions are
crucial for our understanding of the creation of modern chemistry and physics.
Accordingly, this book takes a careful and critical look at how Boyle helps to de-
fine the transition from the physics of the mechanical philosophy to a modern
conception of chemistry.
1
Chemical Philosophy in
the 16th and 17th Centuries
Vitalism, Paracelsian Alchemy,
and Aristotelian Hylomorphism
Vitalism is generally regarded as the view that “vital forces” or “vital spirits”
exist and function as causal agents in nature, and that the presence of vital force
or spirit marks the difference between organic and inorganic matter. Vitalistic
descriptions of natural phenomena tend to be qualitative, and vitalistic pro-
cesses are generally described as holistic and teleological. Most important for our
purposes, vitalism views the causes of motion as inherent in matter and treats
all of nature as if it were intrinsically self-organizing.1 Throughout the history
of both speculative and natural philosophy, vitalistic theories have been over-
laid with theological overtones of one sort or another, and the vitalistic theories
discussed herein are no exception. Medieval and Renaissance metaphysicians
and natural philosophers believed themselves to be living in an enchanted uni-
verse, in which matter was not inert but either was itself animate (i.e., it contained
a “world soul” or anima mundi) or was inhabited by vital forces and spirits that
played a causal role in the occurrence of natural phenomena. Depending on the
theological proclivities of these philosophers, the presence of a world soul or of
vital forces and spirits was to be ultimately attributed either to divine emanation
or to divine action.
Another characteristic of Renaissance vitalism was its affirmation of a funda-
mental correspondence between what is above and what is below. This micro-
cosm/macrocosm analogy “was at the center of a group of ideas derived from the
[. . .] mystical-alchemical tradition crossed with themes common to Neoplatonic
mysticism. The vital substances of objects [were] made up of invisible spirits or
forces of nature.”2 Natural philosophers, or natural magicians as they were called
in the Renaissance, were individuals who not only studied these vital forces and
correspondences but also learned how to deploy them for controlling or altering
natural phenomena. Vitalism dominated natural philosophy during the 15th
1 Bloor, “Durkheim and Mauss Revisited,” 78–80.

2 Rossi, The Birth of Modern Science, 141.
University Press.
DOI: 10.1093/oso/9780197502501.001.0001
Chemical Philosophy in the 16th and 17th Centuries 9
and 16th centuries due to the Renaissance revival of Neoplatonism and hermeti-
cism. These ideas infused the work of such dominant thinkers as Marsilio Ficino,
Tommaso Campanella, Cornelius Agrippa, and Giordano Bruno and they con-
tinued to influence natural philosophy well into the 17th century.
The late 16th to 17th centuries, however, witnessed a significant transition
from the vitalistic ontology that had dominated medieval and Renaissance
natural philosophy to the early modern mechanistic paradigm endorsed by
Cartesian and Epicurean natural philosophers, who rejected the presence of vital
forces in nature and regarded matter as inert rather than as inherently active.
This transition was most strongly felt in the context of early modern chymistry,
which gradually severed its ties to Scholastic and spagyric alchemy, thereby
opening the way for mechanistic approaches to chemical philosophy.3 It is point-
less to speculate about what chemistry might have become had this ontological
shift never occurred, but one cannot deny that the change from a vitalistic to a
mechanistic ontology significantly shaped the evolution of early modern chem-
ical philosophy. The transition from vitalism to mechanicism was not abrupt,
however, and close examination reveals that 16th-and 17th-century chemical
philosophies involved very complex and nuanced ontologies that often straddled
the line between vitalism and mechanicism. I will argue that some of these hy-
brid ontologies helped to shape early modern chemical philosophy and, more
specifically, the chemical ontology of Robert Boyle.
The ontological shift from vitalism to mechanicism occurred concurrently
with the revival and reappraisal of Epicurean atomism, which in its ancient form
had been strictly materialistic and deterministic. However, although Epicurean
atomism was revived in the 17th century, corpuscularian theories of matter
had coexisted quite comfortably with vitalism throughout the Middle Ages and
the Renaissance. It would, therefore, be erroneous to claim that particulate or
3 I wish to clarify some of the terminology that I will be using in the rest of this book. The terms
“chymist,” “chymical,” and “chymistry” are employed in this book, as in recent historiography of
science, to indicate the transitional phase between the alchemy of the 15th, 16th, and early 17th
centuries, an empirical practice with obvious magico-theological characteristics, and the modern
chemistry that emerged out of the 18th-century Chemical Revolution. In the early 17th century,
chymists tended to hybridize the magico-theological elements of alchemy with the naturalism char-
acteristic of modern chemistry. Although the mid-17th century saw chymists slowly shedding the
mythopoetic conception of nature, their work is still considered transitional because it retained many
strongly speculative elements that were later rejected by Lavoisier. On the other hand, the modern
terms “chemist,” “chemical,” and “chemistry” are employed to refer to chemical practice more gen-
erally or to refer to those practitioners who embraced the mechanistic philosophy that rejected
the vitalistic or magico-theological perspective embraced by alchemists and chymists. The terms
“chymist,” “chymical,” and “chymistry” are also employed when quoting directly from early modern
sources, since these were the terms commonly used in the 16th and 17th centuries with regard to
chemical practice. The terms “corpuscularian” and “corpuscularianist” will be employed to refer to
the corpuscular philosophy and its adherents, while the terms “mechanicism” and “mechanicist” will
be employed to refer to the mechanical philosophy and its adherents, except of course when quoting
directly from other sources that employ different terms, such as “corpuscularism” or “mechanism.”
10 The Chemical Philosophy of Robert Boyle
corpuscularian theories of matter imply a mechanistic hypothesis. More will be

said in the next chapter about the revival of Epicurean atomism, particularly as
it affected chemical philosophy. However, at this point, it will suffice to say that,
as late as the early 17th century, many of the most significant chemical philoso-
phies were those that reconciled a corpuscularian theory of matter with a vital-
istic ontology.
As already mentioned, the shift from a vitalistic to a mechanistic chemical on-
tology was not sudden and radical but, rather, followed a subtle, nuanced, and
gradual path from the 16th to the 17th century. In fact, aspects of vitalism and
mechanicism coexisted in interesting ways in the chemical ontologies of the
early modern period, with corpuscularianism as an important commonality
shared by these competing perspectives. Ultimately, the gradual demise of vi-
talism in the context of chemical philosophy resulted not from the victory of
mechanistic reductionism but from a physicalistic and naturalistic rationali-
zation of chemical qualities. In other words, the dominant chemical paradigm
shifted from a dualistic ontology of matter and spirit to an ontology in which
the only causal agents in nature were entirely material principles. In this shift,
vital spirit itself was eventually regarded as having a physical ontological status
since it was ultimately identified with aerial niter or potassium nitrate. However,
this physicalistic and naturalistic approach to the articulation of vital spirit and
of other formerly immaterial concepts did not entail a commitment to an en-
tirely reductionist mechanicism. In fact, one finds that residues of vitalism
such as the concept of “seminal reasons” linger within early modern chemical
ontologies, even in Robert Boyle’s own work. One also finds that Boyle’s chem-
ical ontology is naturalistic and physicalistic but remains, to a large extent, non-
reductionist regarding chemical qualities. Thus, I would argue that it is not the
victory of mechanicism over vitalism that ushers in modern chemistry. Rather, it
is the gradual naturalization of chemical qualities and processes and of concepts
such as ferment and spirit that leads ultimately to the modernization of chemical
practice and to the quantification of chemical theory.
In this chapter, I will examine many of these ideas in detail by focusing on
the more significant elements of vitalism that influenced 16th-and 17th-century
chymistry, as well as some of the key concepts of Scholastic and spagyric alchemy
that would be later rejected by the mechanistic chemical philosophy. I will also
focus on some of the key figures in this transition, whose hybrid ontologies
paved the way for the work of early modern chemists such as Boyle. These figures
are Paracelsus, Daniel Sennert, and Jan Baptista van Helmont, whose alchem-
ical ideas and experiments provided the concepts and methods that would have
the greatest impact on Boyle’s work. In addressing the chemical ontologies of
these predecessors, I will focus on the key notions that served to inform Boyle’s
chemical practice. I will, however, also discuss the Scholastic and spagyric ideas
the rejection of which helped to define Boyle’s commitment to the mechanical

hypothesis, these being Scholastic hylomorphism and the Paracelsian notion of
the tria prima. This chapter will, thus, provide the background for the discussion
of Boyle’s complex corpuscularian ontology that is discussed in Chapter 3. I will
first begin by examining vitalism to set the context for the discussions later in
this chapter.
1.1 The Vitalistic Character of Renaissance Alchemy
One discipline within the Renaissance tradition of natural magic that was espe-
cially receptive to vitalistic ontology was alchemy and, even during the transition
from alchemy to chymistry, lingering elements of the vitalistic standpoint con-
tinued to resonate within chemical philosophy. As well, to the extent that alchem-
ical ontology impacted developments in iatrochemistry and medicine, vitalism
continued to influence these disciplines throughout the 16th and into the 17th
century. However, one of the misconceptions informing much of the literature
on this subject is that vitalism ended with the advent of corpuscularian and at-
omistic theories of matter. At the heart of this misconception is the idea that par-
ticulate theories of matter imply mechanicism. This is indeed a misconception.
The historical process whereby ancient atomism was revived in the 17th century
was in fact quite complex and, although “most historians [and philosophers] of
science have considered early seventeenth-century atomism as preparatory to
the mechanical theory of matter,”4 there is strong evidence that for much of the
17th century many chymists adhered to a particulate theory of matter while also
embracing a vitalistic ontology. Two chemically important concepts within this
ontology were the notions of “spirit” and of “ferment” conceived vitalistically. As
Antonio Clericuzio explains, “the doctrine of spirit played a substantial part in
seventeenth-century natural philosophy and medicine. The Neoplatonic spirit of
the world was widely adopted by chemical philosophers as a principle of motion
and life [. . .] This notion became central to chemistry and medicine thanks to
Ficino’s De vita Philosophicae (1571).”5
To understand the complex ontologies that informed the chemical philoso-
phies of the 16th and 17th centuries, one must realize that chemical philosophy
had been involved in a process of self-definition and development since it was
first articulated as such in the work of the alchemist Paracelsus. Chemistry, as a
scientific discipline, did not enjoy the long and venerable historical tradition that
was enjoyed by astronomy, mechanics, mathematics, and physics. In fact, up to
4 Clericuzio, Elements, Principles, and Corpuscles, 37.

5 Ibid.
the 16th century, chemistry “had no organized structure whatsoever, no theories

of change and reactions, and no clearly defined tradition. Like geology and mag-
netism, chemistry became a science between the seventeenth and eighteenth
centuries. Unlike mathematics, mechanics, and astronomy, it was itself a product
of the Scientific Revolution.”6 As Paolo Rossi explains, “there is no figure like
a Euclid, Archimedes, or Ptolemy in the history of chemistry. Instead, modern
chemists find themselves in the somewhat disconcerting company of alchemists,
druggists, iatrochemists, sorcerers, astrologers, and other sundry figures.”7 In
fact, one of the reasons for which chemistry was long considered the redheaded
stepchild in the family of science is that, for much of its history up to the 17th
century, it was primarily a practical enterprise that did not seem to be anchored
upon a solid theoretical and philosophical foundation.
As My Giyung Kim recently noted, the discourse of chemical philosophy
did exist in the 16th and 17th centuries.8 However, 16th-century chemical phi-
losophy was grounded primarily in a vitalistic worldview heavily tinged with
mystical and theological overtones. Therefore, as Kim points out, 16th-century
“alchemical and Paracelsian linguistic heritage came under increasing attack
from natural philosophers seeking to domesticate this rich empirical field in
order to refurbish their systems of philosophical knowledge [. . .] The institution-
alization of philosophical chemistry required a social evolution of chemists from
the ‘sooty empirics’ who prepared medicaments in the basement into learned
savants who could converse on equal terms with the scholarly world.”9 By the
18th century, what was lacking was a philosophical chemistry, that is, an organ-
ized and systematized field of knowledge that would qualify chemistry as an
independent and autonomous natural philosophy.10 Thus, the transition from
practical art to institutionalized natural philosophy required the successful de-
velopment of “a new philosophical language, one less hermetical and more in
tune with natural philosophy, rather than a disciplined practice.”11
In fact, the 17th century represents that crucial period in chemical history
during which this transition played itself out, and Robert Boyle stands out as one
of the more notable figures in this process since his aim was precisely to promote
“chemistry as the ‘key’ to ‘true’ natural philosophy. [He] endeavored to dissociate
the laboratory practice of chemistry from the discourse of [Paracelsian] Chemical
Philosophy and to forge a philosophical chemistry [chemia philosophica] that
would translate the laboratory culture of chemistry for a learned audience.”12

7 Ibid., 139.
8 Kim, Affinity, That Elusive Dream, 3.
9 Ibid.
10 Ibid., 3–4.
11 Ibid., 4.
12 Ibid., 3.
Given these considerations regarding the scientific status of early modern

chymistry and its close historical ties to vitalism and to alchemy, it would be-
hoove us to examine closely the context in which early modern chymistry
evolved, both by examining the ideas of 16th-and 17th-century alchemists and
the ontological framework that informed 17th-century chemical philosophy,
particularly regarding the relationship between vitalism and corpuscularian the-
ories of matter. In terms of ontology, it is important to realize that many 16th-
century chymists still found themselves somewhat shackled to certain aspects
of Aristotelian science as it had been revised by the Scholastics. One of the most
problematic concepts with which they had to contend was the theory of substan-
tial form, and much of Boyle’s own chemical work is invested in providing evi-
dence against the existence of substantial form.
1.2 The Scholastic Theory of Substantial Form
The Scholastic theory of substantial form has its roots in Aristotle’s theory of four
causes, the two external causes being efficient (or first) cause and final cause and
the two internal causes being material cause and formal cause. Though Aristotle
ultimately concludes that form is substance, the substantial conception of form
was developed by the Scholastics in ways that Aristotle could not have antici-
pated. It is thus important to examine the role played by form within Aristotelian
metaphysics. I will frame my discussion of Aristotelian forms in mereological
terms, both because form is a mereological concept for Aristotle and also to link
this discussion with that of Boylean mereology in Chapter 5.
For Aristotle, forms are organizing principles that account for the unity, iden-
tity, order, and properties of things, whether simple wholes or composite wholes.
Mereologically speaking, Aristotle is concerned with what makes composite
wholes distinct from mere heaps. In other words, what is the something “over
and above” the parts or “elements,” to use Aristotle’s term, that is present in the
whole but not in the heap.13 That “something” is form or essence, and it accounts
for the unity that is present in wholes but that is absent in heaps. One of the is-
sues that Aristotle addresses in Metaphysics Z.17 is mereological hylomorphism,
that is, the idea that the form is itself a part of the whole that it unifies. Aristotle
rejects mereological hylomorphism because it would ensue in an unacceptable
regress. Referring to the form of composite wholes, Aristotle says, “And this is
the substance [ουσια] of each thing; for this is the primary cause of its being;
and since, while some things are not substances, as many as are substances are
13 For a detailed and fascinating discussion of this topic, see: Kathrin Koslicki, “Aristotle’s
Mereology and the Status of Form,” 715–736.

formed naturally and by nature, their substance would seem to be this nature
[φυσιζ], which is not an element but a principle [αρχη]. An element is that into
which a thing is divided and which is present in it as matter [υλη].”14
To maintain mereological consistency, however, Aristotle also states that
“both the form and the matter are proper parts of a matter/form-compound.”15
So, to resolve the problem of potential regress, Aristotle takes the form of a com-
posite whole to be ontologically distinct from the material parts or elements of
that whole. Aristotle’s concept of form is, therefore, distinctly metaphysical to the
extent that “the unity of a substance is a product not just of its elements, but of
some further unifying principle, the form, that is not itself an element.”16 Thus,
the theory of forms is a “top-down” theory of organization within Aristotelian
metaphysics, in that forms are of a distinct ontological kind from the wholes that
they unify. The elements that are part of the unified whole belong to the onto-
logical category of matter, while the form belongs to the ontological category of
immaterial principle, so that unified wholes are both mereologically and onto-
logically complex.17 However, as Kathrin Koslicki points out, “the mere recog-
nition of a particular kind of ontological complexity within a genuinely unified
whole by itself does not yet solve the mystery of why it is that these entities of
distinct ontological types (in Aristotle’s case, form and matter, or principle and
element) can come together to produce a single genuinely unified thing.”18
Scholastic philosophers undertook to clarify the question of what a form is
exactly, and 12th-century metaphysicians proposed several answers to this
question. In part, the Scholastics were interested in spelling out the distinction
between substantial forms and accidental forms, and a lot of their theorizing
revolved around solving this question. By the 13th century, no consensus had
been reached regarding this issue and, as John Duns Scotus pointed out, there
was only one universally accepted conclusion regarding substantial forms, that
is, “that the substantial form plays a part in the essence of a thing.”19 This was a
universally accepted idea primarily because of a statement made by Aristotle in
Physics 2.3 194b27, in which he claims that the formal cause of a thing constitutes
its essence. There was, indeed, a great deal of debate among the Scholastics re-
garding exactly what it means to say this, though an account of these debates
would take us beyond the scope of this chapter.20 However, it must be noted that,
to the extent that form is an internal cause of being, it would seem to play a role
14 Aristotle, The Complete Works of Aristotle, Z.17, 1041b32–33.

15 Koslicki, “Aristotle’s Mereology and the Status of Form,” 727.
16 Pasnau, “Form, Substance, and Mechanism,” 33.
17 Koslicki, “Aristotle’s Mereology and the Status of Form,” 723.
18 Ibid., 727.
20 Robert Pasnau provides a very detailed and meticulous discussion of these debates in his “Form,
Substance, and Mechanism.”

“very much like the role of internal efficient cause, sustaining and regulating the
existence of that which the efficient cause originally produced.”21
As Robert Pasnau affirms, by the middle of the 13th century, Scholastic
metaphysicians conceived of substantial form in a much more concrete way
than as a merely metaphysical principle. In fact, for these Scholastics, substan-
tial form is that which accounts for the stability of material things, that is, it is
that which sustains a thing’s essential properties or the “constant set of properties
that are characteristic of that thing.”22 Thus, in De mineralibus 3.1.7, Albert the
Great states that “there is no reason why the matter in any natural thing should
be stable in its nature, if it is not completed by a substantial form. But we see that
silver is stable, and tin, and likewise other metals. Therefore, they will seem to
be perfected by substantial forms.”23 Therefore, the substantial form is not to be
identified with the properties or the set of properties but, rather, it is the “some-
thing further that explains their enduring presence.”24 These points will become
particularly significant in my discussions of Boyle’s theory of essential form and
essential properties in Chapter 3 and of Boylean mereology in Chapter 5.
Despite the turn toward a concrete conception of substantial form, one does
find some Scholastics affirming the more metaphysical understanding of form.
Aquinas, for example, claims that substantial form is that which accounts for the
individuation of a substance. Thus, in Summa contra gentiles, he states that “this
is clear from the fact that both the whole and the parts take their species from it,
and so when it leaves, neither the whole nor the parts remain the same species.”25
However, this metaphysical understanding of substantial form, which clearly
seems more faithful to the Aristotelian account, was not dominant among the
Scholastics. Pasnau explains that
it is one of Aristotle’s most cherished ideas that material and efficient

explanations must be supplemented by a further level of formal analysis.
Scholastic authors seemed to be sliding ever farther back toward the materi-
alism Aristotle sought to refute, as if they could not resist the temptation to
ground formal explanation on material and efficient causes at a deeper level.
In turn, as the Scholastic conception of form grew increasingly remote from
its metaphysical roots in Aristotle, it became at the same time increasingly
naturalistic.26
22 Ibid.
23 Albert the Great, Book of Minerals, 173.
25 Aquinas, Liber de veritate Catholicae fidei contra errores infidelium, seu Summa contra Gentiles,
2.72.1484. “Quod autem sit forma substantialis totius et partium, patet per hoc quod ab ea sortitur
speciem et totum et partes. Unde, ea abscedente, neque totum neque partes remanent eiusdem
speciei.”
We find that it is this increasingly naturalistic and wholly concrete conception

of substantial forms that is handed down to Renaissance metaphysicians and
to 16th-and 17th-century natural philosophers, and it is this conception that
comes under virulent attack from mechanistic thinkers, whether they be of
a Cartesian or an empiricist mindset.27 A detailed account of these critiques is
given in Chapters 2 and 3. Therefore, I now turn to another important concept
that informed the natural philosophical ontologies of the 16th and 17th centu-
ries, that is, the Paracelsian concept of the tria prima.
1.3 Paracelsian Spagyria and the Tria Prima
Theophrastus von Hohenheim, a.k.a. Paracelsus (ca. 1493–1531), is a sem-

inal figure for both early modern spagyrists and chymists. Most important,
Paracelsus helps to transform 16th-century alchemy by giving it an essentially
medical identity, and he makes this the basis for the development of an alchem-
ical epistemology.28 Paracelsus makes significant advances in medicine by
arguing against the Galenic and Scholastic view that disease is caused by an im-
balance of the four bodily humors, which might be cured by bleeding or herbal
remedies. Anticipating the modern bacteriological and viral conceptions of
disease, Paracelsus claims that the cause of disease is the presence of external
agents attacking the body and that this condition can be cured through “chem-
ical” remedies. He identifies the characteristics of many illnesses, such as goiter
and syphilis, and treats them with sulfur and mercury compounds. He is, there-
fore, “the first to introduce the medicinal use of mineral substances to the prac-
tice of medicine. Alchemy, or ‘the spagyric art’ as Paracelsus called it, became the
cornerstone of medicine.”29
By all accounts, however, the popular characterization of Paracelsus as a
mystic is accurate. He fits very comfortably within the Renaissance tradition
of natural magic, to the extent that the theoretical framework upon which his
work relies is staunchly theological and vitalistic. Walter Pagel, Allen Debus,
and other scholars30 suggest that Paracelsus is influenced in these regards by
Rabbinic and Kabbalistic sources, particularly the Shemoth Rabba, that provide
“a religious background for the homeopathic principle”31 that he employs in
his iatrochemical preparations. Paracelsus not only posits the existence of vital
forces and spirits, but he also fully affirms the theory of correspondence between
27 See Ibid.
28 See Alchemy and Chemistry in the 16th and 17th Centuries.
30 See Pagel, Paracelsus; Forshaw, “Cabala Chymica or Chemia Cabalistica.”
31 Pagel, Paracelsus, 217.
microcosm and macrocosm and, in this affirmation, he is once again inspired

by his intention “to unravel the occult—‘kabbalistic’ and symbolical—meaning
of phenomena by visualising concordances everywhere.”32 Not surprisingly, this
mingling of theology and mysticism with chemical philosophy shapes his inter-
pretation of Genesis in chemical terms as the separation of the elements by God.
For Paracelsus, “since the divine creation [is] best understood as a chemical pro-
cess, then nature must continue to operate in chemical terms. Chemistry [is] the
key to nature—all created nature.”33
Paracelsus’ chemical philosophy is based upon three fundamental prin-
ciples: his theory of prime matter, his theory of elements, and his theory of
principles. His theory of water as prime matter is based upon his chemical in-
terpretation of Genesis in which God is a divine alchemist who spontaneously
creates the world ex nihilo with water as the first element.34 His theory of the
elements includes prime matter (i.e., water), fire, earth, and air, all of which he
also considers to be matrices. “Plants, minerals, metals, and animals were the
fruit of the four elements.”35 Although Paracelsus inherits his theories of prime
matter and of the elements from ancient sources, it is with his theory of princi-
ples that he makes truly original contributions to Renaissance alchemy and to
later chymistry.
According to Paracelsus, the principles of chemical reaction, which he calls the
tria prima, are salt, sulfur, and mercury. “This tria prima also consist[s]‌of spir-
itual substances and correspond[s] to the Body, Soul, and the Spirit. Salt makes
bodies solid, Mercury makes them fluid, and Sulfur makes them inflammable.”36
Although the theory of the tria prima is a modification of earlier sulfur-mercury
theories of metals, it has a special significance in the rise of modern science be-
cause it represents a broadening of these theories “to provide an explanation for
all of nature.”37 In addition to these fundamental principles, Paracelsus and later
Paracelsians also believe that vital spirit is essential for both the organic and the
inorganic worlds. “Spirits were conceived as the active agents, upon which all the
principal operations in nature and in the human body depended.”38 In his De
natura rerum (1537), Paracelsus states that “the life of things is none other than a
spiritual essence, an invisible and impalpable thing, a spirit and a spiritual thing.
On this account there is nothing corporeal, but has latent within himself a spirit
and a life, which, as just now said, is none other than a spiritual thing.”39
32 Ibid.
33 Debus, The Chemical Philosophy, 86.
35 Ibid.
36 Ibid.
37 Debus, The Chemical Philosophy, 78–79.
38 Clericuzio, “The Internal Laboratory,” 52.
39 Paracelsus, as cited in Ibid.
Paracelsus’ chemical philosophy is, however, not merely mystical but also
contains the fundamental elements of what would later become early modern
chymistry and modern scientific method. For example, Paracelsus contributes
to the later chemical interpretation of spirit by identifying the celestial vital sub-
stance or anima mundi, which he believes to be contained in the air, with the
aerial niter (also called saltpeter, i.e., potassium nitrate). This choice is not arbi-
trary. His reason for identifying the vital spirit with aerial niter is that potassium
nitrate gives off life-sustaining oxygen when heated. Paracelsus reflects on what
he considers to be the unique properties of potassium nitrate and claims that “no
other salt in the world is like [saltpeter] [. . .] The part played by Saltpeter in gun-
powder is one reason why it is different from all other salts, and Paracelsus re-
peatedly explains thunder and lightning in terms of an aerial, windy, or aetherial
nitre and sulphur.”40 Later Paracelsians contend that, in addition to salt, there are
two volatile parts in saltpeter, these being sulfur and mercury, which represent
for them the soul and the spirit of the aerial niter.
Based upon Paracelsus’ own investigations into the nature of the aerial
niter and upon the Neoplatonic belief in the microcosm-macrocosm analogy,
Paracelsians develop the theory that the vital spirit originates in the celestial
sphere and is carried in the air. It is then inhaled by human beings and reaches
the heart, from which it is “carried through the body in a circular motion, imi-
tating the divine circularity [of the celestial bodies]. This motion impressed on
the blood relates not only to the spirit of the blood in the heart but to all of the
spirit of the blood in the body.”41 It is interesting to note that the interest in the
properties of potassium nitrate and speculation regarding its identity as the vital
spirit of the world continued well into the 17th century, during which time
a powerful stream of speculation wondered whether saltpeter contained the sal

nitrum or spiritus mundi, the “nitrous universal spirit,” that would unlock the
secrets of nature. If so, as Robert Boyle observed, the subject “may well deserve
our serious enquiries” . . . Francis Bacon himself, Lord Chancellor and Privy
under James I, dismissed “crude and ignorant speculations” about gunpowder
and its “spirit,” yet continued to identify saltpeter as the energizing “spirit of the
earth.”42
Paracelsus and the later Paracelsians also hold a persistent interest in the
blood and its relation to the vital spirit. Like Helmont’s identification of vital
spirit with volatile alkaline salt, which will be discussed later in this chapter, the
40 Debus, “The Paracelsian Aerial Niter,” 47.

41 Debus, The Chemical Philosophy, 235.
42 Cressy, Saltpeter, 13–14.
Paracelsian identification of anima mundi with saltpeter and the theory that this
aerial niter penetrates the body through the blood are important first steps to-
ward the late-17th century naturalization of vital spirit and the reinterpretation
of this notion in physicalistic terms. For 17th-century Paracelsians, “the doctrine
of the world soul, in its most theoretical formulations, was rooted in an inter-
pretation of chemical operations. Far from being merely a recovery of a notion
inherited from Neoplatonic Renaissance philosophy, this doctrine was justified
by its formulation in chemical terms, which show that what is being referred to is
material, albeit a very tenuous conception of matter that is called spirit.”43
Another aspect of Paracelsus’ work that contributes to the later development
of early modern chymistry is his solidly empirical approach to medicine and
spagyria. According to him, “to attain true knowledge one must abandon the
surface of bodies, penetrate their inner nature and break them up into their con-
stituent parts until each of these is accessible to sight and touch.”44 To make the
constituent parts of bodies empirically accessible, Paracelsus emphasizes anal-
ysis, although he and later Paracelsians “[lay] the foundation for viewing anal-
ysis as only half of the equation—as a necessary preliminary to resynthesis”45 or
redintegration, as later chemists such as Boyle would call it. Paracelsus and his
followers thus [change] the practice of alchemy by emphasizing the twin pro-
cesses of analysis and synthesis (spagyria), to penetrate the true inner nature of
bodies.
When discussing this inner nature, Paracelsus is heavily influenced by Ficino
and “place[s]‌special emphasis on semina, which he [also] consider[s] as invisible
spiritual forces and as archetypes.”46 There are clearly echoes of the Augustinian
doctrine of seminal reasons in Paracelsus’ conception of semina. For him,
“Semina, which originate in the Word [or Logos] are contained in the Yliaster [the
universal matrix of the cosmos] and are prior to chemical principles and to elem-
ents. For Paracelsus, nature as a whole is a panspermia,”47 meaning that seeds
imbue the entire universe with life. In this regard, Paracelsus is deeply influenced
by the ancient Aristotelian doctrine of semina rerum, which contributes inval-
uably to the persistence of vitalism in the 16th and even into the 17th century.
To the extent that the concept of seminal reasons and that of minima naturalia
43 Joly, “Les alchimistes étaient-ils des matérialistes? Quelques remarques sur le psychisme
humain et l’esprit du monde,” 61: “On voit alors à quel point la doctrine de l’esprit du monde, dans ce
qu’elle pourrait avoir de plus théorique, s’enracine dans une interprétation des opérations chimiques.
Loin de n’être que la reprise conceptuelle d’une notion héritée de la philosophie néoplatonicienne de
la Renaissance, cette doctrine se justifie par sa formulation en des termes chimiques qui montre bien
que ce dont il est question relève de la matière, même s’il s’agit de cette matière extrêmement tenue
que l’on nomme esprit.”
44 Bianchi, “The Visible and the Invisible: From Alchemy to Paracelsus,” 18.
45 Newman and Principe, “Alchemy and the Changing Significance of Analysis,” 79.
47 Ibid.
play such an important role in 17th-century chymistry, I will now discuss these
concepts in greater detail. Together these notions occupy an invaluable place in
the development of vitalistic corpuscularianism and continue to impact particu-
late theories of matter well into the 17th century.
1.4 Semina Rerum, Minima Naturalia, and

Vitalistic Corpuscularianism
The doctrine of semina rerum has a long and complex history and received con-
flicting interpretations as it was adopted by different philosophical traditions.48
Lucretius and other Epicureans conceive of semina rerum, or the “seeds of
things,” in entirely physical terms as atoms, while the Stoics conceive of semina
as immaterial active and formative principles having nothing to do with atoms.
Inspired by the Stoic interpretation of semina, Neoplatonic philosophers argue
that the ordering principle of the universe, or Logos, contains within itself active
constituents that they likened to “seeds.” Hence, Neoplatonists refer to these ac-
tive constituents as logoi spermatikoi (λόγοι σπερματικοὶ), a term later Latinized
as rationes seminales, that is, “seminal reasons.” For Neoplatonists, seminal
reasons contain a rational “program” that specifies the creative power that is im-
manent in nature. In its Neoplatonic iteration, this idea plays a central function
in Augustinian ontology and Augustine uses it to reconcile apparent exegetical
tensions between the account of creation in Genesis 1 and that in Ecclesiasticus
18:1 (Book of Sirach).49
The notion of semina rerum interpreted in Stoic terms ultimately occupies a
prominent role in the natural philosophies of the 15th and 16th centuries, which
are heavily influenced by the Renaissance revival of Platonism and Neoplatonism.
In particular, as already pointed out in the previous section, Paracelsus makes
considerable use of the notion of seminal reasons, which he interprets as the
forces and active powers in any object, and this idea resonates in the work of
later Paracelsians. However, the writings of the French Paracelsian Pierre-Jean
Fabre (1588–1658) indicate that, by 1629, the notion of semina acquires clearly
48 Hiro Hirai provides an extensive, detailed, and definitive history of this concept as it evolved
from the Renaissance to the early modern period, and I am heavily indebted to his work and to the
work of Antonio Clericuzio for my discussion here. See Hirai, Le concept de semence dans les théories
de la matière à la Renaissance, and Clericuzio, Elements, Principles, and Corpuscles.
49 Augustine identifies a tension between the account found in Genesis 1, according to which God
created the universe in stages, and the account found in the Book of Sirach (also called Ecclesiasticus),
according to which God created everything at once. Augustine reconciles this apparent conflict by
arguing that, although God did not create all species of things at once, he implanted in nature the
seminal reasons or “seeds” of all things that will ever exist, even though each of these species of things
is materially realized at different points in time.
physicalistic connotations, albeit still within the context of a generally vitalistic

conception of nature. In his Traicté de la peste, Fabre states that “seeds are in the
heavens and descend into the inferior elements, as if within their matrix, in order
to become elementary mixts and material bodies. God wanted it and disposed it
thus through his mighty power and wisdom.”50
As already mentioned, for Paracelsus semina are the causal agents respon-
sible for the generation of natural bodies, including metals in the bowels of the
earth, and this idea continues to influence Paracelsians throughout the 16th
and 17th centuries. Paracelsus’ contemporary Girolamo Fracastoro takes this
Neoplatonic interpretation of semina rerum a step further by combining it with
Lucretian atomism, reinterpreting it “in terms of invisible units of matter”51 and
arguing that semina are also the causes of communicable diseases when they
propagate through the atmosphere and penetrate a host organism. In this way,
both Paracelsus and Fracastoro greatly advance the medical theory of diseases
by moving away from the strictly Galenic theory of humoral imbalance and to-
ward the ontological theory of pathology and contagion. However, influenced
by the revival of Epicurean atomism, 17th-century chemists would eventually
both incorporate the concept of semina rerum into their chemical ontologies and
interpret this concept in much more naturalistic and physicalist terms than the
Paracelsians ever did.
Semina rerum become particularly relevant for 17th-century mineralogy,
with many chymists and metallurgists attributing the birth and growth of
metals in the bowels of the earth to specific “seeds” that endow these metals
with an internal vegetative principle. For metallurgists and chymists, it is this
internal principle that permits the transmutation of a base metal into a noble
metal. “The relation between the doctrine of metallic seeds and that of trans-
mutation resides in the theory of one type of seed for all metals, rather than spe-
cific seeds for different metals, which means that the seed can produce different
metals according to the degrees of maturation that it has reached.”52 Among the
chymists who sustain this seminal conception of the generation of metals are
many corpuscularians such as John Webster, Daniel Sennert, Jan Baptista van
Helmont, Pierre Gassendi, Walter Charleton, and the young Robert Boyle.53
50 Fabre, Traicté de la peste, selon la doctrine des médecins spagyriques, 19–20: “Les semences de
toutes les choses sont dans les Cieux, & descendent ça bas dans les éléments inférieurs, comme dans
leur matrice, pour y estre faictes mixtes élémentaires, & corps matériels. Dieu la voulu & la disposé
ainsi par sa toute puissance & sagesse.”
52 Clericuzio, “Alchimie, philosophie corpusculaire, et minéralogie dans la Metallographia de John
Webster,” 298. “Le rapport entre la doctrine de la semence des métaux et celle de la transmutation
réside dans la théorie d’une seule semence pour tous les métaux, mais pas de semences spécifiques, ce
qui signifie que la semence peut produire différents métaux selon les degrés de maturation auxquels
elle parvient.”
53 Ibid., 297–298.
In addition to the concept of semina rerum, the notion of minima naturalia

also plays a significant role in the 16th-century revival of atomism and in the de-
velopment of 17th-century corpuscularianism. Although the modern revival of
Epicurean atomism dates to Poggio Bracciolini’s 1417 rediscovery of Lucretius’ De
rerum natura,54 what makes it possible for Epicurean atomism to be resurrected
in its modern version is the survival of non-Epicurean corpuscularian theories
of matter through late antiquity and into the Middle Ages. The notion of minima
naturalia can be traced at least as far back as Aristotle, for whom minima are
the minimum amount of matter necessary to instantiate a specific substantial
form.55 However, by interpreting minima naturalia as the smallest possible and
irreducible particles of reagents, many medieval and Renaissance alchemists
develop their own type of corpuscularianism or particulate matter theory. This
“alchemical atomism” functions as a qualitative version of Epicurean atomism,
and the concept of minima thus interpreted allows Scholastic and Renaissance
chymists to embrace a corpuscularianism in which the fundamental particles
are endowed with substantial form and in which the properties of said particles
are chemical and not strictly mechanical.56 Thus, although the atomism of
Democritus, Epicurus, and Lucretius had been materialistic, mechanistic, and
deterministic, the many advocates of atomism during the Renaissance seek to
reconcile Epicurean atomism with a non-mechanistic conception of the universe
and employ a qualitative conception of minima for the purpose of developing a
vitalistic atomism.
One of these Renaissance philosophers is Giordano Bruno who, in his De
triplici minimo et mensura, cleverly combines the medieval concept of minima
with the Neoplatonic doctrine of anima mundi to develop a notion of atoms as
enlivened by soul. For Bruno, “the atom itself is to be understood as one aspect
only of [. . .] a triple minimum. The primary minimum is the monad: the first
principle of quantity and as such the basis of metaphysics. The next type of min-
imum [. . .] is the mathematical point: the first principle of extension and the
basis of geometry. The atom is the minimum of body, or three-dimensional min-
imum: and as such the basis of physics.”57
In a manner that seems to provide an anticipatory reply to 17th-century
debates regarding the incompatibility of the mathematical and physical
conceptions of atoms, Bruno argues that there is no incompatibility in consid-
ering the minimum both as geometrical point and as three-dimensional body or
corpuscle.58 For Bruno, the physical atom is an indivisible and weightless sphere
54Gatti, “Giordano Bruno’s Soul-Powered Atoms,” 163.
55Murdoch, “The Medieval and Renaissance Tradition of Minima Naturalia,” 91–97.
Webster,” 300.
57 Gatti, “Giordano Bruno’s Soul-Powered Atoms,” 166.
58 Ibid.
that has no substantial form but whose principle of form and motion is the soul,
that is, “a spiritual substance hidden deep within all atomic minimi.”59 The soul
of atoms functions much like seminal reasons in that it directs, structures, and
coordinates atomic concretions, thereby “transforming them into live, moving
and organic bodies.”60 Thus, the soul of atomic minima performs the function of
“bonding” physical atoms to each other, in accordance with rational structures.
For Bruno, “the minimum was clearly a relational notion that referred to the pro-
cess of composing and decomposing, an idea that allowed him to distinguish
mathematical and physical minima, at the price, however, of a somewhat bizarre
mathematics.”61
As will be pointed out in later chapters, one of the weaknesses of the mechan-
ical philosophy, a weakness that also infuses the work of Boyle, is that it does
not recognize the concept of force and, thus, cannot account for how physical
atoms or corpuscles are held together to form stable corpuscular concretions
and macroscopic bodies. As well, the mechanical philosophy can only account
for the mobility of particles by attributing this to the action or will of God.
Bruno’s vitalistic atomism, on the other hand, “[avoids] a number of the diffi-
culties which would beset the mechanical philosophy of the following century
by introducing an element of soul or energy into his atoms, thereby proposing a
vitalistic concept of self-moving and self-coordinating matter.”62 Bruno’s specific
conception of atoms as ensouled illustrates the compatibility of the particulate
or corpuscularian theory of matter with vitalistic ontology and serves as one ex-
ample of what may be called “vitalistic corpuscularianism,” a theory of matter
that would later be embraced by the influential 17th-century Paracelsians Daniel
Sennert and Jan Baptista van Helmont.
Despite the continued influence of vitalistic corpuscularianism, the interpre-
tation of minima gradually begins to change, as early as the 1650s, by taking on
more materialistic connotations. For example, Giulio della Scala (a.k.a. Scaliger)
interprets minima as particles and Daniel Sennert unequivocally interprets
minima as atoms, thus paving the way for the materialistic corpuscularianism
of many early modern chemists and of Robert Boyle. It is also interesting to note
that corpuscularian chymists often couple the seminal conception of metals
discussed earlier with a mechanistic conception of minima naturalia, to explain
how seeds affect the birth, growth, and transmutation of metals. John Webster,
for example, explains the properties of gold by appeal to the compacting of its
minimal particles: “Gold is more dense and compacted then [sic] any other of the
59 Ibid., 173.
60 Ibid., 174.
61 Meinel, “Early Seventeenth-Century Atomism,” 75.
62 Gatti, “Giordano Bruno’s Soul-Powered Atoms,” 180.
Metals, that is, it is less porous then [sic] any of the rest being so closely joyned
per minima, that little or none of the air, or globuli aetherei (as Cartesius calls
them) can lodge within its particles. And this is the cause of two other of its prop-
erties: to wit its heaviness, and power of extension, both of which are far beyond
either of those qualities in other Metals.”63 According to Webster, this close com-
paction of minima is also responsible for the transmutation of less noble metals
into gold. He states:
there is a radical Solution and Penetration of all the small parts of atoms of the
metal to be changed, by the subtile penetrability and ingression of their so much
purified and exalted Tincture, and thereby all things in it whatsoever that are of
an Heterogeneous nature, are separated and extruded, and the Homogeneous
Particles joined together per minima as much as Nature can admit of and so
must needs be of less bulk, and possesses less room or place, which is manifest
in Gold.64
Paracelsians in the 17th century, however, continue to consider vitalism to be

compatible with the corpuscularian theory of matter, and one of the Paracelsians
whose work is particularly intriguing in this regard is Sebastien Basso (a.k.a.
Basson or Bassonus), whose Philosophia naturalis (1621) “stands out as one of the
earliest and most articulate expositions of the corpuscular theory of matter.”65 In
fact, along with Isaac Beeckman, Basso is considered one of the early inventors of
molecular theory. Basso strongly affirms that all natural phenomena are caused
by the motions and rearrangements of atoms and, in this regard, he has much in
common with later mechanistic atomists. Yet, Basso’s work also stands out as a
prime example of Neoplatonic vitalism, since he understands motion in terms
of sympathy and antipathy, that is, in terms of internal forces of attraction and
repulsion with which atoms are endowed by the actions of the world soul and,
ultimately, of God.
Although Basso’s endorsement of a corpuscularian theory of matter signif-
icantly contributes to the eventual development of early modern chymistry,
other Paracelsians take even further steps toward the development of a theory
of vital spirits that would, eventually, lead to the 17th-century reinterpreta-
tion of these spirits in chemical and naturalistic terms. “From the 1650s the
notions of spirit (and of fermentation) became central issues [. . .] [Many
English chymists and physiologists] shared the view that matter was endowed
63 Webster, Metallographia, or an History of Metals, as cited in Clericuzio, “Alchimie, philosophie
corpusculaire, et minéralogie dans la Metallographia de John Webster,” 299–300.

Webster,” 301.
with an internal principle of organization, life and sensibility, namely, the

spirit which they described in terms of particles having specific chemical
properties.”66 This explains why “the distillation of spirits became an impor-
tant component of seventeenth-century chemistry and medicine [. . .] [for
the purpose of] identifying and manipulating the spiritual essences extracted
from natural bodies by means of distillation [. . .] [and for] ‘capturing’ the
spirit of the world, which Paracelsians conceived as the celestial vital sub-
stance contained in the air.”67
We see this same preoccupation with conceiving spirit in chemical terms
among the French Paracelsians. The work of Duschesne and Croll, for example,
develops the view “that medical spirits and spirits extracted by chemists [have]
the same source, namely, the spirit of the world. On this basis they stated that
the only active remedies [are] those prepared by using spirits extracted by
distillation.”68 Additionally, although the 17th-century French Paracelsians
continue to appeal to God as the creator of matter and compare the lapis
philosophorum to the Trinity in its expression of divine unity, their chemical
philosophy is materialistic and naturalistic, and their central focus on cor-
poreal substances is founded upon the operations and manipulations of the
chemical laboratory. Thus, in spite of the symbols and analogies employed, the
principles invoked, including that of the world soul, are always conceived as
being material.69
Such unambiguously chemical interpretations of vital spirit and world
soul are even more evident in the chemical philosophy of Jan Baptista van
Helmont, for whom vital spirit (archeus) is conceived as an alkaline salt that
moves through the body. As we shall see presently, there is a strong relation-
ship between Helmont’s theory of vital spirits and his corpuscularian theory of
matter. At this point, therefore, I will proceed to examine the work of the iatro-
chemist Daniel Sennert and, after this, of the alchemical atomist Jan Baptista
van Helmont, both of whom serve as transitional figures between the vitalistic
corpuscularians and the early modern Epicurean atomists and as influential
predecessors of Robert Boyle.
66 Clericuzio, “The Internal Laboratory,” 59.

67 Ibid., 53–54.
68 Ibid., 53.
69 Joly, “Les alchimistes étaient- ils des matérialistes? Quelques remarques sure le psychisme
humain et l’esprit du monde,” 62: “Les principes qu’elle invoque, y comprit l’Esprit du Monde, sont
toujours matériels [. . .] au-delà des symboles et des analogies qu’elle déploie dans ses discours [. . .]
elle forge sa doctrine au contact des opérations de laboratoire et de la manipulation des diverses
substances corporelles qui font de la matière, vivante ou minérale, l’objet central de sa réflexion.”
1.5 Daniel Sennert’s Structural Hylomorphism

and Atomicity as a Negative-Empirical Concept
There are many reasons to be grateful for the work of 20th-century historians and
historiographers of science in reappraising the contributions of the alchemical
tradition to the development of early modern chemistry. One of the figures who
has benefited from such reappraisal is the renowned Wittenberg iatrochemist
Daniel Sennert (1572–1637), whose work was once regarded as insignificant at
best and regressive at worst. Marie Boas, for example, was quite censorious in
her assessment that “Sennert contributed nothing new to the development of a
mechanical philosophy based upon a theory of atoms.”70 What recent historiog-
raphy shows, however, is that, far from being tangential, Sennert’s theories and
experimental methods significantly impacted the development of later atomism.
Although Sennert’s theories are transitional between Scholastic ontology and
mechanistic atomism, he has been characterized as an archetypal transitional
figure precisely because his work is significant in several ways. As of 1619,
Sennert embraced a conception of matter that hybridized the Scholastic theory
of substantial forms with classical atomism. He also rejected the Scholastic con-
ception of homogeneous mixts and endorsed an early version of chemical struc-
turalism. Thus, his work marks an early attempt at conceptualizing atomism in
the context of post-Paracelsian alchemy. Secondly, his work anticipates the im-
portant role that structuralism would later play in modern chemistry. Finally,
Sennert’s negative-empirical conception of “atomicity” influences Boyle’s con-
ception of “chymical atoms” and anticipates Antoine-Laurent de Lavoisier’s un-
derstanding of elementarity as determined by the limits of chemical analysis.
Sennert uses a variety of experiments to support his commitment to atomism,
the most notable of which is his “reduction to the pristine state,” which eventually
becomes an important experimental procedure for later chemists such as Boyle.
In this experiment, the limits reached by chemical analysis play a notable role in
determining what constitutes “atomicity.” In fact, the aspect of Sennert’s empir-
ical work that most impacts later chemistry is his conception of “atomicity” as
grounded in the limits attained by the analytical method of the laboratory. This
notion of atomicity focuses on what cannot be achieved in the context of exper-
imental practice and with the available methods of analysis. This way of under-
standing “atomicity” is, therefore, considered to be a negative-empirical concept.
The term “negative-empirical concept” first appeared in the work of David
Knight71 in 1967, in his book on 19th-century English matter theories. The term
was then appropriated by Arnold Thackray72 in his 1970 work on 18th-century
70 Boas, “The Establishment of the Mechanical Philosophy,” 429.
71 Knight, Atoms and Elements.
72 Thackray, Atoms and Powers.
Newtonian matter theory, and it was later resurrected in Bernadette Bensaude-

Vincent and Isabelle Stengers’s 1993 book on the history of chemistry73 to de-
scribe Boyle’s concept of element. Bensaude-Vincent and Stengers explain that
“the idea of a negative-empirical concept is . . . a purely epistemological no-
tion. It represents a new type of argument that locates the authority of proof
not within reason but within experimental practice.”74 However, although the
term “negative-empirical concept” may have been coined in the late 1960s, the
practice of defining either atomicity or elementarity operationally as the limits
of experimental analysis can be traced at least as far back as the 13th century,
where it played an important role in Scholastic alchemy. As William Newman
has emphasized:
already in the High Middle Ages (if not earlier) alchemists were employing this
approach . . . According to this analytical ideal, a substance is viewed as ele-
mentary if it cannot be decomposed by the tools of the alchemist. In fact, an
atomism based on the resistance of materials to laboratory operations such as
sublimation and calcination can already be found the Summa perfectionis of
Geber. As strikingly powerful analytical agents were [later] discovered, these
provided chymists with the means to identify “atoms” that could withstand dis-
solution into [their] components.75
Negative-empirical principles that defined the temporary limits of analysis were,

therefore, used from the Middle Ages to the 18th century to provide operational
definitions of chemical elements or substances.
Sennert’s use of negative-empirical concepts is particularly important because
these form a part of a larger theoretical program, in which his compositional
theory of matter serves to provide an account of substances and mixts that he
considers superior to the Scholastic accounts. More specifically, Sennert rejects
the Scholastic idea that genuine mixts are endowed with distinctive substantial
forms. The Scholastics inherit their conception of mixts from Aristotle, but they
modify the Aristotelian view in significant ways. Regarding the formation of
new compounds, Aristotle believes that “the form of the product appear[s]‌out
of nothing, ex nihilo, whereas the forms of the original bodies [disappear] into
nothing, in nihilum.”76 A true mixt, then, is one in which “the ingredients act
to change each other so that they cease actually to exist.”77 However, Aristotle
qualifies this position by adding that the ingredients within the mixt “continue to
73 Bensaude-Vincent and Stengers, A History of Chemistry.

74 Ibid., 51–52.
75 Newman, “What Have We Learned from the Recent Historiography of Alchemy?,” 320.
77 Wood and Weisberg, “Interpreting Aristotle on Mixture,” 682.
exist potentially, and [so] they can be separated out again.”78 Thus, for Aristotle,
synkrisis (synthesis) does not preclude the possibility of diakrisis (analysis).
Unlike Aristotle, the medieval Scholastics do not consider the four elem-
ents to be fundamental but instead posit an undifferentiated prime matter (the
πρϖτον δεκτικόν or proton dektikon) that is endowed with qualities by imma-
terial substantial forms. Each of the four elements is generated and determined
in its essential properties by the presence of a distinctive substantial form. This
view hylomorphism impacts on the Scholastic theory of mixts. According to
Scholastics such as Aquinas, when two substances are mixed, the substantial
forms of the individual ingredients are destroyed and replaced by the substantial
form of the mixt, which determines its distinctive properties as well as its homo-
geneity and uniformity. This view implies that the ingredients of a genuine mixt
cannot be recovered, since their substantial forms no longer exist. Therefore, ac-
cording to Scholastics such as Aquinas, true synthesis precludes the possibility
of analysis.
Sennert realizes that this conclusion presents a problem for the Scholastic
theory and thus rejects the idea that homogeneous mixts are endowed with their
own distinctive substantial forms. However, given the lack of viable alternatives
to the Scholastic theory, Sennert is reluctant to give it up entirely, preferring to
salvage some of its important features. He does so by developing a hylomorphic
corpuscularianism that rejects undifferentiated prime matter and, instead, posits
fundamental, indivisible, and immutable particles called atoms, each endowed
with a distinctive substantial form.79 Sennert’s ontology is hierarchical and plu-
ralistic since “there are various grades of atoms and each higher grade of atom
is composed of an organization (or structure) of the lowest grade of atoms.”80
The simplest atoms are the minima naturalia and the first composed atoms (or
corpuscles) immediately above the minima are the prima mixta. The prima mixta
themselves can compose higher orders of compounded corpuscles.81
Thus, in Sennert’s account, it is not chemical substances but, rather, their atoms
that are endowed with form. According to him, the distinctive substantial forms
of a material body’s atoms, coupled with the unique microstructure according
to which the corpuscles are arranged, account for the chemical properties that
are peculiar to that material body. In his view, when two or more substances are
synthesized, their respective corpuscles are rearranged spatially to form a new
microstructure so that the new compound has properties that differ from those
of its ingredients. For Sennert, the components of the mixt can be recovered and
78 Ibid.
79 Michael, “Daniel Sennert on Matter and Form,” 286.
80 Ibid.
81 Ibid.
returned to their pristine state when the original microstructural configuration

of their respective corpuscles is restored via analysis with aqua fortis, aqua regia,
or fire.
Sennert believes that this type of atomism provides a better theory of mixture
and one that can be supported by experimental evidence. Sennert’s atomism,
however, diverges significantly from classical atomism. Classical atomism
endorses the view that atoms are substantially uniform and are distinguished only
by the mechanistic properties of shape, size, and spatial orientation. In Sennert’s
hylomorphic atomism, the different types of atoms are endowed with distinc-
tive immaterial substantial forms that define their essence and determine their
distinctive properties. I would venture here to say that Sennert’s hylomorphic
atomism constitutes an attempt at explaining the existence of natural kinds from
within an atomistic perspective, albeit one that is not mechanistic. In this re-
gard, I would also add that, although his theory retains elements of the Scholastic
theory of substantial forms, Sennert’s rejection of undifferentiated prime matter
in favor of hylomorphic atomism represents an early attempt at articulating a dis-
tinction later made by Diderot and the French materialists between “la matière”
and “les matières,” that is, between theoretical and abstract prime matter and the
diverse and concrete substances of the chemist’s laboratory.82
Sennert’s theory is also an early version of a type of chemical structuralism
and there is an interesting interplay between essence and structure in his theory.
Although he considers simple atoms to be elementary particles, Sennert also
believes that the observable properties of different substances are determined by
the structure according to which their atoms are arranged, along with the sub-
stantial forms of the atoms. Sennert regards such corpuscular aggregations as
non-elementary atomic species that have specific chemical properties observable
at the macro-level. The chemical synthesis of substances simply involves altering
the structural arrangement of the elementary corpuscles of those substances in
order to form a new structural composition. When such microstructures are al-
tered, the new structural aggregation of the mixt constitutes a different material
species with different observable chemical properties.
Sennert’s structural explanation of mixture is that it involves a rearrangement
of the elementary “atoms” of the components so that the chemical properties of
the mixt are distinct from the chemical properties of the components. Therefore,
while the atoms of Substance A and of Substance B retain their substantial
forms when A is mixed with B, the resulting Substance C will have properties
that are distinct from A and B precisely because the structural arrangement of
82 For a fascinating discussion of the distinction between “la matière” and “les matières” in 18th-
century French materialist thought, see Pépin, La philosophie expérimentale de Diderot et la chimie,
and Pépin, “Quelques perspectives chimiques pour le matérialisme?”
the constituent atoms has been altered. For Sennert, although the structural ar-
rangement of corpuscles is altered during synthesis, the substantial forms of the
elementary particles remain unchanged, making it possible to recover the orig-
inal substances through analysis. The importance of these ideas for Sennert is
that they invalidate the Thomistic conception of mixts, particularly the claims
that the forms of the components are destroyed in the synthesis, that the mixed
substance has a distinctive form of its own, and that the components are not re-
coverable from the uniformity and homogeneity of the mixt.
As an experimental chymist with little sympathy for pure speculative philos-
ophy, however, Sennert recognizes the difficulty of positing any sort of immate-
rial substantial form to explain the essential properties of a material substance,
whether that substantial form is posited as belonging to the material substance as
a whole or to each of its elementary particles. The difficulty, of course, arises from
the fact that substantial form is regarded as being immaterial, that is, “completely
insensible, it is considered to be a causal terminus post quem [a causal limit] from
which perceptible qualities arise without revealing the nature of their source.
Substantial form, therefore, is a sort of ‘black box’, that is, a device or system that
can be viewed only in terms of its inputs and outputs or transfer characteristics.
from which qualities emerge. The unknowable nature of Sennert’s substantial
form is an Aristotelian empiricist’s statement of nescience.”83
In spite of these problems, Sennert believes that he can experimentally cor-
roborate his hylomorphic atomism by demonstrating, against the Scholastic
view of true synthesis, that substances are recoverable from mixts via chemical
analysis. Once this has been experimentally demonstrated, it will prove both that
the particles of the component substances are “atomic” and that these particles
retain their substantial forms in the mixt. Sennert uses several variations of the
“reduction to the pristine state” to establish that substances are recoverable from
homogeneous mixts.
Although earlier alchemists and iatrochemists had used the reduction to the
pristine state for strictly pragmatic purposes, Sennert uses this analytical proce-
dure for primarily theoretical purposes, to provide empirical support for the idea
that the atomic ingredients of a mixt are recoverable and that these atoms retain
their distinctive substantial forms within the mixt. For Sennert, “the validity of
chymical analysis is borne out by the Scholastic axiom ‘The things into which
composites can be dissolved are the things out of which they are made’—based
on Aristotle’s De caelo 3 302a 15–18.”84 It is interesting to note that Boyle would
later appropriate this same experimental procedure but he would use it for the
83 Newman, Atoms and Alchemy, 138–139.

84 Ibid., 97.
exact opposite theoretical purpose, that is, to undermine the idea of substantial
form in favor of the notion of mechanistic form.
Although Sennert performed reductions to the pristine state using various
mixtures, I will discuss only one of these, which is considered the most influ-
ential of his reductions, in which he dissolves silver in aqua fortis (nitric acid),
resulting in a clear blue liquid. The resulting liquid is filtered to show that no solid
residue remains and that the mix is homogeneous. Sennert then precipitates the
silver from the liquid by pouring salt of tartar (potassium carbonate) into the
solution. After cleaning, drying, and heating the precipitate, he observes that a
satisfactory quantity of the silver has been recovered from the solution, thereby
invalidating the Scholastic claim that the components of a homogeneous mixt
are not recoverable.
Sennert describes the results of this experiment as follows: “If aqua fortis is
poured on, the silver is so thoroughly dissolved that no metal can be detected in
the water by sight. But since it is really present, it can emerge thence in segregated
form, and certainly in such a way that . . . the silver retains its own nature.”85 He
explains that, although the liquid into which the silver is dissolved may seem to
be completely clear so that it can be filtered with no residue, nonetheless
the silver preserves its own nature in it . . . thus also if a single mass be made
by fusion of gold and silver together . . . the gold and other metals retain their
own nature in aqua fortis and can be precipitated again in the form of a powder,
sometimes dark, sometimes yellow, sometimes of another color.86
Sennert believes that the atomistic implications of this experiment are clear.
Despite having been completely dissolved in the acid, the silver can be recov-
ered to a great degree of quantitative precision. For Sennert, this means that the
corpuscles of the silver have retained their essential identity and their distinc-
tive substantial forms in the liquid, although these corpuscles have been clearly
restructured so that the qualities of the original silver are no longer observable in
the liquid. The action of the salt of tartar works to restore the original structural
composition of the silver atoms, which is made manifest when the original silver
precipitates out of the solution. Sennert, therefore, interprets this experiment as
confirmation of hylomorphic atomism.
85 Sennert, De chymicorum cum Aristotelicis et Galenicis consensu ac dissensu, 213: “Si vero possea
aqua fortis affundatur, ita solvitur argentum, ut ullum metallum in ea aqua deprehendi visu non
possit: cum tamen revera insit & hinc segretatum emergat . . . & argentum suam naturam retineat.”
86 Ibid.: “Argentum suam naturam retineat . . . quae in aurum & argentum purissimum fusione
iterum reducitur . . . Aurum, ut & alia metalla, retinent suam naturam in aquis fortibus in forma
pulveris, nunc atri, nunc flavi, nunc alterius.”
The fact that the silver can be recovered by precipitation confirms that its
corpuscles are atomic since the acid is able to break down their structural com-
position but not the corpuscles themselves. “Sennert’s explanation of the ap-
parent disappearance of silver in aqua fortis involves the necessary assumption
that the particles into which it was divided were extremely small”87 and this as-
sumption is reinforced when
the minuteness of the silver atoms is confirmed by passing them through filter
paper before their precipitation from the acid-silver solution. From the per-
spective of the “negative-empirical” principle, the corpuscles of the silver are
therefore operationally a-tomos— indivisible— since they have resisted all
efforts at laboratory decomposition into [their] components . . . the precipitated
silver particles are also so small that they satisfy another canonic criterion of
atomism—the requirement of minute size.88
Following this and similar experiments, including ones in which gold and
silver are alloyed and then recovered via analysis and precipitation, Sennert
concludes that 1) the analytical tools have not succeeded in breaking down the
particles of the metal, therefore these particles are operationally “atomic” and
2) the recovery of the metal to its original state and with great quantitative preci-
sion proves that the atoms of the component retain their substantial forms in the
mixt. Following these experiments, Sennert believes that he has ample empirical
evidence “for the two theses that he had wanted to prove—that a metal seemingly
mixed with an acid or with another metal retains its own nature intact, and that
it is composed of extremely tiny corpuscles. In doing this, he has simultaneously
shown the inadequacy of the . . . Scholastic theories of mixture while also pro-
viding a convincing demonstration of the reality of semi-permanent atoms that
experience no substantial modification.”89
It is important to emphasize that Sennert is not using the term “atom” in its lit-
eral sense, that is, to mean an ontologically indivisible particle. Rather, by “atom,”
Sennert means “a particle of matter that cannot easily be divided or decomposed
into other substances.”90 Thus, for Sennert, the term “atom” simply means “oper-
ationally indivisible,” thereby rendering atomicity as a negative-empirical con-
cept. Sennert also believes that there are different genera of atoms that reflect
different stages of corpuscular composition or structure. Those atoms of silver
and gold that Sennert claims to observe as the products of the “reduction to the
87 Newman, Atoms and Alchemy, 122.

88 Ibid., 99–100.
89 Ibid., 123.
90 Newman, “What Have We Learned from the Recent Historiography of Alchemy?,” 319.
pristine state” are “what he takes to be atoms of higher compositional stage, not
the tiny elemental corpuscles out of which the former are composed.”91 However,
although the visible atoms of silver that result from the reduction are at a higher
compositional stage, they are not further reducible via analysis and are, there-
fore, considered as operationally elementary. “In the context of his complex at-
omism, the analytical agents of the chymist are used by Sennert to define the
constitution of ‘indivisibility’.”92
From a historical and philosophical standpoint, Sennert’s work is interesting
for several reasons. First, although the reduction to the pristine state cannot ul-
timately prove Sennert’s compositional theory of matter, Bensaude-Vincent and
Stengers point out that “the operations that brought reversibility to the fore in-
spired a new classification of chemical procedures that theoretically favored pu-
rification processes.”93
Second, Sennert’s hylomorphic atomism serves as an important transi-
tional hypothesis from the Scholastic theory of substantial form to mechanistic
corpuscularianism, and he develops an innovative compositional theory whose
structural focus anticipates not only the work of later chemists such as Boyle but
also the eventual structural focus of modern chemistry.
Third, Sennert’s experimental work involves “both a synthetic and an an-
alytic stage . . . [this] paired cycle of synthesis and analysis used in a demon-
strative proof ”94 would have great repercussions in the early modern period,
such as, for example, when Boyle and Spinoza use experiments such as the red-
integration of potassium nitrate to provide empirical support for mechanistic
corpuscularianism.
Finally, Sennert’s negative-empirical approach and operational understanding
of atomicity would be important for future chemists “in considering the limits
of technical analysis to provide the natural philosopher with a working ‘atom,’
namely, any substance that was resistant to dissolution in the laboratory.”95 The
negative-empirical approach to elementarity “resurfaces most importantly in the
work of Lavoisier, who claims that the term ‘element’ should be restricted to the
‘last point which analysis is capable of reaching’.”96
As Newman points out, Sennert’s work shows “the interplay between the
two types of qualitative explanation [. . .] the structural and the substantial [. . .]
Sennert uses a generalized microstructural explanation in combination with
chymical properties originating in the substantial form to explain the origin of a

92 Ibid., 97.
93 Bensaude-Vincent and Stengers, A History of Chemistry, 33.
94 Newman, “What Have We Learned from the Recent Historiography of Alchemy,” 317–318.
96 Antoine-Laurent de Lavoisier, Traité élémentaire de chimie, xvii.
macrolevel effect.”97 It must certainly be recognized that Sennert is a man of his

time and that, despite the prescience of his atomism, of his analytical methods,
and of his theory of microstructure, his conception of matter is hylomorphic and
thus still firmly grounded in the Aristotelian and Scholastic notions of substance
that would later be supplanted by the mechanistic philosophy. In addition to this,
Sennert’s theory of matter retains significant elements of vitalism that ground it
in the traditions of Paracelsian alchemy. For example, Sennert believes that sym-
pathy or affinity and, very rarely, antipathy play an important role in the dissolu-
tion and precipitation of substances.
Finally, although he turns to experiment to support his hylomorphic at-
omism, Sennert fails to sustain the notion of corpuscular substantial form within
chemical explanations. This is because, as Bensaude-Vincent and Stengers have
pointed out, Sennert’s reduction to the pristine state is “not sufficient to support
his atomism, since the Scholastic idea of the mixt would be perfectly capable of
explaining reversibility.”98 For example, a Scholastic alchemist could argue that
the clear blue solution that results from pouring aqua fortis on silver is not a gen-
uine mixt precisely because the silver is recoverable by analysis, so Sennert’s ex-
periment does not prove his atomistic theory. In fact, this problem is an instance
of the underdetermination of theory by evidence, since the recovery of silver
from the nitric acid solution in which it has been dissolved can be explained by
several incompatible theories such as, in this case, Sennert’s hylomorphic at-
omism as well as the Scholastic theory of mixts. As we shall see in Chapter 3,
Boyle’s theory of mechanistic form represents yet another attempt at explaining
the result of this same experiment, using an entirely different paradigm and one
that provides a much more satisfactory heuristic alternative to the theory of sub-
stantial form in any of its iterations. It is indeed ironic that Sennert’s emphasis on
microstructure as the source of chemical qualities and chemical stability would
later be used by Boyle to significantly reinforce his arguments against the very
concept of substantial form. I will now turn to Jan Baptista van Helmont (1579–
1644), another important Paracelsian whose reconceptualization of the notions
of semina, minima, spirit, and ferment had great impact on the ideas of Boyle.
1.6 Jan Baptista van Helmont and the Chemical

Interpretation of Spirit and Ferment
The bulk of Helmont’s chemical and medical philosophies are concerned with
the activity of vital spirit in nature. He believes that all things in nature arise

98 Bensaude-Vincent and Stengers, A History of Chemistry, 32–33.
from spiritual seeds, that is, from semina that possess the life force of all animals,
vegetables, and minerals. By means of a ferment, semina mingle with water to
become individual entities. These ideas are interdependent with what Newman
calls Helmont’s “vitalistic corpuscularianism.” Helmont’s complex chemical
ontology integrates the non-vitalistic corpuscularianism that he inherits from
Geber with Paracelsian vitalism. He, then, also applies the Scholastic theory of
minima naturalia to his idea of chemical combination. Pre-Helmontian natural
philosophers had tended to embrace either the concept of minima naturalia or
the concept of semina rerum, depending on their own tendencies either toward
materialism or toward vitalism, since minima were associated with material-
istic ontologies while semina were associated with vitalistic conceptions of the
universe. Unlike his predecessors, however, Helmont does not consider these
concepts as incompatible and, instead, embraces both since each plays a distinct
role in his hybridized chemical ontology.
Helmont follows Paracelsus in interpreting semina as “the main agents in na-
ture [and as] spiritual non-corporeal entities,”99 while minima remain for him
strictly physical corpuscles. In fact, “Helmontian atoms are identical with the
minima naturalia, i.e., the smallest particles into which a substance may be di-
vided. There is little doubt that for van Helmont minima naturalia are actual
physical units. [However,] it is also apparent that they have qualitative deter-
minations, not mechanical properties.”100 These qualitative and non-mechanical
properties are accounted for by semina, which work together with minima to
bring about changes in nature by providing the spiritual force of action that
brings about qualitative chemical alterations. Helmont modifies the Geberian
and Scholastic theories by postulating that spiritual semina serve as the center for
the actualization of forms.
Like Severinus before him, Helmont rejects the Aristotelian theory of ele-
mental qualities and maintains that it is semina rerum that function as the causes
of all generation. He considers semina to be the link between the visible and
the invisible realms, that is, between the material and the spiritual domains.101
Following Paracelsus in favoring the Stoic over the Epicurean interpretation of
semina, Helmont believes that these are not corporeal or visible seeds. Rather,
they are spiritual substances “whose active ordering principle is an architec-
tonic spirit, endowed with scientia”102 or knowledge. Helmont calls this vital
spirit the archeus, and it is safe to say that the archeus fulfills the same function in
Helmontian ontology as the Logos had fulfilled in Neoplatonic ontology.
99 Clericuzio, Elements, Principles and Corpuscles, 56.

100 Ibid.
101 Clericuzio, “From van Helmont to Boyle,” 307.
102 Ibid.
According to Helmont, if natural substances did not contain a vital and sem-
inal principle, they would be “dead and inert.”103 For Helmont, semina contain
“information” that regulates the way a material substance will develop, in a pro-
cess that he regards as analogous to fermentation. The development of semina
occurs by way of an “idea” that is impressed upon the seed and that originates in
the divine mind. God’s agent in this process is the archeus, which transmits ideas
to semina.104 The efficient agency of the archeus is reminiscent of the role played
by the agent intellect in medieval Aristotelian hylomorphism. However, the dif-
ference is that for Helmont the archeus is an internal efficient cause, rather than
an external principle as the agent intellect was for the Scholastics.105
Helmont explains that all semina have their own archei, by which they are
directed to specific goals. The archeus that governs each semen is a spiritual sub-
stance directing the course of generation or, in the case of disease, that of cor-
ruption. Like Paracelsus, Helmont regards all organic processes as based upon
chemical events that are always presided over by the archeus, the vital spiritual
force that is the embodiment of energy. As well, like Paracelsus, Helmont rejects
the humoral conception of disease in favor of a pathogenic conception. So, for
him, a body is healthy when all of the different chemical processes are governed
correctly by the archeus. Disease, then, is understood as the failure of the archeus
to govern correctly, and this occurs as a result of germs entering the body and
bringing about chemical changes that the archeus cannot master. Finally, death
results when the archeus or vital spirit is lost, and natural chemical changes are
left entirely to themselves without any restraining influence,106 thereby resulting
in the decomposition of the body.
The ontological status occupied by the archeus is that of an interface between
the corporeal and the incorporeal. Though it is not essentially material, it does
exercise a profound effect on material processes. For example, in the human
body, the archeus directs morphogenesis and the generation of specific indi-
vidual organs. Given his belief in the universality of the archeus, therefore, it is
not surprising that Helmont would claim that all things arise from the action of
the internal archeus, the semina, and the resultant ferments.107 Thus, Helmont’s
comprehensive and influential worldview is clearly an attempt to unify chymical,
medical, and theological ideas.
According to Helmont, semina and archeus are the real internal princi-
ples directing both animate and inanimate things in their generation and
103 Hirai, Le concept de semence dans les théories de la matière à la Renaissance, 451.
104 Oldroyd, “Some Neo-Platonic and Stoic Influences on Mineralogy in the Sixteenth and
Seventeenth Centuries,” 140–141.
105 Hirai, Le concept de semence dans les théories de la matière à la Renaissance, 451.
106 Moon, “Van Helmont, Chemist, Physician, Philosopher and Mystic,” 25.
107 Debus, The Chemical Philosophy, 340–343.
development. In a 1631 letter to Marin Mersenne, he accuses Aristotle of having

simple-mindedly overlooked the important principle of the archeus in favor
of the very crude and mechanical notion of external efficient causes.108 Thus,
Aristotelian natural philosophy, in its ignorance of semina and archeus, reduces
much of nature to an artifact that is devoid of internal principles of motion or
change.
For Helmont, semina are indispensable self- moving principles and the
archeus, or vital spirit, functions as the “internal efficient cause” of all substances.
Helmont goes so far as to identify the archeus with “gas” and, thus, believes to
have made it empirically accessible. Semina are immaterial principles that are
hidden deep within bodies and, unlike material substances, “semina operate
by means of a ‘radial activity’ that need not involve physical, bodily contact.”109
Because semina are the origin of all genuine and real change, that is, of all chem-
ical change, their “deep internal workings . . . constitute the ultimate object of
practical laboratory studies”110 in Helmontian chymistry.
One important point to keep in mind is that one of Helmont’s major goals,
which underlies much of his chymistry, is the attempt to distinguish superficial
physical changes from the intimate chemical interactions that result in substan-
tial change, that is, in transmutation. Helmont makes a significant effort in trying
to distinguish superficial from substantial change and, to do so, he grafts his vi-
talistic interpretation of semina, archeus, and ferment onto the materialist and
corpuscular notion of minima naturalia.111 Helmontian chymistry distinguishes
the mere mechanical division and spatial displacement of minima, conceived as
corpuscles or “atoms,” from the “deep connection” and alteration of substances
that results from seminal interactions. As will be stressed, for Helmont, super-
ficial changes are merely physical and due to mechanical operations, such as
the division and spatial rearrangement of minima, while genuinely substantial
change is chemical and, therefore, involves the transformation of the original
corpuscles of a substance into corpuscles of a different type through the action
of ferments guided by semina. Therefore, although the aspects of Helmontian
chymistry dealing with changes of state can be correctly associated with “me-
chanical corpuscularianism,” those aspects of his chymistry that deal with
reactions, mixtures, and transmutations are more correctly associated with “vi-
talistic corpuscularianism.”112
108 van Helmont, “Lettre à Marin Mersenne,” 13: “Archeus sive causa efficiens interna (quam
Aristoteles ignoravit; omnem causam efficientem externam indigetans, rustico ac plane mechanico
intellectu).”
109 Newman and Principe, Alchemy Tried in the Fire: Starkey, Boyle, and the Fate of Helmontian
Chemistry, 62.
110 Ibid., 295.
111 Ibid., 64.
112 Newman, Gehennical Fire: The Lives of George Starkey, an American Alchemist in the Scientific
Revolution, 148–149.
Helmont appeals to the distinction between merely physical superficial

changes and truly chemical substantial changes to account for the various
alterations undergone by water, since water plays a central role in his chymical
philosophy. Ontologically, Helmont rejects the four elements of Aristotelian
cosmology and embraces, instead, an elementary monism. Inspired in part by
Thales, he believes that water is the universal element that, through transmuta-
tion, transforms into the various substances that exist in the world. Some of the
changes undergone by water are merely mechanical and superficial alterations.
These include the changes of state from liquid to solid, vapor, and gas (a term
coined by Helmont from the word “chaos”). These superficial changes do not in-
volve the action of semina but simply the spatial rearrangement of minima. Thus,
when explaining the superficial transformation of water into vapor, Helmont
relies on his theory that the corpuscles of water are composed of shells that cor-
respond to the three Paracelsian principles or tria prima, that is, the three princi-
ples of mercury, sulfur, and salt. The mercury and salt are found in the two outer
shells, while the sulfur inhabits the core. He claims that these three principles
cannot be separated in water, but they can exchange places. This is what Helmont
refers to as the “extraversion,” or the “turning inside out,” of parts.
Based on these assumptions, Helmont argues that the normal vaporization, or
sublimation, of water takes place as follows: Upon heating, the three principles
of water undergo “extraversion.” This exchange of place results in the extenua-
tion,113 or “thinning out,” of the particles of water into “atoms.” Since they are so
small and light, these “atomic” corpuscles are “driven up by the heat,”114 resulting
in a change of state from liquid to vapor. Helmont also explains the process of
freezing (or solidification) by employing a similar, albeit reversed, account of the
internal spatial rearrangement of the minima or corpuscles. His chief purpose in
both cases is precisely to explain, in purely physical and mechanical terms, the
striking changes observed when water vaporizes or freezes. The main point, in
Helmont’s own words, is that “It is not a new substantial generation when vapor
is elevated from water, since it is only an extenuation, due to the extraversion
of the parts. No mutation of essence occurs where there is only local division
and extraversion of parts.”115 Since these are not genuine chemical changes, they
cannot be accounted for by chemical explanations.
As mentioned earlier, however, Helmont considers water to be the funda-
mental element from which all other substances are formed. Thus, it can undergo
changes other than merely superficial changes of state, that is, it can undergo true
113 From the Latin extenuare, “to make thin” (based on tenuat, “it thins”).
114 Newman and Principe, Alchemy Tried in the Fire, 65.
115 van Helmont, Gas aquae, 75: “Non est itaque nova, ac substantialis generatio, dum ex aqua
vapor elevatur, cum sit tantum extenuatio, propter partium extraversionem. Non intercedit enim
essentiae mutatio, ubi sola est localis divisio & partium extraversio.”
“mutation of essence.” Such “mutation of essence” or transmutation of water into

different substances is a chemical change that is governed by the active ordering
principles of archeus and semina. These active principles radically convert water
into all the various substances that exist in the world.116 The complex process
by which the semina induce the passive material of water to take on the qual-
ities of other substances is fermentation. For Helmont, then, archeus, semina,
and ferment are closely related principles upon which all natural phenomena
ultimately depend. He also believes that most of these substances can eventu-
ally be transformed back into primordial water by using heat and cold, thereby
establishing a continuous cycle of creation and destruction.
Helmont tells us that “nothing doth arise anew in Nature, without a seed. In
the next place, Every seed operates by dispositions . . . which it propagates in the
matter for its intended [purpose].”117 Helmontian chymistry, however, is not the
only context in which the notion of semina is important for explaining gener-
ation and transmutation. To the extent that his chymistry is closely tied to his
medical ideas, Helmont also attempts to create medicines based on his concep-
tion of transmutation via semina. He argues that any substance heated with the
alkahest118 will be first decomposed into its proximate ingredients (that is, salt,
sulfur, and mercury) and, by being further heated, will be reduced to water. If
this process is stopped at the correct point and the alkahest is distilled off, the
“first essence” (ens primum) of the dissolved substances would be left behind as
a crystalline salt. This essence was said to contain the concentrated medicinal
powers of the dissolved substance, free from any noxious properties, much like a
spagyric preparation but, according to Helmont, far easier to prepare.
In these chemical and iatrochemical processes, Helmont tells us that “the
mediating Instruments, by which the semina dispose materials, [are] Ferments”
(Van Helmont 1662, 859). In De lithiasi, Helmont states that “ferments . . . are
instruments by which the indwelling seeds of all things, or semina, go about their
business . . . it is the ferments that allow these seeds to work.”119 For example, if
we examine Helmont’s explanation of the development of metals and minerals,
he claims that water interacts with a ferment (in a manner that he does not
specify) to generate the physical seeds that, in time, develop to produce a metal
or a mineral. “The ferment prepares, arouses and produces the seeds of corporeal
substances.”120 Thus, ferments serve as the mediating agents between the semina
116 Newman, “The Corpuscular Transmutational Theory of Eirenaeus Philalethes,” 65–66.

117 van Helmont, De lithiasi, in Oriatrike Or, Physick Refined (1662), 859.
118 In Paracelsian alchemy, the alkahest is a hypothetical universal solvent with the power to dis-
solve every substance, including gold.

119 van Helmont, De lithiasi, 29.
120 van Helmont, as quoted in Hirai, Le concept de semence dans les théories de la matière à la
Renaissance, 455: “Le ferment prépare, suscite et produit les semences des choses corporelles.”
rerum, as immaterial principles, and the physical seeds that produce the specific
substances in nature.
In his Imago fermenti, Helmont further explains that “fermentation
is the predecessor wholly necessary to every transmutation”121 and to
every real chemical and substantial change. Helmont believes that, when a
body is reduced into smaller atoms than its substance can stand, the body
will be transmuted. He explains that this reduction to smaller particles is
carried out by ferments: “The imbibed ferment, seizing the foresaid atoms,
imbues them with its own alien character, during the reception of which
there occur divisions of particles. The resolution of the matter follows these
heterogeneities and divisions of particles . . . into minima.”122 Therefore, a
ferment is a substance with the ability to divide matter into its most minimal
particles. Such division is the key to any genuine chemical reaction. Thus, for
Helmont, true mixture of substances must also be explained in these same
terms. After two substances have been divided into minimal particles, their
respective semina are free to act on one another. Only in this fashion can two
substances be truly mixed into a new, third substance. Once again, the cen-
tral agent in all these chemical processes is the archeus, which Helmont calls
the Archeus faber, the “Master-workman” and “Governor of generation.” He
explains that “any and all material bodies in the world require a beginning
to their movements, something that ‘excites and internally directs genera-
tion’ ”123 and his is the archeus.
Helmont’s complex chemical ontology is, therefore, the hybridization of the
“mechanistic” conception of corpuscles capable of undergoing spatial inver-
sion with the “vitalistic” conception of substantial chemical transformation via
ferments governed by the active principles of archei and semina rerum. Helmont’s
corpuscularianism, unlike the materialist atomism of antiquity, denies that
substances can genuinely form compounds by a mere juxtaposition of their
minute particles. As established earlier, Helmont distinguishes between “mere
apposition” of particles and true “wedlock.” The first of these is a “bare commin-
gling,” which Helmont contrasts to the genuine “marriage” that occurs when
substances are deeply connected. As in modern chemistry, Helmont asserts that
what we would call today a “mechanical mixture,” in which particles are merely
juxtaposed, is not a compound at all. Helmont considers this another example of
121 van Helmont, Imago fermenti, in Ortus medicinae, 69.

122 Ibid.: “Quaternus haustum fermentum, arripiens praefatos atomos, eos alieno sui charactere
imbuit, in cuius susceptione fiunt divisiones partium, quas partium heterogeneitates & divisiones,
resolutio materiae consequitur. Hactenus nimirum Chymia digerit, ac putrefactiones praemittit, ut
accepto fermento partes dehiscant in minima.”
123 Hirai, Le concept de semence dans les théories de la matière à la Renaissance, 457: “Van Helmont
explique ensuite que n’importe quel corps dans le monde a besoin d’un commencement de ses
mouvements, ‘excitateur et directeur interne de la génération’.”
a mere superficial change, not real or substantial and, therefore, not genuinely
chemical.
But here, the similarity between Helmontian chymistry and modern
chemistry obviously ends. Where modern chemistry speaks of compounds,
in which recoverable elements are held together by chemical bonds, Helmont
speaks of “indissoluble marriages.” As Newman and Principe explain,
Helmont’s idea of a “chemical change” is “considerably more restricted than
our own analogous division between physical and chemical change . . . [This
is because Helmont views] all processes in which the initial ingredients
are recoverable as examples of superficial, rather than of genuine substan-
tial change. Only when the change is so radical that the original substances
cannot be recovered”124 does Helmont consider that a true chemical reaction
has taken place. Such radical and irreversible transformations occur only
when substances are irrecoverably altered by the intervention of new semina
or the mortification of old semina. For example, Helmont does not consider
the dissolution of gold in aqua regia to yield a transparent yellow liquid to be
a genuine chemical change, “because the original gold may be recovered un-
changed from the solution by precipitating it with salt of tartar (or potassium
carbonate).”125
In his Progymnasma meteori, Helmont explain that only substances that can
“indissolubly marry,” and only in a very restricted way, are volatile substances
or “spirits.” If one tries to mix grosser matter, one will not achieve a permanent
coalescence (coalitus) but merely a juxtaposition of particles. Therefore, ac-
cording to Helmont, substances must be volatilized or subtilized before a true
compound can be formed or even before a transmutation can occur. Helmont,
however, does not believe that a mere mixture of spirits constitutes proper and
indissoluble chemical “marriage.” For such a “marriage” to occur, the substances
must be subtilized “to the point that they cannot be further [reduced].”126 If the
subtilizing continues beyond that point, the substances will irreversibly coalesce
and “will finally pass into another substance.”127 In Imago fermenti, it becomes
clear that Helmont’s concept of transmutation is based upon the fundamental
Scholastic principle, regarding minima naturalia, that “there is a natural limit
to divisibility, beyond which a substance, qua substance, cannot pass.”128 Thus,
he affirms that, if a substance should pass this limit, then it will become another
substance. Helmont expresses this idea once again, this time using the language
of atomism, by saying that: “I have found that so often as a body is divided into
124 Newman and Principe, Alchemy Tried in the Fire, 66.

125 Ibid.
126 van Helmont, Progymnasma meterori, 42.
127 Ibid.
128 Newman, Gehennical Fire, 142
smaller atoms than the necessity of its substance can stand, a transmutation of
that body will immediately follow, except in the case of an element.”129 There is
therefore, in Helmont’s theory, no intermediate state between mechanical mix-
ture and transmutation.
Although his theory of chemical change radically differs from our own
modern understanding, Helmont’s claim that non-substantial changes of state
are due to the spatial rearrangement of corpuscles ultimately had the inter-
esting consequence of facilitating the corpuscularian explanations for many of
the alterations observed in the 17th-century chemist’s laboratory. To elaborate
further, since Helmont considers all laboratory processes in which the initial
ingredients are recoverable as examples of superficial changes rather than gen-
uine substantial mutation, it follows that chemistry actually leaves considerable
room for explanations in terms of the spatial rearrangement of particles and,
thus, in terms of “mechanistic” corpuscularianism. By Helmont’s own standard,
short of genuine irreversible mixtures or of genuine transmutations, very few
laboratory processes would actually require explanations in terms of archeus,
semina, and ferments. Thus, although mere juxtaposition of parts, spatial rear-
rangement, and mechanical processes are relegated, by Helmont, to the realm of
the not genuinely chemical, they are nonetheless of key importance in his own
experimental work.130
As with Sennert, when we study Helmont, we must keep in mind that he is
a transitional figure and that his language and metaphors are imbued with vi-
talism and anthropomorphism, using such terms such as “marry,” “conceive,”
and “impregnate” to describe chemical processes. For example, he describes
the process of mixture very poetically as two substances losing their own being
to one another and, through the action of semina, giving rise to a third but dif-
ferent substance. In effect, the use of such metaphors illustrates the manner in
which Helmont has grafted the Geberian theory of minima naturalia with the
hylozoism of Paracelsus to arrive at a unique type of vitalistic corpuscularianism
that blurs the distinction of matter and spirit, thereby allowing him to attribute
to ferments guided by immaterial principles the power to carry out chemical
transformations. Helmont’s spiritualistic characterization of the fermentation
process contrasts sharply with the more materialist conceptions of fermenta-
tion of his time, such as that of Thomas Willis. However, in its various forms, the
fermentational program found powerful followers in the second half of the 17th
century.
Although interesting, Helmont’s theory of genuine transmutation as an ir-
reversible change of substance is in tension with his account of water as the
129 van Helmont, Progymnasma meterori, 72.

130 Newman and Principe, Alchemy Tried in the Fire, 66–67.
universal element from which all other substances are formed. He explains that,
in order to form all of the substances that exist, water undergoes a true “muta-
tion of essence” or transmutation, involving the active guidance of archeus and
semina. He adds that most substances can be transformed back into primor-
dial water, thereby establishing a continuous cycle of creation and destruction.
However, this latter claim conflicts with Helmont’s own claims about gen-
uine transmutation, since this process is supposedly irreversible. Nowhere in
Helmont’s writings does he either acknowledge or resolve this conflict between
his elementary monism and his theory of chemical transmutation.
Furthermore, although Helmont’s chemical interpretation of spirit and fer-
ment affirms that genuine chemical transformation requires the action of non-
corporeal semina rerum, guided by the spiritual archeus and acting by means
of fermentation, the great majority of his laboratory observations do not con-
form to his definition of genuine transformation or “true mixtures.” Most of
his experimental work involves substances that are ultimately recoverable
following analysis and synthesis. Although for Helmont only the immaterial
archeus, semina rerum, and ferments have genuine chemical functions and are
the true object of chemical study, he concedes that the majority of changes that
he observed in his own experimental work requires mechanical explanations
in terms of the extenuation of parts. However, despite its close ties to medieval
and Paracelsian alchemy, Helmont’s work goes far beyond that of Paracelsus
and his other predecessors, insofar as he takes one of the first steps toward the
naturalization of chemical explanations by employing the notion of a phys-
ical particle as the object of chemical change. Though his explanations con-
tinue to be steeped in the vitalistic ontology that dominated both medieval
and Renaissance natural philosophy, Helmont significantly contributes to the
naturalization of chemical ontology and of chemical philosophy and, thus, to
its modernization.
After Helmont’s efforts to provide a chemical interpretation of spirits and
ferments, it would not be long before other chemists completed the naturaliza-
tion of these two notions. Eventually, the mechanical philosophy championed by
Descartes, Spinoza, and Boyle would dominate the discourse regarding chem-
ical transformations and would explain the chemical activities of spirits and
ferments in terms of the mechanistic properties of fundamental particles. Thus,
although Sennertian and Helmontian chymistry made contributions that would
influence the chemical philosophy of Robert Boyle, the vitalism that lingered in
their chemical philosophies was a liability in an age that was quickly being over-
taken by the mechanistic philosophy. For many chemists, mechanicism provided
the only viable alternative to the theory of substantial form, and much more will
be said about this in the next chapter. However, that chapter will also show that
the Cartesian attempt to resolve the problem of substantial form by appealing
to a strict mechanicism entailed sacrificing the status of chemistry as systematic

natural philosophy. It would be up to Robert Boyle to find a way to reconcile
the mechanistic philosophy with experimental chemistry and he would do this,
in part, by embracing some of the key notions of Sennertian and Helmontian
thought.
2
Chemical Philosophy vs.
Rationalistic Mechanicism
The Heuristic Limits of Cartesianism for Chemistry
“Mechanicism,” or the mechanical philosophy, is the view “according to which

matter is inert and all interactions in nature are produced by the impact of
particles.”1 The mechanical philosophy, whether in its Cartesian or in its em-
piricist iteration, became quite critical of the Scholastic theory of substantial
form which, by the 17th century, had acquired a wholly concrete interpretation
as a power or cause “over and above” the material constitution of bodies from
which all their properties flow. Robert Pasnau mentions that the two chief 17th-
century criticisms against the notion of substantial form were that it was an ob-
scure notion and that forms should not be treated as substances.2 Despite their
many other disagreements, one thing about which both Cartesian and empiricist
mechanicists could agree was that the concept of substantial form, understood
as “something” over and above the material constitution of bodies that served
as the source of their phenomenal properties, had absolutely no heuristic or
explanatory value.
One of the questions that this book addresses in detail, in this and in the fol-
lowing chapters, is whether the mechanical corpuscularian hypothesis could
provide some other explanatory alternative to account for the properties of ma-
terial bodies or whether it could simply re-conceptualize the notion of form to
give it a mechanistic and corpuscularian interpretation. Here, I will show that
Cartesian mechanicists preferred the former option while, in the next chapter,
I will establish that Boyle chose the latter option.
2.1 The Cartesian Rejection of Substantial Forms
Descartes’ interpretation of substantial forms is that they are meant to be some-

thing subsistent in material bodies and, in a letter to the Dutch philosopher and

University Press.
DOI: 10.1093/oso/9780197502501.001.0001
physician Henricus Regius, Descartes states that “to prevent any ambiguity of ex-
pression, it must be observed that when we deny substantial forms, we mean by
the expression a certain substance joined to matter, composing with it a merely
corporeal whole, and which no less than or even more than matter, is a true sub-
stance or a thing subsisting per se, since it is said to be an actuality, and matter
only a potentiality.”3 According to Pasnau, Descartes’ criticism is based on a mis-
understanding of the Scholastic view of substantial forms, perhaps due to a con-
flation of the Scholastic view with some of Aristotle’s own pronouncements in
De anima. This being said, Descartes finds the notion of substantial forms to be
incompatible with a truly mechanistic corpuscularian view of inanimate mate-
rial bodies and believes that the mechanical philosophy can dispose of this idea,
since there is only one res extensa that is shared by all material bodies and its
mechanical affections fully account for the phenomenal qualities that we ob-
serve. Because he believes that the qualities and behaviors both of natural bodies
and of artifacts are fully accounted for by their mechanical affections, Descartes
rejects the Scholastic distinction between natural and artificial bodies according
to which the former are endowed with substantial forms while the latter lack
substantial forms.
Although Descartes believes that the mechanistic hypothesis can fully explain
both the properties and the behavior of all material bodies without recourse to
any notion of form, substantial or otherwise, his formless mechanicism does
not help him resolve the mereological question with which Aristotle was clearly
concerned in his Metaphysics. That is, how does one account for both the syn-
chronic and the diachronic unity that distinguishes composite (i.e., divisible)
wholes from mere collections or heaps. I specify composite wholes here because
Descartes has no trouble accounting for the synchronic and diachronic unity of
simple (i.e., indivisible) wholes such as res cogitans since, in this unique case, he
willingly appeals to substantial form, as did his Scholastic predecessors. The dif-
ficulty for composite wholes, however, also translates into a difficulty for realism
about natural kinds and for the notion that extended bodies can legitimately and
non-arbitrarily be classified as belonging to distinctive material species. Because
Descartes, like Hume, denies the presence of any sort of metaphysical “glue” that
holds material bodies together, in space and time, he must find some other way
of explaining unity. His mechanistic commitments preclude endorsing any sort
of force or occult quality that holds particles together to account for a body’s
3 Descartes, The Correspondence between Descartes and Henricus Regius /De briefwisseling tussen
Descartes en Henricus Regius, 106: “Ne enim aliqua sit ambiguitas in verbo, hic est notandum, nomine
formae substantialis, cum illam negamus, intelligi substantiam quondam materiae adiunctam, et
cum ipsa totum aliquod mere corporeum componentem, quaeque non minus, aut etiam magis quam
materia, sit vera substantia, sive res per se subsistens, quia nempe dicitur esse Actus, illa vero tantum
Potentia.”
Chemical Philosophy vs. Rationalistic Mechanicism 47
unity, nor does he wish to re-conceptualize the notion of form in mechanistic

terms. Thus, the mereological question of what accounts for the unity of com-
posite wholes remains unanswered.4
In his Principles of Philosophy, Descartes argues that particles are separated
by their motions and are joined together by their state of rest. His argument
suggests that appealing to another substance beyond that of the particles to ex-
plain unity would lead us into a regress when we try to explain how this other
substance holds the particles together. On the other hand, appealing to a mode
other than rest to explain unity would violate the principles of mechanicism by
invoking a non-mechanistic principle. Yet, the idea that composite wholes are
unified simply because their particles are at rest is an unsatisfactory explanation,
if for no other reason than that it cannot account for the stability of material
bodies. Thus, as already stated, the mereological question remains unanswered
as does the question of what accounts for the stability of material bodies that
warrants their classification into natural kinds or material species. The inability
to accommodate any notion of form or to re-conceptualize this notion in mech-
anistic terms is one of the weaknesses of this ontology. For this, among other
reasons, Boyle finds it difficult to fully reject the notion of form and, instead, tries
to reconceptualize it in mechanistic terms. As we shall see in the next chapter,
Boyle’s chemical philosophy represents one of several attempts to develop a con-
ception of form that resolves the questions of unity and identity, while remaining
completely committed to the mechanistic corpuscularian hypothesis.
2.2 Pierre Gassendi and the Reformation

of Epicurean Atomism
Pierre Gassendi (1592–1655) was arguably the strongest proponent of mech-

anistic Epicurean atomism in early 17th-century France although we now
know that, far from simply being a “rationalistic mechanicist,” Gassendi was
also influenced by Paracelsians such as Peter Severinus and Jan Baptista van
Helmont.5 From the point of view of chemical philosophy, his Philosophia
Epicuri syntagma (1649) is significant because it marks the shift from the vi-
talistic corpuscularianism of Sennert and Helmont to the mechanistic
corpuscularianism of the sort later defended by Walter Charleton, Robert
Boyle, and Isaac Newton. Gassendi contributes to the acceptance of Epicurean
atomism by making this view compatible with Christian theism, and he does
this by revising classical atomism in significant ways. Before discussing these
4 For a detailed discussion, see Pasnau, “Form, Substance, and Mechanism.”

5 Hirai and Yoshimoto, “Anatomie du chymiste sceptique,” 93.
revisions, however, it is important to point out that atomism is one among sev-
eral corpuscularian theories, not all of which accept the existence of atoms as in-
divisible particles. One such non-atomistic view is Cartesian corpuscularianism,
and it is important to examine this in some detail to understand where and why
Gassendi chooses to break with Cartesianism in this regard.
Descartes rejects atomism because he believes matter to be infinitely divisible
due to its essential property of extension. Thus, although the minutest detectable
particles may be called “atoms” for brevity’s sake, these particles are not truly a-
tomos, that is, not truly indivisible. According to Descartes, even if God wished
for some particles to be indivisible in nature, it does not follow that these particles
are indivisible in principle and that God could not divide them if he wished to do
so. In Principles of Philosophy (Principia philosophiae), he states “even if we sup-
pose that God wished to reduce some part of matter to a minuteness so extreme
that it could not be divided into smaller parts, we could not conclude from this
that it is indivisible. For if God rendered this part so small that no creature could
divide it, he certainly could not deprive himself of the power to divide it: because
it cannot be that he could diminish his own power.”6 Thus, Descartes reaches the
conclusion that “there cannot be any atoms or particles of bodies that are indi-
visible, as some philosophers have imagined. No matter how small we imagine
particles to be, since they must be extended, we cannot conceive that there is not
one among these that cannot be further divided in two or more even smaller
particles [. . .] We shall say then that the smallest possible extended particle that
can exist can always be divided.”7
Gassendi, however, rejects the Cartesian claim that the essential property
of matter is extension and that this makes the notion of indivisible material
particles incoherent. Instead, in the Syntagma philosophicum (1658), Gassendi
claims that matter is that which has dimensions and is capable of resistance,
which means that prime matter is constituted of solid and indivisible particles.
Because Gassendi wishes to attribute both empirical reality and causal power to
material atoms, he also rejects the Cartesian view that only geometrical points
can be considered indivisible and atomic, since these have no empirical reality
or causal power.
6 Descartes, Principia philosophiae, 51: “Quin etiam, si fingamus, Deum efficere voluisse, ut
aliqua materiae particula in alias minores dividi non possit, non tamen illa prioprie indivisibilis erit
dicenda. Ut etenim effecerit eam a nullis creaturis dividi posse, non certe sibi ipsi ejusdem dividendae
facultatem potuit adimere: quia fieri plane non potest, ut propriam suam potentiam imminuat.”
7 Descartes, Les principes de la philosophie, 74: “il ne peut y avoir des atomes, ou des parties de
corps qui [. . .] soient indivisibles, ainsi que quelques Philosophes ont imaginé. D’autant que, si petites
qu’on suppose ces parties, néantmoins, parce qu’il faut qu’elles soient estenduës, nous concevons qu’il
n’y en a pas une entr’elles qui ne puisse estre encore divisée en deux ou plus grand nombre d’autres
plus petites [. . .] nous dirons que la plus petite partie estenduë qui puisse estre au monde, peut tous-
jours estre divisée.”
Gassendi also disagrees with Descartes’ complete denial of form, since this
is equivalent to denying substantial identity. Gassendi points out that, when
material bodies undergo changes, there is something that undergoes those
changes but also that endures in time. This, however, does not mean that he
endorses the Scholastic notion of substantial form, an idea that all mechanistic
corpuscularians rejected. In fact, Gassendi does not purport to answer the ques-
tion of what the subject of change is. He simply limits himself to pointing out
that Descartes’ ontology is incomplete in this regard. Although we may never be
able to answer the question of what the subject of change is and although our on-
tology may always be incomplete, Gassendi claims that we should at least admit
this and not pretend that there is no problem, as Descartes seems inclined to do.
I wish to pause here and point out that Boyle attempts to pick up precisely where
Gassendi leaves off by attempting to resolve the question of the enduring subject
of change in mechanistic terms. Much more will be said about Boyle’s proposed
solution in the next chapter.
Returning to Gassendi, another disconcerting problem that he detects in
Cartesian corpuscularianism is its failure to explain an important mereological
“element in all natural development, for completed wholes [attain] to natures
that their parts [do] not possess.”8 Like Sennert before him and Boyle after him,
Gassendi opts for the hypothesis that “the primordial atoms combine with one
another to form compound corpuscles,”9 which he calls “molecules” (moleculae).
These molecules are mereologically distinct from atoms in that they are com-
posite and stable wholes that possess qualities not possessed by their constituent
atoms. To properly address the issue of whether wholes acquire properties that
are “over and above” those of their discrete parts, one must engage in a detailed
discussion of mereology and, in particular, of chemical mereology. I will engage
in such a discussion in Chapter 5 but, for now, suffice it to say that, although
Gassendi offers no mereological explanation for the stability of molecules or for
the relation between molecules and their constituent particles, he finds such an
explanation to be the only satisfying alternative to the reduction of all higher-
level properties to the mechanistic properties of fundamental particles.
Although, as an atomist, Gassendi believes in the ontological dependence
of the higher-level properties of molecules upon the lower-level mechanistic
properties of atoms, he also recognizes that this ontological dependence does
not entail explanatory reductionism. In fact, explanatory reductionism has
little heuristic value for Gassendi, at least when one is confronted with the task
of explaining the stability and chemical properties of substances. For him, be-
cause molecules are produced by chemical resolution, they are in a certain sense
8 Gregory, “The Animate and Mechanical Models of Reality,” 311.

“elementary,” although they are not simple or fundamental particles as such.

Though Gassendi does not speculate about what it is that makes molecules
stable, that is, what “holds” them together, he does hypothesize that molecules
are compound corpuscles that cannot be further analyzed and that serve as the
intermediaries between indivisible atoms and tangible perceptible bodies. In this
sense, Gassendi’s “molecule” is very close to Sennert’s “chymical atom” minus the
hylomorphism.
Although it must be stressed that, for Gassendi, the only true elements are
atoms since they are the only particles that are completely indivisible, he also
believes that there are several intermediary levels of compounded corpuscles
between fundamental atoms and concrete bodies, and it is these molecules that
compose the traditional chemical “elements” (sulfur, mercury, salt, earth, water).
Clearly anticipating the work of Boyle, Gassendi believes that textural alterations
to molecules produce new qualities in substances, and that such changes in qual-
ities can be induced by chemical operations. He suggests that the molecules of
chemical principles characterize the various species of bodies, depending on
their proportion and composition. He finds it difficult, however, to distinguish
homogeneous bodies with identical molecules from mixed bodies, especially
when determining the nature of metals.10
As mentioned at the beginning of this section, Gassendi contributes to the
acceptance of Epicurean atomism by revising classical atomism in important
ways to make it compatible with Christian theism. Significantly, Gassendi argues
against Epicurus that motion is not inherent in matter and that, therefore, an
external cause is required to impress motion upon atoms. This external cause is
God as the prime mover. Gassendi’s arguments in this regard were crucial to the
success of mechanistic atomism in Europe and had a great influence on the work
of Walter Charleton (1619–1707). Charleton’s work was, in turn, the primary ve-
hicle for the acceptance of mechanistic atomism in England and for the influ-
ence of Gassendi’s ideas on Robert Boyle. In fact, Boyle is one of the first English
scientists to embrace “purified” Epicurean atomism as advocated by Gassendi
and Charleton, though he adjusts Epicurean theory to make it compatible with
the Cartesian view of matter.
As we shall see in Chapter 3, Boyle attempts to bring the atomic and mechan-
ical philosophies within the compass of experiment. In fact, one of the main
reasons for Boyle’s influence on the Royal Society’s acceptance of mechanical at-
omism was that his experiments were indeed qualitatively and quantitatively su-
perior to those that had been previously performed by other chemists. However,
it will be shown that Boyle did not endorse a reductionistic chemical ontology
10 See Pinet, “La philosophie de la matière de Galilée à Newton,” 67–82.

and that much of his chemical work consisted precisely in rendering his mecha-
nistic and corpuscularian theory of matter compatible with the need for chem-
ical explanations that did not simplistically reduce higher-level properties to the
lower-level mechanical properties of corpuscles. In part, this may have been due
to the recognition that the sort of explanation championed by Spinoza and other
Cartesian mechanicists suffered from a lack of heuristic power and could, theo-
retically, be hostile both to the autonomy of chemistry and to chemical ontology.
2.3 The Limitations of the Cartesian Project for Chemistry

and Chemical Philosophy
Although not all mechanistic philosophers were Cartesians, mechanicism lent

itself well to the Cartesian project of building scientific theories in a deductive
manner, since “deduction allowed finding general principles under which elem-
ents of knowledge were to be structured.”11 Thus, when applied to chemistry, the
Cartesian project required that general deductive principles be assumed and then
employed to interpret the result of chemical experiments. It is not surprising,
therefore, that 17th-century chemistry textbooks typically endorsed a theory of
matter that was grounded in mechanicism but “very often defined chemistry as
an art, not as a purely deductive science in the manner of the Cartesian project,”12
precisely because chemistry did not lend itself to the sort of deductive derivation
favored by Cartesians.13
In fact, Alexis Smets identifies a tension, in the 17th century, “between chem-
ical and physical theories, and between practice and theory.”14 The controversies
that erupted due to this tension, such as that between Leibniz and Stahl, raised
“the crucial questions of the status and role of theories . . . as well as that of the re-
lations that chemistry should or should not keep with other disciplines, and no-
tably with mechanics and medicine.”15 Smets emphasizes, however, that “in spite
of the eventual difficulty of erecting chemistry on its own principles, there existed
a real need, internal to chemistry itself, for a theory that gave a solid account of
practice.”16 In the chapters to follow, I will demonstrate how Boyle attempted
to develop a chemical philosophy that could give such a solid account of prac-
tice within the context of a mechanistic theory of matter and that, in spite of his
11 Smets, “The Controversy between Leibniz and Stahl on the Theory of Chemistry,” 291.
12 Ibid.
13 Ibid.
14 Ibid.
15 Ibid.
16 Ibid.
failure in this regard, he laid the groundwork for Lavoisier’s later contributions to
the modernization of chemistry.
Returning to the Cartesian project, this project attempted to ultimately ex-
plain all qualities in terms of the quantitative mechanical affections of size,
shape, and mobility of fundamental particles and to formulate explanations of
all material phenomena, including chemical phenomena, in such mechanistic
terms. For Cartesians, all higher-level phenomena, properties, and processes
were deducible from mechanistic lower-level properties. In this view, “the phys-
ical world is represented by particles of matter in motion and can be interpreted
by the laws of motion determined by statistics [. . .] dynamics [and] mechanics
[. . .] Natural phenomena such as air resistance, friction, the different behaviors
of individual bodies, the qualitative features of the physical world were now con-
sidered irrelevant to the discourse of natural philosophy or viewed as disturbing
circumstances which were not [. . .] to be taken into account in an explanation of
the physical world.”17 For Descartes and his followers, “any explanation of natural
events requires the building of a mechanical model as a ‘substitute’ for the actual
phenomena being studied.”18 Mechanical philosophy is, therefore, anti-vitalistic
and anti-teleological, since it assumes that “nature is not the manifestation of a
living principle but is a system of matter in motion that follows [mathematically
precise] laws [. . .] the explanation of natural phenomena excludes all reference to
vital forces or final causes.”19
In addition to accounting for all the properties of matter in terms of the quan-
titative and mechanical affections of particles, Cartesian mechanicism accounts
for all changes in inorganic and organic material bodies through deterministic
and mechanistic laws of motion that are external to matter itself. According to
this view, “all interactions in nature are produced by the impact of particles”20 in
accordance with mechanical principles. As mentioned in the earlier discussion
of Gassendi, Cartesians sought to ensure that their affirmation of mechanistic
principles would be compatible with Christian doctrine by arguing that matter
is not intrinsically active or self-organizing and that its motions are not self-
determined. Instead, the motions and impact of particles are entirely determined
by mechanistic causal chains and external laws of motion determined by God.
Additionally, Descartes proposes to explain the variety of material bodies in the
universe, as well all qualities and phenomena, via the mechanical affections of
particles of universal matter.
17 Paolo Rossi, The Birth of Modern Science, 122.

18 Ibid., 125.
19 Ibid.
This discussion explains why the ultimate goal of the Cartesian project was
to erect all natural philosophical theories from a priori first principles, while
using deduction to structure those theories. As already suggested, this approach
engendered a certain hostility, on the part of Cartesians, toward the concep-
tion of chemistry as a theory of chemical qualities and substances and toward
the idea that chemical explanations were autonomous from mechanistic phys-
ical explanations. I will argue, in the chapters to follow, that this is precisely why
Boyle felt the need to develop a complex chemical philosophy and ontology that,
while grounded in the mechanistic theory of matter, could retain its explanatory
autonomy. As Mi Gyung Kim has recently pointed out and as I will elaborate in
greater detail in Chapter 3, “it is entirely possible that Boyle designed his ‘corpus-
cular’ philosophy as a hybrid of the alchemical particulate theories and the me-
chanical philosophy. The two different orders of particles, linked by elaborate yet
fragile connections, could meet the demands of chemical practice and system-
atic philosophy simultaneously,”21 thereby beginning to forge the link between
laboratory practice and philosophical chemistry that was eventually completed
via the 18th-century Chemical Revolution.
To understand the reason for which many 17th-century chemical philos-
ophies rejected strictly mechanistic explanations of chemical phenomena and
processes, one must closely examine the implications of Cartesian mechanicism
for chemical theory and practice. It must first be clarified that Descartes did take
an interest in the experimental sciences and believed that much practical know-
ledge could be acquired from them. In fact, Descartes was deeply interested in
the subject matter of the sciences and he offered explanations for many different
kinds of empirical phenomena, from the motions of the planets to the passage
of light from the sun to the earth and other phenomena. However, as Chalmers
explains, Descartes ultimately hoped to subsume these sciences under “his
mechanical account of the universe involving nothing other than portions of
matter/extension in motion . . . Insofar as basic processes . . . came to be regarded
as physical processes subject to mathematical laws, the characterization of phys-
ical causes came to be seen as intrinsically mathematical . . . A term that had
come into use in the early seventeenth century to accommodate these trends was
‘physico-mathematics’.”22
In the second part of his Principles of Philosophy (Les principes de la philosophie,
1647), Descartes posits that “there is only one material substance in the entire
universe, and we know it only because of its extension; and all the properties that
we distinctly experience in this substance are due to its divisibility into parts and
it receives the various dispositions that we experience by the movement of said
21 Kim, Affinity, That Elusive Dream, 42.

22 Chalmers, One Hundred Years of Pressure, 60.
parts.”23 In the fourth part of this massive work, Descartes attempts to apply the
principles of the mechanistic hypothesis to the objects of chemistry. However,
as Bernard Joly points out, in spite of addressing himself to the subject matter
of chemistry, Descartes has no intention of developing a specifically chem-
ical discourse.24 Rather, Descartes’ aim is to provide mechanistic accounts for
all properties of matter, including properties such as magnetism. For example,
he explains the phenomenon of magnetic attraction by appealing to differently
shaped particles that fit and lock into each other, very much like the pieces of a
jigsaw puzzle.
Although Descartes’ explanations are not all as simplistically mechanistic as
the example of magnetic attraction might imply, he does ultimately attempt to
explain even seemingly non-mechanical properties, such as fluidity and solidity,
in terms of the motion of corpuscles relative to each other. Liquid substances
like the universal ether are thought to be composed of corpuscles that are in
constant motion in relation to each other, while solid substances are com-
posed of corpuscles that are at rest with respect to each other. Even the “cos-
mical quality” of light, which is a seemingly non-mechanistic phenomenon, is
explained in terms of the linear movement of ether corpuscles away from the
sun. Furthermore, although he believes that such corpuscles are stable, Descartes
explains this stability by once again appealing to the relative rest of the parts of
each corpuscle in relation to each other.
Even when the young Descartes briefly wrote on hydrostatics, his interest in
the topic was ultimately absorbed within his “ambitious plan to construct an
outline of the ultimate corpuscular structure of the universe in its entirety.”25 As
Chalmers explains, “The corpuscular system of the world outlined in Descartes’
Principles of Philosophy treated pressure in liquids and the transmission of light
in analogous ways. Both were seen as resulting from the transmission of pushes
from corpuscle to corpuscle. This had the result that the linear transmission of
hydrostatic forces in liquids was as fundamental as the linear transmission of
light.”26 Besides hydrostatics, Descartes also demonstrated an interest in chem-
istry and anatomy, which he began studying in 1630 and from which he believed
much could be learned, as he admitted in a letter to Marin Mersenne dated April
23 Descartes, Oeuvres philosophiques de Descartes, 311: “Il n’y a donc qu’une même matière en
tout l’univers, et nous ne la connoissons que par cela seul qu’elle est étendue; et toutes les propriétés
que nous apercevons distinctement en elle se rapportent à cela seul qu’elle peut être divisée et mue
selon ses parties, et partant qu’elle peut recevoir toutes les diverses dispositions que nous remarquons
pouvoir arriver part le mouvement de ses parties.”
24 Joly, Descartes et la chimie, 106: “si Descartes aborde ici l’examen des objets de la chimie, il
n’entend pourtant pas développer un discours spécifiquement chimique.”

26 Ibid.
15 of that year.27 However, the problem for Descartes consisted in whether chem-
istry could qualify as a systematic natural philosophy or, as we would say today,
a theoretically autonomous science. This problem, in turn, related to the ques-
tion of whether chemistry could provide explanations that appealed primarily to
mechanistic causes.
In his Principles of Philosophy, Descartes certainly attempts to formulate what
could be regarded as a mechanistic approach to chemical explanations that
“proposes to explain the properties of chemical substances by invoking the con-
figuration of the ‘small particles’ of which they are formed; with regard to chem-
ical operations, one can hypothesize that they result from the manner in which
those small particles come into contact with one another through the incessant
movement in which they have been agitated by the flux of subtle matter.”28 For
example, his account of how different metals form in the earth is given exclu-
sively in terms of the shapes and sizes of the particles that compose the dif-
ferent metals: “According to the different sizes and shapes of these particles of
matter [. . .] they compose the different kinds of metals, which I would have
here explained in greater detail had I the convenience to conduct all of the
experiments that would be required to verify my reasoning on this subject.”29
In this work, Descartes also presents mechanistic accounts of chemical op-
erations such as distillation, which he explains as the forced escape of small
particles by means of an alembic. In his view, “the particles that can be easily
chased out of terrestrial bodies by the action of fire are of different types, as one
can experience quite easily through chemistry [. . .] there are those [. . .] that es-
cape quite easily from those bodies; that is to say, those that, being collected and
joined together through the means of an alembic, compose the aqua vitae, such
as we are accustomed in extracting from wine, from wheat and from numerous
other materials.”30 The caustic nature of acids is attributed to the pointy or spiky
27 Descartes, Letter to Mersenne (April 15, 1630), 21: “I am now studying chemistry and anatomy
simultaneously; every day I learn something that I cannot find in any book.”
28 Joly, “Chimie et mécanisme dans la nouvelle Académie royale des sciences: les débats entre
Louis Lémery et Etienne-François Geoffroy”: “Descartes propose de rendre compte des propriétés
des substances chimiques en invoquant la configuration des ‘petites parties’ dont elles sont formées;
quant aux opérations de la chimie, on peut faire l’hypothèse qu’elles résultent de la manière dont ces
petites parties entrent en contact avec les autres dans le mouvement incessant dont elles sont agitées
par les flux de la matière subtile.”
29 Descartes, Oeuvres philosophiques de Descartes, 380: “Selon les diverses grandeurs et figures
qu’ont ces parties du corps [. . .] elles composent diverses espèces de métaux, lesquelles j’aurait peut-
être ici plus particulièrement expliquées si j’avais eu la commodité de faire toutes les expériences qui
sont requises pour vérifier les raisonnements que j’ai faits sur ce sujet.”
30 Ibid., 393: “Les parties qui peuvent être chassées hors des corps terrestres par l’action du feu sont
de divers genres, comme on expérimente fort clairement par la chimie [. . .] il y en a [. . .] qui sortent
fort aisément hors de ces corps; à savoir celles qui, étant ramassées et jointes ensemble par le moyen
d’un alambic, composent des eaux-de-vie, telles qu’on a coutume de les tirer du vin, du blé et de
quantité d’autres matières.”
particles of acidic substances, and the neutralization of acids is explained as the

spiky particles fitting into the pores of the alkaline particles, thereby locking
them together and hiding the acidic particles’ points. The Cartesian account
does not suggest that chemical qualities are higher-level properties that have any
causal power in chemical processes, because causal efficacy is attributed only to
mechanistic properties and to primary causes.
In the preface to The Principle of Philosophy, Descartes faults the chemists of
his time for basing their conclusions on principles that are not primary and self-
evident by stating that “all conclusions that are deduced from a principle that
is not evident cannot themselves be evident [. . .] when one begins with flawed
principles, inasmuch as one cultivates them and applies them carefully to ex-
tract diverse consequences from these, thereby thinking that one philosophizes
well, then one will just as much distance oneself from the knowledge of truth and
from wisdom.”31 It is ironic that Descartes should say this, however, since the
veracity of the mechanistic corpuscularian hypothesis is simply assumed by him
and never proven, which would be an impossible task since these principles are
themselves neither self-evident nor empirically confirmable.
The French Cartesian Jacques Rohauld (1618–1672) admits that Cartesian
natural philosophy is just as guilty of postulating hidden causes for observed phe-
nomena as Paracelsian chymistry, though in the case of Cartesian mechanicism
these causes are corpuscles so small that they resist any possible detection.
Thus, Rohault poses the rhetorical question, “What good do those long and nice
Disputes do, about the Divisibility of Matter? For though it could not be accu-
rately determined, whether it be infinitely divisible or no; it would be sufficient
to know that it can be divided into Parts small enough to serve for all Purposes
that can be.”32
Yet, to fully understand the limitations of Cartesian mechanicism with regard
to chemical philosophy, one must fully grasp the rigidity of its position con-
cerning explanation in natural philosophy. All of the French Cartesians shared
the view that “phenomena have to be explained by the motion, figure, and size
of particles of a universal matter—even if one can distinguish different sorts of
particles. Secondly, the only admissible explanatory causes are the motion, size,
and figure of particles, because only these fit the nature of bodies [. . .] and are
able to explain their production. To explain a phenomenon is to explain this me-
chanical genesis.”33 For example, Géraud de Cordemoy is almost dogmatic in
31 Descartes, Les principes de la philosophie, 17–18: “toutes les conclusions que l’on déduit d’un
principe qui n’est point évident ne peuvent pas être évidentes [. . .] lorsqu’on a de mauvais prin-
cipes, d’autant qu’on les cultive davantage et qu’on s’applique avec plus de soin à en tirer diverses
conséquences, pensant que ce soit bien philosopher, d’autant s’éloigne- t-
on davantage de la
connoissance de la vérité et de la sagesse.”
32 Rohault, System of Natural Philosophy, preface.
33 Peterschmitt, “The Cartesians and Chemistry,” 194–195.
his appeal to strictly mechanistic explanations. In Le discernement du corps et

de l’âme en six discours (1666), he states, “I have made a little reflection upon the
notions that are held regarding bodies and matter, and I have seen that bodies
can be conceived only as indivisible substances and that matter can be conceived
only as a heap of these same substances.”34
As Luc Peterschmitt notes, there is no place for chemistry in this rigidly mech-
anistic atomism, claiming that “chemistry presents a problem for mechanism;
indeed, chemistry turned mechanism into a problem, because chemistry is a
limit for mechanism [. . .] chemistry reveals that mechanism is a scientific ide-
ology, extending concepts and models of explanation outside of their realm.”35
Peterschmitt illustrates this point by showing that the most strictly Cartesian
mechanicists of the 17th century either completely ignored chemical phe-
nomena in their explanations of processes such as dissolution and digestion
and attempted to explain these processes via strictly mechanical means or they
attempted to actually provide a critique and refutation of chemistry as a legiti-
mate experimental practice. Cordemoy favored the first approach, while Rohault
favored the second.
The basis of Rohault’s critique is the fact that, since analysis could not yield
fundamental particles, even mechanistic chemists had to treat the final products
of analysis as operationally “elementary” for heuristic purposes. Rohault
disapproves of this method used by chemists and claims it to be defective be-
cause it succeeds only in isolating “the sensible parts of which a body is com-
posed”36 while the fundamental principles escape it. “In his view, if experiment
and observation seem to lead to some states of matter (principles or elements),
what chymists present as their principles cannot be properly described as such.
A principle is not an empirical fact, but part of the theory.”37
As an aside, it is interesting to note that the methodological practice that
Rohault finds most objectionable, that is, the appeal to operational elementarity
rather than to fundamental principles, is precisely the method that was later
advocated by Antoine Lavoisier as a way of modernizing chemical practice
and of quantifying chemical explanations, thereby distancing chemistry from
its former more speculative and metaphysical concerns regarding the ultimate
34 de Cordemoy, Le discernement du corps et de l’âme en six discours, pour servir d’éclaircissement
à la physique, 15: “J’ay seulement fait un peu de reflection sur les notions que l’on a des corps, & de la
matière, & j’ay reconnu que l’on ne sauroit concevoir les corps que comme des substances indivisibles,
& que l’on ne sauroit concevoir la matière que comme un amas de ces mesmes substances.”
35 Peterschmitt, “The Cartesians and Chemistry,” 201.
36 Rohault, Traité de physique, 153: “Ce qui faite donc que je n’approuve pas la méthode des
Chymistes, c’est premièrement parce qu’elle est défectueuse; d’autant qu’il est certain qu’en travaillant
le plus exactement qu’il est possible, ils ne sauroient recueillir & ramasser que les parties sensibles
dont un corps est composé.”
37 Dobre, “Cartesianism and Chymistry,” 129.
nature of matter. Nonetheless, since Rohault favors a hypothetico-deductive

approach to science, the methods employed by chemists are unsatisfactory
and “their so-called principles are not well founded.”38 Thus, to the extent that
chemists’ access to true principles is operationally and empirically restricted,
“the whole of their knowledge amounts to being able to give names to things that
they do not know at all; thus it is impossible to predict what would result when
they are mixed, which is however one of the main conditions demanded of the
elements.”39 Consequently, by denying to chemical theory its predictive power,
Rohault is denying to chemistry the status of a legitimate empirical science.
It is important to note that, although Cordemoy, Rohault, and other strict
Cartesian mechanicists deny the scientific status of chemistry, they still consider
the activities of chemists to be useful for the progress of science. For example,
though Nicolas Lémery and Pierre-Sylvain Régis argue that the foundational
principles of chemistry must always be limited to negative-empirical concepts
and that only mechanistic physics can achieve a complete knowledge of fun-
damental principles, they also recognize the albeit limited heuristic value of
chemistry for understanding the nature and properties of composed and mixed
substances. In his Cours de chymie (1683), Lémery claims that
in chymistry, the term “principle” is not to be taken in an exact sense; because

the substances which are so named are principles only with respect to us, and
insofar as we cannot go further in the division of bodies; but one understands
that these principles are still divisible into an infinity of parts which may be
more justly termed principles [. . .] But if one wants to approach as near as pos-
sible to the true principles of nature, there is no more assured way than that of
chymistry [. . .] This art [. . .] will give a very great idea of the nature and figure of
the first little bodies that enter into the composition of mixed bodies.40
In his Cours entier de philosophie (1691), Régis states, in a very similar vein to
Lémery, that
38 Ibid.
39 Rohault, Traité de physique, 155: “Si bien que toute leur science se borne à savoir donner des
noms à des choses qu’ils ne connoissent point du tout, & du mélange desquelles il est part conséquent
impossible de prévoir ce qui en pourra résulter; que est cependant une des conditions principales que
l’on demande dans les Élemens.”
40 Lémery, Cours de chymie, 5: “Le nom de Principe en Chymie, ne doit pas être pris dans une sig-
nification tout à fait exacte; car les substances qu’on appelle ainsi, ne sont principes qu’à nôtre égard,
& qu’en tant que nous ne pouvons point aller plus avant dans la division des corps: mais on comprend
bien que ces principes sont encore divisibles en une infinité de parties qui pourroient, à plus juste
titre, estre appellées Principes [. . .] mais si l’on veut approcher autant qu’il se pourra des véritable
principes de la nature, on ne peut prendre une voye plus assurée que celle de la Chymie: cet Art [. . .]
donnera une fort grande idée de la nature & de la figure des premier petits corps qui ont entré dans la
composition des mixtes.”
among the chymists, the term principle is not to be taken in an exact sense;
because the substances to which they give this name are principles only with
respect to us, and insofar as art cannot go any further in the division of bodies,
although one can be sure that these principles are themselves composed of an
infinity of parts of the first element, which may be more justly termed princi-
ples. One must however grant that if one wants to approach as near as possible
to the true principles of nature, there is no more assured way than chymistry;
because [. . .] it gives nevertheless a very great idea of the nature and figure of
the insensible particles that enter into the composition of mixed, gross, and pal-
pable bodies.41
Peterschmitt convincingly argues that, for these Cartesian mechanicists,

“chemists did not develop a science, but they have worked diligently and have pro-
vided the materials for a science. Chemistry is at best an experimental art which
displays the phenomena that are to be explained [. . .] Without the chemists, it is
not possible to be a good and productive mechanist.”42 Ultimately, however, the
only genuinely scientific explanation is the one provided by the mechanicists,
that is, an explanation in terms of the mechanical affections of particles and
of the mechanistic principles that govern their motions. Unfortunately for the
Cartesians, however, they were unable to demonstrate the truth of their hypo-
thesis that chemical principles are equivalent with mechanistic affections of fun-
damental particles.
In a sense, such reasoning (i.e., proving an identity of properties) is not strictly

possible from a logical point of view. It would require showing that all the prop-
erties are the same, which is absurd in a practical sense, since it would require
an exhaustive enumeration of all properties [. . .] the question that remains
is whether it is really possible to give a mechanical explanation of chemical
matters of fact.43
Ultimately, in the mid-18th century, Lavoisier would circumvent these questions

entirely by forsaking all metaphysical speculations about the nature of ultimate
41 Régis, Cours entier de philosophie ou Système général selon les principes de M. Descartes, 333–
334: “Il est vray que le nom de Principe ne doit pas être pris chez les Chymistes dans une signification
tout à fait exacte; car les substances qu’ils appellent de ce nom, ne sont Principes qu’à nôtre égard &
en tant que l’Art ne peut pas aller plus avant dans la division des Corps, bien que nous soyons assurés
que ces Principes soient eux-mêmes composés d’une infinité de parties du premier Élément qui
pourroient à plus juste titre estre appellés Principes. Il faut avoüer pourtant que si l’on veut approcher
autant qu’il se pourra des véritables Principes de la Nature, on ne peut prendre une voye plus assûrée
que celle de la Chymie; car [. . .] elle donne néanmoins une fort grande idée de la Nature & de la figure
des particules insensibles qui entrent dans la composition des Corps Mixtes, grossiers & palpables.”
42 Peterschmitt, “The Cartesians and Chemistry,” 198.
43 Ibid., 200–201.
matter and by restricting his chemical theorizing to that which could be em-
pirically observed, measured, and classified. In his Essays de chimie (1702), the
chemist Wilhelm Homberg anticipates Lavoisier’s rejection of metaphysics when
he claims agnosticism regarding the nature of fundamental principles and states
that “we have not yet been able to establish anything incontestable as to the figure,
order and motion of the primary matters”44 and by concluding that chemistry
must restrict itself to material and sensible principles. Thus, as already noted, it
is precisely by embracing the negative-empirical conception of elementarity and
rejecting the appeal to fundamental principles that Lavoisier was ultimately able
to modernize chemistry as a purely empirical and quantitative science.
Due to its inability to account for chemical qualities and phenomena,
Cartesian mechanicism, as well as Descartes’ insistence that matter is inherently
inert, met with some degree of resistance in the context of 17th-century chem-
istry, physiology, and medicine. In fact, as was discussed in Chapter 1, for many
chymists and physiologists, organization and spirit were inherent in matter al-
though, in this transitional period, spirit was conceived in chemical terms
rather than as an immaterial substance. However, even for those chemists who
embraced the mechanistic corpuscularian philosophy as a theory of ultimate
matter, mechanicism had to be made compatible with the existence and causal
efficacy of chemical properties, as well as for the existence of stable chemical
substances. “In spite of the eventual difficulty of erecting chemistry on its own
principles, there existed a real need, internal to chemistry, for a theory that gave
a solid account of practice.”45 It is, therefore, in this context that we must under-
stand Robert Boyle’s chemical philosophy as an attempt to reconcile the mecha-
nistic hypothesis of fundamental particles with the empirical reality of causally
efficacious chemical properties, substances, and reactions.
2.4 Mechanistic Corpuscularianism and Experimental

Natural Philosophy
Although early modern particulate matter theorists clearly believed that atoms
and corpuscles have empirical reality, actual empirical support for atomism and
corpuscularianism was lacking in the early 17th century. As Christoph Meinel
aptly points out, “in atomism [. . .] there was no experimental proof possible, al-
though most corpuscular theories of the seventeenth-century explicitly claimed
to be based upon experience.”46 The (mostly unconvincing) arguments that were
44 Homberg, “Essays de chimie,” 33: “Nous n’avons pas encore pû déterminer rien d’incontestable
sur la figure, sur l’arrangement & sur le mouvement des premieres matières.”
45 Smets, “The Controversy between Leibniz and Stahl on the Theory of Chemistry,” 291.
presented in support for atomism fell into five general categories. The first group
of arguments, which can also be found in the writings of ancient Epicurean
atomists, involved analogically extrapolating the existence of atoms, as well as
their minute dimension, from macroscopic phenomena. Daniel Sennert, in fact,
performed an experiment in 1636 in which he “described a distillation in which
a stream of alcohol vapor passed through a sheet of paper, the density of which
was supposed to give an idea of how small the atoms really were [. . .] However,
it is clear that there was no quantitative methodology behind these indications.
Sennert used the language of the laboratory in a merely figurative and persuasive
manner, appealing to the imagination of the reader.”47
The second type of argument in favor of atomism was based on microscopic
observations. Since the invention of the compound microscope in the 1590s,
early modern scientists had been fascinated with “the worlds to be found in a
drop of water,”48 although it wasn’t until the publication of Giambattista Odierna’s
L’occhio della mosca (The Fly’s Eye) in 1644 that the first detailed account of micro-
scopic observations became available to the larger scientific community. It was
also not until the 1660s that the microscope became widely used for scientific re-
search. Gassendi was one of the first natural philosophers to realize the potential
of this instrument for research, although this “was presented as an empirical fact
by Henry Power a few years later.”49 However, as Power’s own attempts demon-
strated, the potential of the 17th-century microscope as a research tool was more
likely to be fulfilled in areas such as biology than in the theory of matter. The
best one could do, once again, was to extrapolate the existence and dimension
of atoms from what was observed via the microscope. Since [atoms] are so small
that they could be inferred only rationally, methodological difficulties [. . .] arose
if one attempted to model the real after the visible.”50
The third type of argument in favor of atomism and corpuscularianism con-
cerned “transport phenomena in which material substances appear or disappear
invisibly [. . .] [For example,] [t]‌he drying of bread or the slow evaporation of
liquids are material processes, although the flux of material cannot be observed.
In all these cases, quantitative change can be recorded, and from this the ex-
istence of invisible parts of matter could be inferred.51 However, although this
type of naively inductive argument rendered plausible the existence of atoms or
corpuscles, it could neither prove their existence nor establish the material iden-
tity of these particles.
47 Ibid., 78–79.
48 Ibid., 80.
49 Ibid.
50 Ibid., 84.
51 Ibid., 81.
The fourth type of argument, based on physical experiments involving

changes such as condensation and rarefaction, attempted to provide more
convincing empirical evidence for the existence and behavior of atoms. These
experiments remained inconclusive, however, because their results were open
to interpretation regarding the existence or non-existence of the vacuum. The
ancient atomists had posited a vacuum, to explain the mobility of hard and im-
penetrable atoms and, thus, the possibility of change. This vacuum could be
either a continuous void or an interparticulate void. However, the existence of
the vacuum was extremely controversial in the 17th century. Even after the first
vacuum was produced by Torricelli in 1643, in a column of mercury and under
controlled experimental conditions, it was not clear what the implications of this
experiment were for atomist and corpuscularian matter theories.
In fact, no clear consensus could be reached on this issue, with some
philosophers arguing for the continuous universal vacuum, others arguing
for interparticulate vacua, and others still arguing against the existence of any
vacuum at all and in favor of a substance, such as the ether, that both filled the
spaces between atoms and held the particles of bodies together.52 However, as
Meinel points out, “the re-introduction of an active spirit or ether into atomism,
aimed at explaining how the atoms interact and how their actions are transmitted,
undermined the theoretical consistency of the mechanical corpuscularianism.
[. . .] Plenist corpuscular theories such as Basso’s and Descartes’ exemplify that
it was entirely acceptable to assume corpuscles without admitting the void.”53
Still, the physical experiments themselves could not settle this question, since the
very results of these experiments could be interpreted to support any of the com-
peting hypotheses regarding the vacuum. It is no wonder, then, that Cartesian
mechanical philosophers such as Spinoza were suspicious of the usefulness of
experiment for confirming the mechanical hypothesis and ultimately sought to
replace the uncertainty of laboratory practice with the rigorously mathematized
discourse of physics.54
It seems then that the fifth type of argument, based upon chemical processes
such as the reduction and redintegration of substances, had the greatest poten-
tial for providing empirical support to atomistic and corpuscularian theories of
matter. As will be discussed in more detail in the next chapter, one of the impor-
tant experiments that Robert Boyle appropriates in this regard is Daniel Sennert’s
reduction to the pristine state. There is, however, another important experiment
that Boyle uses for lending empirical support to mechanistic corpuscularianism,
52 Ibid., 82.
53 Ibid., 83.
54 Joly, “Le cartésianisme de Boyle,” 155: “Les pratiques incertaines du laboratoire doivent [. . .]
laisser place à la rigueur d’une physique rationnelle qui se justifie dans la mathématisation de son
discours.”
that experiment being the redintegration of potassium nitrate. To the extent that
Spinoza disagreed with Boyle’s interpretation of this experiment, it became the
focus of a very lively epistolary debate between Boyle and Spinoza, a debate that
was mediated by their mutual friend Henry Oldenburg who was responsible both
for translating and for transporting the letters between the two philosophers
from England to the Netherlands and back. The historical significance of this ex-
periment is that it illustrates the distinction between those natural philosophers
who were committed to the value of experiment for confirming or disconfirming
hypotheses regarding the nature of matter and those philosophers for whom ex-
periment had little heuristic power and who relied instead on a priori arguments
to support their theory of matter.
Before I discuss the Boyle-Spinoza controversy over the interpretation of the
redintegration of potassium nitrate, however, I will describe Boyle’s experiment
and its results. Boyle’s reasons for choosing to work with potassium nitrate are
worth mentioning. Seemingly in agreement with Paracelsus that “no other salt
in the world is like [saltpeter],”55 Boyle may have chosen saltpeter for his exper-
iment because he believes it to be “ ‘the most catholic of salts,’ a most puzzling
concrete, ‘vegetable, animal, and even mineral,’ both acid and alkaline, ‘partly
fixed and partly volatile.’ The knowledge of it, Boyle opined, ‘may be very con-
ducive to the discovery of the nature of several other bodies, and to the improve-
ment of divers parts of natural philosophy.’ ”56 It is, perhaps, because of these
properties of potassium nitrate that Boyle believes he can analyze it into “fixed
nitre” and “spirit of nitre,” which have seemingly contradictory properties. He,
thus, sets out to analyze a sample of potassium nitrate and then to resynthesize
the products of analysis to yield the original saltpeter.
The experiment proceeds as follows: during the process of analysis, Boyle
throws hot coal into a measured sample of potassium nitrate (KNO3), also
called niter or saltpeter, thereby decomposing the saltpeter and yielding “fixed
nitre” or potash (K2CO3 or potassium carbonate) and nitrogen dioxide gas
(NO2). The nitrogen dioxide gas is then condensed on the glass and mixed with
water to produce “spirit of nitre” (HNO3, aqua fortis or nitric acid). The analysis
complete, Boyle proceeds to redintegrate the saltpeter by synthesizing the “fixed
nitre” or potash with “spirit of nitre” that has been sourced separately, thereby
obtaining a good amount of potassium nitrate of negligibly lesser weight than
the original sample.57 According to Boyle, this experiment shows that potas-
sium nitrate or saltpeter is a chemical compound, rather than a mixture, be-
cause the properties of the saltpeter are distinct from those of fixed niter and
55 Paracelsus, as cited in Clericuzio, “Carneades and the Chemists,” 52.

56 Cressy, Saltpeter: The Mother of Gunpowder, 14.
57 Duffy, “The Difference between Science and Philosophy,” 126.
the spirit of niter, fixed niter being alkaline and spirit of niter being volatile and
acidic. Boyle states that
when salt-petre is distilled, the volatile liquor and fixed salt, into which it is
reduced by the fire, are endowed with properties exceedingly different both
from each other, and from those of the undissipated concrete. For the spirit
of nitre is (as we formerly have observed) a kind of Acetum Minerale, and
possesses the common qualities to be met with in acid spirits [. . .] whereas the
fixed nitre is of an alkalizate nature, and participates [in] the qualities belonging
generally to lixiviate salts; and salt-petre it self is a peculiar form of salt, dis-
criminated by distinct properties both from those salts, that are eminently acid
[. . .] and from those, that are properly alkalizate [. . .] And whereas salt-petre
it self is partly fixed, and partly volatile, the acid ingredients of it are altogether
volatile, [and] the alkalizate fixed.58
Boyle concludes that the parts constituting saltpeter are of “different specific
and chemical natures,” although made of the same universal matter as saltpeter
itself. These constituent parts exhibit properties that are distinct from those of
the potassium nitrate: alkalinity and fixedness for the fixed niter, acidity and
volatility for the spirit of niter. From this reconstitution, Boyle infers that salt-
peter is a heterogeneous compound in which the fundamental corpuscles of the
component parts remain unchanged, but the chemical properties are altered and
transformed because of the structural change that brings about the re-synthesis.
[Boyle’s] explanation of the experiment was not grounded on the primary and
mechanical affections of particles. His interpretation of the “redintegration” of
nitre was based on the consideration that nitre was a compound body and that
the parts into which it was analyzed were not “volatile and fixed parts of that
concrete” but two distinct substances of different nature, which are obtained
from nitre by altering its texture. From this alteration, it follows that corpuscles
of spirit of nitre and those of fixed nitre “are enabled to disband from concrete
and associate themselves with those of their own nature.”59
For the sake of accuracy, it is important to note that Boyle’s account of the
analysis and redintegration of saltpeter does not correctly render what actually
occurs in the experiment. The reason for this error is that Boyle was neither aware
of the contributions made by the coal and the water to the reactions nor was he
aware that these contributions account for the differences in properties between
58 Boyle, Certain Physiological Essays, in The Works of Robert Boyle, 105.

saltpeter and those of fixed niter and spirit of niter. To better understand the ac-
tual chemical process involved, the experiment can be illustrated as follows, with
the relevant substances in the analysis and synthesis indicated in bold:
1) Analysis of potassium nitrate:
4KNO3 + 3C → CO2↑ + 2NO2↑ + N2↑ + 2K2CO3 (K2CO3 — alkaline

and fixed)
2) Reaction with water:
2NO2 + H2O → HNO3 + HNO2 (HNO3 —acidic and volatile)
3) Redintegration (or re-synthesis) of potassium nitrate:
K2CO3 + 2HNO3 → 2KNO3 + H2O + CO2↑
In step one of the analysis (4KNO3 + 3C → CO2↑ + 2NO2↑ + N2↑ + 2K2CO3),

coal contributes carbon to form fixed niter or potassium carbonate, which is
strongly alkaline due to its carbonate ion. In step two of the analysis (2NO2
+ H2O → HNO3 + HNO2), water contributes hydrogen to form spirit of niter
or nitric acid, which is strongly corrosive. In the process of re-synthesis or
redintegration (K2CO3 + 2HNO3 → 2KNO3 + H2O + CO2↑), the potassium
and oxygen combine with the nitrogen of spirit of niter to form potassium
nitrate once again, while the remaining oxygen combines with hydrogen to
yield water and with carbon to yield carbon dioxide, which is released into
the air. However, because he is unaware of the actual chemical reactions in
this experiment, Boyle does not attribute the alkalinity of the fixed niter
to the carbonate ion contributed by the coal and the acidity of the spirit of
niter to the hydrogen contributed by the water. Instead, Boyle mistakenly
attributes the changes in properties to changes in the textures of the cor-
puscular concretions and uses this experiment to support his structural
corpuscularian theory of chemical substances, which will be discussed in de-
tail in Chapter 3.
Nevertheless, notwithstanding the errors in Boyle’s interpretation of the
reactions, what makes his interpretation philosophically interesting is that, de-
spite his commitment to the mechanical philosophy, Boyle does not invoke
the mechanical affections of fundamental particles but appeals, instead, to the
chemical properties of the substances involved in the experiment. Bernard Joly
makes an important point that bears repeating: for Boyle, although mechanical
explanations sustain the chemical explanations, the latter remain autonomous
and cannot be replaced or clarified by the former.60 The reasons for this will be-
come clear once Boyle’s complex corpuscularian theory of matter is discussed in
the next chapter.
As mentioned earlier, Boyle’s interpretation of this experiment is at the center
of a series of epistolary exchanges with Spinoza, via Henry Oldenburg’s medi-
ation. Spinoza specifically rejects Boyle’s argument that the analysis of the salt-
peter is a decomposition of a chemical compound into two different substances
with distinctive chemical properties. Spinoza also rejects the claim that the re-
composition of the two separate substances alters their texture and, thereby,
their properties to transform them once again into saltpeter. Instead, Spinoza
argues that the experiment is consistent with the Cartesian mechanistic view that
differentiations among extended substances are always due to different quanti-
ties of motion and rest of the particles. Spinoza argues that the different proper-
ties of the fixed niter, the spirit of niter, and the potassium nitrate (or saltpeter)
are due to a difference in the motion of the basic particles of universal matter,
rather than to an essential difference in structure or texture as Boyle surmises.
For Spinoza, the a priori approach to the theory of matter succeeds in
uncovering the real nature of matter, and he does not believe that Boyle’s ex-
periment either amplifies or improves on the Cartesian approach. As well, for
Spinoza, the only criterion that accounts for the variety of material bodies and
their properties is the quantifiability of the mechanical properties of fundamental
particles.61 In his correspondence with Oldenburg, Spinoza declares that “Boyle
infers from his experiment concerning the reconstitution of Niter that Niter
[potassium nitrate] is something heterogeneous, consisting of fixed and volatile
parts, whose nature (so far as the Phenomena are concerned, at least) is nonethe-
less very different from the nature of the parts of which it is composed, though it
arises solely from the mixture of these parts.”62 Spinoza, however, rejects the view
that a whole compound could possibly display novel or distinct properties from
those of its parts, and this is one reason for his rejection of Boyle’s explanation.
Spinoza proposes a different interpretation of the experiment that he him-
self reproduces several times. He tells Oldenburg that “in order [. . .] to explain
Louis Lémery et Etienne-François Geoffroy,” 4: “Boyle expose la suite des opérations par lesquelles il
divise le salpêtre en esprit acide et sel alcali puis le recompose à partir des deux substances que la dis-
tillation avait permis de séparer, la réversibilité des opérations constituant la preuve que la composi-
tion du salpêtre, alors considéré comme une substance hétérogène, est bien connue. Or, pour rendre
compte de ces opérations, Boyle n’invoque pas tant le mouvement et la forme des corpuscules que les
qualités chimiques des diverses substances qui ont la propriété de dissoudre, de précipiter ou de fixer.
Bref, pour Boyle, l’explication chimique garde son autonomie et ne peut être réduite à une explication
mécanique qui sans doute la sous-tend mais ne peut jamais être explicitée en tant que telle [emphasis
mine].”
61 See Gabey, “Spinoza’s Natural Science and Methodology,” 177–179.
62 Spinoza, The Collected Works of Spinoza, 173–174.
this phenomenon in the simplest possible way, I will suppose no other distinc-
tion between spirit of nitre and nitre itself than that which is sufficiently man-
ifest: namely, that the particles of the latter are in a state of rest, while those of
the former are swiftly moved with respect to one another.”63 Spinoza concludes
that the spirit of niter is simply the volatile state of pure, crystallized saltpeter
and he dismisses the fixed niter as an impurity that was mixed in with the orig-
inal sample of potassium nitrate. Thus, the only difference between saltpeter and
spirit of niter is a difference in the motion of their basic particles, whose basic
shapes are the same, rather than to a structural difference in the texture of cor-
puscular concretions.
The Boyle-Spinoza debate regarding the redintegration of potassium nitrate
is a concrete example of the fundamental disagreement over the role of exper-
iment in adjudicating between different explanatory models. Although Boyle’s
interpretation of the experimental results is no less theory-driven than Spinoza’s,
Boyle considers these results to serve the heuristic purpose of settling the ques-
tion in favor of corpuscularianism over the theory of substantial form. For Boyle,
the experiment clearly demonstrates that hylomorphism is mistaken because, if
the substantial form of saltpeter were destroyed in the analysis, its redintegration
would not have been possible. Although Spinoza agrees with Boyle’s rejection of
hylomorphism, he rejects Boyle’s chemical account of the reactions and insists
on imposing an a priori explanatory framework upon the experimental results.
Thus, Spinoza’s interpretation of the experiment is not only theory-driven, but it
is driven by an a priori framework that nullifies the role of experiment in adjudi-
cating between competing explanatory models.
Mechanistic theoreticians such as Spinoza do not completely reject the role
of experiment in establishing a theory of matter. However, Spinoza is intent on
upholding the primacy of rational intuition over sensory evidence by forcing
an a priori mechanistic explanation of experimental results that might other-
wise suggest a different understanding of the nature of matter and its properties.
Therefore, he does not consider a priori hypotheses regarding the nature of fun-
damental matter to be refutable by experimental evidence. In fact, he claims that
experiment “cannot uncover the nature of things: sensory knowledge belongs to
the imagination, the knowledge of essences and causes to the intellect alone.”64
In order to experimentally disprove a conclusion that is dictated by rational in-
tuition, one would have to establish its logical impossibility, which is something
that no experiment can do. For Spinoza, while experimental results that are con-
sistent with an a priori theory can be taken to confirm the theory, results that
are incompatible with a rationally determined theory cannot weight against that
63 Oldenburg, The Correspondence of Henry Oldenburg, 449–459.

64 Gabey, “Spinoza’s Natural Science and Methodology,” 177–179.
theory. Thus, since Boyle’s experiment does not establish the logical impossibility
of the Spinozist account, then there is no reason to interpret this experiment in
any other way than in the fashion proposed by Spinoza.
Although Spinoza’s position may seem eccentric, “alleged experimental
confirmations of mechanical hypotheses were far from being as conclusive as the
theoreticians and experimentalists of the mechanical philosophy pretended.”65
As Thomas Kuhn points out, and as was discussed earlier in this chapter, the
same mechanistic philosophy that helped Boyle to argue against the theory of
substantial form “proved a sterile and occasionally adverse intellectual climate
for an understanding of the processes underlying chemical change.”66 An ex-
perimental chemist such as Boyle certainly used the mechanical philosophy to
buttress his understanding of the fundamental nature of matter, but he was not
interested in subordinating chemical explanations to mechanistic ones.
By its very nature, the chemical approach was pragmatic, realistic, and eclectic.
Most chemists worked on real matter and real properties in a purposeful way
[. . .] They simply could not afford to rely too closely on a rigid theory [. . .]
Hence, it was not the mechanical philosophy that was to succeed in chemistry,
but a noncommittal, substance-oriented notion of the corpuscle as something
closer to an elementary particle or a small amount of substance, corresponding
to something real in the chemists’ vessels and furnaces and endowed with sen-
sible properties.67
As a chemist, Boyle was strongly committed to explaining chemical phe-

nomena in a manner that accounted for the qualitative changes that occur in
chemical processes. He was, therefore, not satisfied with simply explaining all
chemical reactions and transformations by appealing to the motions and other
mechanical affections of fundamental particles. Boyle clearly understood that
strict mechanicism held negative implications for considering chemistry as a sys-
tematic natural philosophy independent of mechanics or, to use more contem-
porary terminology, for considering chemistry as an autonomous science with
its own distinctive explanations. He did not favor the kind of mechanistic in-
terpretation of all phenomena that was advocated by Descartes in the Principles
of Philosophy precisely because such an interpretation comes at the price of
being unable to predict novel chemical properties of bodies. In fact, it is not sur-
prising that “the most ardent defenders of Cartesian mechanism, at the end of
the 17th and the beginning of the 18th centuries were [. . .] virulent adversaries
65 Ibid., 180.
66 Kuhn, “Robert Boyle and Structural Chemistry,” 15
of chemistry,”68 precisely because they understood that the requirement for ul-
timately mechanistic explanations leaves little room for any heuristically inde-
pendent science other than physics.
2.5 Boyle’s Relation to the Cartesian Project

in Natural Philosophy
Despite Boyle’s resistance to the project advocated in Descartes’ Principles of

Philosophy, many early modern scholars considered Boyle to be an advocate not
only of the mechanical philosophy but of the very strict type of mechanistic ex-
planation favored by Cartesians. According to Joly, this attribution can be traced
at least as far back as Fontenelle who, in his History of the Royal Academy of
Sciences (Histoire de l’Académie Royale des Sciences) of 1733, not only identifies
corpuscularianism with Cartesian mechanicism but also attributes these undif-
ferentiated positions to Boyle.69 Joly explains that, after having opposed the “spirit
of chemistry” to the “spirit of physics,” “Fontenelle, in a manner that may seem
rather abusive, effects a double operation. First, he identifies corpuscularianism
with Cartesian mechanism, in other words, with the idea that the explanation of
all natural phenomena must be ‘reduced’ to the invocation of mechanical rela-
tions among ‘small particles’; there are, therefore, only two variables: the config-
uration of these small particles (their size and shape) and their motion. Secondly,
[Fontenelle] attributes this position to Boyle.”70
Gabriel François Venel and several other 18th-century chemists went so far
as to regard Boyle as more of a physicist than a chemist, precisely because of his
association with what they called “the corpuscular-mechanical philosophy.” In
his article titled Chymistry or Chemistry, written for Diderot and d’Alembert’s
Encyclopédie, Venel says of Boyle that he “is ordinarily counted among the
chemists, and he has indeed written much on chemistry. However, he is too
much of a corpuscularian-mechanistic physicist or, strictly speaking, a physi-
cist, so that we have contrasted him with the chemist at the beginning of this
68 Joly, “Le cartésianisme de Boyle du point de vue de la chimie,” 55: “Une fois ses opérations
expliquées, la chimie disparaît pour laisser la place à une science mécaniste qui se reconnaît sous
le nom de physique. On ne s’étonnera pas, dans ces conditions, que les plus ardents défenseurs
du mécanisme cartésien, à la fin du XVIIe et au début du XVIIIe siècle aient été [. . .] de farouches
adversaires de la chimie.”
Louis Lémery et Etienne-François Geoffroy,” 3.

70 Ibid.: “Fontenelle effectue, d’une manière qui peut sembler abusive, une double opération. D’une
part il identifie le corpuscularianisme au mécanisme cartésien, c’est-à-dire à l’idée selon laquelle

l’explication de tous les phénomènes naturels doit se ‘réduire’ à l’invocation de relations mécaniques
entre les ‘petits corps’; il n’y a donc que deux variables: la configuration de ces petits corps (leur taille
et leur figure) et leur mouvement. D’autre part [Fontenelle] fait de cette position celle de Boyle.”
article.”71 Several of Boyle’s works do indeed contain very clear statements in

favor of mechanicism, especially The Excellency and Grounds of the Corpuscular
or Mechanical Philosophy, the title of which clearly conflates the notion of
corpuscularianism with that of mechanicism as though these notions im-
plied each other.72 However, as Chapter 1 extensively argued, not only does
corpuscularianism not entail mechanicism but, especially in the 16th and 17th
centuries, corpuscularian theories of matter were often accommodated within
vitalistic ontologies. Thus, “the acceptance of a particulate theory of matter very
rarely involved the idea that all natural phenomena could be accounted for by
means of particles endowed only with mechanical properties.”73
Despite Boyle’s unfortunate conflation of corpuscularianism with
mechanicism in the title Of The Excellency and Grounds of the Corpuscular or
Mechanical Hypothesis, the evidence shows that he does not embrace the strictly
Cartesian mechanicism attributed to him by Fontenelle, Venel, and some con-
temporary scholars and that his corpuscularianism was instead influenced by the
work of Helmont, Sennert, and other vitalistic corpuscularians. As argued in this
and in the following chapters, Boyle’s complex chemical ontology clearly resists
any simplistically mechanistic interpretation and more recent scholarship rejects
traditional readings of Boyle, supporting instead the kind of non-reductionist
interpretation of his chemical philosophy that I am defending in this book.74
Boyle does affirm the mechanical philosophy and accepts that the mecha-
nistic properties of corpuscular concretions serve as the general causes of nat-
ural phenomena. However, this does imply for him that one must invoke such
causes to explain the behavior and structure of different chemical bodies, and he
finds fault with the “Epicureans and Cartesians” who attempt to establish a direct
71 Venel, Chymie ou Chimie, 299: “[Boyle] est ordinairement compté parmi les chimistes; & il a
effectivement beaucoup écrit sur la chimie. Mais il est trop exactement physicien corpusculaire-
mécanicien, ou physicien proprement dit, tel que nous l’avons mis en contraste avec le chimiste au
commencement de cet article.”
72 Several other authors have argued against the conflation of corpuscularianism and mechanicism,
and further argument on this topic extends beyond the scope of this chapter. For a more detailed, al-
though not exhaustive, discussion of this topic, I refer the reader to: Rattansi, Alchemy and Chemistry
in the 16th and 17th Centuries; Clericuzio, “A Redefinition of Boyle’s Chemistry and Corpuscular
Philosophy” and Elements, Principles, and Corpuscles; Pagel, Joan Baptista van Helmont: Reformer
of Science and Medicine; Debus, The Chemical Philosophy: Paracelsian Science and Medicine in the
Sixteenth and Seventeenth Centuries.
73 Clericuzio, “A Redefinition of Boyle’s Chemistry and Corpuscular Philosophy,” 563–564.
74 See, for example: Principe, “Newly Discovered Boyle Documents in the Royal Society
Archive” and The Aspiring Adept; Clericuzio, “A Redefinition of Boyle’s Chemistry and Corpuscular
Philosophy” and Elements, Principles, and Corpuscles; Anstey, The Philosophy of Robert Boyle, “Boyle
on Seminal Principles,” and “Robert Boyle and the Heuristic Value of Mechanism”; Chalmers, “The
Lack of Excellency of Boyle’s Mechanical Philosophy,” “Boyle and the Origins of Modern Chemistry,”
and The Scientist’s Atom and the Philosopher’s Stone; and Newman, “The Significance of ‘Chymical
Atomism’,” “The Alchemical Sources of Robert Boyle’s Corpuscular Philosophy,” and Alchemy Tried
in the Fire.
connection between all observed phenomena and the mechanical affections of

corpuscles.75 In An Essay on Various Degrees or Kinds of the Knowledge of Natural
Things, Boyle states that Epicureans and Cartesians “are so charm’d with ye clear-
ness & pleasure of Theorys & explications, yt are deriv’d immediately from met-
aphysical and mathematical notions & theorems; yet they oftentimes give forced
and unnatural accounts of things, rather than not to be thought to have deriv’d
them immediately from these highest principles. And, wch is much worse, they
despise & perhaps too condemn or censure all yt knowledge of the works of na-
ture yt Physicians, Chymists & others pretend to, because they cannot be clearly
deduc’d from the Atoms, or ye Catholic Laws of motion.”76
In fact, Boyle embraces many of the ideas that informed the natural philos-
ophies of his vitalistic predecessors, albeit interpreted in chemical terms and
stripped of their explicitly vitalistic content. Thus, “although Boyle often repeated
that the mechanical properties of corpuscles were to be regarded as the most ge-
neral notions of natural philosophy, a closer analysis of his natural philosophy
reveals a number of agents not operating according to the principles of the me-
chanical philosophy. These agents are seminal principles, spirits, and ferments—
which Boyle conceived as corpuscles endowed with the power of fashioning
other parts of matter [. . .] Boyle’s conclusions are explicitly anti-reductionistic.”77
In his chemical explanations, Boyle does not appeal to any of the properties
of fundamental matter precisely because the limits of chemical analysis do not
permit the isolation of the most elementary corpuscles. Instead, he intends to
use analytical methods to identify homogeneous chemical substances con-
stituted by combinations of aggregate corpuscles.78 Boyle’s rejection of strictly
mechanistic explanations of chemical phenomena contributed importantly to
the 17th-century debate over the status of chemistry that raged on among many
natural and speculative philosophers. Boyle’s insistence on developing chemical,
rather than strictly mechanical, explanations of his experimental work played a
significant role in the recognition of chemistry as a science proper, independent
75 Joly, 2: “Certes, Boyle affirme que les effets que nous observons dans la nature résultent des
diverses tailles, formes, mouvements et concrétions d’atomes. Ce sont là les causes générales de toutes
les choses. Mais il n’en résulte pas qu’il soit possible de rendre compte de tel ou tel phénomène en
invoquant la figure, la taille et le mouvement des corpuscules dont la concrétion constitue la struc-
ture des différent corps chimiques. Boyle s’en prend alors explicitement aux ‘Épicuriens et Cartésiens’
qui prétendent établir un lien direct entre les phénomènes observés et les propriétés mécaniques des
atomes”
76 Boyle, An Essay of Various Degrees or Kinds of the Knowledge of Natural Things, in Royal Society
Boyle Papers, Vol. 8, folio 166r.

77 Clericuzio, Elements, Principles, and Corpuscles, 103–109.
78 Bernard Joly, “Le cartésianisme de Boyle du point de vue de la chimie,” 152: “Boyle ne prétend
pas descendre jusqu’à ce niveau ultime de la structure de la matière, puisque l’analyse chimique
n’isole pas réellement les corpuscules élémentaires dont l’existence est affirmée. Boyle entend plutôt
s’arrêter au niveau des agrégats de corpuscules composés selon diverses combinaisons, qui sont en
quelque sorte de véritables substances chimiques homogènes.”
from physics. “The specificity of chemical objects, of the operations that involve
them, and of the science that studies them is thus preserved. Thus, unlike what
is concluded by Descartes, mechanistic explanation does not substitute chemical
explanation.”79
Yet, Boyle’s rejection of strictly mechanistic explanations is not intended as
an attack against the mechanistic hypothesis. In fact, quite the opposite is true.
As Joly emphasizes, mechanical explanations sustain chemical explanations “and
guarantee their coherence, while at the same time preserving a certain autonomy
for chemistry since both levels of explanation are maintained.”80 Therefore, Boyle
hoped both to reconcile chemistry with ontological mechanicism and, through
experimental means, to provide support for mechanistic corpuscularianism. In
the preface of A Physico-Chymical Essay, he explicitly expresses the hope that his
experimental work
may conduce to the Advancement of Natural Philosophy, if, as I said, I be

so happy as, by any endeavours of mine, to possess both Chymists and
Corpuscularians of the advantages that may redound to each Party by the
Confederacy I am mediating between them, and excite them both to enquire
more into one anothers Philosophy, by manifesting, that as many Chymical
Experiments may be happily explicated by Corpuscularian Notions, so many
of the Corpuscularian Notions may be commodiously either illustrated or con-
firmed by Chymical Experiments.81
Interestingly, Boyle’s conception of spirit and ferment is clearly influenced by

the work of Basso, Sennert, and Helmont. As Clericuzio points out, a transfor-
mation of the notion of spirit by 17th-century English chemical physiologists
permitted them to view vitality as resulting from the chemical activity of
substances rather than arising from a homogeneous spirit or soul, regarded as
an immaterial substance.82 Boyle’s theory on the general nature and classifica-
tion of spirits was, in part, developed during his studies of the spirit of blood.
In Memoirs for the History of Blood, he “recorded that from the distillation of
blood he had obtained, besides oily and phlegmatic parts, a clear liquor which,
though probably it contained some phlegm, might be called spirit, because ‘it
is fully satiated with saline and spirituous parts’ [. . .] Boyle’s researches on the
79 Ibid.: “La spécificité des objets chimiques, des opérations qui les concernent et de la science qui
s’en occupe se trouve ainsi préservée. De ce fait, à la différence de ce qui se produit chez Descartes,
l’explication mécaniste ne vient pas se substituer à l’explication chimique.”
80 Ibid., 152–153: “elle la maintient et en garantit la cohérence, conservant ainsi à la chimie une
certaine autonomie puisque les deux niveaux d’explication sont maintenus.”

81 Boyle, Certain Physiological Essays, 92.
82 See: Antonio Clericuzio, The Internal Laboratory.
spirit of blood [. . .] brought about the abandonment of the belief that spirit as
such—a homogeneous and vaguely defined [non-physical] substance—had to
be regarded as the origin of vital spirit.”83 What I will argue in the next section is
that Boyle’s position on these matters is not inconsistent with his ultimate onto-
logical commitment to the mechanical philosophy. Rather, his resistance to strict
explanatory reductionism can best be explained by examining more closely the
two distinct functions of the mechanical philosophy in Boyle’s scientific research
programme.
2.6 The Negative and Positive Heuristic Functions

of the Mechanical Philosophy in Boyle’s Scientific
Research Programme
In spite of the fact that Boyle’s approach to chemical explanation does not rely
on primary causes at the level of fundamental particles and appeals instead to
higher-level chemical properties, his commitment to the mechanical philosophy
remains an important constant in all of Boyle’s work. I argue in this section that
the reason for this commitment is that the mechanical philosophy fulfills two
distinct functions for Boyle, and I thus hope to resolve the apparent tension be-
tween his ontological commitment to mechanistic corpuscularianism and his
resistance to mechanistically reductive explanations. In order to develop my ar-
gument, I will examine Boyle’s commitment to the mechanical philosophy from
the perspective of Imre Lakatos’s distinction between the positive and negative
heuristics of scientific research programmes.84
For Lakatos, a scientific research programme is constituted neither by pure
theory alone nor by empirical research divorced from theory. Rather, scientific
research programmes involve both components, a theoretical “hard core” and an
experimental “protective belt.” The “hard core” is constituted by theories that are
deemed irrefutable by methodological fiat, while the “protective belt” is consti-
tuted by auxiliary hypotheses that can be either confirmed or falsified by exper-
imental observations, and each of these components fulfills a different function.
The hard core of theoretical hypotheses grounds the research programme and
defines what Lakatos calls the “negative heuristic” of the programme. The neg-
ative heuristic is a limiting principle that forbids scientists from directing “the
modus tollens [of falsification] at this ‘hard core’.”85 Lakatos’s classic example of a
83 Ibid., 64.
84 Lakatos, “Falsification and the Methodology of Scientific Research Programmes”; Lakatos,
The Methodology of Scientific Research Programmes; Lakatos, “Criticism and the Methodology of
Scientific Research Programmes.”
85 Lakatos, “Falsification and the Methodology of Scientific Research Programmes,” 133.
“hard core” is Newton’s gravitational theory. “In Newton’s programme the neg-
ative heuristic bids us to divert the modus tollens from Newton’s three laws of
dynamics and his law of gravitation. This ‘core’ is ‘irrefutable’ by the methodo-
logical decision of its protagonists: anomalies must lead to changes only in the
‘protective’ belt of auxiliary, ‘observational’ hypotheses and initial conditions.”86
On the other hand, the “protective belt” of the scientific research programme
defines its positive heuristic, which “consists of a partially articulated set of
suggestions or hints on how to change [and/or] develop the ‘refutable variants’
of the research programme [and] how to modify [and/or] sophisticate, the ‘re-
futable’ protective belt.”87 Lakatos explains that, in developing the protective belt,
“we must use our ingenuity to articulate or even invent ‘auxiliary hypotheses,’
which form a protective belt around this core, and we must redirect the modus
tollens to these. It is this protective belt of auxiliary hypotheses which has to bear
the brunt of tests and get adjusted and re-adjusted, or even completely replaced,
to defend the thus-hardened core.”88 For this reason, Lakatos maintains that the
positive heuristic is “more flexible than the negative heuristic . . . Thus, the meth-
odology of scientific research programmes accounts for the relative autonomy of
theoretical science . . . Which problems scientists working in powerful research
programmes rationally choose is determined by the positive heuristic of the
programme.”89
Lakatos also discusses what he calls “metaphysical research programmes” and
explains that the hard core of such research programmes is irrefutable because
it is non-empirical, while the auxiliary hypotheses of the protective belt are re-
futable, and it is towards these that experiments are directed. The “metaphys-
ical” hard core of such research programmes forms the background to scientific
theories and provides “the main guide for directing the scientists’ attention to
certain problems rather than to others.”90 Thus, the negative heuristic of the
“metaphysical” hard core serves as a limiting and directive principle. In a manner
that is very relevant for our purposes here, Lakatos discusses Cartesian meta-
physics as a classic example of a “metaphysical research programme.” He states
that “Cartesian metaphysics, that is, the mechanistic theory of the universe—
according to which the universe is a huge clockwork with push as the only cause
of motion—functions as a powerful heuristic principle: excluding, on the neg-
ative side, all scientific theories . . . which are inconsistent with it (negative heu-
ristic) and implying, on the positive side, a ‘metaphysical’ research programme to
86 Ibid.
87 Ibid., 137.
88 Ibid., 133.
89 Ibid., 137.
90 Lakatos, “Criticism and the Methodology of Scientific Research Programmes,” 178.
look behind all phenomena (and theories) for explanations based on clockwork
mechanisms (positive heuristic).”91
If one can conceive of Boyle’s natural philosophy and experimental work as a
scientific research programme in the Lakatosian sense, then one can argue that
the mechanical philosophy fulfills important functions both as a negative and as
a positive heuristic. I believe that Lakatos’s comments on Cartesian metaphysics
cited here provide us with a useful guide to help us unpack these ideas. When
the mechanical philosophy is considered as the theoretical grounding of Boyle’s
research programme, it functions as its irrefutable “metaphysical” hard core. It is
a deep philosophical theory about the nature of matter that can be neither con-
firmed nor falsified by experiment precisely because it is non-empirical. This is
what is defended in Boyle’s theoretical writings as an alternative to the theory of
substantial forms and to the spagyric principles. The term “mechanical,” when
used in this context, is used in the stricter ontological sense and refers to mecha-
nistic corpuscularianism.
Following Lakatos’s remarks about the relation between the hard core and
the protective belt, one can say that the “metaphysical” hard core of mechanistic
corpuscularianism is relatively autonomous from Boyle’s experimental work in
the sense that it neither stands nor falls on the basis of specific experimental dis-
coveries. As the “metaphysical” hard core of Boyle’s research programme, the
mechanical philosophy functions as a negative heuristic or as a limiting prin-
ciple that excludes or forbids any explanations that are inconsistent or incom-
patible with it. Borrowing from the vocabulary of apophatic theology, Andrew
Pyle considers this sense of the mechanical philosophy as a via negativa that “can
be construed as the fourfold denial of (1) action at a distance, (2) spontaneity,
(3) immanent or irreducible teleology, and (4) incorporeal causes.”92 Thus, ac-
cording to the mechanical philosophy’s negative heuristic, a necessary condition
91 Ibid., 168.
92 Pyle, “Boyle on Science and the Mechanical Philosophy,” 181. Although recent scholarship
would seem to contradict Pyle’s claim regarding the mechanical philosophy’s rejection of teleology,
it is important to emphasize that Pyle is specifically referring to the rejection of immanent teleology,
as distinguished from extrinsic teleology. Immanent teleology locates final causes within nature it-
self, so that final causes are intrinsic to natural phenomena. Extrinsic teleology, on the other hand,
identifies the finality of nature with God’s purposes. What recent scholarship shows is that many
adherents of the mechanical philosophy (including Boyle himself in his Disquisitions about the Final
Causes of Natural Things of 1688) embraced extrinsic teleological explanations to counteract atheistic
Epicureanism. But this in no way contradicts Pyle’s claim that the mechanical philosophy implied a
denial of immanent teleology.
For more on this topic, see Boyle, A Disquisition about the Final Causes of Natural Things;
Lennox, “Robert Boyle’s Defense of Teleological Inference in Experimental Science”; Osler, “From
Immanent Natures to Nature as Artifice,” “How Mechanical was the Mechanical Philosophy?,” and
“Whose Ends?”; Carlin, “The Importance of Teleology to Boyle’s Natural Philosophy” and “Boyle’s
Teleological Mechanism and the Myth of Immanent Teleology”; and Shanahan, “Teleological
Reasoning in Boyle’s Disquisition about Final Causes.”
for an explanation to qualify as mechanical is that it must exclude all reference to

substantial forms, sympathies and antipathies, vital forces, the anima mundi, or
other immaterial causes.
Boyle’s writings support this conception of the mechanical philosophy as a
negative heuristic. In the Hydrostatical Discourse, he explains that, although he
believes in the existence of an incorporeal God who created the world and the
laws that govern it, what happens to inanimate bodies does not require appeal to
immaterial agents and is strictly attributable to mechanical causes, even if these
causes are not fully known or understood. Boyle then invokes Occam’s razor to
support this claim by saying that “this, being agreeable to the generally own’d
rule about Hypotheses, that Entia non sunt multiplicanda absque Necessitate, has
been by almost all the modern Philosophers of different Sects thought a suffi-
cient reason to reject the agency of Intelligences.”93 When used in his chemical
writings, the mechanistic reference functions in the same way and “allows Boyle
to reject the obscurities and the contradictions of traditional doctrines that were
coming into question in his own time without thereby condemning chemistry
to a reabsorption by physics.”94 One can argue that Boyle remains committed
throughout his entire scientific career to the negative heuristic of mechanistic
corpuscularianism, that is, to eschewing any explanatory principle that is incom-
patible with the mechanical philosophy, such as substantial forms, vital spirits,
spagyric principles, action at a distance, and anima mundi.
On the other hand, when considering Boyle’s research programme from the
point of view of its protective belt of auxiliary empirical and refutable hypoth-
eses regarding chemical, pneumatic, and hydrostatic phenomena, the mechan-
ical philosophy takes on different connotations. To the extent that the positive
heuristic concerns experimental work designed to test empirical hypotheses, the
mechanical philosophy serves to provide a model for how all natural phenomena
behave, that is, as clockwork mechanisms in which the various connected parts
transmit motion to each other. Alan Chalmers points out that that the term
“mechanical,” when used in Boyle’s experimental writings on hydrostatics and
pneumatics, does not refer to mechanistic corpuscularianism but is used in the
more common sense of this word, that is, as referring to the science of simple
machines. He explains that “machines bring about their effects by virtue of the
way in which pushes and pulls are transmitted through them via the connec-
tion of their parts. Boyle also made explicit the point that an understanding of
machines involved taking properties such as weight and elasticity for granted,
93 Boyle, An Hydrostatical Discourse, in The Works of Robert Boyle, 159.

94 Joly, “Le cartésianisme de Boyle du point de vue de la chimie,” 154: “La référence mécaniste
permet ainsi à Boyle de rejeter les obscurités et les contradictions des doctrines chimiques finissantes
de son temps sans pour autant condamner la chimie à une résorption dans la physique.”
without requiring that they be explained at some deeper, corpuscular or other,

level.”95 It is interesting to see that Chalmers’s point here concurs with Lakatos’s
claim that, in its positive heuristic, the mechanical philosophy provided the im-
petus “for explanations based on clockwork mechanisms.”96
It is also interesting to note that the negative and positive heuristics work
hand in hand for, if physical phenomena are described as being analogous to
clockwork mechanisms in which connected parts transmit motion to each other,
then action at a distance is perforce rejected as an explanatory principle. In ex-
perimental writings such as An Hydrostatical Discourse, Of the Systematical or
Cosmical Qualities of Things, the appendix to Final Causes, and the Workdiaries,
“Boyle arrived at and utilised explanations that were innovative but which were
not ultimate.”97 However, to the extent that the non-mechanistic properties, sub-
ordinate causes, and cosmical qualities that Boyle invokes are subject to the laws
of mechanics, these explanations abide by the negative heuristic of eschewing ap-
peal to non-material principles. Thus, although these explanations do not make
explicit reference to mechanistic corpuscularianism, the limiting principle of the
mechanical philosophy is built into them.
Boyle’s conception of the mechanical philosophy is not one that requires
replacing chemical discourse with mechanistic discourse and it is certainly not
one that implies the subversion of chemistry as an autonomous science. In fact,
Boyle believes that, by establishing the uselessness and lack of heuristic power
of substantial forms and of the Paracelsian tria prima, the mechanical philos-
ophy can actually serve to justify chemical practice and the explanatory appeal
to chemical properties as causal agents, as well as the discovery of new homoge-
neous substances that enter into the composition of mixed bodies.98
As this book ultimately demonstrates, Boyle endorses a conception of chem-
ical properties as emergent, dispositional, and relational and he takes this po-
sition to be perfectly compatible with his commitment to the mechanical
philosophy. The following chapter establishes that his conception of chemical

96 Lakatos, “Criticism and the Methodology of Scientific Research Programmes,” 168.
97 Chalmers, “The Lack of Excellency of Boyle’s Mechanical Philosophy,” 545.
98 Joly, “Le cartésianisme de Boyle du point de vue de la chimie,” 154–155: “Il ne s’agit donc pas,
comme chez Descartes, de remplacer le discours des chimistes par un discours mécaniste, mais au
contraire de montrer que l’hypothèse mécaniste facilite le développement de la chimie en lui offrant
de nouvelles perspectives: guidés par les principes du corpuscularianisme, les chimistes devraient
pouvoir découvrir de nouvelles substances homogènes qui entrent dans la composition des corps
mixtes [. . .] Alors que Descartes se livre à une réduction mécaniste de la chimie qui implique sa
disparition en tant que telle et le rejet de sa tradition, Boyle veut au contraire la conserver, y compris
dans sa dimension alchimique, en la justifiant par une philosophie mécaniste dont la fonction
n’est pas de rendre compte par le menu détail des diverse propriétés des corps chimiques, mais de
montrer l’inutilité des théories principielles traditionnelles, qui n’apportent finalement rien à la
compréhension des opérations chimiques.”
properties is buttressed by a complex corpuscularian ontology that allows him to

formulate autonomously chemical explanations in which structural differences
account for chemical differences, chemical properties have causal power, and
material bodies with different essential properties belong to different material
species. Boyle’s complex ontology thus accommodates mechanicism, structural
realism, and realism about natural kinds, while remaining committed to the me-
chanical philosophy as the best available alternative to Scholastic hylomorphism.
3
The Ontological Complexity of Boyle’s
Corpuscularian Theory
Microstructure, Natural Kinds, and Essential Form
One of Boyle’s principal heuristic reasons for preferring mechanistic particulate

theories to the Scholastic theory of substantial form or to the Paracelsian no-
tion of the tria prima is that neither of these theories could provide satisfactory
explanations for chemical phenomena and processes. Thus, above all things, The
Sceptical Chymist represents an extended critique of these two unsatisfactory
theories. I propose first to review Boyle’s critique and then move on to a detailed
account of his mechanistic corpuscularianism as an alternative to the notion of
substantial form and to the spagyric theory of principles.
3.1 The Sceptical Chymist: Against Scholastics

and Spagyrists
By the mid-17th century, it is evident to natural philosophers that most chemical

reactions do not involve the active role of some form of salt, sulfur, or mercury.
Eventually, therefore, the notion of the tria prima quickly loses credibility for
early modern chemists such as Boyle. In the preface to The Sceptical Chymist,
Boyle explains that “there are a thousand Phaenomena in Nature [. . .] which will
scarcely be clearly & satisfactorily made out by them that confine themselves to
deduce things from Salt, Sulphur and Mercury, and the other Notions peculiar to
the Chymists, without taking much more Notice than they are wont to do, of the
Motions and Figures, of the small Parts of Matter, and the other more Catholick
and Fruitful affections of Bodies.”1 Boyle appeals to corpuscularian principles to
argue against some of the fundamental assumptions of the Paracelsian doctrine
of the tria prima or hypostatical principles. According to this doctrine, because
salt, sulfur, and mercury are the principles of mixed bodies, these ingredients
can always be extracted via analysis by fire. However, Boyle remarks that “it does
not appear, that Three is precisely and Universally the Number of the Distinct
1 Boyle, The Sceptical Chymist, 208.
University Press.
DOI: 10.1093/oso/9780197502501.001.0001
Substances or Elements, whereinto mixt Bodies are resoluble by Fire; I mean that
’tis not prov’d by Chymists, that all the Compound Bodies, which are granted to
be perfectly mixt, are upon their Chymical Analysis divisible each of them into
just Three Distinct Substances, neither more nor less, which are wont to be lookt
upon as Elementary.”2
After noting this, Boyle presents two arguments against the hypostatical prin-
ciples, the first of which is speculative and relies upon the corpuscularian princi-
ples, while the second is “drawn from Experience,” to use Boyle’s own words. In
the first argument, Boyle posits that those substances that are considered elem-
ents by the spagyrists may themselves be composed of numerous coalitions of
minute particles. “It will not be absurd to conceive, that such primary Clusters
may be of far more sorts than three or five; and consequently, that we need not
suppose, that in each of the compound Bodies we are treating of there should be
found just three sorts of such primitive Coalitions.”3 This argument is a priori and
depends upon assuming the truth of corpuscularianism. As such, it presupposes
that corpuscles are elementary and that salt, mercury, and sulfur are themselves
coalitions of elementary particles. If so, it is reasonable to suppose that there may
be more than merely three of such coalitions for, if we grant that the so-called
tria prima or hypostatical principles are not elementary, then we have no reason
to believe that all compound bodies should be reducible to one of these three
substances.
Boyle is saying that, in fact, salt, mercury, and sulfur are not principles.
However, because this is an a priori argument that depends upon assuming the
truth of corpuscularianism, Boyle’s argument will only be acceptable to those
who embrace this same assumption. The argument itself does nothing to es-
tablish the truth of corpuscularianism over that of the tria prima hypothesis.
Boyle seems aware of this weakness and refrains from presenting this argument
as being conclusive in favor of corpuscularianism. Instead, he limits himself to
claiming that it is at least plausible to conceive that the spagyric elements may be
reducible to more fundamental particles.
Boyle’s second argument, on the other hand, attempts to use empirical evi-
dence to establish that the tria prima are not elementary principles. Here, Boyle
relies on several experiments in which he purports to have established, by way
of analysis, that many bodies “in their Resolution Exhibite more principles than
three.”4 From these considerations, he concludes that
2 Ibid., 278.
3 Ibid., 279.
4 Ibid., 284.
Boyle’s Corpuscular Theory 81
the Ternary Number is not that of the Universal and Adequate Principles of
Bodies. If you allow of the Discourse I lately made You, touching the primary
Associations of the small Particles of matter, You will scarce think it improb-
able, that of such Elementary Corpuscles there may be more sorts than either
three, or four, or five [. . .] confining our selves to such wayes of Analyzing mix’d
Bodies [. . .] it may without Absurdity be Question’d, whether besides those
grosser Elements of Bodies, which they call Salt Sulphur and Mercury, there
may not be Ingredients of a more Subtile Nature, which being extreamly little,
and not being themselves Visible, may escape unheeded at the Junctures of the
Destillatory Vessels.5
Given these considerations, Boyle places the burden of proof on those

chymists who embrace the Paracelsian theory of the tria prima and adds that
they have failed to meet this burden of proof because they have not provided
credible support for the existence of the hypostatic principles. First, regarding
any of the purportedly empirical evidence provided by spagyrists, Boyle argues
that experiments “wont to be brought, whether by the common Peripateticks,
or by the vulgar Chymists, to demonstrate that all mixt bodies are made up pre-
cisely either of the four Elements, or the three hypostatical Principles, do not
evince what they are alleg’d to prove.”6 He adds that
if what those patriarchs of the Spagyrists, Paracelsus and Helmont, do on di-

vers occasions positively deliver, be true; namely that the Alkahest does Resolve
all mixt Bodies into other Principles than the fire, it must be decided which of
the two resolutions (that made by the Alkahest, or that made by the fire) shall
determine the number of the Elements, before we can be certain how many
there are . . . although we should acquiesce in that resolution which is made by
fire, we find not that all mixt bodies are thereby divided into the same number
of Elements and Principles; some Concretes affording more of them than
others do; Nay and sometimes this or that Body affording a greater number
of Differing substances by one way of management, than the same yields by
another . . . Nor does it appear more congruous to that variety that so much
conduceth to the perfection of the Universe, that all elemented bodies be
compounded of the same number of Elements, then [sic] it would be for a lan-
guage, that all its words should consist of the same number of Letters.7
5 Ibid., 284–285.
6 Ibid., 342–343.
7 Ibid., 343.
Boyle also rejects the a priori arguments in favor of spagyrist theories by saying
that these arguments, which pretend
to be drawn from Reason [. . .] are Commonly grounded upon such unrea-

sonable or precarious Suppositions, that ’tis altogether as easie and as just for
any man to reject them, as for those that take them for granted to assert them,
being indeed all of them as indemonstrable as the conclusion to be inferr’d from
them; and some of them so manifestly weak and prooflesse; that he must be a
very courteous adversary that can be willing to grant them; and as unskilful a
one that can be compelled to do so.8
Satisfied that he has thus dispatched the tria prima hypothesis, Boyle turns his
critical attention to the theory of substantial forms. Boyle attacks the theory of
substantial forms while critiquing what he considers to be a flawed conception
of chemical qualities in the theories of chymists and alchemists such as William
Sennert and other Paracelsians. In Of the Imperfection of the Chemists’ Doctrine
of the Qualities, Boyle states that one of the main suppositions of these chemists is
that this or that quality must have its πρϖτον δεκτικόν9—as Sennertus, the
learnedest champion of this opinion, calls it—or some particular material prin-
ciple, to the participation of which, as of the primary, native and genuine sub-
ject, all other bodies must owe it [. . .] I think it will be hard to show what is
the πρϖτον δεκτικόν of gravity, volatility, heat, sonorousness, transparency,
and opacity, which are qualities to be indifferently met with in bodies whether
simple or mixed.10
To understand the role of the πρϖτον δεκτικόν within the theory of substan-
tial forms, it is important to recall from Chapter 1 that Sennert had attempted to
resolve the problems with the Scholastic theory of substantial forms by adopting
corpuscularianism and by attributing such forms to the fundamental particles
of substances, rather than to the substances themselves, thereby allowing for the
possibility of both analysis and synthesis. However, in spite of Sennert’s persis-
tent efforts to save the theory of substantial forms by rendering it compatible with
corpuscularianism, this theory had run its course by the mid-17th century and
needed to be replaced with less problematic and more informative explanations
for chemical and other physical processes and phenomena.
8 Ibid.
9 proton dektikon or “prime matter.”
10 Boyle, Of the Imperfection of the Chemists’ Doctrine of Qualities, 121–123.
For Boyle, substantial form could not provide adequate explanations for the
chemical transformations that occurred via analysis and synthesis, in part be-
cause such explanations either did not explain anything or explained things in
a circular fashion. I elaborate further. If one simply claims that analysis destroys
the form of a substance while synthesis restores it, then this by itself is not enough
to tell us what this form is. On the other hand, if one claims that substantial form
is something non-physical that disappears during analysis and then reappears
during synthesis, then this does not explain how a chemical procedure can caus-
ally affect a non-material “form.” Lastly, if by “substantial form” one simply refers
to the substance itself, then the explanation for what happens during analysis
and synthesis is rendered circular. That is, the substance has been altered be-
cause the substance has been altered. Any of these options leaves us epistemically
unsatisfied.
To be fair, there is another and more interesting way of understanding sub-
stantial form that can serve as a precursor to Boyle’s structural conception. That
is, substantial form can be understood as referring to the inner structure that
gives a substance its essential properties. Of course, if one were to opt for this
interpretation of substantial form, then one would still have to explain what this
inner structure is. This, however, is clearly not what the Scholastics meant by
“form.” It is also not what Sennert meant by “form” since he took “substantial
form” and microstructure to be two different things, the former being the im-
material and inherent essence of distinctive corpuscles and the latter being the
spatial arrangement of these to form specific material bodies. Boyle, however,
latches on to the conception of form as inner structure and seeks to understand
it in strictly material and corpuscularian terms, without any reference to the du-
bious notion of “substantial form” or any of its problematic connotations.
Boyle believes that mechanistic corpuscularianism provides a more reason-
able alternative to the theories of the tria prima and of substantial form and that
this serves as a compelling reason for favoring it. However, an even more com-
pelling reason for preferring mechanistic corpuscularianism is that it explains
the phenomena of nature by appealing to entirely physical principles rather
than to immaterial notions, although Boyle rejects the more extreme materialist
tenets of ancient and modern atomism. In fact, in Of the Excellency and Grounds
of the Corpuscular or Mechanical Hypothesis, he makes it clear that his intent is
not to embrace all facets materialistic Epicurean atomism but
[to] plead onely for such a Philosophy, as reaches but to things purely Corporeal,
and distinguishes between the first original of things; and the subsequent course
of Nature [. . .] the Phaenomena of the world [. . .] are Physically produc’d by
the Mechanical affections of the parts of Matter, and what they operate upon
one another according to Mechanical Laws. And now having shewn what kind
of Corpuscular Philosophy ’tis that I speak of I proceed to the particulars that

I thought the most proper to recommend it.11
After explaining this, Boyle goes on to enumerate those aspects of

corpuscularianism that, in his view, make it superior to competing hypotheses.
The first aspect “is the Intelligibleness or Clearness of Mechanical Principles and
Explications”12 in contrast to the unintelligibility of Peripatetic, Scholastic, and
Spagyric notions such as archeus, tria prima, and so forth. Secondly, mechanical
corpuscularian principles enjoy “great Comprehensiveness”13 in that they pro-
vide the foundation for explanations of all sorts of phenomena, whether these are
physical, chemical, pneumatic, static, or hydrostatic in nature. He indicates that
“the Mechanical Principles are so universal, and therefore applicable to so many
things, they are rather fitted to include, than necessitated to exclude, any other
Hypothesis that is founded on Nature, as far as it is so.”14 Thus, Boyle believes
that mechanical corpuscularianism is compatible with any other explanatory
hypothesis, once that hypothesis is equally naturalistic and physicalistic in its
principles. This is an important point since, as we shall see both in this and in
the following chapters, Boyle’s chemical ontology and his brand of mechanistic
corpuscularianism are most compatible with an emergentist conception of
chemical qualities.
3.2 Boyle’s Corpuscularian Theory of Matter
As was discussed in detail in the previous chapter, there was considerable dis-
agreement among speculative and natural philosophers regarding the role of
chemistry in providing empirical support for the mechanical hypothesis. On
the other hand, there was also considerable debate regarding whether the me-
chanical hypothesis could ever provide satisfactory explanations for chemical
phenomena, for the causal efficacy of chemical properties, and for the stable na-
ture of chemical substances. There is no doubt that Boyle remains firmly com-
mitted to the mechanical hypothesis as the soundest available alternative to the
theory of substantial form and to the spagyrist theory of principles. But, he also
believes that chemistry is able to provide empirical support for the mechanical
hypothesis.
11 Boyle, Of the Excellency and Grounds of the Corpuscular or Mechanical Hypothesis, 105.
12 Ibid., 104.
13 Ibid., 106.
14 Ibid., 109.
Peter Anstey has discussed in detail the three types of empirical arguments
in favor of mechanical corpuscularianism that Boyle derived from his chem-
ical experiments, and I rely here on his excellent discussion for my own exposi-
tion. First, Boyle uses the results from redintegration experiments, particularly
from the redintegration of niter, to support the view that this chemical process is
better explained in corpuscularian terms than by appealing to substantial forms.
Second, he argues from the reduction to the pristine state, which I discuss later
in this chapter, for the same conclusion. Third, he argues from the mechanical
production and alteration of qualities for the superiority of the corpuscularian
theory and, therefore, for the hypothesis itself.15
These ideas will be examined more closely in the following sections of this
chapter. Suffice it to say at this point that the results of chemical experiments
can, at the very least, provide empirical evidence against hylomorphism, thereby
indirectly supporting the mechanical philosophy. However, to the extent that
the Cartesian mechanistic hypothesis is unable to accommodate the notion of
chemical properties or of stable chemical substances, the challenge for Boyle is to
provide empirical chemical support for mechanistic corpuscularianism without
sacrificing chemistry itself. To succeed in this task, Boyle revises the mechan-
ical hypothesis to accommodate ontological realism about chemical properties
and about their causal agency in nature. Thus, he develops a nuanced theory of
matter that combines a mechanistic conception of fundamental particles with a
realist ontology of chemical qualities, causes, and substances.
Boyle’s clearest statement of his theory is found in The Origin of Forms and
Qualities (1666). Here, he takes a position that mediates between those of
Gassendi and Descartes. Although Boyle rejects classical Epicurean atomism,
he does find certain aspects of Gassendian atomism to be especially compel-
ling. One of the things that makes Gassendi’s revision of classical atomism at-
tractive for Boyle is the insistence that nature should not be treated as an agent.
Boyle agrees with Gassendi that nature as such is inanimate and devoid of pur-
pose, volition, and sentience and that the only source of agency is God.16 Yet,
against Gassendi and in agreement with Descartes, Boyle denies the presence
of a God-given internal energy or “motive virtue” within material corpuscles.
He argues this point again in Of the Excellency and Grounds of the Corpuscular
or Mechanical Hypothesis, in which he distances himself in this regard both
from ancient atomists and from some modern atomists such as Gassendi by
specifying that
15 Anstey, “Essences and Kinds,” 22.

16 Shapin and Schaffer, Leviathan and the Air-Pump, 202.
when I speak of the Corpuscular or Mechanical Philosophy, I am far from

meaning with the Epicureans, that Atoms, meeting together by chance in
an infinite Vacuum, are able of themselves to produce the World, and all its
Phaenomena; nor with some Modern Philosophers, that, supposing god to
have put into the whole Mass of Matter such an invariable quantity of Motion,
he needed do no more to make the World, the material parts being able by their
own unguided Motions, to case themselves into such a System (as we call by
that name).17
Although he agrees with Descartes that all particles may be infinitely divisible
by God, Boyle also believes that corpuscles are impenetrable and indivisible in
nature and that they resist any attempt to penetrate or divide them by chemical
means. Thus, for him, corpuscles are infinitely divisible in theory but are impen-
etrable and indestructible in practice. In his early manuscript Of the Atomicall
Philosophy (1651–1653), Boyle had already distanced himself from Cartesians
by equating atoms with minima naturalia, although his conception of minima
differs from that of the Scholastics by ruling out substantial form.18 In this work,
Boyle states that “by Atoms the Assertors of them understand not indivisible
Mathematical points [. . .] but minima naturalia [. . .] Because tho they may be
further divided by Imagination yet they cannot by Nature.”19 Furthermore, and
again disagreeing with Descartes, Boyle believes that matter conserves shape and
size and cannot, therefore, be reduced to pure geometrical extension. Like visible
bodies, insensible corpuscles have three essential mechanical affections: shape,
size, and mobility. According to Boyle, although God did not endow corpuscles
with internal “motive virtue,” God furnished corpuscles externally with various
motions and directed their various movements and compositions to form the va-
riety of inanimate and animate bodies that exist in nature.
In The Origin of Forms and Qualities, Boyle describes in detail his complex
theory of matter. Note that Boyle’s first point in the following list is a reitera-
tion of what has already been said regarding minima naturalia, and it is also a
statement of his agreement with Gassendi and disagreement with Descartes re-
garding the divisibility of fundamental particles. Boyle states:
1. That there are in the World a great store of Particles of Matter, each of which
is too small to be, whilst single, Sensible: and being Entire, or Undivided,
must needs both have its Determinate Shape, and be very Solid. Insomuch,
that though it be mentally, and by Divine Omnipotence divisible, yet by
17 Boyle, Of the Excellency and Grounds of the Corpuscular or Mechanical Hypothesis, 104–105.
18 See Clericuzio, Elements, Principles, and Corpuscles.
19 Boyle, Of the Atomicall Philosophy, 227.
reason of its Smallness, Nature doth scarce ever actually divide it; and these
may in this sense be call’d Minima or Prima Naturalia.
2. That there are multitudes of Corpuscles, which are made up of the Coalition
of several of the former Minima Naturalia; and whose Bulk is so small,
and their Adhesion so close and strict, that each of these little Primitive
Concretions or Clusters (if I may so call them) of Particles is singly below
the discernment of Sense, and though not absolutely indivisible by Nature
into the Prima Naturalia that compos’d it [. . .] they very rarely happen to be
actually dissolv’d or broken, but remain entire in a great variety of sensible
Bodies, and under forms and disguises.20
For Boyle, the mechanical philosophy is a superior hypothesis in part be-

cause it is a “bottom-up” theory of organization in which minima occupy the
lowest level and corpuscular concretions occupy progressively higher levels of
complexity. Primary concretions represent a second order of complexity above
fundamental mechanistic particles and form clusters of even higher degrees of
complexity, such as that which occurs when different homogeneous substances
are mixed together.
As discussed later in this chapter, Boyle attributes the real constitution of a
material body and its essential properties to the underlying corpuscular struc-
ture, in which the fundamental particles fit so closely together as to form stable
chemical species that resist analysis, even by the most corrosive agents available.
Microstructure creates stable entities that are of a higher order of complexity
than the minima and that have non-mechanical properties, in addition to the
mechanical properties of fundamental particles. Some of these non-mechanical
properties are specifically chemical and have the power to affect other parts of
matter in various ways. In some of his writings, Boyle describes chemical qual-
ities by alluding to seminal powers, that is, to the power of generation and the
power to “fashion other parts of matter.” Unlike the Scholastics, however, he
regards seminal powers not as spiritual but as material and as originating from
the specific structures of corpuscular concretions.
Boyle’s notion of compounded corpuscles is influenced by Gassendi and, al-
though he does not adopt the Gassendian term “molecule,” Boyle does avail him-
self of the concept that is attached to this term. Both in his early works and in his
later writings, Boyle explains chemical reactions by means of the compounded
corpuscles or corpuscular concretions that I am here discussing, since he
considers that chemical reactions take place at this higher level of corpuscular
organization. Boyle refers to this first order of compounded corpuscles as the
20 Boyle, The Origin of Forms and Qualities, 325–326.

“prima mixta,” but the prima mixta are themselves compounded to form higher
and higher orders of complex particles. In The Sceptical Chymist, he explains that
it does not at all appear, that all Mixtures must be of Elementary Bodies; but it
seems farr more probable, that there are divers sorts of compound Bodies, even
in regard of all or some of their Ingredients, consider’d Antecedently to their
Mixture. For though some seem to be made up by the immediate Coalitions
of the Elements, or Principles themselves, and therefore may be call’d Prima
Mista or Mista Primaria, yet it seems that many other Bodies are mingl’d (if
I may so speak) at the second hand, their immediate Ingredients being not
Elementary, but these primary Mixts newly spoken of; And from divers of these
Secondary sort of Mixts may result, by a further Composition, a Third sort, and
so onwards.21
Although the concept attached to the term “prima mixta” is the same as that
attached to the term “molecule,” William Newman has suggested using the term
“chymical atoms” in reference to the prima mixta. The term “chymical atom”
is an appropriate name for these compounded corpuscles, more so than the
term “molecule,” because the term emphasizes their operational atomicity. For
Boyle, chymical atoms are not analyzable by any of the ordinary chemical means
available in his time, such as nitric acid, hydrochloric acid, and fire. He, thus,
considers them to be highly stable entities. To the extent that chymical atoms
are stable, homogeneous regarding their essential properties, and operation-
ally fundamental in that they cannot be further chemically analyzed, they can
be legitimately considered as chemically “elementary” entities. In fact, “within
the framework supplied by Boyle’s Essay of the Atomicall Philosophy, [the terms]
‘elementary’ and ‘atomic’ can be seen as coextensive terms—both imply resist-
ance to decomposition.”22 According to Boyle, chymical atoms endow homoge-
neous substances with the properties that define them as members of specific
material species, that is, essential properties. Thus, although these compounded
corpuscles are hierarchically secondary with respect to primary corpuscles, they
are primary with respect to homogeneous substances and, when combined, they
form primary mixtures that can in turn be combined to form different degrees
of mixtures. In the context of chemical decomposition, “the identity between the
smallest available particles and the substance as a whole provides direct evidence
for such atomism, since an obvious consequence of this uniformity is that no de-
composition products are present . . . homogeneity after attempted analysis is a
warrant for the claim that the material at hand is atomic.”23
21 Robert, The Sceptical Chymist, 296–297.
22 Newman, “The Significance of ‘Chymical Atomism’,” 254.
23 Ibid., 254–255.
It must be acknowledged at this point that, according to Boyle, prima mixta or

“chymical atoms” may be reduced in the context of chrysopoetic transmutation,
which is not an ordinary chemical process. Transmutation is indeed of central
importance in Boyle’s work and he does believe that, unlike ordinary chemical
agents, the alkahest (i.e., the universal solvent described by Paracelsus) and the
Philosopher’s Stone act upon the prima mixta, either reducing them into pri-
mordial water or converting base metals into gold. Indeed, Boyle implies that
his own menstruum, peracutum,24 does this as well when it transmutes gold into
silver and he claims to have witnessed the Anti-Elixir doing the same. His de-
sire to see what such experiments can reveal about the composition of matter is
undoubtedly a key reason for his fervent search for these arcana throughout his
career.25 However, Boyle’s belief that the prima mixta can be analyzed by the al-
kahest, the Philosopher’s Stone, or the peracutum does not invalidate the claim
that, in the context of ordinary chemical operations, the prima mixta is opera-
tionally elementary.
Although there seems to be a paradox between Boyle’s chymical atomism and
his belief in chrysopoetic transmutation, Newman explains that
chymical atomism and chrysopoeia were anything but incompatible. The fact
that metals stubbornly retained their own identity in the face of analytical
assaults by chymists was a knowledge born of hard experience. Hierarchical
theories of matter like those of Boyle, Becker, and Stahl originated out of the
alchemical tradition’s attempts to circumvent the obstinate refusal of nature
to yield up the secret of transmutation. Since chymical atoms were not the
solid and impenetrable units of Democritus and Leucippus, but structured
composites made up of smaller particles, there was every reason to imagine that
a sufficient powerful solvent [such as the alkahest or Boyle’s peracutum] should
be able to penetrate and break them in to their components.26
Thus, although chymical atoms cannot be reduced by ordinary analytical means,

they are in principle analyzable if a powerful enough menstruum is discovered.
Therefore, Boyle’s chymical atomism is precisely what allows him to explain
the chemical stability of substances, while also believing in the possibility of
transmutation.
24 Boyle prepares this highly corrosive solvent by distilling a mixture of nitric acid (aqua fortis)
and antimony trichloride (butter of antimony).

25 Principe, The Aspiring Adept.
26 Newman, “Robert Boyle, Transmutation, and the History of Chemistry before Lavoisier,” 77.
3.3 Composition vs. Microstructure
One of several important issues that inform contemporary philosophy of

chemistry is that of structural explanation, precisely because modern chem-
istry is primarily concerned with microstructure. The importance of structural
explanations in chemistry arose in the 19th century from the discovery of iso-
meric molecules, that is, molecules that are composed of the same elements in
the same proportions but that display different macroscopic and, more specifi-
cally, different chemical properties. In the 1860s, A. M. Butlerov explained isom-
erism as based on molecular structure. Thus, the compositional formulae that
had been used since the 18th and early 19th centuries had to be supplemented
with structural formulae. This is what distinguishes contemporary chemistry
from chemistry as it was practiced and theorized prior to the early 19th century,
when the focus was placed on elemental and atomic composition. However, one
of the points that I argue in this section is that concern over microstructure, al-
beit understood in a radically different way than it is today, also informs Boyle’s
chemical philosophy.
Boyle’s conception of chymical atoms cannot simply be regarded as composi-
tional, because it is not only their composition but also the structure in which the
minima are arranged that account for the essential properties of chymical atoms.
I sustain that the essential property of chymical atoms is their texture, which is
determined by the microstructural arrangement and coordinated movements of
the minima that compose them. The chemical properties of a stable substance are
due to the properties of the chymical atoms that compose it but “Invisible changes
made in the minute and perhaps undiscernible parts of a stable Body may suffice
to produce such alterations in its Texture, as may give it new Qualities.”27 In some
ways, this view is similar to Sennert’s theory of microstructure but without the
appeal to “substantial forms” of corpuscles. Genuine chemical changes can alter
the texture of chymical atoms in such a way as to alter the essential properties of
a substance. Mere mechanical changes, however, will simply alter the spatial rela-
tions between chymical atoms without altering their texture and will, thus, only
yield a change in the extra-essential (or non-essential) properties of a substance.
From this discussion, one can conclude that the idea of microstructure fulfills
two tasks for Boyle: 1) To the extent that Boyle embraces a mechanistic concep-
tion of fundamental particles, microstructure helps him reconcile the idea that
minima are endowed only with mechanical affections with the idea that different
species of matter exist in nature and have distinctive non-mechanical properties
that differentiate them from other species, and 2) microstructure helps to explain
27 Boyle, History of Particular Qualities, 279.

Boyle’s experimental observation that some properties of substances are stable

and immune to merely mechanical changes, while other properties are less stable
and are susceptible to alteration by such mechanical changes. Thus, and as pre-
viously mentioned, while it is generally believed that our contemporary concern
over structural explanation is a function of modern chemistry’s emphasis on mi-
crostructure, Boyle’s structural realism illustrates the way in which many of our
contemporary concerns have deeply historical origins.
Boyle’s spatial, geometrical, and mechanistic conception of structure allows
him to appeal to form to explain chemical stability while, at the same time,
rejecting the Scholastic conception of substantial form. Because mechanistic
form accounts for the texture of chymical atoms and the essential properties
of material bodies, he refers to this deep microstructure as the essential form.
It is evident that essential form fulfills, for Boyle, the role that substantial form
had fulfilled for the Scholastics and for vitalistic corpuscularians such as Daniel
Sennert but without any of its problematic connotations. In fact, in The Origin of
Forms and Qualities, Boyle explicitly states “that which he [Sennert] ascribes to
the dominion of the specific Form, I attribute to the structure and especially to
the connexion of the parts of the compounded body.”28
Again, and as previously established, although the extra-essential properties
of a chemical substance may be altered by analysis, the texture of chymical atoms
remains stable within the products of analysis and this explains why the original
substance can then be re-synthesized or redintegrated. Since chymical atoms are
operationally irreducible, resistant to the most corrosive analytical tools avail-
able, and operationally “elementary,” Boyle refers to them as “minima of their
own genus,” a term that he appropriates from Sennert. In The Origin of Forms and
Qualities, Boyle explains this.
That there are multitudes of Corpuscles, which are made up of the Coalition of
several of the former Minima Naturalia; and whose Bulk is so small, and their
Adhesion so close and strict, that each of these little Primitive Concretions or
Clusters (if I may so call them) of Particles is singly below the discernment of
Sense, and though not absolutely indivisible by Nature into the Prima Naturalia
that compos’d it [. . .] they very rarely happen to be actually dissolv’d or broken,
but remain entire in a great variety of sensible Bodies, and under forms and
disguises.29
Elsewhere, Boyle states “there are divers Concretes [. . .] wherein the noblest
properties [. . .] depend immediately upon the form (or if you will result from the
28 Boyle, The Origin of Forms and Qualities, 459.

29 Ibid., 325–326.
determinate structure of)—the Whole Concrete; and consequently they that got
about the Vertues of such bodies, by exposing them to the Violence of the Fire,
do exceedingly mistake.”30 The only thing that such powerful analytical tools
can do is rearrange the spatial relationships between chymical atoms, via com-
bination, separation, and transposition of these corpuscular concretions. This,
however, results in a merely mechanical alteration of the relationship between
chymical atoms, rather than in a genuine alteration of their microstructure, and
thus only affects the extra-essential properties of the substance, leaving the es-
sential properties untouched. This means, therefore, that chemical mixts can be
re-synthesized after being analyzed into the homogeneous substances of which
they are composed because the chymical atoms of the homogeneous substances
have not themselves been texturally altered in the process of analysis. To support
this hypothesis, Boyle conducts numerous experiments through which he hopes
to empirically establish the stability and operational irreducibility of chymical
atoms, the most notable of which is the redintegration of potassium nitrate,
which will be discussed at the end of this chapter.
At this point, however, I wish to address some claims that have been recently
made regarding the stability of chymical atoms. In a recent article, Kleber Cecon
argues that, although Boyle’s aggregates “are relatively stable entities . . . [they]
can be considered only semi-permanent, and still not enough to establish proper
intermediate causes as required for the origins of modern experimental chem-
istry.”31 However, Cecon adds that “this semi-permanent nature could reflect
a certain stability present in chemical substances as defined in common sense
terms, which would result in observable chemical behavior. In this sense, the
chemical behavior of specific substances can be considered an intermediate
cause.”32 Although I agree with Cecon that Boyle appeals to subordinate and in-
termediate causes in chemistry, I wish to take a position stronger than Cecon’s
regarding the stability of chymical atoms. In this book, I defend the view that
Boyle’s notion of “atomicity” is a negative-empirical concept defined by the limits
of chemical analysis. Thus, although chymical atoms are ontologically and theo-
retically analyzable into their constituent fundamental particles, they are opera-
tionally irreducible and thus, for all practical purposes, they can be considered as
chemically elementary entities.
As already established, the stability of chymical atoms derives not from any
substantial form but from their microstructural form, which Boyle considers to
be a form so stable that it resists corrosive analysis and, as discussed in the next
section, it provides distinct substances with operationally unalterable essential

31 Cecon, “Robert Boyle’s Experimental Programme,” 87
32 Ibid.
properties that warrant their classification within given natural kinds. In fair-
ness to Cecon’s point, it is true that Boyle ultimately fails to provide a satisfactory
theory of microstructure because he cannot explain what accounts for its sta-
bility. I address this problem at the end of this chapter. However, Boyle does en-
dorse the existence of operationally stable entities, that is, of entities that cannot
be further reduced by the then available means of analysis, such as aqua regia,
aqua fortis, or even fire.
Certainly, one can justifiably argue that to remain a consistent mechanistic
corpuscularian, Boyle cannot regard chymical atoms as theoretically indivisible.
As a chemist and natural philosopher, he affirms the existence of stable natural
kinds and stable chemical properties but, to the extent that he rejects substan-
tial forms as the agents of stability and embraces a mechanistic corpuscularian
theory of matter, he must reject the theoretical indivisibility of corpuscular
aggregates or chymical atoms. There is no tension here, however, because it is
the operational indivisibility of chymical atoms, the stability and predictability
of chemical reactions, and the causal efficacy of chemical properties that have
methodological functionality and true heuristic power in the context of his ex-
perimental chemistry. In this sense, one could say that Boyle is fundamentally a
pragmatist who does not allow his theoretical and ontological commitment to a
given theory of matter stand in the way of the heuristic power of experimental
science.
The Sceptical Chymist is, thus, not merely a polemic against Scholastics and
spagyrists, nor is it simply a defense of a generic conception of mechanistic
corpuscularianism. It is also, first and foremost, an exposition and defense of
a very specific type of corpuscularian theory, that is, one that both posits fun-
damental mechanistic particles as the ontological terminus of material bodies
and posits stable corpuscular concretions (chymical atoms) as the operational
terminus of chemical analyses. Such chymical atoms retain their texture and,
thus, their identity and integrity when they enter more complex compositions
with other types of chymical atoms, and this can be demonstrated because the
substances that are composed of such chymical atoms are recoverable through
analysis. As an example, Boyle discusses how gold and quicksilver will either dis-
solve or precipitate, change into a variety of different colors, become malleable or
hard, form a salt or an oil, and so on, depending upon the reagent used in each
experiment. He explains that
the Reason [. . .] that I have represented these things concerning Gold and
Quicksilver, is, That it may not appear absurd to conceive, that such little primary
Masses or Clusters, as our Proposition mentions, may remain undissipated,
notwithstanding their entering into the composition of various Concretions,
since the Corpuscles of Gold and Mercury [. . .] are able to concurre plentifully
to the composition of several very differing Bodies, without loosing their own
Nature or Texture, or having their cohaesion violated by the divorce of their as-
sociated parts or Ingredients.33
Earlier in the same work, Boyle defines the term “element” to mean “those
primitive and simple Bodies of which the mixt ones are said to be composed, and
into which they are ultimately resolved.”34 As William Newman points out,
within the framework supplied by Boyle’s Essay of the Atomicall Philosophy, “el-
ementary” and “atomic” can be seen as coextensive terms—both imply resist-
ance to decomposition [. . .] The identity between the smallest available particles
and the substance as a whole provides direct evidence for such atomism, since
an obvious consequence of this uniformity is that no decomposition products
are present [. . .] homogeneity after attempted analysis is a warrant for the claim
that the material is atomic.35
Thus, in the context of chymical reactions and processes, gold and quicksilver
can be considered elementary although, ontologically, they are composed of fun-
damental particles with strictly mechanical affections. Again, there is no tension
between Boyle’s mechanistic corpuscularianism and his belief in operationally
irreducible concretions because, for all practical, pragmatic, methodological, ex-
planatory, and heuristic purposes, chymical atoms are indeed stable and opera-
tionally irresolvable elementary entities.
At the end of the citation given earlier, Cecon states that Boyle’s semi-
permanent aggregates are “still not enough to establish proper intermediate
causes as required for the origins of modern experimental chemistry.”36 Cecon’s
tacit allusion to the Chemical Revolution here seems to suggest that Lavoisier’s
successful chemical reforms were predicated upon a commitment to completely
stable and irreducible elements. However, a commitment to the theoretical ir-
reducibility of elements was not a requirement for the development of modern
chemistry.
Lavoisier’s proposed reform of the nomenclature of compound substances
in terms of their elementary composition forces the question of what is to
be regarded as an “element.” Lavoisier proposes to answer this question by
restricting himself to what can be determined via experiment. He explains that,
“if by the name of element, we mean the simple and indivisible molecules that

34 Ibid., 220.
35 Newman, “The Significance of ‘Chymical Atomism’,” 254–255.
36 Cecon, “Robert Boyle’s Experimental Programme,” 87.
compose bodies, it is probable that we do not know them: if, on the contrary, we
attach the name of element or principle of bodies to the idea of the last point at
which analysis arrives, all of the substances that we have not yet been able to de-
compose by any means are, for us, to be considered elements.”37
This operational definition provides an experimental criterion for deciding
when a substance should be regarded as an element, but it does not tell us what
the term “element” means.38 Lavoisier understands this and admits that the table
of elements derived by applying this criterion is entirely open to revision. He
states, “we cannot assure that the substances that we regard as simple are not
themselves composed of two or perhaps a greater number of principles. However,
since these principles cannot be separated or, rather, since we have no means of
separating them, they behave for us in the manner of simple substances, and we
must not assume them to be composed until experience and observation prove
otherwise [emphasis mine].”39
As already established, however, Lavoisier’s analytical definition is not entirely
original with him. In fact, according to Paul Needham, this type of definition
dates even further back than Sennert or Boyle and goes as far back as the writings
of Aristotle, although the latter’s conception of elements is obviously quite dis-
tinct from that of Lavoisier.40 More interesting however is the fact that, opera-
tionally, Lavoisier’s “element” is the same as Boyle’s “chymical atom.” “Lavoisier,
like Boyle before him, conceptually distinguishes between ultimate particles and
the undecomposed substances of the laboratory.”41 However, although Boyle ac-
counts for his inability to analyze such substances by theorizing that their mi-
crostructure cannot be decomposed into their constituent primary corpuscles,
Lavoisier refuses to theorize about what might be occurring at the micro-level
to explain why a substance cannot be further analyzed. Thus, by Lavoisier’s own
admission, his list of elements is open to revision, if future methods of analysis
37 Lavoisier, Traité élémentaire de chimie, xii: “Si par le nom d’élémens, nous entendons désigner les
molécules simple & indivisibles qui composent les corps, il est probable que nous ne les connoisons
pas: que si au contraire nous attachons au nom d’élémens ou de principe des corps l’idée du dernier
terme auquel parvient l’analyse, toutes les substances que nous n’avons encore pu décomposer par
aucun moyen, sont pour nous des élémens.”
38 Hendry, “Antoine Lavoisier (1743–1794),” 66.
39 Lavoisier, Traité élémentaire de chimie, xii–xiii: “Non pas que nous puissions assurer que ces
corps que nous regardons comme simples, ne soient pas eux-même composes de deux ou même d’un
plus grand nombre de principes, mais puisque ces principes ne se séparent jamais, ou plutôt puisque
nous n’avons aucun moyen de les séparer, ils agissent à notre égard à la manière des corps simples, &
nous ne devons les supposer composes qu’au moment où l’expérience & l’observation nous en auront
fourni la prévue.”
40 Needham, “An Aristotelian Theory of Chemical Substance.” See also Newman, “What Have We
Learned from the Recent Historiography of Alchemy?”; Newman, Atoms and Alchemy; Bensaude-
Vincent and Stengers, A History of Chemistry.
41 Levere, Transforming Matter, 81.
reveal that these “elements” are further operationally reducible to more funda-
mental entities.
I cannot, therefore, agree with Cecon that a fixed or stable conception of
elementarity is what lies at the origins of modern experimental chemistry since,
as mentioned, Lavoisier admits that what is considered elementary at one time
is open to revision as analytical methods improve. Rather, Lavoisier’s reform and
modernization of experimental chemistry lies in his staunch empiricism, in his
emphasis on a quantitatively oriented chemical methodology, and in his rejec-
tion of philosophical theories of matter, which he faults for being metaphysically
“suspect.”
Another recent claim regarding Boyle’s chemical ontology with which I must
disagree was made by Alan Chalmers, who has explicitly stated that “Boyle’s me-
chanical or corpuscular hypothesis is spelt out in most detail in The Origin of
Forms and Qualities According to the Corpuscular . . . a key feature of it is its re-
ductionist character. All the phenomena of the material world are to be reduced
to the action of matter in motion.”42 Elsewhere, Chalmers reiterates again that
“A key feature of Boyle’s mechanical philosophy that needs to be stressed is the
extreme reductionism that it involved.”43 Although Chalmers concedes that, for
Boyle, “bodies and substances will have qualities other than the primary [me-
chanical] ones, both those detectable by the senses and those involved in the in-
teraction of bodies, such as elasticity and magnetic properties,”44 he adds that
“for Boyle, the secondary qualities are to be reduced to, that is, explained in
terms of, the primary ones. More specifically, all the phenomena of the material
world are to be explained in terms of the shapes, sizes and mobility of corpuscles
together with the spatial arrangement of those corpuscles amongst themselves.
These fundamental explanations will involve appeal to the laws of nature that
govern the motions of corpuscles.”45
I disagree with Chalmers’s conclusion that Boyle’s mechanistic
corpuscularianism entails such an extremely reductionist explanatory approach.
It is obvious that Boyle’s chemical explanations are not reductionist in the strict
sense implied by Chalmers since, as Chalmers himself admits, Boyle does not
invoke mechanistic corpuscularianism as an explanatory principle in his ex-
perimental writings. As was discussed at the end of the previous chapter, the
mechanical philosophy as the “metaphysical” hard core of Boyle’s research pro-
gramme implies an ontological reductionism of all phenomena to the mechan-
ical affections of corpuscles and thus limits permissible explanations to those that

43 Chalmers, “Boyle and the Origins of Modern Chemistry,” 2.
45 Ibid.
are consistent with the mechanical philosophy. However, although the mechan-
ical philosophy provides such a limiting principle, its ontological reductionism
has no heuristic value or explanatory power with regard to chemical reactions
and transformations. Therefore, explanations of such phenomena must invoke
higher-level chemical causes for which a reductionist account would neither be
accessible nor informative. As I will argue in the next chapter, Boyle holds a con-
ception of higher-level qualities as emergent, relational, and dispositional and,
as the final chapter will show, the whole-parts relation in chymical atoms cannot
be captured with the kind of classical and summative mereology that would be
required by a strictly reductionist mechanicism.
At this point, however, I wish to address the intimate connection between
Boyle’s structural realism, his realism about natural kinds, and his ideas re-
garding the taxonomical classification of material species. As pointed out in the
previous chapter, Boyle is fully aware of the limits of Cartesianism for chemistry
and clearly prefers to formulate explanations that rely on chemical properties,
cosmical qualities, and other subordinate causes, as the experiment involving the
redintegration of potassium nitrate clearly illustrates. Further evidence of Boyle’s
predilection for such explanations is found in most of his experimental writings.
For example, in the Experiments Touching Colours (1664), Boyle describes and
explains one experiment involving a common sublimate (probably mercuric
chloride [HgCl2] and water),46 oil of tartar (potassium carbonate [K2CO3]), and
oil of vitriol (sulfuric acid [H2SO4]) in the following manner:
The Experiment is very easie, and it is thus perform’d: Take good common
Sublimate, and fully satiate with it what quantity of Water you please, Filtre the
Solution carefully through clean and close Paper, that it may drop as Clear and
Colourless as Fountain water. Then when you’l [sic] shew the Experiment, put
of it about a Spoonfull into a small Wine-glass, or any other convenient Vessel
made of clear Glass, and droping in three or four drops of good Oyl of Tartar,
per Deliquium, well Filtered that it may likewise be without Colour, these two
Limpid Liquors will in the twinkling of an Eye turn into an Opacous mixture
of a deep Orange Colour [. . .] And when the Spectators have a little beheld this
first Change, then you must presently drop in about four or five drops of Oyl of
Vitriol [. . .] the whole Colour, if you have gone Skillfully to work, will immedi-
ately disappear, and all the Liquor in the Glass will be Clear and Colourless as
before, without so much as a Sediment at the Bottom.47
46 It is safe to speculate that this was the common sublimate used in this experiment since mer-
curic chloride with water was a common sublimate that Boyle used, with potassium hydrogen tartate
and sulfuric acid, in a similar experiment described in History of Colours. This latter experiment is
discussed in Clericuzio, Elements, Principles, and Corpuscles, 144–145.
47 Boyle, Experiments and Considerations Touching Colours, in The Works of Robert Boyle, 150–151.
Boyle explains the change of the solution from transparent to orange, then back
to transparent, as being due to purely chemical processes so that
the Yellowness of the Mercurial Solution and the Oyl of Tartar is produc’d by
the Precipitation occasion’d by the affusion of the latter of those Liquors, and
that the destruction of the Colour proceeds from the Dissipation of that Curdl’d
matter, whose Texture is destroy’d, and which is dissolv’d into Minute and
Invisible particles by the potently Acid Menstruum, which is the reason, why
there remains no Sediment in the Bottom [. . .] this seems to be the Chymical
reason of this Experiment, that is such a reason, as [. . .] may give such an ac-
count of the Phaenomena as Chymical Notions can supply us with.48
Although Boyle’s chemical explanation relies on the notion of texture, he does

not renounce his commitment to the mechanistic corpuscularian ontology
and reminds his readers that he is “sufficiently aware of the difference betwixt a
Chymical Explication of a Phaenomenon, and one that is truly Philosophical or
Mechanical.”49
Boyle’s commitment to chemical explanations that eschew any reference
to the mechanical affections of minima naturalia is even more pronounced in
his physiological essays, such as his Memoirs for the Natural History of Human
Blood. In fact, the idea that natural phenomena must be explained by appealing
to causes other than the mechanical affections of minima naturalia is consistent
with Boyle’s approach in most of his experimental writings. For example, in the
appendix to Final Causes, he states that
there are a great many things of wch we may have some knowledge, and dis-
course to one another rationally & usefully, wch yet cannot with any conven-
ience be immediately deduc’d from the First and simplest Principles; namely,
Corpuscles and Motion; but must be deriv’d from subordinate Principles;
such as ye Greate Systeme of ye World, Gravity, Fermentation, Springiness,
Magnetism, &c.50
Boyle not only affirms the usefulness of such subordinate causes, which include
chemical properties as well as cosmical qualities such as magnetism and gravity,
but he also claims these to be the most useful of all explanatory principles in nat-
ural philosophy. Immediately after making the just-cited remark, he states that
48 Ibid., 152.
49 Ibid.
50 Boyle, Title Material Relating to Appendix to “Final Causes,” Vol. 9, folio 40v.
the most useful Notions we have both in Physick, Mechanicks, Chymistry, and
the Medicinal Art, are not deriv’d Immediately from the First Principles; but
from—Intermediate Theories, notions, and Rules:—it being often sufficient
for very useful Purposes, to terminate our researches into some settled rule or
equivalent Axiome of Nature, and thence derive—Practical conclusions by way
of Inference or Application. Of the Subordinate or Intermediate—causes or
Theories of natural things—there may be many; some more and some less re-
mote from the First Principles, and yet each of them capable to afford a just de-
light & usefull Instruction to the mind. And these we may—call for distinction
sake, the Cosmographical, the Hydrostatical, the Anatomical, the Magnetical,
the Chymical, and other causes or reasons of—Phaenomena, as those wch are
more Immediate (in our way of estimating things) than ye general and primor-
dial causes of natural effects.51
These remarks, however, may seem to be at odds with some of the things ex-
plicitly stated in theoretical works, such as The Excellency of the Corpuscular
or Mechanical Philosophy, to the effect that “if the Agents or active Principles
resorted to, be not Immaterial, but of a Corporeal Nature, they must either in
effect be the same with the Corporeal Principles above-nam’d; or, because of the
great Universality & Simplicity of ours, the new ones propos’d must be less ge-
neral than they, and consequently capable of being subordinated or reduc’d to
ours.”52 There seems to be a tension between what is said here and what Boyle
claims in the appendix to Final Causes. However, as I argued in the previous
chapter, this apparent tension may be resolved once one understands the dis-
tinction between the mechanical philosophy’s negative heuristic and its positive
heuristic and the ensuing semantic shift in his use of the term “mechanical” in
the experimental writings.
3.4 Taxonomical Classification, Natural Kinds, and

Essential Form
To situate Boyle both historically and philosophically regarding the issue of nat-
ural kinds, I will briefly examine the distinction between his views and those of
John Locke, whom he influenced but with whom he was also engaged in lively
debates. Both Locke and Boyle reject the Aristotelian conception of natural
kinds, which was predicated on the possession of distinctive substantial forms.
As discussed in Chapter 1, Aristotle considers substantial form to be “responsible
51 Ibid., folios 40v–41r.

52 Boyle, Of the Excellency and Grounds of the Corpuscular or Mechanical Hypothesis, 117.
for making naturally occurring materials what they are.”53 One of the reasons
for Locke’s and Boyle’s rejection of this Aristotelian view is that it is untenable
both from an empiricist and from a mechanistic perspective, because it invokes a
principle other than matter itself to account for the identity of material bodies as
members of distinctive species.
However, an underlying concern for both men is that of how to justify the
taxonomical classification of material species, in the absence of substantial form.
Locke wishes to preserve the distinction between nominal essence and real es-
sence, that is, between the ideas that we acquire about substances based on their
observable properties and the real nature of things, which cannot be known due
to our intellectual and perceptual limitations. Due to these epistemic limits, our
classifications are based on nominal essences rather than on real essences, since
the latter are not empirically accessible to observation. Even though our taxo-
nomical practices are clearly not completely arbitrary, since they are based on the
observed properties of things, these observed properties reveal nothing about
real essences. Thus, although there is debate among scholars about whether
Locke was an ontological realist about natural kinds, his theory of taxonomical
classification is strongly conventionalist.54
As any student of early modern British empiricism knows, the distinction
between nominal essence and real essence was difficult to sustain within any
kind of empiricist position unless that empiricism was imbued with elements
of Cartesian rationalism, as was the philosophy of John Locke. George Berkeley
and David Hume would jointly close the lid on the notion of real essences, along
with that of abstract ideas and other “priestly dogmas” as Hume would call
them, by arguing that one must not embrace any notion whose origin cannot
be traced back to direct observation. Robert Boyle, whose natural philosophy is
strongly committed to empiricism, is also concerned with avoiding any appeal
to notions that cannot be supported by the evidence of observations. Thus, his
approach to taxonomical classification is undoubtedly imbued with elements of
conventionalism.
In The Origin of Forms and Qualities, Boyle clearly displays conventionalist
leanings when he states that “men having taken notice, that certain conspicuous
Accidents were to be found associated in some Bodies, and other Conventions
of Accidents in other Bodies, they did for conveniency, and for the more expe-
ditious Expression of their Conceptions agree to distinguish them into several
Sorts, which they call Genders or Species.”55 Boyle agrees with Locke that, in the
53 Chalmers, “Klein on the Origin of the Concept of Chemical Compound,” 45.
54 Some of the scholars who reject the view that Locke was an ontological realist about natural
kinds are Margaret Atherton, Lisa Downing, and Jan-Erik Jones. See Atherton, “Locke on Essences
and Classification”; Lisa Downing, “Locke’s Ontology”; and Jones, “Lockean Real Essences and
Ontology.”
absence of substantial form, there must always be an element of convention in

classification. He also agrees that the boundaries between material species may at
times remain unclear, because taxonomical classifications are “made on the basis
of phenomenal qualities [Accidents] and not on the basis of an understanding
of their underlying structure.”56 However, it would be an error to conclude that
such conventionalism is Boyle’s ultimate position.
Robert Pasnau unfortunately errs in this direction by attributing to Locke and
Boyle the same final position regarding taxonomical classification. Here, I will
discuss and respond to Pasnau’s claim, and use this as a springboard for what
I take to be a more accurate account of Boyle’s views. Pasnau states that
for the scholastics, our classification of individuals into species tracks the
essences of things. Though we may not have a direct or comprehensive grasp
of what those essences are [. . .] we know enough to sort individuals into their
true species. Like Boyle before him, Locke utterly rejected this optimistic view,
and took our distinctions between species to be the product of haphazard and
highly fallible groupings on the basis of superficial resemblance [. . .] Locke
argues that without knowing the essence, we have no way of knowing which
properties point toward the nature of the species and which are merely acci-
dental. Hence, our distinctions between species can be grounded only on our
haphazard conceptions of the nominal essences of things.57
Although I fully agree with Pasnau’s description of Locke’s view of classifica-

tion, I believe that it is a mistake to equate this view with Boyle’s final position. As
has been extensively argued already, Boyle does believe that one can distinguish
between those properties of a material body that are accidental and those that are
essential, that is, those properties that derive from the stable and operationally
irreducible microstructure or essential form of the corpuscular concretions. He
also believes, as I will show presently, that it is because of said essential properties
that we can establish the species to which a material body belongs. Furthermore,
to the extent that Boyle identifies an empirical basis for understanding the real
essence of material bodies, his conventionalism is muted by the fact that he does
not regard the process of classification as being purely arbitrary.
For Boyle, there is an empirical relationship between the real essence of a ma-
terial body and its observable properties. Those properties are essential that are
jointly necessary and sufficient for a material body to belong to a specific natural
kind or species. We can know these essential properties by observing that they are

not mechanically alterable by ordinary chemical reagents.58 Thus, we can classify

material bodies into kinds by identifying the necessary and sufficient properties
that define those kinds. Boyle states that “since to every Determinate Species of
Bodies, there doth belong more then [sic] One Quality, and for the most part a
concurrence of Many is so Essential to That sort of Bodies, that the want of any
of them is sufficient to exclude it from belonging to that Species: there needs no
more to discriminate sufficiently any One kind of Bodies from all the Bodies of
the World, that are not of that kind.”59
Although, in the earlier cited passage from The Origin of Forms and Qualities,
Boyle refers to phenomenal qualities as “accidents,” we should not extract from
this that such properties are merely contingent to a particular body’s member-
ship within a species. In fact, Boyle does not distinguish the term “accident” from
the term “essence” but, rather, distinguishes it from the term “substance.” Simply
put, Boyle uses “accident” to mean “quality.” He explains that, although specific
accidents are never essential to catholic matter as such, some accidents are essen-
tial to specific natural kinds. In his own words, “Nor need we think that Qualities
being but Accidents, they cannot be essential to a Natural Body; for Accident,
as I formerly noted, is sometimes oppos’d to Substance [. . .] and though an
Accident can be but accidental to [universal] Matter [. . .] yet it may be essential
to this or that particular Body.”60 Thus, Boyle’s conventionalism regarding taxo-
nomical classification is tempered by realism regarding the essential properties
of natural kinds, which serve as guides for such classification. He firmly believes
that “the real constitution of a body, along with its constituent properties, mind-
independently makes it the kind of thing that it is.”61 Key passages in The Origin
of Forms and Qualities indicate that Boyle’s realism about natural kinds compels
him to formulate an empirically sound theory of taxonomical classification
that explains the existence of natural kinds through a material principle, rather
than through a substantial principle that is distinct from matter. He, therefore,
replaces the notion of substantial form with one of mechanical form that is
grounded in his structural realism and, as previously established, because me-
chanical form determines the essential properties of a substance, Boyle refers to
it as “essential form.”
For Boyle, essential forms are “naturally repeated material structures of
bodies that determine their kind. Matter constitutes a ‘form’ [. . .] when it is or-
dered in specific structures.”62 Boyle clarifies that the term “form,” as he employs
58 We are excluding from our discussion those extraordinary chemical processes such
as chrysopoeisis, which involve unusually powerful reagents such as the alkahest or the
Philosopher’s Stone.
60 Ibid., 324.
61 Jones, “Locke vs. Boyle,” 660.
62 Jones, “Boyle, Classification, and the Workmanship of the Understanding Thesis,” 177.
it in this context, does not mean a substance separate from or independent of

matter. In The Origin of Forms and Qualities, he states, “though I shall for brev-
ities sake retain the word Forme, yet I would be understood to mean by it, not a
Real Substance distinct from Matter, but onely the Matter it self or of a Natural
Body.”63As Peter Anstey concurs, “it is clear that, for Boyle, form can be defined
in terms of phenomenal qualities or underlying micro-structure, though the
weight of textual evidence suggests that it was the qualities rather than mechan-
ical affections which are the primary referent of the term.”64 However, I would
add that, although essential qualities of a substance are the primary referent of
the term, it is the microstructure that accounts for the essential qualities. Thus,
essential form is microstructure. Since the qualities are observable, while the es-
sential form is not, qualities are what reveal the form to the observer.
Boyle holds that the mechanical form is the essential form of material bodies
because, without it, bodies would not have the essential properties that they do
in fact have. The peculiar mechanical form of a body is sufficient to denominate
it as a body of a certain kind. Thus, “so long as bodies possess the right struc-
ture, they are part of that species.”65 For this reason, Boyle also disagrees with
the Scholastic distinction between natural substances and artifacts, a distinc-
tion that is grounded in the theory of substantial forms. In fact, all mechanistic
philosophers rejected this distinction precisely because they rejected substantial
forms. For Descartes, this meant that neither artificial nor natural bodies have
form of any sort. However, for Boyle, though neither natural nor artificial bodies
have substantial forms, they both have essential mechanical forms that qualify
them as members of particular species, thus rendering the natural/artificial sub-
stance distinction irrelevant “for there is not always such a difference as many
imagine between the one and the other.”66 As will be seen in the final chapter, the
unity afforded to chymical atoms by essential form also provides mereological
reasons for rejecting the natural/artificial substance distinction.
Boyle’s conception of form can do all the work for the mechanical philosophy
that substantial form did for the Scholastics. That is, it can explain species mem-
bership, “account for the unity of the body as a single entity, and ground the prop-
erties of the body.”67 He also indicates his preference for structural essential form
over substantial form in The Sceptical Chymist, stating that “This Aggregate or re-
sult of Accidents you may, if You please, call either Structure or Texture [. . .] Or if,
retaining the Vulgar Terme, You will call it the Forme of the thing it denominates,
I shall not much oppose it; Provided the word be interpreted to mean but what

66 Robert Boyle, Letters and Papers of Robert Boyle, 300.
I have express’d, and not a Scholastick Substantial Forme, which so many intelli-
gent men profess to be to them altogether Un-intelligible.”68 The unintelligibility
of substantial form is, in no small part, due to its immateriality and empirical
inaccessibility. However, although essential form is also empirically inaccessible
as such, something about it is revealed by phenomenal and chemical properties.
Thus, although we have no direct perceptual access to essential form or deep
structure, the results of specifically designed chemical experiments give us indi-
rect epistemic access to it and I discuss the most famous of these experiments in
the following section.
3.5 The Empirical Nature of Essential Form: The Reduction

to the Pristine State
It is true that Boyle’s chymical atoms and their microstructure are entirely hypo-
thetical posits, since they are not empirically accessible as such just as substan-
tial form had not been empirically accessible. Unlike substantial form, however,
chymical atoms are postulated as material entities with causal properties, which
implies that experiments can be conducted to confirm hypotheses about these
entities. To be more precise, because the microstructure or essential form of
chymical atoms accounts for the stability of chemical substances and of chemical
properties, essential form retains an empirical connection to the world that the
epistemically suspect notion of substantial form did not have. Thus, a chemist
can conduct experiments to demonstrate that microstructure is stable in chem-
ical processes such as combination, separation, and transposition of chymical
atoms, which are merely mechanical processes that alter only the extra-essential
properties of substances.
Of the many experiments that Boyle uses for these purposes, the one that
he considers most effective for establishing the stability of microstructure is
“the reduction to the pristine state,” an experiment that he appropriates from
Daniel Sennert. As Christoph Meinel points out, however, Sennert himself had
inherited this procedure from iatrochemists and pharmacists who had used it
for pragmatic and atheoretical purposes, such as describing quantitatively the
formation and decomposition of metallic substances.69 Sennert had both appro-
priated and further developed the reduction to the pristine state for theoretical
reasons, that is, to provide empirical support for vitalistic corpuscularianism and
for the theory of substantial form as an immaterial principle of identity. As al-
ready explained in Chapter 1, however, Sennert regarded the minute particles

69 Meinel, “Empirical Support for the Corpuscular Theory in the Seventeenth Century,” 85.
or atoms of substances, rather than the substances themselves, as endowed with

form. Thus, for Sennert, the purpose of reduction to the pristine state was to es-
tablish empirically that the corpuscles of a chemical substance retained their dis-
tinct substantial form even when that specific substance was fused with other
substances.
Sennert eventually developed different types of reduction to achieve different
theoretical purposes, and Boyle appropriated these methods selectively to pro-
vide empirical support for his conception of mechanical form. “The first type
[of reductions] were simple distillations and sublimations of substances such as
alcohol, sulphuric acid, and sulphur, which he regarded to be merely mechanical
operations by means of which bodies were mashed into their atoms.”70 This par-
ticular type of reduction to the pristine state was not useful for Boyle’s purposes,
and he thus shunned it in favor of the second type of reduction developed by
Sennert, which attempted to show that it is the actual atoms of a substance that
retain their form, rather than the substance as such. For this experiment, Sennert
had fused “gold and silver together to obtain an entirely homogeneous alloy.
Then he poured aqua fortis or nitric acid on it. The silver was dissolved, whereas
the gold particles settled to the bottom. He separated the two phases and precip-
itated the silver from the solution to obtain another fine sediment. Eventually, he
melted both powders and obtained, quantitatively, gold in the first case, silver in
the latter.”71
Boyle appropriates this type of reduction because he believes that it can es-
tablish the stability of chemical substances and can, thus, indirectly serve to es-
tablish 1) the existence of stable chymical atoms, 2) a mechanical explanation
for essential properties, and 3) the empirical grounds for a non-arbitrary taxo-
nomic classification of natural kinds. Although Boyle carries out the experiment
with various metals, the basic procedure of “reduction to the pristine state” is the
same throughout. It basically involves dissolving metallic atoms in acid, indu-
cing precipitation of those atoms by adding an alkali, and reducing the precipi-
tate thereby restoring it to its former state. For Boyle, the conclusion to be drawn
from this experiment is that the metal had “simply been hidden within the solu-
tion all along in the form of indissoluble”72 corpuscular aggregates or chymical
atoms and that the metal’s essential properties had not been altered, which is
what allows for their recovery via precipitation and reduction.
William Newman describes one such experiment in detail. “One of Boyle’s
most important reductions to the pristine state involves the dissolution of cam-
phor in nitric or sulfuric acid. If sulfuric acid is used, the camphor forms a deep
70 Ibid.
71 Ibid., 85–86.
72 Newman, “Newton’s Early Optical Theory and Its Debt to Chymistry,” 291–292.
reddish solution and loses its color. Hence, the camphor becomes unrecognizable
as camphor and seems to be perfectly mixed in the solution. But the mere addi-
tion of water causes the camphor to precipitate and to return to its former state,
including the reacquisition of its powerful scent.”73 In The Sceptical Chymist,
Boyle describes a solution to the pristine state involving gold and concludes that
the metal is made up of “minute Masses or Clusters” that are aggregates of more
minute particles74 and that the aggregate corpuscles are “not easily dissipable
into such Particles as compos’d them.”75 Here, Boyle is “making a strong claim
for the semi-permanence of second-order or aggregate corpuscles”76 or chymical
atoms and, thus, for the existence and stability of essential form.
Boyle concludes that non- essential phenomenal changes, such as those
observed in the reduction to the pristine state, are the result of alterations in the
spatial relations between aggregate corpuscles that leave the microstructure of
these corpuscles untouched. Once again, he reaches this conclusion due to his
inability to alter the camphor’s essential properties, which accounts for his re-
covery of the camphor in its original state. It is clear from this that Boyle’s con-
ception of chymical atoms is a negative-empirical concept, that is, it is a concept
that reflects “the limits of the technique of analysis”77 and is defined by those
limits. Thus, although he embraces the theoretical existence of more funda-
mental entities, Boyle defines atomicity and elementarity in operational terms,
that is, in terms of what remains as the homogeneous final product of analysis.
Newman explains how this operationally defined chymical atomism fits within
the context of Boyle’s chemical philosophy:
According to this analytical ideal, a substance is viewed as elementary if the

tools of the chemist cannot decompose it. Hence, the concept is negative in that
it defines an element solely in terms of what chemistry cannot do, and empir-
ical in that it relies on the experience of the laboratory. For this reason, then,
the ability of aggregate corpuscles [or chymical atoms] to withstand the cor-
rosive menstrua of the seventeenth century makes it probable that they are
operationally indissoluble, and hence atomic. For Boyle, the identity between
the smallest available particles and the substances as a whole provides direct
evidence for such atomism, since an obvious consequence of this uniformity
is that no decomposition products are present. Homogeneity after attempted
analysis is a warrant for the claim that the material at hand is atomic.78
73 Ibid., 292.
75 Ibid.
76 Newman, “The Alchemical Sources of Robert Boyle’s Corpuscular Philosophy,” 583.
77 Bensaude-Vincent, Bernadette, and Stengers, A History of Chemistry, 37.
Thus, “the irreducible constituents that function in chemical reactions, that is,
‘chymical atoms,’ are themselves concretions of an underlying homogeneous
matter that retain their structural arrangement through analysis by fire or chem-
ical corrosives. These allow Boyle to retain an essential form within matter and
thus to avoid the consequences of a completely permissive mereology.”79 This last
point will be examined in more detail in the final chapter of this book.
The success of these experiments convinced Boyle of the truth of his hypothesis
regarding the stability of chymical atoms and their deep structures. However, the
burden for Boyle is that of explaining why the deep structure of chymical atoms
would resist penetration by air or alteration by fire or any of the other extremely
caustic analytical tools of the chymical laboratory. Let us remember that Boyle
considers microstructure to result from the adhesion of minima in a very close
spatial and geometrical juxtaposition, which brings these particles so close to-
gether that even air cannot traverse them. Why would such powerful analytical
tools as nitric acid, sulfuric acid, or fire be unable to break through the extremely
close and tight juxtaposition of minima that forms the structure of chymical
atoms? Let us remember that Boyle embraces mechanicism and would, there-
fore, reject the possibility that the deep structures of chymical atoms are held
together by the presence of bonding forces, even if such a hypothesis had been
available to him as an alternative to that of geometrical structure. The failure to
explain how the tight spatial juxtaposition of minima could possibly result in
the stability and operational irreducibility of chymical atoms is one of the many
weaknesses of Boyle’s structural hypothesis.
Given Boyle’s adherence to the notion of aggregate corpuscles or chymical
atoms with essential form, I would agree with recent authors who emphasize that
Boyle’s mechanistic corpuscularianism incorporates a compositional theory of
matter that allows for a non-reductionist account of chemical properties within
the context of a mechanistic conception of minima naturalia. However, I argue
that it is not simply the composition of aggregate corpuscles that does the work.
Rather, it is the composition coupled with structure or texture that accounts for
the chemical properties of substances. I contend that Boyle’s notion that struc-
ture, rather than mere composition, defines chemical identity anticipated by sev-
eral centuries the concept of chemical isomerism.
Boyle believes that the microstructural nature of matter is not merely phil-
osophical speculation but is supported by empirical evidence of the sort pro-
vided by the reduction to the pristine state. His chymical atomism and structural
realism are operational notions, rather than purely metaphysical ones. Though
these theories are grounded in the mechanistic hypothesis, they emphasize

aggregate corpuscles “that retain their substantial identity rather than deriving
all of their macro-level properties from [. . .] mechanical characteristics”80 of
prima naturalia. For Boyle, the chymical atom is “an operationally indestruct-
ible bit of matter [. . .] not a putative entity merely and strictly composed of
uniform catholic matter.”81 The determinate and essential properties of chem-
ical substances, arising from the microstructure of chymical atoms, function
as causal agents in the laboratory and in nature. Essential form also provides a
non-arbitrary method for classifying material species as natural kinds, leaving
their substantial identity intact without the need for the dubious concept of sub-
stantial form, at least as it was altered from its original Aristotelian purpose and
handed down from the medieval tradition and through the Renaissance.
In the following chapter, I will argue that, although the microstructure or tex-
ture of chymical atoms determines the disposition of a chemical substance to
display certain non-mechanical properties, it is the substance’s relation to other
substances in the context of specific chemical processes and procedures that
allows dispositional properties to be actualized and to causally affect changes in
other material bodies. As well, I will also argue that Boyle conceives of chem-
ical qualities as emergent properties, to the extent that he regards chemical
qualities as novel properties that supervene on the texture or essential form of
chymical atoms.

81 Ibid., 263–263.
4
Boyle’s View of Chemical Properties
as Dispositional, Relational, and
Emergent Properties
As indicated in the previous chapter, Boyle considers texture or microstructure

to be a mechanical, albeit not fundamental, feature of chymical atoms. Therefore,
to establish that Boyle favors a non-reductionist conception of chemical quali-
ties, it is not enough to show that these qualities are a function of microstructure
and are, therefore, not directly reducible to the mechanical affections of funda-
mental particles. If we merely stop at establishing this, then we will simply have
shown that Boyle’s views regarding chymical atoms are comparable to those of
Gassendi regarding molecules. If all we can determine is that chemical proper-
ties are reducible to texture, then we would still have to consider Boyle’s chemical
philosophy as mechanistically reductionist, although not in the strict sense. To
establish that Boyle holds a non-reductionist chemical ontology, we must go fur-
ther than this and argue that he considers chemical qualities to be dispositional,
relational, emergent, and supervenient properties. This is precisely what I intend
to do in this chapter.
To accomplish this goal, I first argue that Boyle considers chemical qualities to
be dispositional and relational. Peter Anstey has already shown that, for Boyle,
sensible qualities are dispositional and relational. I will examine in detail Anstey’s
arguments and show that these arguments can be extended to Boyle’s conception
of chemical qualities. However, although this will constitute a first step toward
establishing a non-reductionist interpretation of Boyle’s chemical ontology, we
must go a step further and argue that Boyle conceives of chemical qualities as
emergent properties that supervene upon microstructure. This means showing
that he considers chemical qualities to be something novel, “over and above” the
chymical atoms from which they emerge. I believe that there is enough textual
evidence in Boyle’s chemical writings to sustain such an argument and I will,
thus, refer extensively to these writings. My conclusions in this chapter will serve
as a foundation for my discussion in the final chapter, in which I will continue
this line of argument by fleshing out the mereological theory that is most con-
sistent with this non-reductionist interpretation of Boyle’s chemical ontology.
University Press.
DOI: 10.1093/oso/9780197502501.001.0001
4.1 The Hierarchy of Properties in Boyle’s

Chemical Ontology
Before I begin discussing the dispositionality and relationality of chemical qual-

ities, I wish to delineate in detail the hierarchy of properties in Boyle’s ontology
to establish exactly where chemical properties fit in relation to mechanical prop-
erties. To do this, I wish to first distinguish between Boyle’s conception of qual-
ities and those of Galileo Galilei, who first articulated his distinction between
primary and secondary qualities in Il Saggiatore (1623). Though Boyle is some-
what indebted to Galileo’s views, there are important differences between them
regarding the primary/secondary quality distinction. After this, I will provide
a brief overview of Boyle’s conception of qualities1 and then proceed to a dis-
cussion of Anstey’s arguments for the dispositionality and relationality of sen-
sible qualities. I will then extend Anstey’s arguments to defend the notion of
dispositionality and relationality of chemical qualities in Boyle’s ontology.
As is well known, Galileo was the first natural philosopher to draw an explicit
distinction between primary and secondary qualities. However, we shall see that
there is a sharp difference between what Galileo means and what Boyle means
when each of them appeals to this distinction. Although Galileo articulates
the distinction between primary and secondary qualities for reasons similar to
Boyle’s, that is, as an alternative to Aristotle’s unsatisfactory qualitative concep-
tion of properties, Galileo’s account is distinct from Boyle’s in that Galileo does
not take secondary qualities to be qualities of the object at all but locates them in
the perceiver instead. In Il Saggiatore, he states:
I say that, as soon as I conceive of a material or corporeal substance, I feel myself

pulled by the need to conceive that it is limited and configured by this or that
shape, that it is large or small in relation to other substances, that it is in this or
that place, in this or that time, that it is either in motion or at rest, that it either
touches or does not touch another body, that it is one, few or many, and I cannot
separate it in my imagination from these conditions; but that it must be white or
red, bitter or sweet, loud or mute, sweet smelling or foul smelling, I do not feel
mentally compelled to apprehend it as being necessarily accompanied by these
conditions [. . .] I think, therefore, that these flavors, odors, colors, etc. [. . .] hold
residence only in the sensitive body so that if the animal is removed, so will all
these qualities also be removed and annihilated.2
1 I am particularly indebted for this discussion to Peter Anstey’s very clear schematization of
Boyle’s division of qualities in The Philosophy of Robert Boyle, 29.

2 Galileo Galilei, Il Saggiatore, 370: “Per tanto io dico che ben sento tirarmi dalla necessità, subito
che concepisco una materia or sostanza corporea, a concepire insieme ch’ella è terminata e figurata di
questa or quella figura, ch’ella in relazione ad altre è grande or piccola, ch’ella è in questo or quel luogo,
in questo o quel tempo, ch’ella si muove o sta ferma, ch’ella tocca o non tocca un altro corpo, ch’ella
Boyle’s View of Chemical Properties 111
The only relationality in Galileo’s ontology is that which exists between objects
and perceivers, and this is what constitutes the ontological difference between
the two types of qualities. As Filip Buyse points out in a recent paper on the topic,
“the difference between the two types, for Galileo, is that primary properties be-
long necessarily to the body in itself . . . and exist, as a consequence, independ-
ently of the observer. In short, they are mind-independent properties. Secondary
affections, on the contrary only exist . . . in the mind of an observer. Thus, they
are mind-dependent qualities.”3 It follows that, for Galileo, secondary qualities
are relational, while primary qualities are non-relational. Therefore, he finds
it impossible to fit solubility, malleability, fixedness, volatility, or magnetism
into his arrangement of qualities. The reason for this is that these qualities are
both clearly relational (though in a different sense of relationality) but also in-
dependent of observers. Thus, these qualities seem to fit neither his description
of primary qualities nor his description of secondary qualities. Yet, since many
of these qualities are chemical qualities, they are precisely the types of qualities
that Boyle wishes to account for. In fact, Boyle is concerned not only with “the
qualities which corporeal objects possess in themselves and as they are related
to percipients, but [with] the qualities which corporeal objects possess as they
are related to other corporeal objects and the universe of which they are a part
as well.”4
For Boyle, the distinction between primary and secondary qualities is inher-
ently connected to his mechanistic corpuscularian ontology. As we have already
established, for him, the most fundamental and truly inherent properties of cor-
poreal objects are the mechanical qualities of fundamental particles (minima
naturalia), which consist of the shape, size, and mobility of these particles. These
are the qualities that Boyle considers to be primary. Texture is also considered
a mechanical primary quality, but it is a quality that is inherent in corpuscular
concretions, rather than in fundamental particles. The mechanical primary
properties are considered inherent to corporeal objects and, as such, they are
considered non-relational. That is, even if a material body were alone in the uni-
verse, unaccompanied by any other body or any perceiver, that corporeal ob-
ject would still have its inherent mechanical primary properties, that is, the
è una, poche o molte, né per veruna imaginazione posso separarla da queste condizioni; ma ch’ella
debba essere bianca o rossa, amara o dolce, sonora o muta, di grato o ingrato odore, non sento farmi
forza alla mente di doverla apprendere da cotali condizioni necessariamente accompagnata [. . .] Per
lo che vo io pensando che questi sapori, odori, colori, etc. [. . .] tengano solamente lor residenza nel
corpo sensitivo, sì che rimosso l’animale, siano levate ed annichilate tutte queste qualità.”
3 Buyse, “The Distinction between Primary Properties and Secondary Qualities in Galileo Galilei’s
Natural Philosophy,” 23.

4 O’Toole, “Qualities and Powers in the Corpuscular Philosophy of Robert Boyle,” 303.
mechanical affections of its minima partes as well as the microstructure of its

corpuscular concretions.
The second type of quality is the non-mechanical secondary quality, which
represents those qualities that are not inherent in corporeal bodies, that derive
from the texture of corpuscular concretions, and that are dispositional. These
dispositional qualities are also considered relational because they are manifested
only when the corporeal object that possesses these dispositions is in the pres-
ence either of another object or of a perceiver. This category of non-mechanical
secondary qualities is itself divided into three sub-categories: manifest qualities,
occult qualities, and sensible qualities. Under the sensible qualities, we find color,
sound, odor, and taste. As shall be argued, these qualities are relational because
the object’s disposition to affect a perceiver in particular ways is not actualized
until the object is in the presence of a perceiver. Under occult qualities, we find
magnetism and electricity, so-called cosmical qualities, which will be discussed
later in this chapter. For Boyle, occult qualities are real because, although they
never become manifest in and of themselves, their effects upon corporeal bodies
are manifested under the appropriate conditions. It will also be argued that, for
Boyle, occult or cosmical qualities are not inherent to objects but, rather, affect
objects to the extent that these objects are part of a universe of objects and laws
of nature. Finally, under manifest qualities, we find a further subdivision be-
tween first, second, and third qualities. The first manifest non-mechanical sec-
ondary qualities are the qualities that Aristotle related to the four elements, that
is, the qualities of hot, cold, wet, and dry. The second manifest non-mechanical
secondary qualities are the chemical qualities, and the third manifest non-
mechanical secondary qualities are the medical qualities. Here is a list of Boyle’s
divisions among qualities:5
Primary qualities:
(mechanical, inherent, non-relational properties)
Minima naturalia
• Shape
• Size
• Mobility
Corpuscular concretions
• Texture (“essential form”)
Secondary qualities:
(non-mechanical, non-inherent, relational, dispositional properties)
1. Sensible
5 Again, I am here indebted to Anstey’s schematization, though my list may look somewhat dif-
ferent from the sketch that he provides on p. 29 in The Philosophy of Robert Boyle.
• Color
• Sound
• Odor
• Taste
2. Occult
• Magnetism
• Electricity
3. Manifest:
a. First
• Hot
• Wet
• Cold
• Dry
b. Second
• Chemical qualities
c. Third
• Medical qualities
Although Galileo based his primary/ secondary quality distinction on

whether the qualities in question were dependent upon the presence of an ob-
server, Boyle bases his primary/secondary quality distinction on whether the
qualities in question are mechanical and inherent or non-mechanical and non-
inherent. Among the secondary, non-mechanical, and non-inherent properties,
the sensible properties are regarded as dispositional and relational in that they
are dependent upon the presence of a perceiver, while the chemical properties
are regarded as dispositional and relational for different reasons. I will discuss
these reasons in Section 4.3 of this chapter. At this point, however, I will present
the arguments for the dispositionality and relationality of sensible qualities and
what these arguments entail for reductionism.
4.2 Sensible Properties as Dispositional and Relational
The most detailed account of sensible properties as dispositional and rela-

tional is found in Peter Anstey’s insightful study of Boyle’s chemical philos-
ophy, titled The Philosophy of Robert Boyle. However, Anstey does not extend
his very well developed and tightly argued discussion to chemical qualities.
I propose to do precisely this in this chapter, that is, extend Anstey’s arguments
for the dispositionality and relationality of sensible qualities to chemical quali-
ties. Following this, I will propose that, for Boyle, chemical qualities are “novel”
properties that emerge from and supervene upon the relation between chemical
substances and their microstructures. The discussion in this chapter will then
serve to buttress the discussion in the final chapter regarding the complex mere-
ology of chymical atoms.
As he begins his thorough and detailed discussion of the dispositionality and
relationality of sensible properties, Peter Anstey points out that the term “dispo-
sitional” can be understood in two different manners. It can refer to the way in
which the parts of “x” are disposed or arranged to form “x,” or it can refer to the
disposition or inclination of “x” as a whole to affect or be affected by other bodies
in particular ways. Anstey clarifies that, for Boyle, these two distinct meanings
are related since the power of “x” to affect or be affected by other bodies is a
function of the texture or structure of the corpuscular concretions of “x.” This is
the reason why, as previously stated, arguing for the dispositionality of proper-
ties may not suffice to establish that Boyle held a non-reductionist view of such
properties.
Before specifically addressing sensible qualities, Anstey acknowledges that
Boyle considers all non-mechanical qualities, including chemical qualities, to
be dispositional and relational. Anstey explains that, for Boyle, “most if not all
non-mechanical qualities are powers. If this is so, they are powers to bring about
certain effects in other bodies and, in the case of the sensibles, in percipients.”6
Briefly addressing chemical qualities, Anstey states that “it is not unusual to find
[Boyle] saying such things as quicksilver ‘has a quality or power [. . .] to dissolve
gold and silver, and a capacity or disposition to be dissolved by aqua fortis’ [. . .] it
is in virtue of the ‘disposition’ [i.e., texture] of its parts that snow has the ‘disposi-
tion’ [i.e., power] to reflect light.”7 After stating this, however, Anstey gives no fur-
ther elaboration regarding chemical qualities and turns his full attention to the
dispositionality and relationality of sensible qualities such as color, sound, taste,
and odor. One of Anstey’s concerns is to use the argument for dispositionality
and relationality to dispel the view that, for Boyle, sensible qualities are entirely
reducible to the mechanical affections of minima naturalia. Although Anstey
is correct in this claim, a reductionist could still argue that dispositional quali-
ties are reducible to texture, which is itself a mechanical property, albeit of cor-
puscular concretions. This is the reason why I believe that the argument for
emergence must also be made, in order both to sustain and to strengthen the
non-reductionist interpretation of Boyle’s chemical ontology.
Anstey begins his discussion by first arguing against Peter Alexander’s re-
ductionist interpretation of Boyle. Alexander bases his interpretation on an
6 Anstey, The Philosophy of Robert Boyle, 87.

7 Ibid. Anstey is here citing from Boyle’s Cosmical Qualities, in The Works of the Honourable Robert
Boyle, 306; italics added by Anstey.
often-cited passage from The Origin of Forms and Qualities.8 The passage reads
as follows:
We may consider then, that when Tubal-Cain, or whoever else were the Smith
that Invented Locks and Keyes, had made his first Lock [. . .] That was onely a
Piece of Iron, contriv’d into such a Shape, and when afterwards he made a Key
to that Lock, That also in it self Consider’d, was nothing but a Piece of Iron of
such a Determinate Figure; but in Regard that these two Pieces of Iron might
now be Applied to one another after a Certain manner, and that there was a
Congruitie betwixt the Wards of the Lock and those of the Key, the Lock and
the Key did each of them now Obtain a new Capacity and it became a Main part
of the Notion and Description of a Lock, that it was made to Lock or Unlok by
that other Piece of Iron we call a Key, that it was Fitted to Open and Shut the
Lock, and yet by these new Attributes there was not added any Real or Physical
Entity, either to the Lock, or to the Key, each of them remaining indeed nothing,
but the same Piece of Iron, just so Shap’d as it was before.9
In this passage, Boyle analogizes the relationship between sensible qualities

and perceivers to the relationship between a lock and key. In this passage, Boyle
states that the lock and the key each have the disposition to obtain a new capacity
that they did not previously have, when one object is brought together with the
other. When these two objects are brought together, the key obtains the capacity
of opening the lock and the lock obtains the capacity of being opened by the
key. These dispositions have nothing to do with any peculiar power residing in
each of these objects independently but are due, instead, to the respective shapes
of the objects in question. In the same way, sensible qualities are not ontologi-
cally distinct entities in material bodies but simply endow material bodies with
the power to affect other bodies in particular ways, when they are brought to-
gether with those bodies under the right conditions.10 Alexander reads the pas-
sage as implying that the respective capacities acquired by the lock and key are
nothing “over and above” their distinctive shapes and, thus, are entirely reducible
to these shapes. Although these capacities are not manifest until the lock and key
are brought together, he considers it wrong “to claim that this is any ontological
addition.”11
Anstey not only disagrees with Alexander’s interpretation of this pas-
sage, but he also wonders why Alexander does not cite other passages that
8 For Peter Alexander’s discussion see his Ideas, Qualities, and Corpuscles.
10 See Ibid., 310.
suggest a non-reductionist interpretation of sensible qualities. For example,

in the An Introduction to the History of Particular Qualities, Boyle says the fol-
lowing: “Especially considering [. . .] that the qualities commonly called sen-
sible, and many others too, being according to our opinion but relative attributes,
one of these now mentioned alterations, though but mechanical, may endow
the body it happens to with new relations both to the organs of sense, and also
to some other bodies, and consequently may endow it with additional qualities
[emphasis mine].”12 As is clear from this passage, Boyle does believe that there is
ontological addition when the dispositional qualities of one body affect another
body that is disposed to be thusly affected. In other words, the affected body will
be endowed with additional qualities that it did not previously possess. For ex-
ample, a body is said to have a particular sensible quality, say redness, when that
body is disposed to affect the visual organs of a perceiver in such a way that,
under the right conditions, the perceiver will experience the sensation of red.
The presence of an idea of redness in the mind of the perceiver is an ontological
addition, since this idea was not present prior to the interaction with the red
object.13 Clearly, for the perception of color to occur, the perceiver must be dis-
posed toward such a perception. Thus, if the material object is in the presence of
a specific organism whose sense organs are not disposed to perceive color, color
will not manifest in the material body with relation to that specific organism. So,
color is both a dispositional and a relational quality, according to Boyle.
Boyle’s belief that all sensible qualities are both dispositional and relational is
grounded in his mechanistic corpuscularian ontology. To explain this, Anstey
refers to the hierarchy of qualities, discussed in the first section of this chapter:
First, there are mechanical properties of matter, then there are the relations
between bodies and finally there are the sensible (and other non-mechanical)
qualities or powers. Now, Boyle constantly says things to the effect that the sen-
sible qualities or powers are “derived from” or “deduced from” the mechanical
affections. Thus, the mechanical affections seem to have some sort of ontolog-
ical priority over the sensibles. Further, the sensible qualities or powers seem
somehow to be dependent upon or identical to relations. So, relations seem to
have some ontological priority over the sensibles. Finally, since it appears that
relations cannot exist in the absence of relata, their relata seem to have some
sort of ontological priority over the relations.14
12 Boyle, An Introduction to the History of Particular Qualities, 115.

13 As Anstey convincingly argues, Boyle accepts a dualist and representational view of percep-
tion, similar to that of Descartes and other 17th-century philosophers. This is the reason why, in
explaining the perception of redness, I have used language consistent with the dualist and representa-
tional theory.
For Boyle, then, the ontological base of qualities consists of the mechanical
affections of fundamental particles. When such particles are structurally ar-
ranged to form corpuscular concretions, they endow the material body with
the disposition to affect and/or to be affected by other material bodies in spe-
cific ways. Thus, when the material body in question (agent) comes into relation
with another body that is appropriately disposed (patient), the sensible qualities
of either one or both material bodies will be altered according to the respective
dispositions of these material bodies.
I agree with Anstey that this account accurately reflects Boyle’s view on the
origin of qualities. However, this account does not settle the question of re-
ductionism/non-reductionism, because it happens to be compatible with both
interpretations of non-mechanical qualities. To settle the question in favor ei-
ther of reductionism or of non-reductionism, Anstey must answer the following
queries: “Are sensible qualities or powers either reducible to or distinct from rela-
tions? And are relations either reducible to or distinct from the mechanical prop-
erties of their terms?”15 To answer these questions, one must examine whether
Boyle regards non-mechanical properties as ontologically distinct from their
mechanistic base. It is in this regard that both Anstey and I would disagree with
Alexander’s reductionist reading, according to which there is no distinctness be-
tween the non-mechanical properties and their mechanistic base. Anstey, in fact,
presents a very convincing argument in favor of a distinction, and I will now ex-
amine his argument in detail.
Anstey begins by listing four logical possibilities regarding the ontolog-
ical status of relations and of sensible qualities or powers, where R represents
the reduction relation and D represents the distinctness relation. These four
possibilities are:
1. R (Quality, Relation) and R (Relation, Relata)

2. R (Quality, Relation) and D (Relation, Relata)
3. D (Quality, Relation) and R (Relation, Relata)
4. D (Quality, Relation) and D (Relation, Relata)16
Keeping in mind that we are operating within a Boylean ontological frame-

work, there are two ways to interpret this list of logical possibilities. The first way
is to read the list as describing the various possibilities of reduction and distinct-
ness with respect to a given material body and its dispositional qualities. Thus,
15 Ibid.
16 Ibid., 90. In my rendition of this list of possibilities, I differ from Anstey by using the term
“quality” instead of “power,” though this is simply a preference and makes no difference to the argu-
ment itself, since both Boyle and Anstey use these terms interchangeably.
to the extent that the disposition to manifest a particular quality is a function of

the microstructure or texture of corpuscular concretions and that this micro-
structure represents a particular relation between fundamental particles, the
terms used must be read in the following manner: “Relata” refers to the funda-
mental corpuscles endowed only with mechanical affections, “relation” refers to
the microstructure or texture in which the fundamental corpuscles are arranged
to form corpuscular concretions, and “quality” refers to the particular non-
mechanical quality that a given material body is disposed to manifest.
Read in this way, possibility (1) tells us that there is nothing distinct from the
mechanical affections at the ontological base and, thus, that qualities are reduc-
ible to relations and relations are reducible to relata. Since reduction is a tran-
sitive property, this would make qualities reducible to relata, in this case to the
mechanical affections of fundamental particles. This is the reductionist account
favored by Alexander, since it “collapses powers into relations which in turn col-
lapse into their relata.”17 Possibility (2) tells us that since relations and relata are
distinct, there is a “two-tiered ontology,” to use Anstey’s term, in which qualities
are reducible to relations but relations are distinct and not reducible to relata.
Possibility (3) tells us that, since qualities are distinct from relations, there is a
“two-tiered ontology” in which qualities are not reducible to relations but rela-
tions are reducible to their relata. Finally, possibility (4) tells us that, since qual-
ities are distinct from relations and relations are distinct from relata, there is a
“three-tiered ontology” in which qualities are not reducible to relations and re-
lations are not reducible to relata. The only possibility that is fully reductionist
is (1) and the only possibility that is fully non-reductionist is (4), while (2) and
(3) each involve some combination of reduction and non-reduction.
A second way to read this list is as describing the various logical possibili-
ties of reduction and distinctness with respect to one body or material substance
(agent) as it affects another body or material substance (patient). Read this way,
the list deals with a higher ontological level of relata and relations and the terms
must be read in the following way: “Relata” refers to the agent and the patient,
“relation” refers to the contact between them, and “quality” refers to the manifest
properties of the agent that affects the patient. After arguing that Boyle endorses
neither possibilities (1), (2), or (3), Anstey defends possibility (4) as being the
most representative of Boyle’s position, and he employs the notion of qualities
as relational as a key element of his argument.18 Read as describing this higher
ontological and relational level, possibility (4) tells us that the relation itself is
17 Ibid.
18 Ibid., 90–108. Explaining Anstey’s arguments against possibilities (1), (2), and (3) would take
me far beyond the scope of this chapter. However, once we accept the relationality of qualities in
Boyle, this suffices to lend the bulk of the support to Anstey’s argument in favor of possibility (4),
thereby making it unnecessary to discuss the arguments against the other possibilities.
“over and above the agent and the patient [which serve] as grounds of the rela-
tion [. . .] it is the distinctness relation (D) between the power and the relation that
resists reduction [emphasis mine],”19 even if one were to concede that the relation
is multilaterally reducible to its relata. With regard to this second reading, Anstey
points out that the relationality of qualities is not “simply dyadic [. . .] but often
involves many more relata than simply the agent and patient.”20
Following this discussion, Anstey cites passages from several works as textual
evidence that Boyle supports this view of relationality. These include The Origin
of Forms and Qualities, the History of Particular Qualities, and the Cosmical
Qualities. In these passages, Boyle stresses repeatedly the point that bodies
should not be considered in isolation but always as parts of the universe, if one
is to understand the origin of sensible qualities. He claims that “every distinct
portion of matter, whether it be a corpuscle of a primary concretion, or a body of
the first, or of any other order of mixts, is to be considered not as if it were placed
in vacuo, nor as if it had relation only to the neighbouring bodies, but as being
placed in the universe, constituted as it is, amongst an innumerable company of
other bodies.”21 He adds that “in reference to the Production of Qualities, a Body
is not to be considered barely in it selfe, but as ’tis placed in, and is a portion of
the Universe.”22
As Anstey points out, however, although this may make Boyle one of the first
proponents of the relational conception of qualities, the question of the ontolog-
ical status of relations is still open. Did Boyle consider relations to be reducible
to relata? We do know that Boyle considered that all the relevant relata must be
present for the relational quality to manifest. He states, “so most of those powers
and attributes, that we call qualities in bodies, depend so much upon the struc-
ture or constitution of other bodies, that are disposed or indisposed to be acted
on by them, that if there were no such objects in the world, those qualities in the
bodies, that are said to be endowed with them, would be but aptitudes to work
such effects.”23
Regarding sensible qualities, the presence of a perceiver is required in order
for the disposition of an object to be manifested: “if there were no Sensitive
Beings, those Bodies that are now the Objects of our Senses, would be but
dispositively, if I may so speak, endow’d with Colours, Tasts, and the like; and
actually but onely with those more Catholick Affections of Bodies, Figure,
Motion, Texture, &c.”24 Thus, the power of a material body to cause a percipient
19 Ibid., 105.
20 Ibid., 108.
21 Boyle, History of Particular Qualities, 298.
22 Ibid.
23 Boyle, Of Mens’ Great Ignorance of the Uses of Natural Things, 479–480.
to experience a particular sensible quality is dependent upon the presence of a

percipient, implying that “the sensible qualities are explicated by a relation to
our senses.”25 Therefore, it is not only the existence of the relata, but also their
presence in some proximity and in some type of relation to one another, that
are required for the manifestation of sensible qualities. As Frederic O’Toole
stresses, for Boyle “the possession of certain primary qualities, although a nec-
essary condition for the possession of a power or capacity, is not, by itself, a
sufficient condition for the possession of a power or capacity.”26 Thus, regarding
sensible qualities, the presence of a perceiver appropriately disposed to be af-
fected by the powers or qualities of a given object is necessary for that power or
quality to be manifested.
As mentioned earlier in this chapter, however, Boyle is not simply con-
cerned with relations between objects and perceivers and with the qualities that
manifest when such relations are actualized. As a chemist, Boyle is primarily
concerned with relations between substances and with how the properties of
such substances affect and are affected by one another. Thus, the question of
whether chemical properties are dispositional and relational is an extremely
relevant one.
Although Boyle does regard chemical qualities as ontologically dependent
upon mechanical properties, he does not believe that chemical reactions occur at
the fundamental level of particles but that they take place between corpuscular
concretions that are differentiated by virtue of the chemical properties to which
they give rise.27 Additionally, Boyle’s many remarks regarding the producibility
of chemical properties in compounds from ingredients that do not bear the
properties in question suggests that Boyle possibly endorses a notion of chemical
qualities as emergent properties, ontologically dependent upon but not deduc-
ible from or reducible to the mechanistic affections of shape, size, and mobility.
Because Boyle refrains “from establishing a direct relationship between a given
quality and a set of mechanical properties of the simplest corpuscles”28 and is
openly critical of Epicureans and Cartesians who “pretend to explicate every par-
ticular Phaenomenon by deducing it from the Mechanicall affections of Atomes
or insensible particles,”29 it is very likely that he also considered chemical proper-
ties to be dispositional and relational.

27 See Newman, “The Alchemical Sources of Robert Boyle’s Corpuscular Philosophy,” 567–585.
29 Boyle, An Essay of Various Degrees or Kinds of the Knowledge of Natural Things, Vol. 8, folio 166r.
4.3 Chemical Properties as Dispositional and Relational
As already stressed, the arguments that Anstey presents in favor of the

dispositionality and relationality of sensible qualities can be used to argue for the
dispositionality and relationality of chemical properties, as well as for the claim
that chemical properties are not entirely reducible to the mechanical affections
of primary particles or to texture tout court. As I will argue in this section, chem-
ical qualities are just as relational as sensible qualities, since they are actualized
only when two or more substances with the appropriate dispositions find them-
selves in each other’s presence and under the right conditions. For example, the
water solubility of sugar is a dispositional property because it is only manifested
when sugar is actually placed in unsaturated water.
As is being stressed throughout this book, although Boyle endorses a mecha-
nistic conception of fundamental matter, he is also a realist about chemical and
other non-mechanical properties and is strongly committed to the development
of chemical explanations for chemical reactions and processes. Therefore, he is
not satisfied with the idea of reducing all qualities and processes to the mechan-
ical affections of fundamental particles. As was discussed in Chapter 2, strictly
mechanistic explanations simply could not satisfactorily account for chemical
phenomena, and early modern chemists, regardless of their commitment or lack
of commitment to the mechanistic hypothesis, recognized the limitations of a
strictly mechanistic philosophy. Thus, Boyle concurs with Paracelsians and other
critics of strict mechanicism that chemical qualities and operations cannot be
explained simply by invoking the mechanical properties of shape, size, and mo-
bility. Therefore, although he was critical of the Paracelsian doctrine of principles
and qualities, Boyle’s criticism “did not entail that all chemical properties were
reducible to [. . .] mechanical attributes [. . .] [He presented] the idea that a quality
had relative character, namely, that it was generated from the constant interac-
tion of different corpuscles, which themselves might not bear the quality in ques-
tion [and] he himself developed new and more sophisticated ways of detecting
the chemical qualities of bodies.”30 For example, in his essay Of the Producibleness
of Chymicall Principles, Boyle examines the various chemical properties of salts.
Regarding solubility, he states that “a disposition to be dissoluble in this or that
liquor may be acquired by mixture, and the new texture of parts.”31
Louis Lémery et Etienne-François Geoffroy,” 4: “Boyle n’invoque pas tant le mouvement et la forme
des corpuscules que les qualités chimiques des diverses substances qui on la propriété de dissoudre,
de précipiter ou de fixer. Bref, pour Boyle, l’explication chimique garde son autonomie et ne peut être
explicitée en tant que telle.”
31 Boyle, Of the Producibleness of Chymicall Principles, 35.
One way in which non- mechanical secondary qualities, such as chem-

ical properties, are dispositional is that they result from the way the primary
particles are structured to form chymical atoms. If the disposition of particles
is altered in any way, different chymical atoms will be produced and this will
result in different chemical properties. Experiments involving analysis, syn-
thesis, and redintegration all involve the restructuring of chymical atoms, that
is, altering the dispositions of the minima naturalia. The redintegration of po-
tassium nitrate, discussed in Chapter 2, can be examined from this perspective.
The experiment illustrates, among other things, that the structural disposition or
texture of concretions accounts for the chemical properties of a body and when
these textures are altered either by breaking down the body into its component
substances or by compounding the body with another body, the interaction of
the different chymical atoms will produce different chemical qualities. Thus, red-
integration requires a reassociation of the fixed and volatile parts according to
their original dispositions.
Another way in which Boyle considers non-mechanical secondary qualities
to be dispositional is that he regards them as powers to affect other bodies in
specific ways. As Clericuzio explains, “Boyle maintained that chemical qualities
depended [. . .] on the way in which the corpuscles that composed a given body
were disposed to act upon, or to be acted on by, those of other bodies [. . .] [and
that they] emerged from the constant interactions of corpuscles passing from
one body to the other [. . .] [thus] he denied that they directly originated from
the mechanical properties of their primary particles.”32 In an important passage
from The Origin of Forms and Qualities, Boyle explains that neither sensible nor
chemical properties have any ontological being of their own apart from the mi-
crostructure of the material body in which they subsist. However, he also adds
that they cannot be said to subsist as such in the microstructure of the material
body. Instead, the specific microstructure or texture endows the material body
with the ability or disposition to affect or to be affected by other material bodies
in specific ways. Boyle states:
Whereas one Body doth often seem to produce in another divers such Qualities,
as we call Sensible, which Qualities therefore seem not to need any reference
to our Senses, I consider, that when one Inanimate Body works upon another,
there is nothing really produc’d by the Agent in the Patient, save some Local
Motion of its Parts, or some Change of Texture consequent upon that Motion;
and so, if the Patient come to have any sensible Quality, that it had not before,
it acquires it upon the same account, upon which other Bodies have it, and it
32 Clericuzio, “A Redefinition of Boyle’s Chemistry and Corpuscular Philosophy,” 588.

is but a consequent to this Mechanical Change of Texture, that by means of its

Effects upon our Organs of Sense, we are induc’d to attribute this or that sen-
sible Quality to it.33
Boyle reiterates this point later by saying that “Bodies of very differing Natures,
being put together, like the Wheels, and other peices [sic] of a Watch, and by their
connection acquiring a new Texture, and so new Qualities, may, without having
recourse to a substantial Form, compose such a new Concrete, as may as well de-
serve to have a substantial Form attributed to it, by virtue of that new Disposition
of its parts, as other Bodies that are said to be endow’d therewith.”34
The dispositions to affect and/or to be affected in specific ways are due to
the microstructures or textures of both the substances in question. However,
such dispositions are manifested when the requisite material bodies come into
contact with one another, under the requisite conditions. In this sense, the
dispositionality of properties is intimately connected to their relationality. For
example, Boyle tells us, some bodies have the disposition to be cathartic or pur-
gative only when mixed with gold. Boyle accounts for the purgative power of
the resulting mixture by appealing to the microstructures of the substances that
have been mixed, which means substance A has the disposition to become pur-
gative when mixed with gold, and that gold has the disposition to make sub-
stance A purgative.
Applying the same analysis that Anstey gives for dispositional sensible prop-
erties, we can say that there is a “three-tiered ontology” here, in which the pur-
gative quality is distinct from the relation between substance A and gold, and
in which the relation between substance A and gold is itself distinct from the
individual relata. Clearly, there is no reducibility here, because neither substance
A nor gold has the purgative quality apart from the relation in which the two
substances are mixed. Furthermore, the mixed purgative substance has the dis-
position to purge the digestive system of an organism and the digestive system
has the disposition to be purged by the mixture. The phenomenon of purging,
however, is manifested only when the mixed substance and the digestive system
of the organism come into contact with one another, under the right conditions.
Thus, the same “three-tiered ontology” analysis can be applied to the relation
between the mixed purgative substance and the digestive system being purged.
One problem with the kind of reductionist interpretation favored by
Alexander and criticized by Anstey is that it would entail collapsing the distinc-
tion that Boyle clearly wants to affirm between primary mechanical qualities and
secondary non-mechanical qualities. We can illustrate this with an example that

34 Ibid., 355.
Boyle discusses in The Origin of Forms and Qualities. This is the example of gold,
which has a disposition to be soluble in aqua regia, a mixture of hydrochloric
acid (HCl) and nitric acid (HNO3), and a disposition to be insoluble in aqua
fortis, or concentrated nitric acid. Regarding these dispositions, Boyle states the
following:
I do not see, why we may not conceive, That as to those Qualities, (for Instance)
which we call Sensible, though by virtue of a certain Congruity or Incongruity
in point of Figure or Texture, (or other Mechanical Attributes,) to our Sensories,
the Portions of Matter they Modifie are enabled to produce various Effects,
upon whose account we make Bodies to be Endow’d with Qualities; yet They
are not in the Bodies that are Endow’d with them any Real or Distinct Entities,
or differing from the Matter it self. Thus though the modern Gold-Smiths and
Refiners reckon amongst the most distinguishing Qualities of Gold [. . .] that it
is easily dissoluble in Aqua Regis, and that Aqua Fortis will not work upon it; yet
these Attributes are not in the Gold any thing distinct from its peculiar Texture
[. . .] There are some Bodies not Cathartick, nor Sudorifick, with some of which
Gold being joyn’d acquires a Purgative Vertue, and with others a power to pro-
cure Sweat [. . .] Nature her self doth [. . .] produce so many things, that have
new Relations unto others; And Art, especially assisted by Chymistry, may, by
variously dissipating Natural Bodies, or Compounding either them, or their
Constituent Parts with one another, make such an Innumerable Company of
new Productions, that will each of them have new operations, either immedi-
ately upon our Sensories, or upon other Bodies, whose changes we are able to
perceive.35
As O’Toole points out, “to identify the powers and capacities of corporeal
objects with certain of their primary qualities is to collapse the distinction be-
tween what an object will do or will have done to it and what is in itself. That is, it is
to assert that what a thing can do or have done to it is nothing distinct from what
it is in itself.”36 This would entail collapsing the distinction between objects and
dispositions, and an absurd conclusion would ensue from this. Since the texture
constitutes what an object is in itself, collapsing the distinction between an object
and its dispositions would mean collapsing the distinction between its texture
and its dispositions. Since identity is a reflexive and transitive relation, if we in-
terpret Boyle’s words in reductive terms and argue that qualities collapse with
texture, this would lead to the conclusion that:
35 Ibid., 310–311.
1. Texture of gold (T) = Disposition to be soluble (S)

2. Texture of gold (T) = Disposition to be insoluble (−S)
/ ∴ 3. T = S and T = −S
The conclusion that texture is equal to two opposite dispositions is clearly absurd.
Further, if we attempt to avoid this conclusion by saying that there is no distinc-
tion between disposition to be soluble and the disposition to be insoluble (S = −S),
then we fall into another absurdity. These problematic conclusions clearly result
from the fact that one is ignoring the relationality of these dispositions and that
new relations lead substances to acquire novel dispositional properties. Once
relationality is introduced, the contradiction is avoided thus:
1. Texture of gold (T) = Disposition to be soluble in relation to aqua regia (Sar)

2. Texture of gold (T) = Disposition to be insoluble in relation to aqua
fortis (−Saf )
/ ∴ 3. T = Sar and T = −Saf
Gold has the disposition to be soluble in relation to aqua regia, and it has the
disposition to be insoluble in relation to aqua fortis. Clearly, it is the texture of
gold that brings about these dispositions. However, the texture of gold must be
brought into relation with the texture of aqua regia in order to be disposed to
solubility. This means that the texture of gold reacts with that of aqua regia in
such a manner that the gold is dissolved. As well, the texture of gold must be
brought into relation to aqua fortis in order to be disposed to insolubility. Again,
this means that, when the texture of gold is brought into such a relation, no re-
action occurs and the gold is not dissolved. Though Boyle might explain this in
mechanical terms by speculating that the texture of the chymical atoms fits dif-
ferently with that of aqua regia than it does with that of aqua fortis, it is still the
case that neither disposition exists apart from the relation that permits its man-
ifestation, since it is within the world as a whole that a body has its secondary
qualities.
Besides leading to absurd conclusions, however, collapsing the distinction be-
tween objects and dispositions would, in turn, also mean collapsing the distinc-
tion between inherent qualities and non-inherent qualities, which is something
that Boyle does not wish to do. If he wished to collapse all qualities thusly, Boyle
would not have bothered drawing distinctions between different types of quali-
ties in the first place. Instead, as O’Toole suggests, Boyle identifies “certain of the
primary qualities of an object [i.e., texture] as the properties in virtue of which
the object can do or have done to it what it does.”37 Thus, the texture of gold is
37 Ibid.
the property in virtue of which gold is disposed to be soluble in aqua regia and
to be insoluble in aqua fortis. Put differently, gold’s disposition to be soluble in
aqua regia and insoluble in aqua fortis is not ontologically the same thing as the
texture of gold but, rather, is a function of this texture. As I will argue later in this
chapter, the disposition to be soluble or insoluble should also be considered as an
emergent property that supervenes upon the texture.
It would also follow from the reductionist collapse of non-mechanical prop-
erties with texture that all non-mechanical properties, including chemical prop-
erties, would simply be epiphenomal and would have no causal power. However,
chemical properties obviously have causal power and must, therefore, not be
simply epiphenomenal. In fact, Boyle stresses the validity and autonomy of
chemistry as natural philosophy precisely because chemical properties not only
have causal power but also have a greater heuristic and explanatory power than
do the mechanical affections of minima naturalia. This is precisely why Boyle
invokes non-mechanical properties almost exclusively in his chemical, hydro-
static, and pneumatic experimental accounts.
As already stated, Boyle stresses that chemical dispositions are actualized
when substances are brought into the appropriate relations with one another. He
illustrates this with an experiment in which sulfur and salt of tartar are mixed
together, thereby acquiring the property of being soluble in spirit of wine, a
property that neither substance manifests prior to the mixture, although Boyle
attributes the acquisition of solubility to an alteration in the structure of the com-
pound corpuscles or chymical atoms of both the sulfur and the salt of tartar that
results from these substances being mixed together. This example shows that sol-
ubility is indeed a novel property that would not be manifested in the absence of
the appropriate chemical relations.
This establishes that, for Boyle, chemical properties such as solubility, fixed-
ness, and others are both dispositional and relational, since they arise from the
relation between appropriately disposed chemical substances. In other words,
the disposition to manifest a specific quality or property is a function of the re-
lation of substance A to substance B. In the absence of this relation, the property
does not manifest. However, we may be tempted at this point to accuse Boyle of
the same circularity of which Aristotle is guilty when discussing dispositional
properties. In the context of Aristotelian science, a body A is said to affect a body
B in specific ways, because A is disposed to affect B and B is disposed to be af-
fected by A precisely in those ways. Such circularity made Aristotelian science
completely ineffective for explaining causal relationships. It might be tempting
to say that Boylean explanations are as circular and vacuous as the Aristotelian
explanations. However, I would argue that Boyle is not guilty of circularity be-
cause, although Aristotle offers no further explanation regarding why bodies are
disposed to affect and be affected by other bodies in specific ways, Boyle does
offer such an explanation beyond merely saying that the dispositions are there.
His explanation, as has already been stated, is that the various dispositions of
bodies to manifest qualities and to either affect or be affected by other bodies
derive or emerge from the microstructure or texture of their corpuscular
concretions and from the way those structures are altered or are affected when
those bodies come into contact. I grant that it is precisely the causal powers of
the corpuscular concretions that are unknown or perhaps even unintelligible.
However, Boyle’s postulation of unobservable theoretical entities and causal
powers in order to explain observed dispositional properties is, in principle, no
more vacuous than any other scientific postulation of theoretical entities to ex-
plain observed phenomena.
Returning to the discussion of relationality, there is another and more complex
way in which chemical properties can be said to be relational. The relationality
of chemical properties is not only a feature of the relation between chemical
substances but is also a feature of the relation between the chemical substances,
the experimental environment in which those substances are placed in relation
with each other, and the actions of the chemist in relation to the goals of the ex-
periment. To examine this aspect of relationality, let us consider Steven Shapin’s
discussion of the generation of “matters-of-fact” in the context of Boyle’s exper-
imental programme. Though Shapin’s discussion focuses on the production of
pneumatic “matters-of-fact” via experiments with the air-pump, the same anal-
ysis can be applied to the generation of chemical “matters-of-fact.”
As Shapin insightfully notes, Boyle and his colleagues at the Royal Society
were profoundly aware that the phenomena and processes that they witnessed
via their experimental work were very much “artifacts” of the experiments. To el-
evate these “artifacts” to the status of “matters-of-fact,” experimenters needed to
use various strategies that would present their experimental results in a manner
that decontextualized them from the experimental conditions within which
they were generated. Specifically focusing on Boyle’s experiments with the air-
pump, Shapin explains that “Boyle’s experimental programme had as its end-
product the generation of indisputable matters of fact [. . .] in the setting of early
Restoration England there was no one solution to the problem of knowledge that
commanded universal assent [. . .] Boyle sought to secure universal assent by way
of the experimental matter of fact [. . .] [However,] Boyle’s matters of fact were
machine-made.”38
As Shapin points out, however, “what makes a fact different from an artifact is
that the former is not perceived to be man-made [. . .] a matter of fact is taken to
be the very mirror of nature. To identify the role of human agency in the making
38 Shapin, “Pump and Circumstance: Robert Boyle’s Literary Technology,” 481–

520. See
also: Shapin and Schaffer, Leviathan and the Air-Pump.
of an item of knowledge is to identify the possibility of its being otherwise. To

shift agency on to natural reality is to stipulate the grounds for universal as-
sent.”39 Shapin then goes on to discuss the various material, literary, and social
technologies employed by Boyle and his assistants to ensure that the man-made
artifacts of the experiment would be perceived as natural matters-of-fact. A dis-
cussion of these various technologies would take us beyond the scope of this
chapter. However, one can extend Shapin’s general analysis of Boyle’s pneumatic
experiments to the generation of matters-of-fact in the chemical laboratory.
Doing this will allow us to understand better the complex layers of relationality
of chemical properties.
Applying Shapin’s analysis to the generation of chemical “matters-of-fact,” the
solubility of the mixture of sulfur and salt of tartar in spirit of wine is an “artifact”
of the actions of the chemist and of the experimental conditions. It is not likely
that such conditions would ever be reproduced in nature, as it is highly improb-
able that such substances would ever be mixed outside of the context of a scien-
tific experiment in which the sole purpose is precisely to render them soluble in
spirit of wine. Thus, the solubility of this mixture in spirit of wine is an artifact
of the experiment and is, therefore, relational in an even more complex manner
than described so far. The quality of being soluble, in this experiment, results not
only from the relation in which the complex corpuscles find themselves but also
from the relation with the actions of the chemist and with the very specific exper-
imental conditions. It is only when the results of this experiment are presented
as matters-of-fact that the chemist’s actions and the experimental conditions in
which these artifacts were generated are “written out” of the account and the
mixture’s solubility is presented as though it were a simple “matter-of-fact” dis-
covered by Boyle.
One of the most interesting concepts used to discuss the relationality of prop-
erties in the contemporary philosophy of chemistry is the concept of affordance.40
The concept of affordance was introduced in contemporary scholarship by psy-
chologist James J. Gibson, who defines it as what an environment offers to an
animal as a possibility for action.41 Affordances are, therefore, dispositional and
relational and characterize the suitability of an environment for a particular ob-
server, although Gibson emphasizes that affordances are observer-independent
and not subjective. Gibson theorizes that human beings and other animals alter
39 Shapin, “Pump and Circumstance,” 507–508.

40 See, for example, Harré, “New Tools for Philosophy of Chemistry”; Earley, “Why There Is No
Salt in the Sea”; Llored, “Emergence and Quantum Chemistry,” “Whole-Parts Strategies in Quantum
Chemistry,” and “Relations, caractérisations, et ‘affordances’ ”; Bensaude-Vincent, “Philosophy of
Chemistry or Philosophy with Chemistry?”; and van Brakel, “Philosophy of Science and Philosophy
of Chemistry.”
41 Gibson, “The Theory of Affordances.”
and modify their environments to increase the action possibilities that the en-
vironment affords to them. Rom Harré, who has also written extensively in the
philosophy of psychology, appropriates the notion of affordance as a conceptual
tool for the philosophy of chemistry. According to Harré, “adopting the concept
of ‘affordance’ to analyze the nature of chemical studies, it becomes clear that
chemical ‘facts’ are attributes not of an independent world revealed by the use of
an apparatus, but are dispositional properties of a hybrid entity—an indissoluble
union of apparatus, experimenter, and world . . . when the apparatus is being put
to use by a chemist it is related to the world in such a way that the phenomena it
displays can exist only as the apparatus in integrated materially into the world.”42
What is interesting about this, however, is that the notion of affordance as a
useful conceptual tool for chemistry does not seem all that new. In fact, Boyle
avails himself quite extensively of the notion of “affordance” in his chemical
writings, as far back as The Sceptical Chymist. Since the concept of affordance is
inherently relational and dispositional, it can be argued that, even in his earlier
works, Boyle appeals to a relational and dispositional notion of chemical qual-
ities. In fact, in The Sceptical Chymist, the relational and dispositional concept
of affordance is used precisely to critique the notion of substantial forms and
spagyric principles. For example, as part of this critique of the Paracelsian no-
tion of the Alkahest, as the substance that is able to “resolve all mixt Bodies into
other Principles than the fire,”43 Boyle states “although we should acquiesce in
that resolution which is made by fire, we find not that all mixt bodies are thereby
divided into the same number of Elements and Principles; some Concretes
affording more of them than others do; Nay and sometimes this or that Body
affording a greater number of Differing substances by one way of management,
than the same yields by another [emphasis mine].”44 Elsewhere in the same
work, Boyle states that “sometimes by corrosive Liquors . . . and sometimes by
the operation of common Sulphur (especially well open’d and associated with
fit Salts) Silver has afforded some grains of very pure Gold [emphasis mine].”45
In another of his chemical writings, while describing an experiment with salt-
peter, Boyle states, “we dissolv’d in fair water as much fix’d Nitre as we could,
and filtrating the Solution through Cap-paper we satiated with Spirit of Nitre,
after the manner above describ’d, and then setting it to evaporate very slowly, and
afterwards suffering it to cool, we obtain’d within some hours after the first mix-
ture of the Liquors, store of fine little Crystalls of Petre, which shot in the Liquor;
the remaining part of which being evaporated afforded more of them.”46 There
42 Harré, “New Tools for Philosophy of Chemistry,” 79.
44 Ibid.
45 Ibid., 58.
46 Boyle, “Some Specimens of an Attempt to Make Chymical Experiments Useful to Illustrate the
Notion of the Corpuscular Philosophy,” 97.

are countless more example in The Sceptical Chymist and other works that are too
many to cite here. The fact that Boyle relies so often on the concept of affordance,
which many contemporary philosophers of chemistry consider to be an invalu-
able tool for conceptualizing relationality and dispositionality, is evidence that
his work is still relevant today for philosophy of chemistry.
It is, indeed, important also to point out that Boyle is one of the first natural
philosophers to recognize the relationality of chemical properties. However, be-
cause many readers of Boyle have limited themselves to The Origin of Forms and
Qualities, they conclude mistakenly that the mechanistic lock-and-key analogy
provided by Boyle in this essay exhausts his account of the relationality of qualities.
However, a more accurate understanding of relationality requires examining other
texts, in which Boyle stresses the role played by the surrounding environmental
conditions in determining the dispositions of both agent and patient. I will explain
this in the next section, in which I propose that Boyle regards chemical qualities as
emergent and supervenient properties.
4.4 Chemical Properties as Emergent and Supervenient
As mentioned earlier in this chapter, to establish Boyle as a non-reductionist, one

must also establish that he regards chemical properties as being emergent, and this
is what I hope to demonstrate in this section. It should be emphasized that the is-
sues of emergence and reducibility are themselves closely related to the more ge-
neral issues of the autonomy of chemical explanations and of chemistry as a science.
The reducibility of chemical laws to physical laws would necessitate that physical
laws be universal and fundamental, and that all laws in the special sciences be non-
fundamental instantiations of these more general and universal physical laws. Given
what has been said so far regarding Boyle’s view of the nature of chymical atoms, it
seems that Boyle would side against the view that chemical laws are reducible to the
laws of mechanics and, thus, in favor of the idea that at least some chemical laws can
be considered operationally fundamental and irreducible. While the strict mecha-
nistic philosopher posits a single fundamental law, or few such laws, the emergentist
allows that there could be a great many laws that are operative in nature.
A strict mechanistic approach to chemistry would conclude that just a few
laws suffice to determine the behavior of all chemical substances and chemical
compositions. Accordingly, it should be possible to deduce the behavior of all
chemical substances simply by turning to the few laws of mechanics that govern
fundamental particles. An emergentist, however, will favor the view that at
least some fundamental laws that describe the behavior of chemical substances
are not merely instances of the more general and fundamental laws of me-
chanics.47 With regard to this issue, I firmly believe that Boyle would side with
the emergentist position and I will base my discussion on what has already been
stated regarding his chemical ontology and on his views of qualities as disposi-
tional and relational. What I hope to show is that his notion of chemical proper-
ties allows Boyle (1) to more clearly conceptualize the result obtained from his
own chemical experiments, (2) to assert the autonomy of chemical explanations
from mechanical explanations, and, therefore, (3) to assert the legitimate status
of chemistry as an autonomous science that is operationally and epistemologi-
cally, if not ontologically, independent from mechanics.
To strengthen the argument that Boyle considers chemical properties to be
emergent, we must examine what the concept of emergence entails and whether
Boyle’s conception of chemical properties conforms to the concept of emergence.
Stated in general terms for our purposes here, the central tenets of emergentism
are the following:
1. Emergence of novel higher-level properties

2. Non-deducibility of emergent properties
3. Irreducibility of emergent properties
4. Causal efficacy of emergent properties
The concept of emergence also entails several interdependent features, which

are related to the general features described earlier. The first of these features is
novelty, that is, the component parts of a substance do not display the emergent
property of that substance. The second feature of emergence is unpredictability,
which is related to the notion of non-summative difference between the super-
venient properties and the “subvenient” properties of the submergent base. In
other words, the supervenient properties are not those of the submergent base
nor can they be additively derived from the properties of the base. The third fea-
ture of emergence is the metaphysical relation of supervenience, that is, “a rela-
tion of determination and dependence of one set of properties on another.”48
The first question for us, then, is whether Boyle regards chemical proper-
ties to be novel features of chymical atoms that are not displayed by the com-
ponent particles. In Of the Imperfection of the Chemists’ Doctrine of Qualities,
Boyle openly critiques the conception of qualities embraced by “chemists,” that
is, by Peripatetics and Paracelsian alchemists and spagyrists. Here, Boyle argues
against the idea that the properties of material bodies, including chemical
47 See Hendry, “Is there downward causation in chemistry?,” 179. I am here referring to Hendry’s
point about downward causation in chemistry to set the context for my discussion of Boyle’s position.
48 Newman, “Chemical Supervenience,” 49.
properties, can be simply deduced from the ingredients of which those bodies
consist. He states:
I shall venture to add that the way employed by the chemists, as well as the
Peripatetics, of accounting for things by the ingredients—whether elements,
principles, or other bodies—that they suppose them to consist of, will often
frustrate the naturalist’s expectation of events, which may frequently prove
differing from what he promised himself upon the consideration of the quali-
ties of each ingredient. For the ensuing notes contain divers instances, wherein
there emerges a new quality differing from, or even contrary to, any that is con-
spicuous in the ingredients: as two transparent bodies may make an opacous
mixture; a yellow body and a blue, one that is green; two malleable bodies, a
brittle one; two actually cold bodies, a hot one; two fluid bodies, a consistent
one, &c. [emphasis mine].”49
I call the reader’s attention, here, to Boyle’s explicit use of the term “emerges” in
reference to qualities or properties. We should also take note of the examples he
uses, of both sensible and chemical qualities, to illustrate the point that some-
thing new is indeed produced when two distinct substances are mixed together,
that is, a new quality is produced that was not possessed by either of the two
substances individually. What accounts for the new quality, as Boyle explains
in the following passage, is the alteration of the corpuscular concretions upon
mixing the two distinct substances:
divers qualities may be changed even in such constant bodies as metals, without
the addition of any considerable proportion of the simple ingredients to which
they [spagyrists] are wont to ascribe those qualities, provided the agent (as an
efficient rather than material cause) be able to make a great change in the me-
chanical affections of the parts whereof the metal it acts on is made up [. . .]
changes it produces in the constituent texture of it [. . .] the chemists’ salt, sul-
phur and mercury themselves are not the first and most simple principles of
bodies, but rather primary concretions of corpuscles or particles more simple
than they, as being endowed only with the first or most radical (if I may so
speak) and most catholic affections of simple bodies, namely bulk, shape, and
motion or rest, by the different conventions or coalitions of which minutest
portions of matter are made those differing concretions that chemists name
salt, sulphur, and mercury.50
49 Boyle, Of the Imperfection of the Chemists’ Doctrine of Qualities, 129.

50 Ibid., 131–132.
There is plenty of textual evidence to demonstrate that, in fact, Boyle does

consider both sensible and chemical qualities to be novel properties that are not
displayed by minima naturalia. Let me add that this view of chemical and other
non-mechanical qualities as “novel” properties has significant mereological
implications that will be discussed in more detail in the final chapter. Now, how-
ever, let us examine whether chemical properties do indeed display novelty. In
the Mechanical Origin and Production of Volatility and Fixedness, Boyle claims
that “the same material parts of a corporeal substance, which, when associated
and interwoven after a determinate manner, constituted a solid and fixed body
[. . .] may by having their texture dissolved, and by being freed from their former
implications or cohesions, become the parts of a fluid body, totally volatile.”51
Elsewhere, he states that “whatever be the number or qualities of the chemical
principles, if they be really existent in nature, it may very possibly be shewn, that
they may be made up of insensible corpuscles of determinate bulks and shapes;
and by the various coalitions and contextures of such corpuscles [. . .] the very
qualities of this, or that ingredient, flow from its peculiar texture [emphasis
mine].”52 Although Boyle clearly affirms the ontological dependence of chemical
properties on mechanical affections, he “also maintains that chemical qualities
depended more on the way in which the corpuscles that composed a body were
disposed to act on, or to be acted on by, those of other bodies [. . .] He believes
that chemical qualities emerge from the constant interaction of corpuscles by
passing from one body to the other.”53 As he states, in The Origin of Forms and
Qualities:
the Colour, Odour, Tast, and other qualities of that Body are to be deriv’d, it will
be easie for us to recollect, That such Changes cannot happen in a portion of
Matter, without so much varying the Nature of it, that we need not deride the
ancient Atomists, for attempting to deduce the Generation and Corruption of
bodies from [. . .] the transposition of their (suppos’d) Atoms [. . .] according
to us, the various manner of the Coalition of several Corpuscles into one vis-
ible Body is enough to give the a peculiar Texture, and thereby fit them to
exhibit diverse sensible Qualities, and to become a Body, sometimes of one
Denomination, and sometimes of another.54
Boyle offers further examples of the novel and emergent character of chemical
qualities when, in discussing salinity, he proposes that:
51 Boyle, The Mechanical Origin and Production of Volatility and Fixedness, 373.
52 Boyle, About the Excellency and Grounds of the Mechanical Hypothesis, 454–455.
whether we allow the Epicurean Hypothesis or the Cartesian; the first Saline
Concretions that were produc’d by Nature must be confess’d to have been made
of Atomes, or of Particles, that before their conjunction, were not Saline [. . .]
small portions of matter may be so broken into minute, and these fragments
may be so shap’d and connected, as, when they are duely associated, to com-
pose a Body capable of being dissolved in water, and of affecting the organs of
Taste.55
Like other chemical properties, the property of salinity cannot simply be

accounted for mechanistically but clearly emerges from the texture of spe-
cific corpuscular concretions. In fact, Boyle’s “aim was to reject the notion that
sensible qualities were reducible to this or that ingredient of a mixed body. He
clearly put forward the idea that a quality had relative character, namely, that it
was generated from the constant interaction of different corpuscles, which them-
selves might not bear the quality in question.”56 As has been argued extensively,
for Boyle, chemical properties are indeed novel properties that are distinct from
the mechanical affections of the submergent base, and these emergent proper-
ties are not merely epiphenomenal since they have causal power. Furthermore,
such properties have an explanatory power in chemical reactions that the me-
chanical affections do not have, which is why Boyle never directly appeals to the
mechanical affections of prima naturalia in his chemical explanations. Thus,
“Boyle maintained chemical qualities depended more on the way in which the
corpuscles that composed a given body were disposed to act on, or to be acted
on by, those of other bodies . . . He thought that chemical qualities emerged
from the constant interactions of corpuscles passing from one body to another.
Accordingly . . . he denied that they directly originated from the mechanical
properties of their primary particles.”57
In his essay on The Origin of Forms and Qualities, Boyle explains that, be-
cause fundamental matter “being in its own nature but one, the diversity we
see in bodies must necessarily arise from somewhat else, than the matter they
consist of.”58 The diversity in chemical and sensible qualities arises from the
specific way the first-order corpuscles coalesce to form primary concretions
or second-order corpuscles. He, in fact, “refrained from establishing a direct
relationship between a given quality and a set of mechanical properties of the
simplest corpuscles.”59 Additionally, his many remarks regarding the produci-
bility of chemical principles in compounds from ingredients that do not bear
55 Boyle, The Producibleness of Chemical Principles, 34.

57 Ibid., 588.
58 Boyle, The Origin and Forms of Qualities, 460.
the qualities in question further supports the view that he endorses a notion of
chemical properties as emergent, that is, as ontologically dependent upon but
not reducible to the submergent base. Boyle clearly states that “things may ac-
quire by mixture very differing qualities from those of any of the ingredients.”60
However, although we have established that Boyle regards chemical properties
to be novel features of chymical atoms that are not displayed by the component
particles, this is not enough to establish that he regards chemical properties to be
emergent properties.
To do so, we must answer a second question, that is, whether Boyle regards
chemical properties as supervenient, that is, whether chemical properties are de-
termined by and dependent upon the mechanistic properties of the lower-level
mechanistic base. To the extent that, for Boyle, any change in chemical qualities
requires a change in the mechanical property of texture or structure of chymical
atoms, it would seem that chemical properties display this crucial feature of
emergence, that is, supervenience. In The Mechanical Origin and Production of
Volatility and Fixedness, Boyle claims that “the same material parts of corporeal
substance, which, when they were associated and contexed [. . .] after such a de-
terminate manner, constituted a solid and fixt body, as a Flint or a lump of Gold;
by having their Texture dissolved, and (perhaps after being subtilized) by being
freed from their former implications or firm cohesions, may become the parts of
a fluid body totally Volatile.”61 For Boyle, chemical qualities clearly supervene on
mechanical qualities, that is, any change in chemical qualities requires a change
in the mechanical properties of the submergent base, namely, a change in the
secondary mechanical property of texture or structure. Thus, if we wish to say
that Boyle regards chemical properties as emergent, we must also establish that
he considers chemical properties to be novel and causally efficacious properties
that cannot be deduced from or reduced to the submergent base of mechanistic
particles.
The third question we must answer is whether, for Boyle, chemical properties
display non-summative difference from the properties of the subvenient base. To
establish that Boyle believes this, we must argue that for him chemical properties
are not properties of minima naturalia nor can they be additively derived from the
mechanistic properties of minima. As established in Chapter 3 and in Section 4.2
of this chapter, to the extent that the same fundamental corpuscles can, in prin-
ciple, be compounded into differently structured chymical atoms whose chem-
ical properties will vary according to these structures, it follows that the chemical
properties of a body cannot simply be deduced from the fundamental mecha-
nistic particles that compose it but are a function of the structure of chymical
60 Boyle, Letters and Papers of Robert Boyle, 270.

61 Boyle, The Mechanical Origin and Production of Volatility and Fixedness, 373.
atoms. Thus, it is the structural disposition or texture of the particles that ac-
counts for the emergence of chemical properties in compounded corpuscles
that are not attributable to the mechanical affections of minima naturalia. Boyle
makes this plainly clear when he states that “we can little better give an account of
the phaenomena of many bodies, by knowing what ingredients composed them,
than we can explain the operations of a watch, by knowing how many, and of
what metals the balance, the wheels, the chain, and other parts are made of,”62
making it also clear that a facile reductionism is not entailed by an adherence to
the mechanical philosophy.
One of Boyle’s reasons for preferring the mechanical hypothesis to the
spagyrist notion of the tria prima is that the mechanical hypothesis provides a
superior explanation for the great variety of material bodies and their qualities
than the idea that there are only three unvarying fundamental principles. Boyle
clarifies this by pointing out that, although shape, size, and mobility refer to
only three principles, each of these can take many distinct forms, thereby giving
rise to a numerous variety of mechanical affections of primary particles. Thus,
the shapes of primary particles are varied, as is their size and the speed of their
motions. When prima naturalia of various sizes and shapes are brought together
into corpuscular concretions, their textures or microstructures will be as nu-
merous and diversified as the possibilities of combinations are numerous. These
varying structures give rise to the great variety of material bodies with their re-
spective variety of sensible, chemical, and other non-mechanical qualities. This
variety of properties, however, would not be possible if the properties of the
whole were simply summatively equal to the properties of the parts. Boyle states,
“if it be said that these ingredients, by the texture resulting from their mixtures,
may acquire qualities that neither of them had before, I shall answer that to allege
this is in effect to confess that they must take in the Mechanical principles (for to
them belongs the texture or structure of bodies) to assist the chemical ones.”63
Thus, far from believing that mechanical principles are incompatible with the
notion of emergent chemical qualities, Boyle is confident that the emergence of
chemical qualities is best explained via the mechanical principle of structure.
This represents both Boyle’s affirmation of and commitment to mechanistic
corpuscularianism and his clear departure from a strictly reductionist approach
to chemical explanations.
Despite some arguments made to the contrary,64 Boyle’s reliance on chem-
ical explanations in terms of higher-level non-mechanical properties does not
62 Boyle, About the Excellency and Grounds of the Mechanical Hypothesis, 454.
63 Boyle, Of the Imperfection of the Chemists’ Doctrine of Qualities, 133.
64 See, e.g., Chalmers, “Boyle and the Origins of Modern Chemistry” and “The Lack of Excellency
of Boyle’s Mechanical Philosophy.”

render him any less committed to the mechanical hypothesis as a grounding

theory. Let us recall here the discussion in Chapter 2 of both the negative and the
positive heuristic role of the mechanical philosophy in Boyle’s scientific research
programme. Thus, his confidence in the emergent character of chemical prop-
erties is, above all else, evidence of his anti-reductionism and commitment to
providing chemical, rather than mechanical, explanations. However, as William
Newman explains, “it does not follow from the fact that these explanations are
not based on the catholic affections of the prima naturalia that they are not me-
chanical.”65 Texture is, after all, considered a “mechanical affection” to the extent
that textural changes are changes in the spatial relations between particles. The
interpretation of Boyle’s explanations as non-mechanistic is based on the erro-
neous assumption that all mechanistic explanations must perforce be reductive.
Newman also points out that
If we look at the vast majority of Boyle’s experimental demonstrations of the

mechanical origin of qualities, it is precisely such aggregate corpuscles that
usually come into discussion [. . .] aggregate corpuscles distinguished by
their chymical properties, not initial atoms having primary qualities alone.
Nonetheless, it is the “mechanical texture” formed by the association of these
finely ground particles that produces the explosiveness of gunpowder [. . .] The
radical disjunction claimed by some scholars to exist between Boyle’s mechan-
ical philosophy and chymistry is in fact illusory.66
For Boyle, a commitment to the mechanical hypothesis is perfectly compatible

with a chemical philosophy grounded in chemical explanations that are not sim-
plistically reductionist, and he defends this compatibility by stating that “the me-
chanical principles are so universal, and therefore applicable to so many things,
they are rather fitted to include, than necessitated to exclude, any other hypo-
thesis, that is founded in nature, as far as it is so.”67
As Joshua Gregory notes, for Boyle “a principle was required to clothe the
naked corpuscles into the richness of nature. if Demokritos were right, physi-
cally real magnitudes might have no sensible qualities [. . .] if fresh corpuscular
dispositions were not connected with new qualities, the originally naked cor-
puscular mechanism was an utterly incompetent conception.”68 What is inter-
esting about Boyle’s conception of chemical qualities, specifically as it relates
to his chemical experiments, is that he does not seem to embrace a simplistic

66 Ibid.
67 Boyle, About the Excellency and Grounds of the Mechanical Hypothesis, 453.
68 Gregory, “The Animate and Mechanical Models of Reality,” 310–311.
dichotomy of reduction/emergence but, rather, is able to adapt the mechanistic

hypothesis to his practical chemical endeavors by working out a hybrid ontology
of fundamental particles with strictly mechanistic properties and corpuscular
concretions from which emerge causally efficacious chemical properties. As
William Newman has so aptly stated, “it is clear that we can no longer accept the
view that Boyle’s corpuscularism was simply a ‘physicist’s theory’ grafted by him
onto a purely qualitative chemistry.”69
4.5 Supervenience, Non-Summative Difference,

and Underdetermination
Because we are arguing that supervenience is a feature of emergence and be-

cause emergence is conceived as an ontological feature of the world, whereas
unpredictability and derivability are epistemic notions, it is best to speak of
supervenience in terms of ontological underdetermination. Following Achim
Stephan, Robert Francescotti advocates replacing “the unpredictability con-
straint with the following underdetermination thesis: the emergent properties
of the whole are underdetermined by the properties of its proper parts [. . .] The
crucial question [therefore] is whether the properties of the parts at one time
underdetermine the properties of the whole at that same time.”70 This underde-
termination is accounted for precisely through the notion of relationality. In fact,
emergence and supervenience, understood in terms of relationality, are keys to
answering the questions of the reducibility of higher-level properties to lower-
level structures. Although the phenomena, reactions, and processes studied by
chemists “are thought to be ontologically dependent upon relationships at the
primary level [. . .] supervenience allows us the virtue of ontological dependence,
without the vice of explanatory reduction.”71
To the extent that chemical species, understood as wholes, display properties
that are underdetermined by their component parts, the fundamental laws that
govern the individual parts cannot, even in principle, permit us to deduce a com-
plete account of the behavior of the chemical species. As Michel Bitbol explains,
“there may exist differences between global states without any corresponding
differences between the local states that are supposed to underpin them [. . .]
Properties and states cannot be treated as preexistent intrinsic features. They
must be construed as relational.”72 To connect this claim to our discussion of
69 Newman, “The Alchemical Sources of Robert Boyle’s Corpuscular Philosophy,” 585.

70 Francescotti, “Emergence,” 52–53.
71 Scerri and McIntyre, “The Case for the Philosophy of Chemistry,” 224.
72 Bitbol, “Downward Causation without Foundations,” 13.
Boyle, it is clear that the few laws supposed by the mechanical philosophy to de-
termine the physical constitution of chemical wholes would not suffice to ac-
count completely (or, perhaps, even partially) for the behavior of such wholes.
Therefore, if such fundamental physical laws do not suffice to explain the be-
havior of chemical species and if only uniquely chemical laws can satisfactorily
provide such explanations, then Boyle is able to affirm the autonomy of chem-
istry from the mechanical philosophy even if chemical wholes are ontologi-
cally dependent upon mechanistic parts. As Mario Bunge has said with regard
to contemporary chemistry, but which also applies to early modern chemistry
even under the umbrella of the mechanical philosophy, “if it were possible to
define every chemical concept in terms of physical concepts, and deduce every
chemical law statement from a set of physical premises, chemistry would still
keep its peculiar referents (chemical systems), methods (e.g., acidity measure-
ment and neutralization), and goals (e.g., understanding and controlling chem-
ical syntheses).”73
What I propose to do in the next, and final, section of this chapter is argue
that Boyle regards not only chemical properties but also cosmical qualities, like
magnetism and gravity, as being dispositional and relational. Although cosmical
qualities are also regarded as ontologically dependent on the properties and be-
havior of corpuscles, the causal power of these properties emerges only when the
proper environmental, relational, and dispositional conditions are in place.
4.6 Cosmical Qualities as Dispositional and

Relational Properties
I have been arguing that there is no tension between Boyle’s commitment to

mechanistic corpuscularianism and his belief that higher-level qualities are rela-
tional and dispositional properties. To further illustrate this point, I now turn to
Boyle’s discussion of cosmical qualities, since he considers these properties both
as having corpuscular origins and as being dispositional and relational prop-
erties. For Boyle, cosmical qualities such as magnetism and gravity are, in fact,
powers in the aerial corpuscular effluvia that penetrate the pores of bodies dis-
posed to be affected by these particles. Catherine Wilson explains that both vi-
talistic and mechanistic corpuscularians often invoked the notion of corpuscular
effluvia to account for apparent action at a distance and for other otherwise un-
explainable phenomena. Corpuscularians made such phenomena “intelligible by
positing invisible, mobile, active particles capable of agitating within corporeal
73 Bunge, “Is Chemistry a Branch of Physics?,” 210–211.

objects and also capable of traveling great distances—perhaps even from celes-
tial objects to the earth.”74 Incidentally, the notion of aerial effluvia is compatible
both with the Paracelsian ontological conception of disease and with the account
of illness via contagion first articulated by Fracastoro in 1546. Although many
vitalistic corpuscularians accounted for the powers of these aerial particles by
appealing to seminal reasons, the inner workings of aerial effluvia were also “a
preoccupation of the mechanical philosophers.”75 For mechanistic philosophers,
corpuscular effluvia provided explanations for phenomena that had previously
seemed mysterious and unexplainable by physicalistic and naturalistic means.
Boyle himself takes the theory of corpuscular effluvia to be perfectly compat-
ible with his version of the mechanistic philosophy, arguing that a body’s dispo-
sition to be affected by the cosmical qualities of aerial particles is itself a function
of the structure of the body in question. To this end, Boyle emphasizes three
principles regarding cosmical qualities, the last two of which are considered the
most important and all three of which he takes to be purportedly confirmed by
experimental observations. These principles are:
1. That there are many Bodies, that in divers cases act not, unless they be acted
on & some of them act, either solely or chiefly as they are acted on by the
Catholick and unheeded Agents, we have been speaking of.
2. That there are certain subtle Bodies in the world, that are ready to insinuate
themselves into the Pores of any Body disposed to admit their action, or by
some other way to affect it, especially if they have the concurrence of other
unobserved Causes and the establisht lawes of the Universe.
3. That a Body by a mechanicall Change of Texture may acquire or loose a
fitnesse to be wrought upon by such unheeded Agents, and also to diversify
their operations on it upon the score of its varying Texture.76
The first principle is Boyle’s affirmation of the existence of cosmical qualities.

The second principle is Boyle’s identification of cosmical qualities with aerial
corpuscular effluvia. The third principle is Boyle’s identification of mechanistic
principles as governing the disposition to be affected by cosmical qualities. As
previously stated, and as he explicitly affirms, Boyle considers the last two princi-
ples as most important. It seems clear to me that the reason for this is that the last
two principles jointly confirm his commitment to the compatibility of cosmical
qualities with the mechanical philosophy. This is the case because the second
principle attributes cosmical qualities to corpuscular effluvia that penetrate the
74 Wilson, Epicureanism at the Origins of Modernity, 71–72.

75 Ibid.
76 Boyle, Of the Systematical or Cosmical Qualities of Things, 289.
pores of those bodies that are disposed to be affected, while the third principle
attributes such dispositions to be affected to the texture or microstructure of the
bodies in question. Boyle then provides an extended defense, as well as obser-
vational evidence, to support these last two principles. This is not enough to es-
tablish compatibility, however, since the first principle requires a more extensive
defense on Boyle’s part.
Such a defense is not forthcoming in Of the Systematical or Cosmical Qualities
of Things since, in this work, Boyle limits himself to two paragraphs in which he
simply provides empirical examples of magnetism and gravity. The more exten-
sive discussion of cosmical qualities occurs in the appendix to this same work,
which is titled Cosmical Suspicions. As the title of this appendix indicates, how-
ever, Boyle’s belief in the existence of such qualities is a mere “suspicion,” and
a faint one at that. Boyle states: “it may now therefore be not unreasonable to
confesse to you, that I have had some faint Suspition [. . .] that there may be,
as I was beginning to say, peculiar sorts of Corpuscles that have yet no distinct
name, which may discover peculiar Faculties, and Ways of working, when they
meet with Bodies of such a Texture as disposes them to admit, or to concur with
the Efficacy of these unknown Agents.”77 He then claims that the qualities with
which these unnamed corpuscles are endowed are probably governed by laws of
the universe that are themselves, as yet, unknown. Although Boyle proceeds to
give empirical examples that convince him that cosmical qualities indeed exist
and are governed by unknown laws of the universe, he does admit at the end of
this appendix that much of what he has discussed therein is conjectural. In spite
of this admission, however, he continues to affirm the existence of such qualities
and their degree of influence upon natural phenomena.
And whereas, Pyrophilus, I have in the former Discourse taken in the Structure
and establisht Lawes of the Universe as an Helpe toward the giving an Account
of the Cosmical Attributes of things; I shall here also ingeniously confesse to
you, that I much feare whether we have yet attentively enough taken notice ei-
ther of the number, or the Kinds of those Lawes. For as I am by some Notions
and Observations inclined to think, that there may be a greater number even
of the more generall Lawes, then have been yet distinctly enumerated [. . .] and
those may have a greater Influence on many Phaenomena of nature then we are
wont to imagine.78
77 Boyle, Cosmical Suspicions (Subjoyned as an Appendix to the Discourse of the Cosmicall Qualities
of Things), 303.
78 Ibid., 305.
As has already been emphatically stressed, Boyle’s reliance upon causes other
than the mechanical affections of primary corpuscles does not signal a rejection
of the mechanical philosophy. To the extent that cosmical qualities ultimately
originate from corpuscular effluvia, his explanatory appeal to such qualities is
consistent with the mechanical philosophy’s negative heuristic as discussed in
Chapter 2. As Boyle states in his Relations betwixt Flame and Air (1672), his ap-
peal to cosmical qualities is in keeping with the mechanical hypothesis because
it relies on physical causes and laws, rather than on the non-material principles
prevalent in the natural philosophies of Paracelsians and spagyrists.79 When
saying this, Boyle is directly responding to Henry More’s criticism of his ear-
lier work, New Experiments Physico-Mechanical, Touching the Spring of the Air,
and Its Effects (1660), in which Boyle had attempted to explain the results of his
experiments with the air-pump, or pneumatic engine, by appealing to strictly
naturalistic principles such as gravitation and elasticity of the air.
More, whose piety eventually led him to recant on his former allegiance to
mechanicism, reinterprets Boyle’s experiments as indicating the presence and
causal activity of “spermatical” and vital principles such as a spirit of nature or
anima mundi. More regards “gravity and the ascent of the piston of the air-pump
as contradictions of mechanism.”80 In the Relations Betwixt Flame and Air, Boyle
directly addresses More’s interpretation by defending the notion that gravity,
air elasticity, and other non-mechanical qualities have mechanistic origins. He
states that
having sufficiently proved, that the Air, we live in, is not devoid of weight, and
is endowed with an Elastical Power or springiness, I endeavour’d by those two
Principles to explain the Phaenomena exhibited in our Engine [. . .] without
recourse to a Fuga Vacui,—or the Anima Mundi,—or any such unphysical
Principle. And since such kind of Explications have been of late generally
called Mechanical, in respect of their being grounded upon the Laws of the
Mechanicks; I, that do not use to contend about Names, suffer them quietly to
be so: And to entitle my now examined Explication to be Mechanical, as far as
I pretend, and in the usual sense of that expression.81
This passage clearly shows that Boyle considers any explanation that eschews
non-physical principles and that affirms strictly physical causes to be a mecha-
nistic explanation or, at least, to be compatible with the mechanical philosophy,
although the ultimate mechanistic principles that govern the spring of the air or
79 Boyle, “Relations betwixt Flame and Air.”

80 Greene, “Henry More and Robert Boyle on the Spirit of Nature,” 468.
81 Boyle, Relations betwixt Flame and Air, 148.
the pressure of water are not included in these explanations and may yet be un-
known. Thus, although the term “mechanical” in these experimental writings
does not explicitly refer to mechanistic corpuscularianism, it remains compat-
ible with the mechanistic theory of matter.
As I’ve suggested in this chapter, the relation between agent and patient is one
that is embedded in a specific environmental context that affects the relational
properties that emerge so that, if both agent and patient were to be placed to-
gether in a vacuum and away from the environmental conditions in which they
are embedded, they would lose many of their dispositions to affect or be affected
by each other. Some of the clearest arguments for the relationality of properties
are found in Boyle’s writings on cosmical qualities, thus we should note that one
of Boyle’s most important discussions of the relationality of properties and their
dependence on surrounding environmental conditions occurs not in The Origin
of Forms and Qualities but in his essay Of the Systematical or Cosmical Qualities of
Things. Here, Boyle clearly specifies that
though in estimating the Qualities of Naturall Bodies we are wont to consider

but the power any particular one has of acting upon, or the capacity it has of
suffering from such and such particular Bodies [. . .] yet there may be some
Attributes, which may belong to a particular Body [. . .] not barely upon the
score of these Qualities that are presumed to be evidently inherent in it, nor of
the respects it has to those other particular Bodies to which it seems manifestly
related, but upon the account of a Systeme so constituted as our World is, whose
Fabrick is such, that there may be divers unheeded Agents, which, by unper-
ceived meanes, may have great Operations upon the Bodie we consider [. . .] So
that although if divers Bodies that I could name were placed together in vacuo
[. . .] they would retaine many of the Qualities they are now endowed with; yet
they would not have them All: but by being restored to their former places in
the World, would regain a new Set of Faculties (or Powers) and Dispositions,
which because they depend upon some unheeded Relations and Impressions,
which these Bodies owe to the determinate Fabrick of the grand Systeme or
World they are parts of, I have [. . .] thought fit to name their Cosmicall or
Systematicall Qualities.82
It has customarily been argued that Of the Systematical or Cosmical Qualities

of Things adds nothing relevant to what is already contained in the more famous
essay on The Origin of Forms and Qualities. However, as John Henry has recently
pointed out, there is more to Boyle’s essay on cosmical qualities than a mere
82 Boyle, Of the Systematical or Cosmical Qualities of Things, 287–288.

reiteration of the mechanistic origins of sensible and chemical properties. Boyle

himself explains the difference between his treatment of relational and disposi-
tional properties in this essay and the account given in The Origin of Forms and
Qualities. He states:
I have in the Origine of Formes touched upon this subject already, but otherwise
then [sic] I am now about to doe. For whereas that which I doe there principally,
(and yet but Transiently,) take notice of, is That one Body being surrounded
with other Bodies, is manifestly wrought on by many of those among whome ’tis
placed: that which I cheifly in This Discourse consider is, the Impressions that a
Body may receive, or the power it may acquire, from those vulgarly unknown,
or at least unheeded Agents, by which it is thus affected, not only upon the ac-
count of its owne peculiar Texture or Disposition, but by virtue of the general
Fabrick of the World.83
As John Henry has claimed, it seems that Boyle is indeed trying to articulate
a notion of non-mechanistic relational qualities that “is rather less compatible
[. . .] with all standard accounts of the mechanical philosophy.”84 For Boyle, the
fabric of the world is such that all bodies within it are affected by other bodies, as
well as by causes of which many may yet be unknown. According to Boyle, these
causes include cosmical qualities, such as gravity and magnetism. In fact, to the
extent that Boyle considers the cosmical qualities of magnetism and gravity to
be “unheeded agents” that work by “unperceived means,” he considers such cos-
mical qualities to be occult qualities that resist mechanistic explanations. Boyle
is a realist about occult qualities precisely because, despite their methodolog-
ical and epistemic irreducibility to mechanistic affections, their causal power
is manifested as one observes their experimental effects. This is another case in
which the existence of unobservable causes is transformed, via experimental ac-
tivities, into a “matter-of-fact.”
John Henry, however, mistakenly considers Boyle’s realism regarding occult
and cosmical qualities as being incompatible with “strictly mechanistic precepts.”
Certainly, if Boyle’s mechanicism were strictly reductionistic, then Henry would
be correct in regarding it as incompatible with the belief in occult and cosmical
qualities. On the other hand, if I am correct that Boyle’s mechanicism is non-
reductionistic, then his belief in the epistemic usefulness of appealing to occult
and cosmical qualities is not necessarily incompatible with his ontological po-
sition regarding fundamental matter. As I have argued extensively, there is ev-
idence that Boyle considers higher-level properties to be emergent, relational,
83 Ibid., 288.
84 Henry, “Boyle and Cosmical Qualities,” 121.
and dispositional properties whose causal powers explain not only chemical, but
also hydrostatic and pneumatic, phenomena without needing to appeal to mech-
anistic corpuscularianism.
In this chapter I have argued that, for Boyle, chemical and other higher-
level properties are supervenient properties that emerge from the dispositions
of chemical substances in relation to other substances. He clearly conceives of
chemical substances as functional wholes whose properties emerge not only
from the structural ordering of their parts but also from their relationship with
other chemical substances in the context of experimental practice. However, as
Isabelle Stengers points out, “as soon as the question of emergence is at stake,
the whole and its parts must thus co-define themselves, and mutually negotiate
what an explanation of the one from the other means.”85 The attribution of non-
summative difference to a property of a chemical whole is itself a mereological
issue, because it implies that the whole possesses properties that are not simply
the summation of the properties of its parts. The next chapter will, therefore, ex-
amine these issues and will argue that the kind of mereology required to make
sense of the notion of supervenience must differ from the widely assumed exten-
sional interpretation of classical mereology, in which any parts whatever auto-
matically compose some whole, regardless of their relation to each other.86
85 Stengers, “La vie et l’artifice,” 207.

5
The Relation between Parts and Wholes
The Complex Mereology of Chymical Atoms
In the last chapter, I argued that Boyle conceived of chemical properties not only
as dispositional and relational but also as emergent and supervenient upon the
submergent base. I also suggested that an accurate account of supervenience
requires a kind of mereology that is distinct from standard extensional mere-
ology in which the properties of the whole are simply a summation of the prop-
erties of the parts. When dealing with emergent and supervenient properties,
there is a non-summative difference of properties between the whole and the
parts. Mereologically, the whole itself is not merely a collection of its parts hap-
hazardly put together but, rather, it is a composite of parts in very specific rela-
tion to one another. Together, the subvenient properties of the constituent parts
and the relation of those parts and properties to one another account for the su-
pervenient properties of the whole, and this is what accounts for the transcend-
ence (i.e., non-reducibility) of the supervenient properties.
In the context of our discussion of Boyle, this point is significant because
there is an often-unstated mereological premise lying at the heart of strict
mechanicism, that is, that natural wholes are analogous to complicated machines
or artifacts, composed of parts that are lacking in any intrinsic relationship to
each other. Thus, the type of mereology that best describes such wholes is a
standard extensional mereology in which wholes are considered as purely sum-
mative aggregates of parts, displaying no novel properties above and beyond the
properties of their distinctive parts. Yet in spite of Boyle’s commitment to the
mechanical philosophy, it is not clear that he would endorse a standard exten-
sional mereology to describe chymical atoms. Although Boyle never addresses
mereological questions directly in his writings, I will argue that the type of mere-
ology that best describes Boylean chymical atoms is a non-extensional and non-
summative mereology that accounts for the fact that chemical wholes display
novel and emergent properties that are not present in their individual parts. Such
a mereology must also take into account the relationality of the parts to one an-
other, since such relations affect what emergent properties the chemical whole
will display.
University Press.
DOI: 10.1093/oso/9780197502501.001.0001
The Relation between Parts and Wholes 147
5.1 Boylean Chemistry as Mereological
It is clear from the discussion of Boyle’s corpuscularian theory of matter that, for
him, the specific particulate composition of corpuscular concretions or chymical
atoms is necessary but not sufficient to determine the chemical properties of a
given substance. In addition to discrete particles with their particular shapes,
sizes, and motions, the specific microstructure in which those particles are coa-
lesced is also a necessary condition for the manifestation of particular chemical
properties. In fact, as Rom Harré and Jean-Pierre Llored have recently pointed
out, “since Robert Boyle’s corpuscularian philosophy, chemistry has been a
mereological science. Displacing the metaphysics of ‘continuous substances’ and
‘qualities’ as the expression of ‘principles,’ chemistry has been built on a ‘part-
whole’ metaphysics. The grammar for the use of ‘part-whole’ concepts is mere-
ology.”1 However, according to Joseph Earley, “an adequate theory of wholes and
parts (mereology) must consider that when individuals enter combinations of
interesting sorts they no longer are the very same individuals that existed prior
to the composition.”2
The mereology involved in chemical supervenience is a non-standard mere-
ology because “the parts undergo changes when they form a whole.”3 The pres-
ence of the supervenient property affects the properties of the constituent parts,
so that the properties of the parts are dependent upon the role that they play
within the whole that expresses the supervenient property. In Earley’s words,
“most philosophers have yet to recognize that, when components enter into
chemical combination, those components do not, in general, maintain the same
identity that they would have absent that combination.”4 Good examples that il-
lustrate Earley’s point are H2O and silver chloride. While the property of being
H2O or of being silver chloride “supervenes on features of the constituent atoms,
the features of the atoms on which it supervenes include features that the atoms
have only by virtue of being parts of that compound. The atomic interrelations
that give rise to the compound would not obtain if the atoms were parts of a
different molecular type.”5 Molecules themselves are defined in accordance with
chemical reaction networks and not vice versa.6 Emergence itself is also a func-
tion of the relationality of the parts within the whole, since “emergent property F
supervenes only given relations [that] the parts bear to one another.”7
1 Harré and Llored, “Mereologies as the Grammars of Chemical Discourse,” 63.

2 Earley, “Why There Is No Salt in the Sea,” 85.
4 Earley, “Varieties of Properties,” 87.
5 Francescotti, “Emergence,” 58.
6 I am grateful to Jean-Pierre Llored for suggesting this useful example.
7 Francescotti, “Emergence,” 61–62.
As discussed in Chapter 1, Aristotle had already understood the importance

of these mereological questions with regard to understanding the nature of
substances. To briefly recapitulate that discussion, Aristotle rejects the idea that
the form of the whole is itself a material part of the whole that it unifies, since
this type of mereological hylomorphism leads into regress. The form, Aristotle
tells us, is not an element but a principle [αρχη].8 However, Aristotle believes
that mereological consistency requires that the form of the whole be both real
and ontologically distinct from the material parts. For the Scholastics, this im-
material Aristotelian principle was conceived much more concretely as “substan-
tial form.” Yet the Scholastics could not precisely explain what substantial form
is except to say that it must exist as that which sustains the constant and stable
properties that identify a particular thing as a member of a particular kind or
species. As explained in Chapter 3, this is one of the reasons for Boyle’s rejec-
tion of substantial form as an immaterial principle. Instead, as that chapter es-
tablished, Boyle favors the notion of essential form, understood mechanistically
as the texture or structure of corpuscular concretions. Although essential form
is a mechanistic structure, it is, nevertheless, also the source of a thing’s iden-
tity and of its emergent and supervenient non-mechanical properties. It was also
stressed that it is not just the composition of aggregate corpuscles that does the
chemical work but, rather, it is also the structure or texture in which the minima
of aggregate corpuscles are tightly coalesced that accounts for the chemical prop-
erties of substances. Thus, we must ask how to understand the notion of mechan-
ical essential form of corpuscular concretions. This understanding is necessary
in order to account for the fact that such form stands mereologically “over and
above” the constituent parts of corpuscular concretions without its being consid-
ered ontologically separate from its parts.
Robert Pasnau’s recent comment about Boyle’s structuralism raises some
doubts about where exactly Boyle stands, from a mereological perspective, with
regard to composite wholes. I will give a reply to Pasnau’s concerns but, first,
I wish to quote his comment in full since it is relevant for the present discussion.
Pasnau states that
if “texture” [. . .] refers only to the corpuscular facts taken as a whole (that is, to a
conjunction of discrete facts), then [Boyle] could really hardly have appealed to
texture as that which unifies a coherent body. There would be no way to distin-
guish between the texture possessed by a “distinct body” and the texture pos-
sessed by, say, all the books and papers on top of my desk. Now perhaps Boyle
would welcome this last result, or perhaps he was in fact torn over just how
8 Aristotle, The Complete Works of Aristotle, Z.17, 1041b32–33.

much weight to put on the texture of the whole as a unifying principle [. . .]

moreover he rejected the entire scholastic conception of substantial identity
and change. So, contrary to the Aristotelian view that the parts of a substance
maintain their identity only as part of that substance, Boyle argues that when a
body is generated, “no new substance is in generation produced, but only that
which was pre-existent obtains a new modification or manner of existence.”9
What Boyle means in the passage cited by Pasnau is that there is no substan-
tial form produced, that is, there is no new immaterial substance that comes into
existence. For him, a change of species or genus includes a real transformation of
structure and qualities but not the generation of a new substantial form. But this
passage does not imply that the notion of diachronic identity is meaningless for
Boyle, for he clearly states at the end of the cited passage that a new manner of
existence is produced. What this means, for Boyle, is that a body belonging to a
different natural kind is generated because, although all of the material particles
that composed the original body are still present and no new substance has been
added, the structure of those material particles has been changed to produce a
material body with different properties and, thus, a material body belonging to
a different species from the original. Thus, the new material body has a different
identity from the original body, because it has a different essential form, that is,
its corpuscles are arranged according to a different texture or structure.
This is true because, as Boyle claims, if the structure is altered though the com-
position remains the same, the chemical properties of the substance will change.
That is, a change in structure, even if not accompanied by a change in composi-
tion, yields new chymical atoms and, thus, substances with different chemical
properties. Thus, Boyle rejects mereological extensionality, since the same group
of fundamental particles arranged according to one structure will produce an
entirely different set of properties from those that it would produce if it were ar-
ranged according to a different structure. Moreover, as explained in Chapter 3
and as Kuhn has also argued,10 Boyle’s chemical philosophy is a “structural re-
alism” that remains situated within an overall mechanistic hypothesis. What this
means is that the chemical identity of a substance is not entirely reducible to its
composition but is also a function of the structure in which its particles are ar-
ranged. This structural realism, which may seem at first to be at odds with the me-
chanical philosophy, is nevertheless mechanistic to the extent that the structure
of chymical atoms is conceived geometrically as a tight juxtaposition of particles.
Though I am simply clarifying the meaning of Boyle’s structural realism here,

10 Kuhn, “Robert Boyle and Structural Chemistry.”
I will address the many problems associated with this geometric conception of
structure at the end of this chapter.
The mereological questions raised by this discussion will also be addressed
in detail later. But, for now, suffice it to say that Pasnau is mistaken in his doubts
regarding Boyle’s position on compositional wholes. As elaborated in Chapter 3,
the texture or essential form of a material body that gives it unity and species
identity is something stable and operationally irreducible. Unlike Gassendi who
offers no account of what holds molecules together, Boyle attempts to explain the
stability of aggregate corpuscles in a mechanistic way by appealing to their tight
spatial juxtaposition. Thus, Boyle does at least offer an account of unity for com-
positional wholes although, as already mentioned, there are serious problems
with this account that I will address more fully later. Boyle would not agree that
the texture of a material body is akin to the texture of the books and papers on
Pasnau’s desk. In fact, he would argue that, although the books and papers are
arranged in a particular way in relation to one another, there is no “texture” or
essential form as Boyle understands these terms because there is no mereological
unity. The books and papers merely constitute a heap, not a compositional whole,
since this pile has no internal unity or stability and can be easily dismantled. In
this sense, Boyle’s notion of essential form as material structure fulfills the same
mereological purpose as the notion of form fulfilled for Aristotle, without any of
its problematic metaphysical commitments.
Although Boyle himself does not address these mereological questions di-
rectly or explicitly, a very distinct mereology is implied by his concept of
chymical atoms as producing emergent, supervenient, and non-summative
chemical properties. Had Boyle adopted a strictly reductionist mechanicism, he
would have fallen prey to the reductionist potential of standard extensionalist
mereology, so that the whole must summatively display only those properties
that are inherent in its parts and the properties of the whole must be entirely de-
ducible from and reducible to the properties of the parts. I believe, however, that
Boyle’s chemical ontology guides him in the direction of what we would call a
non-standard mereology. It is clear from the discussions in the previous chapter
that Boyle does not consider chemical properties to be deducible from the me-
chanical properties of the fundamental particles that constitute these wholes.
Although Boyle does not explicitly articulate these ideas in mereological terms,
mereological concerns are definitely implicit in his theoretical and experimental
writings and, in some sense, anticipate some contemporary concerns regarding
mereology as it relates to chemical wholes.
Although a detailed analysis of the mereological concerns implicit in Boyle’s
writings would greatly contribute to our understanding of his chemical philos-
ophy, it is unfortunate that very little attention has been paid to these ideas. In
fact, the only explicit allusion to the presence of any sort of mereological position
in Boyle’s work is found buried in the pages of Desmond Paul Henry’s Medieval
Mereology. In this work, Henry very briefly alludes to Boyle while clarifying the
mereological theories of Peter Abelard. Henry hopes to elucidate Abelard’s ideas
by comparing them with Boyle’s position regarding the relation between parts
and wholes and by concluding that Abelard’s mereology shares great similarities
with that of Boyle many centuries later. What strikes me the most about Henry’s
brief discussion of Boyle, however, is that his remarks presuppose that Boyle
adopts a reductionist and extensional standard mereology regarding corpus-
cular concretions. Therefore, in the discussion to follow, I will tease out the errors
in Henry’s argument and present my own account, based upon textual evidence
as well as what has already been established regarding Boyle’s chemical ontology.
With this, I hope to provide the kind of detailed analysis of Boyle’s mereology
that has, up to this point, been absent in the literature.
5.2 Continuous vs. Contiguous Integral Wholes
In order to both explain and critique Henry’s conclusions regarding Boylean

mereology, I will begin by discussing some key mereological concepts upon
which his discussion is grounded. To the extent that Henry compares Boyle
to Abelard, I will begin by discussing some general concepts within Scholastic
mereology. Although Scholastic logicians and metaphysicians embrace clas-
sical Aristotelian mereology as foundational, they develop and contribute new
concepts and distinctions to it based upon their own metaphysical and, often, the-
ological concerns. One of these is the distinction between integral wholes, uni-
versal wholes, and potential wholes. Since only integral wholes will be relevant for
this discussion, I shall not delve into the nature of universal or potential wholes.
Integral wholes can be either continuous or contiguous wholes. A continuous
whole is one in which the parts share a common boundary and are “so mutually
admixed that they are believed to hold together, or at least to grip one another,
so that they are [. . .] truly said to be one thing.”11 An example of a continuous
integral whole would be a unified whole such as a stone, a tree, or a human
being. Homogeneous masses, like gold or water, are also considered continuous
wholes. Abelard defines it thus: “A continuous whole is one which is made up of
impredicable parts, in the sense that they cannot have their whole predicated
of them, such that as long as the wholes exists the parts are not susceptible of
change of place, as can be seen in the case of a house.”12 In contrast to continuous
11 Desmond Paul Henry, Medieval Mereology, 124.

12 Peter Abelard, Introductiones dialecticae, in Pietro Abelardo, 193–194: “Continuum totum est
illud quod constat ex partibus impraedicabilibus, id est non recipientibus praedicationem sui totius,
wholes, the parts of non-continuous wholes, such as a flock of birds for example,
are susceptible to change of place even while the whole continues to exist.
Interestingly, although Abelard uses the example of a house to illustrate the
definition of a continuous integral whole, both he and Aquinas believe that,
in a strict sense, no artificial construct can genuinely qualify as a continuous
whole. As mentioned earlier, for Scholastics, the unity of substances that quali-
fies them as continuous wholes is due to the presence of immaterial substantial
forms, which are only possessed by natural substances. Such things as houses
can, therefore, never truly qualify as continuous integral wholes. To the extent
that only natural substances are endowed with substantial form, the unity of ar-
tificial substances, artifacts, or other things lacking substantial form is consid-
ered a lesser kind of unity. More will be said about this later, since I will argue
that Boyle’s rejection of substantial forms and his commitment to mechanistic
corpuscularianism would force him to reject the notion of chemical substances
as continuous integral wholes. Let us now continue our discussion by examining
contiguous integral wholes.
A contiguous integral whole is one in which the parts are discrete but spatially
close together. Such parts “are by nature disjointed, but [. . .] are put into conjunc-
tion by a craft, as in the case of a cuirass or a piece of cloth. In these instances, part
is put close to part and gathered together so that a whole is then effectuated.”13
Contiguous wholes do not have substantial unity in the sense of having substan-
tial form. Hence, for the Scholastics, artifacts can only ever be considered con-
tiguous wholes, since they have accidental form but not substantial form. Thus,
strictly speaking and by Abelard’s own criteria, the house cited in the earlier pas-
sage really only qualifies as a contiguous whole. Interestingly, however, Abelard
considers that any scattered mereological whole, such as a heap of pebbles, can
qualify as a contiguous integral whole even if it lacks real unity of any sort. It is
precisely with regard to this issue that Abelard differs most from Aquinas and
other 12th-century logicians and metaphysicians.
Given his belief in substantial forms, Abelard has no trouble explaining what
it is about a continuous whole that gives it its identity as a unified whole, over
and above its mereological parts. With regard to contiguous wholes, however,
Abelard denies that there is any ontological “increase” in the whole over and
above its parts or that any real novelty ensues upon the combined parts. This
position regarding contiguous wholes is consistent since, for the Scholastics,
non valentibus transmutari secundum localem positionem, manente ipso toto, ut potest videri in
domo.” The only exception to this requirement is in the genitive case “of-a” or “of-the.”
13 Clarembald, Compendium logicae Porretanum, as cited in Henry, Medieval Mereology, 214: “At
totum contiguum est cuius partes naturaliter disiuncte, arte vero sunt coniuncte, ut videtur in lorica
vel in tela; ibi enim pars parti contiguatur aggregatur ut totum inde efficiatur.”
the only wholes that are greater than the sum of their parts are substances, since
these are endowed with substantial forms.
5.3 Integral Parts and Essential Parts
Besides distinguishing between different types of wholes, Scholastic logicians

and metaphysicians also distinguish between different types of parts, more spe-
cifically, between integral parts and essential parts. This will be an extremely im-
portant distinction in my later discussion of Boyle’s mereology. The Scholastic
Walter Burleigh (1275–1344) explains the distinction thus: “Part is sometimes
taken as the essential part which Aristotle and Averroes [Ibn Rushd] call the
‘qualitative part’ in many contexts, and they call the integral part the ‘quantita-
tive part.’ In a corresponding manner, whole is taken as whole in respect of form
and whole in respect of matter, and contrastingly for part there is part in respect of
form and part in respect of matter.”14 Thus, integral parts are associated with the
“quantitative” or material parts of a whole, which are present in only one place
at a time and side-by-side to each other. They are, one might say, the accidental
parts of a whole, so that the loss of particular integral parts does not entail any
loss of identity for the whole. Material parts can increase or decrease without af-
fecting the identity of the whole as a whole. Thus, “material parts [integral parts]
are those which ebb and flow while the whole remains the same.”15
Essential parts, on the other hand, are associated with the “qualitative” or
formal parts of the whole, that is, those parts that are associated with substan-
tial forms. The loss of essential parts destroys the substantial unity of the whole.
“Parts in respect of form [essential parts] are those which always remain the
same, as long as the whole remains the same complete object.”16 Thus, essen-
tial parts can neither decrease nor increase in the whole, without changing the
identity of the whole altogether. The notion that some parts of wholes are essen-
tial parts does not imply the kind of mereological hylomorphism that Aristotle
rejects, since essential parts are not regarded as material. Rather, the essential
parts of a whole are essential because they are united not by a mere principle of
material summation but by a principle of form or, more precisely, substantial
form. In fact, unification by form is precisely that which distinguishes essential
14 Walter Burleigh, De toto et parte, as cited in Henry, “Medieval Mereology,” 407: “Pars enim
aliquando capitur pro parte essentiali quam Philosophus et Commentator multis locis vocant partem
qualitativam et parem inegralem vocant partem quantitativam. Et isto modo accipitur totum pro toto
secundum formam et toto secundum materiam; et per oppositum, pars, pars secundum formam et
pars secundum materiam.”
15 Ibid., 408.
16 Ibid.
parts from accidental parts. Thus, the form is that which unites the essential parts
and renders them essential parts. However, to repeat, Scholastic metaphysicians
do not regard the substantial form of a whole as a material part of that whole.
In the next section, I will discuss how these notions function in the context
of Abelard’s mereology, and I will present an explanation and brief critique of
Henry’s positive comparison of Abelard and Boyle. This will set the context for
the detailed discussion of Boylean mereology in the rest of the chapter.
5.4 Aquinas, Abelard, and Boyle on Substantial Unity
It is important to note here that, despite some agreements between Abelard and
Aquinas, on the issue of substantial unity for example, there are also important
differences in their views regarding the mereological relation between wholes
and parts. Aquinas, like Aristotle, stresses the non-identity of whole and parts,
saying that there is something about the whole that is over and above its parts,
though he never specifies what this “something” is. As Henry explains, Aquinas
is “saddled with the task of explaining how the elements of a whole change their
identity upon incorporation in, or expulsion from, the whole which they com-
pose.”17 After putting forward several arguments that fail to explain how this is
possible, Aquinas settles on the conclusion that the “something over and above”
is the substantial form, that is, the “formal” principle of the substance or the
quiddity, to be distinguished from its material parts.18 This implies that contig-
uous integral wholes are not real wholes at all, in other words, they have no real
integrity since they lack substantial form.
For Abelard, on the other hand, as for David Lewis many centuries later,
“any sort of plurality can be taken to form some whole, even if the parts are dis-
jointed.”19 This point was already made earlier, with regard to Abelard’s claim
that any scattered mereological sum constitutes a contiguous integral whole.
As was also mentioned, Abelard’s contemporaries rejected this view, including
those who shared his nominalist metaphysics. For the purposes of this chapter,
however, it is important to note that one of the consequences of Abelard’s mereo-
logical atomism is that, in effect, “when one thing is conjoined with another [. . .]
neither has more parts than it previously had. Yet neither does the whole which
is composed of these [parts] appear to increase [. . .] there is also the tendency
towards denying that real novelty ensues upon the combination of parts [em-
phasis mine].”20 It is this last implication that will be particularly relevant for our
17 Ibid., 220.
18 Ibid., 224.
19 Ibid., 124.
20 Ibid., 123–124.
discussion later in this chapter. For, although Abelard is obviously not an ad-
vocate of the mechanical philosophy, he embraces the same sort of permissive
mereology that would later be embraced by strict mechanicists, with the same
sort of implications regarding the question of whether any novelty ensues upon
the combination of atomistic parts and, ultimately, whether the properties of
wholes are entirely reducible to the properties of their individual parts.
It is in discussing these issues that Henry draws his misguided analogy be-
tween Abelardian and Boylean mereologies. I attribute Henry’s mistake to his
interpretation of Boyle as a strict mechanicist, and I hope to demonstrate that
Henry’s claim is indeed erroneous. He states that Boyle too “would insist that
the bringing together of formerly scattered parts does not really bring into being
anything which, substantially speaking, was not already there previously [. . .]
previously non-assembled quasi-componential parts are [simply] put together
or appropriately relocated.”21 Henry cites a passage from The Origin of Forms and
Qualities that seems, at first glance, to support the view that Boyle embraces the
same permissive mereology as Abelard: “We say that a body belonging to that
species [. . .] is generated or produced de novo—not that there is really anything
of substantial produced, but that those parts of matter that did indeed before
pre-exist [. . .] are now brought together and disposed after the manner requisite
to entitle the body that results from them to a new denomination, and make it
appertain to such a determinate species of natural bodies.”22
If Henry were correct here, it would follow that Boyle endorses mereological
extensionalism, that is, the thesis that wholes with the same parts are identical to
each other regardless of the arrangement of those parts. Thus, the arrangement
or structure of the parts does not affect the nature of the whole, so that rearran-
ging the parts would simply yield the same whole. I believe, however, that Henry
misinterprets Boyle’s position by misconstruing the purpose of the passage that
he cites from The Origin of Forms and Qualities. I hope to demonstrate, in what
follows, that the true purpose of this passage, for Boyle, is not to argue against
mereological “novelty” but to argue against the presence of substantial forms.
As we shall see presently, to the extent that he believes that the structure or
essential form of chemical wholes imparts identity and integral unity upon
them, Boyle clearly does believe essential form to be a necessary condition for
unity. As Harré and Llored have recently pointed out, “chemical relations are not
merely formal; sums and incomplete fragments lack chemical work. The thesis
that objects with the same parts are identical, say, mereological extensionality,
cannot work in this context.”23 Based upon Boyle’s conception of chymical atoms
21 Ibid., 124–125.
23 Harré and Llored, “Developing a Mereology of Chemistry,” 211.
as stable material species that impart unique chemical properties to substances,

I firmly believe that he would concur with Harré and Llored on this point. Thus,
although Boyle rejects substantial forms, he would also reject the kind of per-
missive mereology implied by mereological extensionalism. What I will argue
later is that Boyle’s rejection of both mereological extremes, that is, his rejection
of substantial forms on the one hand and of mereological extensionalism on the
other hand, implies a rejection also of the distinction between continuous and
contiguous integral wholes. This, in turn, implies a rejection of the distinction
between natural and artificial substances, to the extent that neither has substan-
tial form but that both have the integral unity that is supplied by essential form.
This ties into the discussion of taxonomy in Chapter 3, which also established
Boyle’s rejection of this distinction.
In what follows, I will argue that, although Boyle lacked the vocabulary of con-
temporary mereology, he would most likely concur with Harré and Llored’s re-
cent call for the development of a new set of mereological rules for the logic of
chemical discourses, arguing that chemical “parts and sets are relational. They
are entangled with the whole and its environment and depend on the mode of
access—cognitive or instrumental.”24 Although Boyle did not develop and could
not have developed the kind of relational and dynamic chemical mereology for
which Harré and Llored have recently argued, Boyle would have undoubtedly
sympathized with such efforts by these contemporary philosophers of chemistry.
5.5 The Mereology of Boyle’s Chymical Atoms as Chemically

Elementary Entities
As mentioned in the previous section, what I hope to establish with this analysis is
that Boyle does not embrace the kind of permissive, atomistic, and extensionalist
mereology attributed to him by Henry. In fact, the complex corpuscularian on-
tology discussed in Chapter 3 lends itself to a nuanced mereological analysis
that reconciles the atomism implied by the mechanical hypothesis with Boyle’s
quasi-holistic and non-reductionist conception of chymical atoms as operation-
ally elementary wholes that endow substances with non-mechanistic and causal
chemical properties.
Most scholars would agree that mereological extensionalism is incompat-
ible with Boyle’s mechanistic structuralism, to the extent that the texture of cor-
puscular concretions endows these with unity. Thus, they would agree that two
chymical atoms that have the same composition but different structures must
24 Ibid., 209–210.
produce different chemical properties. However, although they would reject an

mereologically extensionalist conception of Boylean ontology, many scholars
will insist that for Boyle nothing “novel” is produced when the structural dispo-
sition of particles is altered and that the qualities of a substance are nothing other
than the mechanistic texture of its corpuscular concretions.
As Frederick O’Toole has argued, however, one cannot correctly attribute to
Boyle this “reductionist interpretation of powers and capacities.”25 In fact, a re-
ductionist interpretation of powers and qualities leads to absurd conclusions that
Boyle himself would reject. Regarding the attributes of gold to be easily soluble in
aqua regis and to be insoluble in aqua fortis, for example, Boyle explicitly states
that “these attributes are not in the gold anything distinct from its peculiar tex-
ture.”26 A reductionist and non-relational reading of this statement would iden-
tify the proposition “p is not in A anything distinct from q” with the proposition
“p is not distinct from q,” which would amount to saying, “p is q.” Thus, under
this reductionist interpretation, one would have to conclude that, for Boyle, solu-
bility in aqua regis and insolubility in aqua fortis are texture tout court. However,
as O’Toole so aptly points out, to say this would amount to collapsing the distinc-
tion between the qualities of being soluble or insoluble and texture itself. This,
however, is a problem for a number of reasons.
As was discussed in Chapter 3, Boyle believes that distinct natural kinds and
their properties are such by virtue of having distinctively structured or tex-
tured corpuscular concretions. Thus, it would follow that the structure or tex-
ture of their chymical atoms must be distinct, even if the respective and peculiar
structures of the chymical atoms of sugar and of salt render both substances
water-soluble. We must keep this in mind when we examine the implications of
the reductionist interpretation of power and qualities. Let us take the example
of sugar and salt, two distinct natural kinds that are both water-soluble. Let us
use w to represent the attribute of being soluble in water, A to represent sugar, ta
to represent the peculiar texture of sugar, B to represent salt, and tb to represent
the peculiar texture of salt. Let us now see what would follow if we adopt the
reductionist interpretation discussed earlier, this time with regard to the water-
solubility of sugar and salt:
1. If “w is not in A anything distinct from ta” = “w is not distinct from

ta” = “w is ta”
and
2. If “w is not in B anything distinct from tb” = “w is not distinct from
tb” = “w is tb”
then
3. Then, “w is ta” and “w is tb”
4. Thus, “ta = tb”
According to this conclusion, which follows from the reductionist interpre-

tation of powers and qualities, the fact that both sugar and salt are water-soluble
implies that their peculiar textures, ta and tb, are in fact the same since they are
both equal to w, the attribute of being water-soluble. However, if sugar and salt
have the same microstructural texture then all of their properties should be the
same, not just the property of water-solubility. This is the case because the collec-
tion of properties of a substance, according to Boyle, are derived from texture.
However, if we claim that glucose and sodium chloride have the same texture
and the same properties, we would have to conclude that glucose is the same
as sodium chloride. This absurd conclusion follows from the reductionist claim
that water-solubility and all other properties are not anything distinct from tex-
ture. Boyle would obviously not endorse such a nonsensical conclusion and,
thus, would not be likely to endorse the view that powers and qualities are not
anything distinct from texture or microstructure. In fact, I argued in Chapter 4
that such properties are distinct from structure to the extent that they are dis-
positional, relational, emergent, and “novel” properties that supervene upon the
microstructure of corpuscular concretions.
It is important, then, to examine the mereology of chymical atoms and how
it fits within the context of Boyle’s chemical ontology. In Boyle’s ontological hi-
erarchy of minima naturalia and corpuscular concretions, it is the corpuscular
concretions or chymical atoms that endow parts of matter with the specific
faculties to fashion or alter other parts of matter. Although chymical atoms or
corpuscular concretions are definitely to be considered as integral wholes and,
although these wholes enjoy a great degree of stability by resisting chemical
analysis, these should not be considered as continuous wholes because the parts
are not “fused” together, despite Boyle’s use of the term “adhaesion.” As Boyle
conceives of chymical atoms, their adhaesion is not a bond. Rather, it consists of
very “close and strict” spatial proximity so that even air cannot traverse the space
between the particles. Though Boyle does not go into such details, one may im-
agine that the parts “fit together” geometrically, which is the only account that
could explain such close fitting. To the extent that the parts of chymical atoms
are spatially very close but not fused, there may be a temptation to argue that
Boyle’s chymical atoms are contiguous integral wholes. However, as mentioned
earlier and as will be argued further at the end of this section, I believe that Boyle
would reject the continuous/contiguous distinction altogether and simply refer
to chymical atoms and all other material wholes as integral wholes simpliciter.
Regarding the parts of chymical atoms in relation to the whole, the constit-
uent fundamental particles are discrete and integral parts of the chymical atom.
That is, they are the material parts, present only in one place at a time and side-
by-side in a very close spatial proximity. The texture or structure in which these
parts “adhaere,” however, is a substantial part of the chymical atom, since it is
the essential form of this concretion. For medieval metaphysicians, “substantial
part” is analogous to “substantial form.” However, to the extent that for Boyle
essential structural form fulfills the role played by substantial form, then we can
safely say that the substantial part of chymical atoms is their essential form or
texture. While the loss or exchange of one fundamental mechanistic particle for
another does not result in a change of substance, because these are integral parts
of the whole, the loss of structure or essential form does entail a change of sub-
stance. This is because it is precisely this structure that accounts for the specific
non-mechanical essential properties that place a substance within a particular
material species.
To the extent that the perceptible chemical substance or compound body is
made up of chymical atoms and acquires its essential properties from the struc-
ture of these atoms, we can say that the compounded body is also an integral
whole made of discrete chymical atoms as its parts. These chymical atoms are in-
tegral parts of the whole substance, since subtracting some of them from the sub-
stance would not result in the loss of the essential properties that distinguish the
substance. For example, if I have a gram of potassium nitrate and I take away a
half a gram, each of the halves is still potassium nitrate and still displays the same
chemical properties. However, if I were to alter the structure of the chymical
atoms of potassium nitrate, I would give rise to a different substance with dif-
ferent properties.
Boyle gives numerous examples of experiments in which the component parts
of a substance remain the same but the structure is altered, thereby resulting in
a novel set of properties. In Of the Producibleness of Chymicall Principles, Boyle
gives the examples of powdered sulfur and salt of tartar, which will not dissolve
in spirit of wine. He notes, however, that “if this salt [of tartar] and sulphur be
mixt together, [they will] dissolve enough of this matter to be richly colourd by it,
and this without the help of external heat.”27 He concludes from this experiment
that the solubility of this mixture in spirit of wine, which is not a property of the
component parts, is a property that emerges from the compounding of sulfur
and salt of tartar. Regarding the property of salinity, he concludes that “whether
we allow the Epicurean hypothesis, or the Cartesian, the first saline concretions
that were produced by nature, must be confessed to have been made of atoms,
27 Boyle, Of the Producibleness of Chymicall Principles, 35.

or of particles, that before their conjunctions were not saline,” thereby implying
that the property of salinity emerges from the particular texture or structure
of the chymical atoms and from their interactions. It, thus, cannot simply be
accounted for mechanistically. Further in the same essay, Boyle concurs with
both Paracelsus and Helmont that salts are producible from ingredients that are
not salts, when he states that “salt may be made of matter, that was not salt before,
and consequently that salt may be de novo produced.”28 He then devotes the rest
of this essay to discussing specific experiments for the production of acid, vola-
tile, and lixiviate (or alkaline) salts.
Another example of the acquisition of novel properties is discussed in
Experiments and Notes about the Mechanical Origine and Production of Volatility,
in which Boyle claims that “the same material parts of a portion of corporeal
substance, which, when they were associated and contexed (whether by an
archeus, seed, form, or what you please,) after such determinate manner, consti-
tuted a solid and fixt body, as a Flint or a lump of Gold; by having their Texture
dissolved, and (perhaps after being subtilized) by being freed from their former
implications or firm cohesions, may become the parts of a fluid body totally
Volatile.”29
As was argued extensively in Chapter 3, although chymical atoms are chem-
ical composites, it is not simply their composition that determines the properties
and the species of a chemical substance. Rather, the structure in which the parts
are composed determines the properties of the substance. If the same number
and type of minima naturalia were to be arranged differently, that is, according
to a different structure, the chemical and other non-mechanical qualities of the
material body would be different. The structure displays a complex mereology in
which the whole is something more “over and above” its parts and gives rise to
“novel” non-mechanical properties that are no properties of its parts.
The “something more,” however, is not anything itself substantial nor is it a
substantial form. Instead, the “something more” is simply the essential material
form of the microstructure, which is responsible for the properties of the whole
(chymical atom) that are not present in its discrete parts (fundamental mech-
anistic corpuscles). It is, as already established, the essential parts of the whole
that account for the novel properties of the whole, which are not displayed by the
fundamental particles that are its integral parts. The perceptible chemical sub-
stance or compounded body, with its various sensible and chemical properties,
is simply a composite of chymical atoms that has the properties it has because of
these chymical atoms. If and when the structure of these atoms is altered in any
way, the substantial part of the substance will have been changed, giving rise to a
28 Ibid., 378.
29 Boyle, Experiments and Notes about the Mechanical Origine and Production of Volatility, 432.
different substance. But Boyle can explain how this occurs much more satisfacto-
rily than could medieval metaphysicians.
Medieval and Paracelsian alchemists, for example, could not account for how
one immaterial substantial form could “transmutate” into another. Nor could
they explain how one could analyze potassium nitrate and resynthesize it, since
this would imply a loss of substantial form followed by a reacquisition of the
same substantial form. Boyle, on the other hand, would simply say that transmu-
tation is simply the alteration of the geometrical structure of the chymical atoms
that generates a new species of material body. Redintegration occurs because
the original microstructure of a substance is restored. But these are all material,
physical processes that do not depend upon appealing to empirically suspect im-
material forms or principles.
Although, according to Henry, Boyle denies that any real novelty arises in
wholes that is not present in their component parts, the passage cited earlier
clearly conflicts with Henry’s claim. Instead, the passage cited earlier and other
passages from Boyle’s works support my contention that Boyle is simply rejecting
the theory of substantial form but not the idea that mereological wholes can dis-
play properties that are novel “over and above” those of their individual parts. In
fact, the passage from The Origin of Forms and Qualities cited by Henry simply
states, “not that there is really anything of substantial produced,” meaning that
no new matter is produced that did not previously exist in the individual parts.
However, there is nothing in this passage or any other to indicate that Boyle
would endorse the same sort of permissive mereology for which Abelard argued.
As David Armstrong points out, mereological permissiveness “appears to have
no real metaphysical consequences. Mereological wholes supervene on their
parts, as do the parts supervene on the whole. Given one, the other is entailed.
And on the basis of this it may be concluded that the wholes are no increase of
being beyond that of their parts.”30
For Boyle, however, there are definite metaphysical consequences to rearran-
ging the same parts according to a different structure or texture. When this
occurs, a new determinate species of matter is produced, albeit no new matter
as such. Relevantly, this new species of matter has different chemical and other
non-mechanical properties from the original species. As Boyle very explicitly
states in the passage that is misinterpreted by Henry, “those parts of matter that
did indeed before pre-exist [. . .] are now brought together and disposed after
the manner requisite to entitle the body that results from them to a new denom-
ination, and make it appertain to such a determinate species of natural bodies,”
meaning that the new structural arrangement of the parts results in a new natural
30 Armstrong, A World of States of Affairs, 120.

kind or new type of material body. This results from the pre-existing material
parts having acquired a “new modification” and, thus, having become a new type
of whole.
Later in the same passage, in a part not cited by Henry, Boyle adds that what
distinguishes one species of natural body from another are those properties that
are “necessary and sufficient to constitute any one Determinate Species of things
corporeal.”31 As established in Chapter 3, Boyle believes that different species of
bodies are distinguished by different sets of necessary and sufficient properties.
Clearly, then, if a new species of body comes into being when the same parts are
structurally rearranged, then the difference in the properties must be accounted
for by the new structure and clearly not by any difference in parts. This implies,
however, that something de novo is indeed produced from the new structural
arrangement, and what is produced de novo are the new emergent properties
whose presence characterizes a different species of material body. Thus, unlike
the scattered mereological sums discussed by Abelard, whose properties remain
the same after being brought together so that nothing new is added, the mereo-
logical arrangements of material parts referred to as chymical atoms bring about
the existence of emergent properties that were not present prior to that particular
structural arrangement.
All this is explained by the fact that, although Boyle’s chymical atoms lack
substantial form, they are endowed with essential form by way of their texture
or structure. Thus, although they do not qualify as continuous wholes for lack
of substantial form, they have a unity that is not present in contiguous wholes.
Chymical atoms are not simply constituted by the close adhaesion of particles
but by the structure according to which those particles are arranged. It is the
structure that determines the properties of chymical atoms and that determines
the particular material species to which those chymical atoms belong. On the
one hand, Boyle’s rejection of substantial forms implies a rejection of chymical
atoms as continuous integral wholes and, on the other hand, his rejection of
mereological extensionalism implies a rejection of chymical atoms as contiguous
integral wholes. Thus, I wish to argue that Boyle’s chymical atoms are neither
continuous integral wholes nor contiguous integral wholes but, rather, integral
wholes simpliciter.
In fact, Boyle’s non-reductionist mechanical philosophy implies the rejec-
tion of the continuous/contiguous distinction for all material substances, not
just for chymical atoms. As was discussed in Chapter 3, Boyle firmly believes
that specific structure or essential form is that which distinguishes one mate-
rial species from another, even if the simple parts are the same in both material

species. Furthermore, this implies a rejection of the Scholastic distinction be-

tween natural and artificial substances, since neither type of substance has sub-
stantial form and both types of substances have essential form. As well, although
material substances are not endowed with substantial forms, their unity is
not reducible to a mere collection of simple parts but is something “over and
above” those parts. Thus, as already concluded, Boyle’s non-reductionist mech-
anistic corpuscularianism implies that, at least as far as material substances are
concerned, there is no distinction between continuous wholes and contiguous
wholes. If a material substance is a unified whole at all, it is an integral whole sim-
pliciter, whether that substance be natural or artificial.
5.6 A Brief Excursion into the Mereology

of Epicurean Semantics
In An Introduction to the History of Particular Qualities, Boyle explains the ways

in which the various affections of matter produce the varieties of qualities that
exist in nature. In doing so, Boyle appeals to an analogy made by the ancient at-
omist Lucretius, who compares the manner by which the atoms that exist in the
universe come together to produce the variety of things to the manner by which
the twenty-four letters of the alphabet come together to produce the various lan-
guages of the world. Boyle prefaces the discussion of this analogy by first listing
the eleven affections that govern matter in the production of qualities. The first
four are “the most primary and simple affections of matter,”32 which are local
motion, size, shape, and rest. The other seven affections are posture, order, texture,
porousness, effluvia, mixture or composition, “and all of them governed as well
by . . . the universal fabric of things, as by the laws of motion established by the
Author of nature in the world.”33 Boyle then continues:
And now . . . that we have enumerated 11 very general affections of matter,

which with itself make up 12 principles of variation of bodies, let me on the
behalf of the Corpuscularians apply to the origins of qualities a comparison of
the old atomists, employed by Lucretius and others to illustrate the production
of an infinite number of bodies from such simple fragments of matter as they
thought their atoms to be. For since of the 24 letters of the alphabet, associ-
ated several ways as to the number and placing of the letters, all the words of
the several languages in the world may be made, so, say these naturalists, by
variously connecting such and such numbers of atoms, of such shapes, sizes,
32 Ibid., 105.
33 Ibid., 106–107.
and motions, into masses or concretions, an innumerable multitude of different

bodies may be formed [. . .] supposing these eleven principles were but so many
letters of the alphabet that could be only put together in differing numbers and
in various orders, the combinations and other associations that might be made
of them may be far more numerous than you yourself will expect [. . .] of so few
things so many associations may be made, each of which will differ from every
one of the rest, either in the number of the things associated or in the order
wherein they were placed.34
It is fortuitous that Boyle should employ Lucretius’ analogy between words

and things to explain the origin of qualities since, just as the associations of a fi-
nite number of letters can generate an infinity of words, the variety of qualities
and phenomena originate from the combination of a finite set of affections. To
the extent that Boyle uses this Lucretian analogy, it would seem that the eleven
general affections of matter function very much like rules of grammar, in that
these principles govern the multitude of ways in which particles of matter can
associate and combine to give rise to a multitude of qualities.
Although Boyle does not specifically use this analogy with reference to aggre-
gate corpuscles or chymical atoms, the Lucretian analogy between atoms and let-
ters is very apt for our own discussion to the extent that the mereology of chymical
atoms can itself be analogized to the mereology of linguistic expressions. What
characterizes linguistic expressions (words, propositions, etc.) as wholes and
distinguishes them from their discrete parts (i.e., letters) is that expressions
have meaning while their discrete parts do not. Clearly, a linguistic expression
is endowed with something “over and above” its individual parts, that is, the
meaning of the words transcends the individual letters. Yet we can also say that
there is nothing materially present in the whole other than the letters of which
it is made up. But, in coming together in particular configurations, meaning-
less parts (letters) give rise to meaningful simple wholes (words). These words,
in turn, become parts of complex wholes (propositions) whose own meaning
transcends the meaning of the individual words, and so on. We can say that lin-
guistic mereology involves the combinations and recombination of meaningless
parts into meaningful wholes of varying degrees of complexity. But, the meaning
of linguistic expressions, whether they be singular words or complex phrases,
is not simply constituted by bringing the letters together. These letters must be
combined in a particular kind of order, following the specific syntactic and se-
mantic principles that are codified in the given language. The language itself is
34 Boyle, An Introduction to the History of Particular Qualities, 107.

constituted within the context of a particular culture and is heavily dependent

upon practice and tradition.
While embracing an atomistic idea of linguistic expressions as collections
of letters in much the same way that material bodies are collections of atoms,
Lucretius and other Epicureans were also keenly aware of the contextuality
of linguistic meaning. For Epicureans, although expressions are ontological
composites of letters, they acquire their meanings within the context of partic-
ular linguistic and cultural practices, “a notion of context that is not fixed but at
once determining of and determined by the speech act.”35 Thus, the meaning of
linguistic expressions is relational in many different ways. Regarding speech act
theory, Wilson Shearin explains that “of course, Epicurus and his disciples can
hardly be accused of developing in every detail a theory that was not well artic-
ulated until the second half of the twentieth century, but Epicurean thinking on
language anticipates [. . .] the primary contribution of speech act theory [. . .]
the recognition and articulation of the performative. It may also grasp nuances
about context in speech act theory.”36 Within Epicurean semantics, the role of the
speech act is performative, so that Epicurus locates his theory of language within
a theory of action, that is, of practice. And, for him as for Lucretius, the per-
formative role of speech acts fulfills a central function within their therapeutic
conception of philosophy. Indeed, Shearin notes that, for the Epicureans, “a ther-
apeutic philosophy would have great interest in acting upon the world.”37
What is interesting in this analysis of Epicurean semantics in relation to
the mereology of linguistic expressions is that, although Lucretius and other
Epicureans believed that words and other expressions were, materially speaking,
sums of letters, there is something “over and above” the letters that makes the
whole into an expression rather than simply into a collection of meaningless dis-
crete parts. This “something” is meaning. Although the meaning of the whole
transcends the discrete parts, it not an ontologically independent entity in the
Platonic sense. Rather, it emerges from and supervenes upon the collection of
letters, making the whole something greater than the sum of its parts. Secondly,
semantics clearly implies a rejection of mereological extensionalism, since the
same letters arranged differently acquire a completely different meaning, for
example, “end” and “den.” Thirdly, meaning does not emerge simply as a result
of bringing letters together in particular ways. As Edmund Husserl pointed
out, many centuries after the Epicureans, not all collections of letters are mean-
ingful expressions. For Husserl, what is required to make a sum of letters into an
35 Shearin, The Language of Atoms, 16.

36 Ibid.
37 Ibid., 19.
expression is a meaning-intending act.38 For the Epicureans, who clearly could

not have adopted a transcendental phenomenological conception of meaning,
what is required is a performative speech act, to use 20th-century vocabulary,
through which the expression both acquires meaning and acts upon the world in
transformative ways. Such speech acts are meaningful when they abide by partic-
ular contexts of linguistic and cultural practice.
Three ideas stand out from this brief discussion of Epicurean semantics. First,
the meaning of a linguistic expression transcends the sum of the parts (letters)
that constitute the whole (expression). Second, semantics implies a rejection of
mereological extensionalism. Third, meaning emerges from the relation between
the parts, the whole, and the context or environment, which includes linguistic
and cultural practice. Thus, albeit ontologically atomistic, Epicurean semantics
rejects the idea that the meaning of expressions is simply reducible to the parts or
to the sum of the parts of which those expressions are made. In fact, the meaning
is not even entirely reducible to the order in which those parts are structured.
There is also a crucial relational and contextual element in the constitution of
linguistic meaning.
My brief excursion into the mereology of Epicurean semantics was spurred
by Boyle’s direct reference to the Lucretian analogy between words and things
in the History of Particular Qualities to explain that a finite number of princi-
ples can result in an almost infinite number of combinations of material parts,
which themselves produce an almost infinite variety of qualities. Yet the analogy
is also useful to elucidate how, just as linguistic meanings transcend the sum of
letters, chemical properties transcend the sum of primary corpuscles of which
chymical atoms are constituted. We can equally surmise that chemical proper-
ties emerge from and supervene upon the microstructural arrangement of pri-
mary corpuscles that constitute chymical atoms in an analogous manner to that
in which meanings emerge from and supervene upon the sums of letters that
constitute expressions.
This is not to imply that meanings and properties are ontologically of the same
type. Properties are perceivable through our senses and have causal power to act
upon our senses and upon other material bodies. Meanings, on the other hand,
are not sensible properties, although speech acts certainly have causal impact
upon the world. Nevertheless, to the extent that both Lucretius the atomist and
Boyle the corpuscularian both avail themselves of the analogy between words
and things, we can unpack this analogy further and see that atomism, whether
physical or linguistic, does not imply a simplistic extensional mereology in
38 For a detailed account of the relationship between meaning-intending acts, meaning-fulfilling
acts, and linguistic expressions, see: Marina Paola Banchetti-Robino, “Husserl’s Theory of Language
as Calculus Ratiocinator,” 303–321.
which the whole is nothing over and above its parts. Just as rearranging the let-
ters in a word without changing the letters themselves alters the meaning of the
word, altering the microstructure or texture of chymical atoms without altering
their composition will alter the chymical atoms and their properties.
Although Boyle could not possibly have understood the concept of isom-
erism, his notion of substance was clearly not simplistically compositional. Boyle
understood quite clearly that composition by itself is not enough to determine
the properties of a chemical substance, since two chymical atoms with the same
composition but different structures will display different chemical properties.
However, as was argued in Chapter 4, although the microstructure of chymical
atoms determines the non-mechanical properties that these atoms are disposed
to display, it is the chemical substance’s relation to other substances in the context
of specific chemical processes and procedures that allows dispositional proper-
ties to emerge and causally affect changes in other material bodies. For Boyle,
then, chemical substances are functional wholes whose properties emerge not
only from the structural ordering of their parts but also from their relationship
with other chemical substances in the context of experimental practice.
Conclusion
Thomas Kuhn once stated that Robert Boyle “was a man who brought to its
most developed form a type of chemical conceptualization consonant with
a major tendency of the scientific thought of his day. His failure to exert an
important influence upon the future course of chemical theory was due, not
to an inability to ‘fit’ his scientific contributions to ‘the times,’ but to spe-
cific shortcomings of his chemical doctrines themselves.”1 Although recent
authors, such as William Newman, have successfully challenged Kuhn’s view
that Boyle had little influence on the development of chemistry,2 this book
purports to establish that Boyle was more than a historically influential
chemist. He was also an important philosophical figure whose ideas were not
merely consonant with the scientific and philosophical thought of his day but
may have, at times, surpassed it. Boyle anticipated many of the fundamental
questions to be raised by later philosophers of chemistry, questions regarding
ontological and epistemological reduction, emergence, supervenience, and
disciplinary autonomy, as well as questions regarding structural disposition,
the relationality of properties, and the mereology of higher and lower levels
of organization.
Issues regarding chemistry’s scientific autonomy are themselves directly re-
lated to the reducibility of chemistry to physics. In fact, the question of reduc-
ibility has been called “one of the main areas in which philosophical interest
in chemistry should be directed.”3 Since the ontological dependency of chem-
ical properties on fundamental physical states is not at issue, the sort of reduc-
tion being considered is epistemic rather than ontological, and the question is
“whether our current description of chemistry can be reduced to our most fun-
damental current description of physics, namely quantum mechanics—and with
its explanatory consequences.”4 The reducibility of chemical laws to physical
laws would necessitate that physical laws be universal and fundamental and that
1 Kuhn, “Robert Boyle and Structural Chemistry,” 15.

2 See, for example: Newman, “Robert Boyle, Transmutation, and the History of Chemistry before
Lavoisier.”
3 Scerri and McIntryre, “The Case for the Philosophy of Chemistry,” 214.
4 Ibid.
University Press.
DOI: 10.1093/oso/9780197502501.001.0001
Conclusion 169
all laws in the special or secondary sciences be non-fundamental instantiations

of these more general physical laws. The evidence at this point seems to weigh
against the reducibility of chemical laws to the extent that at least some of these
laws can be considered fundamental.
Although some philosophers are firmly convinced that epistemic and explan-
atory reduction is both desirable and possible, at least in principle, chemists and
philosophers of chemistry have serious doubts about whether the ontological
dependency of chemical states upon physical states undermines the epistemic
and explanatory autonomy of chemistry. In this regard, Boyle’s views anticipated
those of contemporary philosophers of chemistry to the extent that he did not
consider chemistry to be a “derivative” science at all but an autonomous practice.
He considered chemistry to be a practice both theoretically and methodologi-
cally independent from mechanics and physics. Although the mechanization of
corpuscularianism in the work of Gassendi and Descartes opened the door for
the naturalization of all causes and principles, at least as far as inanimate bodies
were concerned, Boyle was well aware that the mechanical philosophy could not
successfully provide a complete account of chemical properties and processes.
As this book demonstrates, Boyle’s rejection of reductively mechanistic
explanations for chemistry shows that he “did not consider chemistry as a branch
of physics . . . [and] did not reduce all chemical phenomena to the geometrico-
mechanical affections of inert matter.”5 Boyle did not believe that chemistry
should be subordinated to the mechanical philosophy6 and he distanced him-
self from philosophers such as Spinoza, who attempted explanations of chem-
ical reactions by appealing to the mechanistic properties of fundamental
particles. Ultimately, as an experimental scientist, Boyle could not bring him-
self to adopting any rigid and unrelentingly reductionist approach to chemical
explanations.
In addition to the explanatory autonomy of chemistry, another important
issue in the contemporary philosophy of chemistry is the question of the emer-
gence of chemical properties from relationships at more fundamental physical
levels. When the concept of emergence is analyzed further, it involves a number
of interdependent features. These are novelty of properties, supervenience,
and non-summative difference. There is, in fact, ample support for the notion
of emergence of chemical properties, since they are clearly novel properties not
found at lower ontological levels. However, when discussing emergence and
supervenience, it is important also to include a discussion of relationality, and
these questions are themselves closely linked to the larger question of the afore-
mentioned autonomy of chemical explanations.

6 Ibid.
170 Conclusion
Although this book admits that Boyle lacked the philosophical vocabulary
necessary to speak about emergence and supervenience, his writings display
a very sophisticated understanding of these underlying issues within the con-
text of his chemical philosophy. There is, in fact, evidence in his writings that
he regarded chemical properties as emergent properties that supervene on the
mechanistic property of texture or structure. Boyle was able to accommodate
this emergentist position within a generally mechanistic theory of matter by
developing a hierarchical corpuscularian theory that distinguished between
different orders of compounded corpuscles. Of these compounded corpuscles,
chymical atoms were the operationally elementary and stable chemical wholes
from whose structure chemical properties emerged. By developing this hybrid
chemical ontology, Boyle stressed the importance of chemical microstructure
long before the development of structural chemistry in the mid-19th century.
Once structure is taken into account, mereological questions inevitably arise re-
garding the relations between chemical wholes and their parts, which is why it was
imperative to devote one of the central chapters of this book to the mereological
analysis of chymical atoms. Robert Boyle understood exactly what Robert Mulliken
would state three centuries later, that is, molecules have properties that their com-
ponent atoms lack because a molecule is considered as “a composite in which the
atoms [lose] their singularity.”7 Boyle believed that chemical reactions occurred at
the level of chymical atoms, that is, higher-order corpuscular concretions whose
properties were distinct properties from those of their more fundamental parts.
Given the extensive discussion in this book, it is safe to conclude that Boyle would
be in agreement with Mulliken’s claim that “a molecule has properties which the
atoms do not express and its decomposition gives again the separate atoms . . . This
molecular composite creates new homogeneous bodies starting from heteroge-
neous elements . . . Thus the composite challenges the mere sum of the parts.”8
It is very likely that Boyle would have also agreed with William Newman’s claim
that “the generation of supervenient properties from subvenient ones that is so
characteristic of the subject of chemical studies arises as a result of an interaction
of parts that arises when such parts are in a certain configuration.”9 Furthermore,
Boyle seems to have regarded the relationship between mechanical and chemical
properties as an asymmetric dependence and to have suggested that chemical
properties emerge from and are supervenient upon mechanical properties. In
this regard, Boyle seems to have anticipated contemporary explanations for the
supervenience of chemical properties on subvenient physical properties. As Eric
Scerri and Lee McIntyre assert, “the supervenience argument would entail that
7 Llored, “Mereology and Quantum Chemistry,” 204.

8 Ibid.
Conclusion 171
if two compounds share the same macroscopic property . . . we could not nec-
essarily infer that the microscopic components from which the compounds are
formed would be identical.”10
As this book explains, Boyle also believed that different concretions of pri-
mary particles with quantitatively different mechanical properties could ex-
press the same chemical properties, such as salinity, for example. Thus, Boyle
refrained from “establishing a direct relationship between a given quality and
a set of mechanical properties of the simplest corpuscles.”11 His position aligns
with the understanding of emergent properties as defined by contemporary
philosophers, who establish that a property is emergent “if it is in some way novel
or unpredictable, given the behaviours of the properties displayed by the system’s
elementary components.”12 Boyle’s view that chemical properties emerge from
and supervene upon mechanical properties suggests that he regarded chemical
explanations as irreducible to physical explanations so that chemistry must re-
tain its autonomy from physics and mechanics.
This book’s in-depth analysis of Boyle’s chemical philosophy is, therefore, not
merely historically relevant but also philosophically germane. Once again, it
establishes that Boyle anticipated many of the fundamental questions raised by the
modern practice of chemistry and by contemporary philosophy of chemistry. These
questions regard ontological and epistemological reduction, structural disposition,
the relationality of properties, and the mereology of chemical wholes. Although as a
16th-century chemist, Boyle lacked the appropriate philosophical vocabulary to ex-
plicitly articulate many of these ideas, his writings clearly address the same concerns
as those reflected in the contemporary philosophical literature on chemistry.
In order to address these concerns, he postulated a complex and nuanced hy-
brid and hierarchical chemical ontology, whose purpose was to accommodate
the emergence and supervenience of chemical properties within the frame-
work of the mechanical philosophy. This chemical ontology also allowed him
to defend the autonomy of chemical laws and chemical explanations against
those who would subordinate chemistry to mechanics and physics. Although
chemistry and chemical ontology have advanced quite far since the time of
Boyle, his mereological conception of chymical atoms and his perspective on
the relationality and emergence of chemical properties also seem very much in
tune with the positions reflected in some of the most important contemporary
work in the philosophy of chemistry. In this sense, Boyle’s chemical philosophy
is certainly deserving of an in-depth study as it was both prescient and, in many
regards, well ahead of its time.
10 Scerri and McIntyre, “The Case for the Philosophy of Chemistry,” 225.
12 Newth and Finnigan, “Emergence and Self-Organization in Chemistry and Biology,” 842.
Bibliography
Primary Sources
Abelard, Peter. Pietro Abelardo: Scritti di Logica. Edited by Mario Dal Pra. Firenze: La
Nuova Italia Editrice, 1969.
Albert the Great. Book of Minerals. Translated by D. Wyckoff. Oxford: Clarendon
Press, 1967.
Aquinas, Thomas. Liber de veritate Catholicae fidei contra errores infidelium, seu Summa
contra gentiles. Edited by C. Pera, P. Marc, and P. Carmello. Rome: Marietti, 1961–1967.
Aristotle. The Complete Works of Aristotle: The Revised Oxford Translation, Vols. 1 and 2.
Edited by Jonathan Barnes. Princeton: Princeton University Press, 1984.
Boyle, Robert. About the Excellency and Grounds of the Mechanical Hypothesis. In Selected
Philosophical Papers of Robert Boyle. Edited with an introduction by M. A. Stewart,
138–154. Cambridge: Hackett Publishing Company, 1991.
Boyle, Robert. Certain Physiological Essays. In The Works of Robert Boyle, Vol. 2. Edited
by Michael Hunter and Edward B. Davis, 4–204. London: Pickering and Chatto, 2000.
Boyle, Robert. Cosmicall Suspicions (Subjoyned as an Appendix to the Discourse of the
Cosmicall Qualities of Things). In The Works of Robert Boyle, Vol. 6. Edited by Michael
Hunter and Edward B. Davis, 302–316. London: Pickering and Chatto, 2000.
Boyle, Robert. A Disquisition about the Final Causes of Natural Things. In The Works
of Robert Boyle, Vol. 11. Edited by Michael Hunter and Edward B. Davis, 80–152.
London: Pickering and Chatto, 2000.
Boyle, Robert. An Essay of Various Degrees or Kinds of the Knowledge of Natural Things,
Vol. 8, Fols. 165–170. In Royal Society Boyle Papers, Online: http://www.bbk.ac.uk/
boyle/papers/introduction.
Boyle, Robert. Experiments and Notes about the Mechanical Origine and Production of
Volatility, 422–442. In The Works of Robert Boyle, Vol. 8. Edited by Michael Hunter and
Edward B. Davis. London: Pickering and Chatto, 2000.
Boyle, Robert. Experiments and Considerations Touching Colours. In The Works of Robert
Boyle, Vol. 4. Edited by Michael Hunter and Edward B. Davis, 4–202. London: Pickering
and Chatto, 2000.
Boyle, Robert. History of Particular Qualities. In The Works of Robert Boyle, Vol. 6.
Edited by Michael Hunter and Edward B. Davis, 268–285. London: Pickering and
Chatto, 2000.
Boyle, Robert. An Hydrostatical Discourse. In The Works of Robert Boyle, Vol. 7. Edited by
Michael Hunter and Edward B. Davis, 140–196. London: Pickering and Chatto, 2000.
Boyle, Robert. An Introduction to the History of Particular Qualities. In Selected
Boyle, Robert. Letters and Papers of Robert Boyle, Vol. 2. Collections from the Royal Society.
Bethesda, MD: University Publications of America, 1990.
174 Bibliography
Boyle, Robert. The Mechanical Origin and Production of Volatility and Fixedness. In The
Philosophical Works of the Honourable Robert Boyle, Vol. 1. Edited by Peter Shaw, 369–
387. London: W. Innys, R. Manby, and T. Longman, 1738.
Boyle, Robert. Of the Atomicall Philosophy. In The Works of Robert Boyle, Vol. 13. Edited by
Boyle, Robert. Of the Excellency and Grounds of the Corpuscular or Mechanical Hypothesis.
In The Works of Robert Boyle, Vol. 8. Edited by Michael Hunter and Edward B. Davis,
104–117. London: Pickering and Chatto, 2000.
Boyle, Robert. Of the Imperfection of the Chemists’ Doctrine of Qualities. In Selected
Boyle, Robert. Of the Systematical or Cosmical Qualities of Things. In The Works of
Robert Boyle, Vol. 6. Edited by Michael Hunter and Edward B. Davis, 286–300.
London: Pickering and Chatto, 2000.
Boyle, Robert. The Origin of Forms and Qualities. In The Works of Robert Boyle, Vol. 5.
Edited by Michael Hunter and Edward B. Davis, 282–483. London: Pickering and
Chatto, 2000.
Boyle, Robert. Of The Producibleness of Chymicall Principles. In The Works of Robert Boyle,
Vol. 9. Edited by Michael Hunter and Edward B. Davis, 20–121. London: Pickering and
Chatto, 2000.
Boyle, Robert. The Sceptical Chymist. In The Works of Robert Boyle, Vol. 2. Edited by
Boyle, Robert. Title Material Relating to Appendix to “Final Causes.” In Royal Society Boyle
Papers, Vol. 9, Folio 40v. http://www.bbk.ac.uk/boyle/papers/volume-9/volume-9-
fol.40v-41r.
Boyle, Robert. Of Men’s Great Ignorance of the Uses of Natural Things. In The Works of the
Honourable Robert Boyle, Vol. 3. Edited by Thomas Birch, 470–494. London: Thomas
Birch, 1774.
Bruno, Giordano. On the Infinite Universe and Worlds. In Giordano Bruno, His Life and
Thought. Edited and translated by Dorothea Waley Singer, 229–378. New York: Henry
Schuman, 1950.
de Cordemoy. Le discernement du corps et de l’âme en six discours, pour servir
d’éclaircissement à la physique. Paris: Florentin Lambert, 1666.
de Lavoisier, Antoine-Laurent. Traité élémentaire de chimie. Paris: Cuchet, 1789.
Descartes, René. The Correspondence between Descartes and Henricus Regius /De
briefwisseling tussen Descartes en Henricus Regius. Series: Quaestiones Infinitae:
Publications of the Department of Philosophy, Vol. 37. Utrecht: Utrecht University, 1969.
Descartes, René. Letter to Mersenne (April 15, 1630). In The Philosophical Writings of
René Descartes, Vol. 3. Translated by John Cottingham, Anthony Kenny, Dugald
Murdoch, and Robert Stoothoff, 20–23. Cambridge: Cambridge University Press,
1991.
Descartes, René. Oeuvres de Descartes. Edited by Charles Adam and Paul Tannery.
Paris: Léopold Cerf, 1905.
Descartes, René. Oeuvres philosophiques de Descartes. Edited by M. L. Aimé-Martin.
Paris: Auguste Desrez, 1838.
Fabre, Pierre-Jean. Traicté de la peste, selon la doctrine des médecins spagyriques.
Toulouse: Raimond Colomiez, 1629.
Galilei, Galileo. Il Saggiatore. Rome, 1623.
Bibliography 175
Homberg, Wilhelm. “Essays de chimie. Article premier: Des principes de la Chimie en

général.” In Mémoires de l’Académie Royale des Sciences, 34–46. Paris: Gabriel Martin,
Jean-Baptiste Coignard, & les Frères Guerin, 1702.
Kim, Mi Gyung. Affinity, That Elusive Dream: A Genealogy of the Chemical Revolution.
Cambridge, MA: The MIT Press, 2003.
Lémery, Nicolas. Cours de chymie. Paris: Estienne Michallet, 1683.
Oldenburg, Henry. The Correspondence of Henry Oldenburg. Edited by Rupert A. Hall and
Marie B. Hall. Madison: University of Wisconsin Press, 1965.
Ranelagh, Katherine. Letter to Robert Boyle (29 July 1665). In The Works of the Honourable
Robert Boyle, Vol 6. Edited by Thomas Birch, 525–526. London: Thomas Birch, 1774.
Régis, Pierre-Sylvain. Cours entier de philosophie ou Système général selon les principes de
M. Descartes. Amsterdam: Huguetan, 1691.
Rohault, Jacques. Traité de physique. Paris: Librarie Juré, 1671.
Sennert, Daniel. De chymicorum cum Aristotelicis et Galenicis consensu ac dissensu.
Wittenberg: Zacharia Shurerum, 1619.
Spinoza, Benedictus M. The Collected Works of Spinoza. Edited by Edwin M. Curley.
Princeton: Princeton University Press, 1985.
van Helmont, Jan Baptista. De lithiasis. In Oriatrike Or, Physick Refined. English transla-
tion of Ortus medicinae by John Chandler. London: Lodowick Loyd, 1662.
van Helmont, Jan Baptista. De lithiasis. In Ortus medicinae. Lyon: Huguetan & Guillielmi
Barbier, 1667.
van Helmont, Jan Baptista. Gas aquae. In Ortus medicinae. Amsterdam, 1648. Reprint,
Brussels: Culture et Civilisation, 1966.
van Helmont, Jan Baptista. Imago fermenti. In Ortus medicinae. Lyon: Huguetan &
Guillielmi Barbier, 1667.
van Helmont, Jan Baptista. “Lettre à Marin Mersenne.” In Correspondance du P. Marin
Mersenne: Religieux minime. Edited by Armand Beaulieu, Cornelis De Waard, René
Pintard, and Bernard Rochot. Paris: Presses Universitaires de France & Centre National
de la Recherche Scientifique, 1946.
van Helmont, Jan Baptista. Progymnasma meterori. In Ortus medicinae. Lyon: Huguetan
& Guillielmi Barbier, 1667.
Venel, Gabriel François. Chymie ou Chimie. In Encyclopédie méthodique, ou par ordre de
matières. Edited by Denis Diderot and Jean-Baptiste le Rond d’Alembert. Paris: Henri
Agasse, 1786.
Secondary Sources
Alexander, Peter. Ideas, Qualities, and Corpuscles: Locke and Boyle on the External World.
Cambridge: Cambridge University Press, 1985.
Alexander, Peter. “The Names of Secondary Qualities.” Proceedings of the Aristotelian
Society 77 (1976–1977): 203–220.
Allen, Phyllis. “Scientific Studies in the English Universities of the Seventeenth Century.”
Journal of the History of Ideas 10, no. 2 (April 1949): 219–253.
Anstey, Peter. “Boyle on Seminal Principles.” Studies in History and Philosophy of Biology
and Biomedical Sciences 33 (2002): 597–630.
Anstey, Peter. “Branching Off: The Early Moderns in Quest for the Unity of Knowledge.”
British Journal for the History of Philosophy 19, no. 4 (2011): 819–822.
176 Bibliography
Anstey, Peter. “Essences and Kinds.” In The Oxford Handbook of Philosophy in Early
Modern Europe, edited by Desmond M. Clarke and Catherine Wilson, 11– 31.
Oxford: Oxford University Press, 2011.
Anstey, Peter. “Robert Boyle and the Heuristic Value of Mechanism.” Studies in History
and Philosophy of Science 33 (2002): 161–174.
Anstey, Peter R. The Philosophy of Robert Boyle. London: Routledge, 2000.
Armstrong, David M. A World of States of Affairs. Cambridge: Cambridge University
Press, 1997.
Atherton, Margaret. “Locke on Essences and Classification.” In The Cambridge Companion
to Locke’s Essay, 258–285. Cambridge: Cambridge University Press, 2007.
Ayers, Michael. “Locke Versus Aristotle on Natural Kinds.” The Journal of Philosophy 78,
no. 5 (May 1991): 247–272.
Banchetti-Robino, Marina Paola. “From Corpuscles to Elements: Chemical Ontologies
from Van Helmont to Lavoisier.” In Philosophy of Chemistry: Growth of a New Discipline,
edited by Lee McIntyre and Eric Scerri, 141–154. Boston Studies in the Philosophy and
History of Science 306. Dordrecht: Springer, 2014.
Banchetti-Robino, Marina Paola. “Husserl’s Theory of Language as Calculus Ratiocinator.”
Synthese 112, no. 3 (1997): 303–321.
Banchetti-Robino, Marina Paola. “The Ontological Function of First-Order and Second-
Order Corpuscles in the Chemical Philosophy of Robert Boyle: The Redintegration of
Potassium Nitrate.” Foundations of Chemistry 14, no. 3 (2012): 221–234.
Banchetti-Robino, Marina Paola. “Ontological Tensions in Sixteenth and Seventeenth
Century Chemistry: Between Mechanism and Vitalism.” Foundations of Chemistry 13,
no. 3 (2011): 173–186.
Banchetti-Robino, Marina Paola. “The Relevance of Boyle’s Chemical Philosophy for
Contemporary Philosophy of Chemistry.” In The Philosophy of Chemistry: Practices,
Methodologies, and Concepts, edited by Jean-Pierre Llored, 240–265. Newcastle upon
Tyne: Cambridge Scholars Publishing, 2013.
Banchetti-Robino, Marina Paola. “Van Helmont’s Hybrid Ontology and Its Influence
on the Chemical Interpretation of Spirit and Ferment.” Foundations of Chemistry 18
(2015): 103–112.
Banchetti- Robino, Marina Paola, and Jean- Pierre Llored. “Reality without
Reification: Philosophy of Chemistry’s Contribution to Philosophy of Mind.” In An
Oxford Handbook on the Philosophy of Chemistry, edited by Grant Fischer and Eric
Scerri, 83–110. Oxford: Oxford University Press, 2016.
Ben-Chaim, Michael. “The Value of Facts in Boyle’s Experimental Philosophy.” History of
Science 38 (2000): 57–77.
Bensaude- Vincent, Bernadette. “Philosophy of Chemistry or Philosophy with
Chemistry?” HYLE—Journal for Philosophy of Chemistry 20 (2014): 59–76.
Bensaude-Vincent, Bernadette, and Isabelle Stengers. Histoire de la chimie. Paris: La
Découverte, 1992.
Bensaude-Vincent, Bernadette, and Isabelle Stengers. A History of Chemistry. Cambridge,
MA: Harvard University Press, 1996.
Bianchi, Massimo Luigi. “The Visible and the Invisible: From Alchemy to Paracelsus.” In
Alchemy and Chemistry in the 16th and 17th Centuries, edited by Antonio Clericuzio
and Piyo Rattansi, 17–50. Dordrecht: Kluwer Academic Publishers, 1994.
Bitbol, Michel. “Downward Causation without Foundations.” Synthese 180 (2010):
1–23.
Bibliography 177
Bloor, David, “Durkheim and Mauss Revisited: Classification and the Sociology of
Knowledge.” In Society & Knowledge: Contemporary Perspectives in the Sociology of
Knowledge and Science, edited by Nico Stehr and Volker Meja, 67–92. New Brunswick,
NJ: Transaction Publishers, 2005.
Boas, Marie. “An Early Version of Boyle’s Sceptical Chymist.” Isis 45, no. 2 (July 1954):
153–168.
Boas, Marie. “The Establishment of the Mechanical Philosophy.” Osiris 10 (1952): 412–541.
Boas, Marie. “Quelques aspects sociaux de la chimie au XVIIe siècle.” Revue d’histoire des
sciences et de leurs applications 10, no. 2 (1957): 132–147.
Boas Hall, Marie. “Boyle’s Method of Work: Promoting His Corpuscular Philosophy.”
Notes and Records of the Royal Society of London 41, no. 2 (June 1987): 111–143.
Boas Hall, Marie. “La méthode scientifique de Robert Boyle.” Revue d’histoire des sciences
et de leurs applications 9, no. 2 (1956): 105–125.
Bunge, Mario. “Is Chemistry a Branch of Physics?” Zeitschrift für allgemeine
Wissenschaftstheorie 13 (1982): 209–223.
Buroker, Jill Vance. “Descartes on Sensible Qualities.” Journal of the History of Philosophy
29, no. 4 (October 1991): 585–611.
Buyse, Filip. “The Distinction between Primary Properties and Secondary Qualities in
Galileo Galilei’s Natural Philosophy.” Cahiers du Séminaire québécois en philosophie
modern /Working Papers of the Quebec Seminar in Early Modern Philosophy 1 (2014):
20–43.
Buyse, Filip. “Spinoza and Galileo Galilei: Adequate Ideas and Intrinsic Qualities of
Bodies.” Historia Philosophica 6 (2008): 117–127.
Buyse, Filip. “Spinoza and Robert Boyle’s Definition of Mechanical Philosophy.” Historia
Philosophica 8 (2010): 74–89.
Buyse, Filip. “Spinoza, Boyle, Galileo: Was Spinoza a Strict Mechanical Philosopher?”
Intellectual History Review 23, no. 1 (2012): 45–64.
Carlin, Laurence. “Boyle’s Teleological Mechanism and the Myth of Immanent Teleology.”
Studies in History and Philosophy of Science, Part A 43 (2012): 54–63.
Carlin, Laurence. “The Importance of Teleology to Boyle’s Natural Philosophy.” British
Journal for the History of Philosophy 19 (2011): 663–680.
Carvallo, Sarah. “Chimie et scepticisme: Héritage et ruptures d’une science. Analyse
du Chimiste sceptique, 1661, Robert Boyle.” Revue d’histoire des sciences 55, no. 4
(2002): 451–492.
Cecon, Kleber. “Chemical Translation: The Case of Robert Boyle’s Experiments on
Sensible Qualities.” Annals of Science 68, no. 2 (2011): 179–98.
Cecon, Kleber. “Um exemplo de negação do conceito de elemento na filosofia natural.”
História da ciência e ensino: Construindo interfaces 8 (2013): 68–89.
Cecon, Kleber. A relação entre a filosofia mecânica e os experimentos alquímicos
de Robert Boyle. Universidade Estadual de Campinas: Instituto de Filosofia e
Ciências Huimanas, 2010. Accessed March 1, 2015. http://repositorio.unicamp.
br/jspui/handle/R EPOSIP/2 80332
Cecon, Kleber. “Robert Boyle’s Experimental Programme: Some Interesting Examples
of the Use of Subordinate Causes in Chymistry and Pneumatics.” Intellectual History
Review 25 (2015): 81–96.
Cecon, Kleber. “A tradução química de experimentos alquímicos envolvendo água régia
em Robert Boyle.” Scientiæ Studia 10, no. 4 (2012): 711–732.
178 Bibliography
Chalmers, Alan F. “Boyle and the Origins of Modern Chemistry: Newman Tried in the
Fire.” Studies in History and Philosophy of Science 41 (2010): 1–10.
Chalmers, Alan F. “Klein on the Origin of the Concept of Chemical Compound,”
Foundations of Chemistry 14 (2012): 37–53.
Chalmers, Alan F. “The Lack of Excellency of Boyle’s Mechanical Philosophy.” Studies in
History and Philosophy of Science 24, no. 4 (1993): 541–564.
Chalmers, Alan F. One Hundred Years of Pressure: Hydrostatics from Stevin to Newton.
Archimedes: New Studies in the History and Philosophy of Science and Technology 51.
Dordrecht: Springer, 2017.
Chalmers, Alan F. “Retracing the Ancient Steps to Atomic Theory.” Science & Education 7
(1998): 69–84.
Chalmers, Alan F. “Robert Boyle’s Mechanical Account of Hydrostatics and
Pneumatics: Fluidity, the Spring of the Air, and Their Relationship to the Concept of
Pressure.” Archive for History of Exact Sciences 69 (2015): 429–454.
Chalmers, Alan F. The Scientist’s Atom and the Philosopher’s Stone: How Science Succeeded
and Philosophy Failed to Gain Knowledge of Atoms. Dordrecht: Springer, 2009.
Chang, Ku-ming. “Alchemy as Studies of Life and Matter: Reconsidering the Place of
Vitalism in Early Modern Chymistry.” Isis 102, no. 2 (June 2011): 322–329.
Chang, Ku-ming. “Fermentation, Phlogiston, and Matter Theory: Chemistry and Natural
Philosophy in Georg Ernst Stahl’s Zymotechnia Fundamentalis.” Early Science and
Medicine 7, no. 1 (2002): 31–64.
Christopoulou, Christiana. “Robert Boyle’s Experiments on Cold: A Study of the Role
of Chemical Experiments.” In Neighbours and Territories: The Evolving Identity of
Chemistry, edited by José Ramón Bertomeu-Sánchez, Ducan Thorburn Burns, and
Brigitte Van Tiggelen, 423–431. Louvain-la-neuve: Mémosciences, 2008.
Clericuzio, Antonio. “Alchimie, philosophie corpusculaire, et minéralogie dans la
Metallographia de John Webster.” Revue d’histoire des sciences 49, no. 2 (1996): 287–304.
Clericuzio, Antonio. “Carneades and the Chemists: A Study of The Sceptical Chymist and
Its Impact on Seventeenth-Century Chemistry.” In Robert Boyle Reconsidered, edited
by Michael Hunter, 79–90. Cambridge: Cambridge University Press, 2003.
Clericuzio, Antonio. Elements, Principles, and Corpuscles: A Study of Atomism and
Chemistry in the Seventeenth Century. Dordrecht: Kluwer Academic Publishers, 2000.
Clericuzio, Antonio. “From van Helmont to Boyle. A Study of the Transmission of
Helmontian Chemical and Medical Theories in Seventeenth-Century England.” British
Journal for the History of Science 26 (1993): 303–334.
Clericuzio, Antonio. “The Internal Laboratory: The Chemical Reinterpretation of Medical
Spirits in England (1650–1680).” In Alchemy and Chemistry in the 16th and 17th
Centuries, edited by Antonio Clericuzio and Piyo Rattansi, 51–83. Dordrecht: Kluwer
Academic Publishers, 1994.
Clericuzio, Antonio. “A Redefinition of Boyle’s Chemistry and Corpuscular Philosophy.”
Annals of Science 47 (1990): 561–589.
Clucas, Stephen. “The Atomism of the Cavendish Circle: A Reappraisal.” The Seventeenth
Century 9, no. 2 (1994): 247–273.
Clucas, Stephen. “‘The Infinite Variety of Formes and Magnitudes’: 16th-and 17th-
Century English Corpuscular Philosophy and Aristotelian Theories of Matter and
Form.” Early Science and Medicine 2, no. 3 (1997): 251–271.
Conry, Yvette. “Robert Boyle et la doctrine cartésienne des animaux-machines.” Revue
d’histoire des sciences 33, no. 1 (1980): 69–74.
Bibliography 179
Cook, Margaret G. “Divine Artifice and Natural Mechanism: Robert Boyle’s Mechanical
Philosophy of Nature.” Osiris, 2nd ser., vol. 16, Science in Theistic Contexts: Cognitive
Dimensions (2001): 133–150.
Cressy, David. Saltpeter: The Mother of Gunpowder. Oxford: Oxford University Press,
2013.
Curley, E. M. “Locke, Boyle, and the Distinction between Primary and Secondary
Qualities.” The Philosophical Review 81, no. 4 (October 1972): 438–464.
Daudin, Henri. “Spinoza et la science experimentale: Sa discussion de l’expérience de
Boyle.” Revue d’histoire des sciences et de leurs applications 2, no. 2 (1949): 179–190.
De Mey, Tim, and Markku Keinänen. “Secondary Qualities in Retrospect.” Philosophica
68, no. 2 (2001): 41–59.
De Pace, Anna. Le matematiche e il mondo: Ricerche su un dibattito in Italia nella seconda
metà del Cinquecento. Milan: Francoangeli, 1993.
Debus, Allen G. “Alchemy and Iatrochemistry: Persistent Traditions in the 17th and 18th
Centuries.” Química Nova 12, no. 3 (1992): 262–268.
Debus, Allen G. The Chemical Philosophy: Paracelsian Science and Medicine in the
Sixteenth and Seventeenth Centuries. New York: Dover Publications, 2002.
Debus, Allen G. “Chemists, Physicians, and Changing Perspectives on the Scientific
Revolution.” Isis 89, no. 1 (March 1998): 66–81.
Debus, Allen G. “Mathematics and Nature in the Chemical Texts of the Renaissance.”
Ambix 15 (1968): 1–28.
Debus, Allen G. “The Paracelsian Aerial Niter.” Isis 55 (1964): 43–61.
Debus, Allen G. “Solution Analyses Prior to Robert Boyle.” Chymia 8 (1962): 41–61.
Del Re, Giuseppe. “Technology and the Spirit of Alchemy.” HYLE: International Journal
for Philosophy of Chemistry 3 (1997): 51–63.
Dobre, Mihnea. “Cartesianism and Chymistry.” Society and Politics 2, no. 5, Disciplines
and Forms of Knowledge in Early Modern Thought (2011): 121–136.
Downing, Lisa Jeanne. “Are Corpuscles Unobservable in Principle for Locke?” Journal of
the History of Philosophy 30, no. 1 (January 1992): 33–52.
Downing, Lisa Jeanne. “Locke’s Ontology.” In The Cambridge Companion to Locke’s
“Essay Concerning Human Understanding,” edited by Lex Newman, 352– 380.
Ducheyne, Steffen. “Helmont’s Mechanical Experiments.” In Neighbours and
Territories: The Evolving Identity of Chemistry, edited by José Ramón Bertomeu-
Sánchez, Ducan Thorburn Burns, and Brigitte Van Tiggelen, 253–262. Louvain-la-
neuve: Mémosciences, 2008.
Ducheyne, Steffen. “Joan Baptista van Helmont and the Question of Experimental
Modernism.” Physis: Rivista Internazionale di Storia della Scienza 47 (2005): 305–332.
Duffy, Simon. “The Difference between Science and Philosophy: The Spinoza-Boyle
Controversy Revisited.” Paragraph 29, no. 2 (2006): 115–138.
Earley, Joseph. “Varieties of Properties: An Alternative Distinction among Qualities.” In
Chemical Explanation: Characteristics, Development, Autonomy, edited by Joseph E.
Earley, 80–89. New York: New York Academy of Sciences Publisher, 2003.
Earley, Joseph. “Why There Is No Salt in the Sea.” Foundations of Chemistry 7 (2005): 85–102.
Ehrlich, Mark E. “Mechanism and Activity in the Scientific Revolution: The Case of
Robert Hooke.” Annals of Science 52 (1995): 127–151.
Forshaw, Peter J. “Cabala Chymica or Chemia Cabalistica: Early Modern Alchemists and
Cabala.” Ambix 60, no. 4 (November 2013): 361–389.
180 Bibliography
Forshaw, Peter J. “‘Morbo spirituali medicina spiritualis convenit’: Paracelsus, Madness,

and Spirits.” In Aisthetics of the Spirits: Spirits in Early Modern Science, Religion,
Literature, and Music, edited by Steffen Schneider, 287–306. Göttingen: Vandenhoeck
and Ruprecht, V&R Unipress, 2015.
Francescotti, Robert M. “Emergence.” Erkenntnis 67 (2007): 47–63.
Franckowiak, Rémi. “La chimie du XVIIe siècle: Une question de principes.” Chimie et
mécanisme à l’âge classique 8 (2008): 1–43.
Franckowiak, Rémi. “Eighteenth Century Chemistry: Between Natural Philosophy
without Nature and Physics without Reason.” In Neighbours and Territories: The Evolving
Identity of Chemistry, edited by José Ramón Bertomeu-Sánchez, Duncan Thorburn
Burns, and Brigitte Van Tiggelen, 93–102. Louvain-la-neuve: Mémosciences, 2008.
Fulton, J. F. “Robert Boyle and His Influence on Thought in the Seventeenth Century.” Isis
18, no. 1 (July 1932): 77–102.
Gabey, Allan. “Spinoza’s Natural Science and Methodology.” In The Cambridge Companion
to Spinoza, edited by Don Garrett. Cambridge: Cambridge University Press, 1996.
Gardner, Michael R. “Realism and Instrumentalism in 19th- Century Atomism.”
Philosophy of Science 4, no. 1 (March 1989): 1–34.
Gatti, Hilary. “Giordano Bruno’s Soul-Powered Atoms: From Ancient Sources towards
Modern Science.” In Late Medieval and Early Modern Corpuscular Matter Theories, ed-
ited by Christoph Lüthy, John E. Murdoch, and William R. Newman, 164–180. Leiden:
Brill, 2001.
Gibson, James J. The Ecological Approach to Visual Perception. New York: Psychology
Press, 1979.
Gibson, James J. “The Theory of Affordances.” In Perceiving, Acting, and Knowing: Toward
an Ecological Psychology, edited by Robert Shaw and John Bransford, 67– 82.
New York: Lawrence Erlbaum, 1977.
Greene, Robert A. “Henry More and Robert Boyle on the Spirit of Nature.” Journal of the
History of Ideas 23, no. 4 (1962): 451–474.
Greiner, Frank, ed. Aspects de la tradition alchimique au XVIIe siècle: Actes du colloque in-
ternational de l’Université de Reims-Champagne-Ardenne. Paris: S.É.H.A., 1996.
Gregory, Joshua C. “The Animate and Mechanical Models of Reality.” Journal of
Philosophical Studies 2, no. 7 (1927): 301–314.
Halabi, Selman. “A Useful Anachronism: John Locke, the Corpuscular Philosophy, and
Inference to the Best Explanation.” Studies in the History and Philosophy of Science 36
(2005): 241–259.
Hall, Rupert A., and Marie Boas Hall. “The Intellectual Origins of the Royal Society:
London or Oxford.” Notes and Records of the Royal Society of London 23, no. 2
(December 1968): 157–168.
Hankins, Thomas L. Science and the Enlightenment. Cambridge: Cambridge University
Press, 1985.
Harré, Rom. “New Tools for Philosophy of Chemistry.” HYLE—International Journal for
Philosophy of Chemistry 20 (2014): 77–91.
Harré, Rom, and Jean-Pierre Llored. “Developing the Mereology of Chemistry.” In
Mereology and the Sciences, edited by Claudio Calosi and Pierluigi Graziani, Synthese
Library 371, 189–212. Dordrecht: Springer, 2014.
Harré, Rom, and Jean- Pierre Llored. “Mereologies as the Grammars of Chemical
Discourses.” Foundations of Chemistry 13 (2011): 63–76.
Bibliography 181
Heinecke, Berthold. “The Mysticism and Science of Johann Baptista van Helmont (1579–
1644).” Ambix 42, Part 2 (July 1995): 65–78.
Hendry, Robin Findlay. “Antoine Lavoisier (1743–1794).” In Philosophy of Chemistry, ed-
ited by Andrea I. Woody, Robin Findlay Hendry, and Paul Needham, 63–70. North
Holland: Waltham, MA, 2011.
Hendry, Robin Findlay. “Is There Downward Causation in Chemistry?” In Philosophy of
Chemistry: Synthesis of a New Discipline, edited by Davis Baird, Eric R. Scerri, and Lee
McIntyre, 173–190. Dordrecht: Springer, 2006.
Henry, Desmond Paul. Medieval Mereology. Amsterdam: B. R. Grüner Publishing
Co., 1991.
Henry, John. “Boyle and Cosmical Qualities.” In Robert Boyle Reconsidered, edited by
Michael Hunter, 119–138. Cambridge: Cambridge University Press, 2003.
Henry, John. “Occult Qualities and the Experimental Philosophy: Active Principles in
Pre-Newtonian Matter Theory.” History of Science 24 (1986): 335–381.
Hirai, Hirai. “Anatomie du chymiste sceptique: Robert Boyle et le secret des ses premières
sources sur la croissance des métaux.” In La philosophie naturelle de Robert Boyle, ed-
ited by Myriam Denehey and Charles Ramon, 93–116. Paris: Vrin, 2009.
Hirai, Hiro. “Atomes vivants, origine de l’âme, et génération spontanée chez Daniel
Sennert.” Bruniana & Campanelliana: Ricerche filosofiche e materiali storico-testuali 13,
no. 2 (2007): 477–496.
Hirai, Hiro. Le concept de semence dans les théories de la matière à la Renaissance: De
Marsile Ficin à Pierre Gassendi. Turnhout: Brepols Publishers, 2005.
Hirai, Hiro. “Ficin, Fernel, et Fracastor autour du concept de semence: Aspects
platoniciens de seminaria.” In Girolamo Fracastoro: Fra medicina, filosofia, e scienze
della natura, edited by Alessandro Pastore and Enrico Peruzzi, 245–260. Florence: L.
S. Olschki, 2006.
Hirai, Hiro. “Interpretation chymique de la création et origine corpusculaire de la vie chez
Athanasius Kircher.” Annals of Science 64, no. 2 (April 2007): 217–234.
Hirai, Hiro. “Living Atoms, Hylomorphism, and Spontaneous Generation in Daniel
Sennert.” In Matter and Form in Early Modern Science and Philosophy, edited by Gideon
Manning, 77–98. Leiden: Brill, 2012.
Hirai, Hiro. “Les logoi spermatikoi et le concept de semence dans la minéralogie et la
cosmogonie de Paracelse.” Revue d’histoire des science 61, no. 2 (2008): 245–264.
Hirai, Hiro, and Hideyuki Yoshimoto. “Anatomizing the Sceptical Chymist: Robert
Boyle and the Secret of His Early Sources on the Growth of Metals.” Early Science and
Medicine 10, no. 4 (2005): 453–477.
Holden, Thomas. “Robert Boyle on Things above Reason.” British Journal for the History of
Philosophy 15, no. 2 (2007): 283–312.
Hunter, Michael. “Robert Boyle and the Early Royal Society: A Reciprocal Exchange in
the Making of Baconian Science.” British Journal for the History of Science 40, no. 1
(March 2007): 1–23.
Hunter, Michael. “Robert Boyle’s Early Intellectual Evolution: A Reappraisal.” Intellectual
History Review 25, no. 1 (2014): 5–19.
Hutchison, Keith, “What Happened to Occult Qualities in the Scientific Revolution?” Isis
73, no. 2 (1982): 233–253.
Jacob, James R., and Margaret C. Jacob. “The Anglican Origins of Modern Science: The
Metaphysical Foundations of the Whig Constitution.” Isis 71, no. 2 (June 1980): 251–267.
182 Bibliography
Jacobs, Struan. “Laws of Nature, Corpuscles, and Concourse: Non- Occasionalist

Tendencies in the Natural Philosophy of Robert Boyle.” Journal of Philosophical Research
19 (1994): 373–393.
Joly, Bernard. “À propos d’une prétendue distinction entre la chimie et l’alchimie au XVIIe
siècle: Questions d’histoire et de méthode.” Revue d’histoire des sciences 60 (2007):
167–184.
Joly, Bernard. “Les alchimistes étaient-ils des matérialistes? Quelques remarques sur le
psychisme humain et l’esprit du monde.” In Les matérialismes et la chimie, edited by
François Pépin, 39–62. Paris: Éditions Matériologiques, 2012.
Joly, Bernard. “L’alkahest, dissolvant universel ou quand la théorie rend pensable une pra-
tique impossible.” Revue d’histoire des sciences 49, no. 2 (1996): 305–344.
Joly, Bernard. “Le cartésianisme de Boyle du point de vue de la chimie.” In La philosophie
naturelle de Robert Boyle, edited by Myriam Dennehy and Charles Ramond, 139–155.
Paris: Vrin, 2009.
Joly, Bernard. “Chimie et mécanisme dans la nouvelle Académie royale des sciences: les
débats entre Louis Lémery et Etienne-François Geoffroy.” Methodos 8 (2008). Accessed
April 6, 2009. https://journals.openedition.org/methodos/1403.
Joly, Bernard. Descartes et la chimie. Paris: Vrin, 2011.
Joly, Bernard. “Présences stoïciennes dans les théories des marées aux XVIe et XVIIe
siècles.” Revue d’histoire des sciences 61 (2008): 287–311.
Joly, Bernard. “Un siècle de chimie à l’Académie royale des sciences de sa création (1666)
à l’arrivée de Lavoisier (1768).” Methodos 12. Accessed October 25, 2013. https://
journals.openedition.org/methodos/2937.
Joly, Bernard, and Pierre-François Moreau. “La physique des stoïciens aux XVIe et XVIIe
siècles: Une présence cachée.” Revue d’histoire des sciences 61 (2008): 239–244.
Jones, Richard F. “Science and Language in England of the Mid-Seventeenth Century.”
The Journal of English and Germanic Philology 31, no. 3 (July 1932): 315–331.
Jones, Jan-Erik. “Boyle, Classification, and the Workmanship of the Understanding
Thesis.” Journal of the History of Philosophy 43, no. 2 (April 2005): 171–183.
Jones, Jan-Erik. “Locke vs. Boyle: The Real Essence of Corpuscular Species.” British
Journal for the History of Science 15, no. 4 (2007): 659–684.
Jones, Jan-Erik. “Lockean Real Essences and Ontology.” Southwest Philosophy Review 32,
no. 2 (2016): 137–162.
Kaitaro, Timo. “Can Matter Mark the Hours? Eighteenth-Century Vitalist Materialism
and Functional Properties.” Science in Context 21, no. 4 (2008): 581–592.
Kargon, Robert. “The Testimony of Nature: Boyle, Hooke, and Experimental Philosophy.”
Albion: A Quarterly Journal Concerned with British Studies 3, no. 2 (Summer 1971): 72–81.
Kargon, Robert. “Walter Charleton, Robert Boyle, and the Acceptance of Epicurean
Atomism in England.” Isis 55, no. 2 (June 1964): 184–192.
Kim, Jaegwon. “Making Sense of Emergence.” Philosophical Studies 95 (1999): 3–36.
Kim, Yung Sik, “Another Look at Robert Boyle’s Acceptance of the Mechanical
Philosophy: Its Limits and Its Chemical and Social Contexts.” Ambix 38, Part I (March
1991): 1–10.
Kirschner, Mark, John Gerhart, and Tim Mitchison. “Molecular ‘Vitalism’.” Cell 100
(January 2000): 79–88.
Klein, Joel A. “Corporeal Elements and Principles in the Learned German Chymical
Tradition.” Ambix 61, no. 4 (November 2014): 345–365.
Bibliography 183
Klein, Joel A., and Evan R. Ragland, “Analysis and Synthesis in Medieval and Early
Modern Europe.” Ambix 61, no. 4 (November 2014): 319–326.
Klein, Ursula. “Objects of Inquiry in Classical Chemistry: Material Substances.”
Foundations of Chemistry 14 (2012): 7–23.
Klein, Ursula. “Origin of the Concept of Chemical Compound.” Science in Context 7, no.
2 (June 1994): 163–204.
Knight, David. Atoms and Elements: A Study of Theories of Matter in England in the
Sixteenth Century. London: Hutchinson, 1967.
Koslicki, Kathrin. “Aristotle’s Mereology and the Status of Form.” The Journal of Philosophy
103 (December 2006): 715–736.
Kubbinga, H. H. “Les premières théories ‘moléculaires’; Isaac Beeckman (1620) et
Sébastien Basson (1621): Le concept d’ ‘individu substantiel’ e d’ ‘espèce substantielle’.”
Revue d’histoire des science 37, no. 3 (1984): 215–233.
Kuhn, Thomas. “Mathematical vs. Experimental Traditions in the Development of
Physical Science.” Journal of Interdisciplinary History 7, no. 1 (1976): 1–31.
Kuhn, Thomas. “Robert Boyle and Structural Chemistry.” Isis 43 (1952): 12–36.
Kultgen, J. H. “Boyle’s Metaphysic of Science.” Philosophy of Science 23, no. 2 (April 1956):
136–141.
Lakatos, Imre. “Criticism and the Methodology of Scientific Research Programmes.”
Proceedings of the Aristotelian Society 69 (1968–1969): 149–186.
Lakatos, Imre. “Falsification and the Methodology of Scientific Research Programmes.”
In Criticism and the Growth of Knowledge, edited by Imre Lakatos and Alan Musgrave,
91–196. Cambridge: Cambridge University Press, 1970.
Lakatos, Imre. The Methodology of Scientific Research Programmes. Cambridge: Cambridge
University Press, 1978.
Langermann, Y. Tzvi. “Islamic Atomism and the Galenic Tradition.” History of Science 47,
no. 3 (2009): 277–295.
Leicester, Henry M. The Historical Background of Chemistry. New York: Dover
Publications, 1956.
Lennox, James G. “Robert Boyle’s Defense of Teleological Inference in Experimental
Science.” Isis 74, no. 1 (March 1983): 38–52.
Levere, Trevor H. Transforming Matter: A History of Chemistry from Alchemy to the
Buckyball. Baltimore: The Johns Hopkins University Press, 2001.
Linden, Stanton J., ed. The Alchemy Reader: From Hermes Trismegistus to Isaac Newton.
Llored, Jean-Pierre. “Emergence and Quantum Chemistry.” Foundations of Chemistry 14
(2012): 245–274.
Llored, Jean- Pierre. “Mereology and Quantum Chemistry: The Approximation of
Molecular Orbital.” Foundations of Chemistry 12 (2010): 203–221.
Llored, Jean-Pierre. “Relations, caractérisations, et ‘affordances’.” In Les matérialismes et
la chimie: Perspectives philosophiques, historiques, et scientifiques, edited by François
Pépin, 113–120. Paris: Éditions Matériologiques, 2012.
Llored, Jean-Pierre. “Whole-Parts Strategies in Quantum Chemistry: Some Philosophical
and Mereological Lessons.” HYLE— Journal for Philosophy of Chemistry 20 (2014):
141–163.
LoLordo, Antonia. “Epicureanism and Early Modern Naturalism.” British Journal for the
History of Philosophy 19, no. 4 (2011): 647–664.
184 Bibliography
López Pérez, Miguel, Didier Kahn, and Mar Rey Bueno, eds. Chymia: Science and Nature
in Medieval and Early Modern Europe. Newcastle upon Tyne: Cambridge Scholars
Publishing, 2010.
Lucas, Stephen. “Margaret Cavendish’s Materialist Critique of Van Helmontian
Chymistry.” Ambix 58, no. 1 (March 2011): 1–12.
Lüthy, Christoph. “What to Do with Seventeenth- Century Natural Philosophy?
A Taxonomic Problem.” Perspectives on Science 8, no. 2 (2000): 164–195.
Macintosh, J. J. “Perception and Imagination in Descartes, Boyle, and Hooke.” Canadian
Journal of Philosophy 13, no. 3 (September 1983): 327–352.
Mackie, J. L. “Dispositions, Grounds, and Causes.” Synthese 34, no. 4 (April 1977): 361–369.
Manchak, John Byron. “On Force in Cartesian Physics.” Philosophy of Science 76, no. 3
(2009): 295–306.
McGuire, J. E. “Boyle’s Conception of Nature.” Journal of the History of Ideas 33, no. 4
(October–December 1972): 523–542.
Meinel, Christoph. “Early Seventeenth-Century Atomism: Theory, Epistemology, and the
Insufficiency of Experiment.” Isis 79 (1988): 68–103.
Meinel, Christoph. “Empirical Support for the Corpuscular Theory in the Seventeenth
Century.” In Theory and Experiment, edited by D. Batens and J. P. van Bendegem, 77–
92. Dordrecht: D. Reidel Publishing Company, 1988.
Meynell, Guy. “Locke, Boyle, and Peter Stahl.” Notes and Records of the Royal Society of
London 49, no. 2 (July 1995): 185–192.
Michael, Emily. “Daniel Sennert on Matter and Form: At the Juncture of the Old and the
New.” Early Science and Medicine 2, no. 3 (1997): 272–299.
Michael, Emily. “Sennert’s Sea Change: Atoms and Causes.” In Late Medieval and Early
Modern Corpuscular Matter Theories, edited by Christoph Lüthy, John E. Murdoch,
and William R. Newman, 331–362. Leiden: Brill, 2001.
Moon, Robert Oswald. “Van Helmont, Chemist, Physician, Philosopher and Mystic.”
Proceedings of the Royal Society of Medicine 25 (1931): 23–38.
Moran, Bruce T. Distilling Knowledge: Alchemy, Chemistry, and the Scientific Revolution.
Cambridge, MA: Harvard University Press, 2005.
Murdoch, John E. “The Medieval and Renaissance Tradition of Minima Naturalia.” In
Late Medieval and Early Modern Corpuscular Matter Theories, edited by Christoph
Lüthy, John E. Murdoch, and William R. Newman, 91–131. Leiden: Brill, 2001.
Needham, Paul. “An Aristotelian Theory of Chemical Substance.” Logical Analysis and
History of Philosophy 12 (2009): 149–164.
Newman, Micah. “Chemical Supervenience.” Foundations of Chemistry 10 (2008): 49–62.
Newman, William R. “Alchemical Atoms or Artisanal ‘Building Blocks’? A Response to
Klein.” Perspectives on Science 17, no. 2 (2009): 212–231.
Newman, William R. “The Alchemical Sources of Robert Boyle’s Corpuscular Philosophy.”
Annals of Science 53 (1996): 567–585.
Newman, William R. Atoms and Alchemy: Chymistry & the Experimental Origins of the
Scientific Revolution. Chicago: The University of Chicago Press, 2006.
Newman, William R. “Boyle’s Debt to Corpuscular Alchemy.” In Robert Boyle
Reconsidered, edited by Michael Hunter, 107–118. Cambridge: Cambridge University
Press, 2003.
Newman, William R. “Corpuscular Alchemy and the Tradition of Aristotle’s Meteorology,
with Special Reference to Daniel Sennert.” International Studies in the Philosophy of
Science 15, no. 2 (2001): 145–153.
Bibliography 185
Newman, William R. “The Corpuscular Transmutational Theory of Eirenaeus

Philalethes.” In Alchemy and Chemistry in the 16th and 17th Centuries, edited by Pyio
Rattansi and Antonio Clericuzio, 161–182. International Archives of the History of
Ideas/Archives internationales d’histoires des idées 140. Dordrecht: Springer, 1994.
Newman, William R. Gehennical Fire: The Lives of George Starkey. Chicago: The University
of Chicago Press, 2003.
Newman, William R. “How Not to Integrate the History and Philosophy of Science: A
Reply to Chalmers.” Studies in History and Philosophy of Science 41 (2010): 203–213.
Newman, William R. “Mercury and Sulfur among the High Medieval Alchemists: From
Rāzi and Avicenna to Albertus Magnus and Pseudo-Roger Bacon.” Ambix 61, no. 4
(November 2014): 327–344.
Newman, William R. “Newton’s Early Optical Theory and Its Debt to Chymistry.” In
Lumière et vision dans les sciences et dans les arts: De l’Antiquité au XVIIe siècle, edited by
Danielle Jacquart and Michel Hochmann, 283–307. Geneva: Librarie Droz S.A., 2010.
Newman, William R. “Robert Boyle, Transmutation, and the History of Chemistry Before
Lavoisier: A Response to Kuhn.” Osiris 29, no. 1 (January 2014): 63–77.
Newman, William R. “The Significance of ‘Chymical Atomism’.” Early Science and
Medicine 14 (2009): 248–264.
Newman, William R. “What Have We Learned from the Recent Historiography of
Alchemy?” Isis 102, no. 2 (June 2011): 313–321.
Newman, William R., and Lawrence M. Principe. “Alchemy and the Changing Significance
of Analysis.” In Wrong for the Right Reasons, edited by Jed Z. Buchwald and A. Franklin,
73–89. Dordrecht: Springer, 2005.
Newman, William R., and Lawrence M. Principe. Alchemy Tried in the Fire: Starkey,
Boyle, and the Fate of Helmontian Chymistry. Chicago: The University of Chicago
Press, 2002.
Newman, William R., and Lawrence M. Principe. “Alchemy vs. Chemistry: The
Etymological Origins of a Historiographic Mistake.” Early Science and Medicine 3, no.
1 (1998): 32–65.
Newth, David, and John Finnigan. “Emergence and Self-Organization in Chemistry and
Biology.” Australian Journal of Chemistry 59 (2006): 841–848.
O’Toole, Frederic. “Qualities and Powers in the Corpuscular Philosophy of Robert Boyle.”
Journal of the History of Philosophy 12, no. 3 (July 1974): 295–315.
Oldroyd, David Roger. “The Doctrine of Property— Conferring Principles in
Chemistry: Origins and Antecedents.” Organon 12/13 (1976–1977): 139–155.
Oldroyd, David Roger. “Some Neo-Platonic and Stoic Influences on Mineralogy in the
Sixteenth and Seventeenth Centuries.” Ambix 21 (1974): 128–156.
Osler, Margaret. “From Immanent Natures to Nature as Artifice: The Reinterpretation
of Final Causes in Seventeenth-Century Natural Philosophy.” The Monist 79 (1996):
388–407.
Osler, Margaret. “How Mechanical Was the Mechanical Philosophy? Non-Epicurean
Aspects of Gassendi’s Philosophy of Nature.” In Late Medieval and Early Modern
Corpuscular Matter Theories, edited by Christoph Lüthy, John E. Murdoch, and
William R. Newman, 423–439. Leiden: Brill, 2001.
Osler, Margaret. “Whose Ends? Teleology in Early Modern Natural Philosophy.” Osiris 16
(2001): 151–168.
Pagel, Walter. Joan Baptista van Helmont: Reformer of Science and Medicine. Cambridge:
Cambridge University Press, 2002.
186 Bibliography
Pagel, Walter. Paracelsus: An Introduction to Philosophical Medicine in the Era of the

Renaissance. Basel: S. Karger, 1958.
Pagel, Walter. “Paracelsus and the Neoplatonic and Gnostic Tradition.” Ambix 8, no. 3
(1960): 125–166.
Pagel, Walter. “The ‘Wild Spirit’ (Gas) of John Baptist van Helmont (1579–1644) and
Paracelsus.” Ambix 10, no. 1 (1962): 1–13.
Palmer, David. “Boyle’s Corpuscular Hypothesis and Locke’s Primary-Secondary Quality
Distinction.” Philosophical Studies: An International Journal for Philosophy in the
Analytic Tradition 29, no. 3 (March 1976): 181–189.
Pasnau, Robert. “Form, Substance, and Mechanism.” The Philosophical Review 113
(January 2004): 31–88.
Pauling, Linus. The Nature of the Chemical Bond. Ithaca, NY: Cornell University Press, 1939.
Pépin, François. La philosophie expérimentale de Diderot et la chimie: Philosophie, science,
et arts. Paris: Classiques Garnier, 2012.
Pépin, François. “Quelques perspectives chimiques pour le matérialisme?” In Les
matérialismes et la chimie: Perspectives philosophiques, historiques, et scientifiques, ed-
ited by François Pépin, 8–38. Paris: Éditions Matériologiques, 2012.
Pérez-Pariente, Joaquín. “Alchemical Versus Chemical Use of Distillation Techniques
and Materials: Their Mutual Influences and Divergent Developments.” In
Neighbours and Territories: The Evolving Identity of Chemistry. Proceedings of the
6th International Conference on the History of Chemistry, edited by José Ramón
Bertomeu-Sánchez, Ducan Thorburn Burns, and Brigitte Van Tiggelen, 263–271.
Louvain-la-neuve: Mémosciences, 2008.
Peterschmitt, Luc. “The Cartesians and Chemistry: Cordemoy, Rohault, Régis.”
In Chymists and Chymistry: Studies in the History of Alchemy and Early Modern
Chemistry, edited by Lawrence M. Principe, 193–201. Philadelphia: Chemical Heritage
Foundation and Science History Publications, 2007.
Pinet, Patrice. “La philosophie de la matière de Galilée à Newton.” Revue d’histoire de la
pharmacie 52, no. 341 (2004): 67–82.
Primas, Hans. Chemistry, Quantum Mechanics, and Reductionism. 2nd ed. Berlin: Springer
Verlag, 1983.
Principe, Lawrence M. The Aspiring Adept: Robert Boyle and His Alchemical Quest.
Princeton: Princeton University Press, 1998.
Principe, Lawrence M. “Newly Discovered Boyle Documents in the Royal Society
Archive: Alchemical Tracts and His Student Notebook.” Notes and Records of the Royal
Society of London 49, no. 1 (January 1995): 57–70.
Principe, Lawrence M. The Secrets of Alchemy. Chicago: The University of Chicago Press
2013.
Principe, Lawrence M., ed. Chymists and Chymistry: Studies in the History of Alchemy and
Early Modern Chemistry. Philadelphia: Chemical Heritage Foundation and Science
History Publications, 2007.
Principe, Lawrence M., and Andrew Weeks. “Jacob Boehme’s Divine Substance Salitter: Its
Nature, Origin, and Relationship to Seventeenth Century Scientific Theories.” The
British Journal for the History of Science 22 (1989): 53–61.
Platt, Andrew R. “Divine Activity and Motive Power in Descartes’s Physics.” British
Journal for the History of Philosophy 19, no. 4 (2011): 623–646.
Bibliography 187
Putnam, Hilary, and Paul Oppenheim. “Unity of Science as a Working Hypothesis.” In

Concepts, Theories, and Mind-Body Problem, edited by Herbert Feigl, M. Scriven, and
Grover Maxwell, 3–36. Minneapolis: University of Minnesota Press, 1958.
Pyle, Andrew. “Boyle on Science and the Mechanical Philosophy: A Reply to Chalmers.”
Studies in History and Philosophy of Science 33 (2002): 175–190.
Ramsey, Jeffry L. “Molecular Shape, Reduction, Explanation, and Approximate Concept.”
Synthese 111 (1997): 233–251.
Rattansi, Piyo, and Antonio Clericuzio, eds. Alchemy and Chemistry in the 16th and 17th
Centuries. Dordrecht: Kluwer Academic Publishers, 1994.
Robert, Aurélien. “Le vide, le lieu, et l’espace chez quelques atomistes du XIVe siècle.”
In La Nature et le Vide dans la physique médievale, edited by Joël Biard and Sabine
Rommevaux, 67–98. Turnhout: Brepols Publishers, 2012.
Rogers, G. A. J. “Boyle, Locke, and Reason.” Journal of the History of Ideas 27, no. 2 (April–
June 1966): 205–216.
Rossi, Paolo. The Birth of Modern Science. Translated by Cynthia De Nardi Ipsen. Malden,
MA: Blackwell, 2001.
Rossi, Paolo. La nascita della scienza moderna in Europa. Bari: Editori Laterza, 2000.
Wiley-Blackwell Publishers, 2001.
Sabra, A. I. “The Simple Ontology of Kalām Atomism: An Outline.” Early Science and
Medicine 14 (2009): 68–78.
Scerri, Eric. “Has Chemistry Been at Least Approximately Reduced to Quantum
Mechanics?” Proceedings of the Philosophy of Science Association 1 (1994): 160–70.
Scerri, Eric, and Lee McIntyre. “The Case for the Philosophy of Chemistry.” Synthese 111
(1997): 213–32.
Shanahan, Timothy. “Teleological Reasoning in Boyle’s Disquisition about Final Causes.”
In Robert Boyle Reconsidered, edited by Michael Hunter, 177– 192. Cambridge:
Cambridge University Press, 2003.
Shapin, Steven. “Pump and Circumstance: Robert Boyle’s Literary Technology.” Social
Studies of Science 14, no. 4 (1984): 481–520.
Shapin, Steven, and Simon Schaffer. Leviathan and the Air-Pump: Hobbes, Boyle, and the
Experimental Life. Princeton: Princeton University Press, 1985.
Shearin, W. H. The Language of Atoms: Performativity and Politics in Lucretius’ “De rerum
natura.” Oxford: Oxford University Press, 2015.
Siegfried, Robert. “The Chemical Revolution in the History of Chemistry.” Osiris, 2nd ser.
4, The Chemical Revolution: Essays in Reinterpretation (1988): 34–50.
Smets, Alexis. “The Controversy between Leibniz and Stahl on the Theory of Chemistry.”
In Neighbours and Territories: The Evolving Identity of Chemistry. Proceedings of the
6th International Conference on the History of Chemistry, edited by José Ramón
Bertomeu-Sánchez, Duncan Thorburn, and Brigitte Van Tiggelen, 291–306. Louvain-
la-neuve, Belgium, 2008.
Smith, Crosbie. “‘Mechanical Philosophy’ and the Emergence of Physics in Britain: 1800–
1850.” Annals of Science 33, no. 1 (1976): 3–29.
Solt, Leo F. “Puritanism, Capitalism, Democracy, and the New Science.” The American
Historical Review 73, no. 1 (October 1967): 18–29.
Stengers, Isabelle. La vie et l’artifice: Visages de l’émergence. Cosmopolitiques VI. Paris:
Éditions La Découverte /Les Empêcheurs de Penser en Rond, 1997.
188 Bibliography
Stuart, Matthew. “Locke on Superaddition and Mechanism.” British Journal for the History
of Philosophy 6, no. 3 (1998): 351–379.
Tammone, William. Science in Ferment: The Doctrine of Fermentation in Seventeenth-
Century British Science and Medicine. Bloomington: Indiana University Press, 1996.
Teller, Paul. “A Contemporary Look at Emergence.” In Emergence or Reduction? Essays on
the Prospects of Nonreductive Physicalism, edited by Ansgar Beckerman, Hans Flohr,
and Jaegwon Kim, 139–153. Berlin: De Gruyter, 1992.
Thackray, Arnold. Atoms and Powers: An Essay on Newtonian Matter-Theory and the
Development of Chemistry. Cambridge, MA: Harvard University Press, 1970.
Thagard, Paul. “The Conceptual Structure of the Chemical Revolution.” Philosophy of
Science 57, no. 2 (1990): 183–209.
Trevisani, Francesco. “La teoria corpuscolare in cartesio dal Traité du monde ai Principii.”
In Ricerche sull’atomismo del Seicento: Atti del Convegno di studio di Santa Margherita
Ligure (14–16 ottobre 1976), edited by Ugo Baldini, Giancarlo Zanier, Paolo Farina, and
Francesco Trevisani, 179–223. Florence: La Nuova Italia Editrice, 1977.
van Brakel, Jaap. “Philosophy of Science and Philosophy of Chemistry.” HYLE—Journal
for Philosophy of Chemistry 20 (2014): 11–57.
Viano, Cristina, ed. L’Alchimie et ses racines philosophiques: La tradition grecque et la tradi-
tion arabe. Paris: Librarie Philosophique J. Vrin, 2005.
Villani, Giorgio. La chiave del mondo. Dalla fisolofia alla scienza: l’onnipotenza delle
molecole. Naples: Editore CUEN, 2001.
Villani, Giorgio. Complesso e organizzato: Sistemi strutturati in fisica, chimica, biologia, ed
oltre. Milan: Franco Angeli Editore, 2008.
Walmsley, Johathan. “‘Morbus,’ Locke, and Boyle: A Response to Peter Anstey.” Early
Science and Medicine 7, no. 4 (2002): 378–397.
Walton, Michael T. “Boyle and Newton on the Transmutation of Water and Air, from the
Root of Helmont’s Tree.” Ambix 27, Part I (March 1980): 11–18.
Webster, C. “Water as the Ultimate Principle of Nature: The Background to Boyle’s
Sceptical Chymist.” Ambix 13, no. 2 (1996): 96–107.
Weininger, Stephen J. “The Molecular Structure Conundrum: Can Classical Chemistry Be
Reduced to Quantum Chemistry?” Journal of Chemical Education 61 (1984): 939–44.
Westfall, Richard S. The Construction of Modern Science: Mechanism and Mechanisms.
Wiener, Philip Paul. “The Experimental Philosophy of Robert Boyle.” The Philosophical
Review 41, no. 6 (November 1932): 594–609.
Wilson, Catherine. Epicureanism at the Origins of Modernity. Oxford: Oxford University
Press, 2008.
Wilson, Margaret D. “Superadded Properties: The Limits of Mechanism in Locke.”
American Philosophical Quarterly 16, no. 2 (April 1979): 143–150.
Wood, Joshua M. “On Grounding Superadded Properties in Locke.” British Journal for the
History of Philosophy 24, no. 5 (2016): 878–896.
Wood, Rega, and Michael Weisberg. “Interpreting Aristotle on Mixture: Problems about
Elemental Composition from Philoponus to Cooper.” Studies in History and Philosophy
of Science 35, no. 4 (2004): 681–706.
Yost, R. M., Jr. “Locke’s Rejection of Hypotheses about Sub-Microscopic Events.” Journal
of the History of Ideas 12, no. 1 (January 1951): 111–130.
Bibliography 189
Zanier, Giancarlo. “Il macrocosmo corpuscolaristico di Sebastiano Basson.” In Ricerche

sull’atomismo del Seicento: Atti del Convegno di studio di Santa Margherita Ligure (14–
16 ottobre 1976), edited by Ugo Baldini, Giancarlo Zanier, Paolo Farina, and Francesco
Trevisani, 77–118. Florence: La Nuova Italia Editrice, 1977.
Zaterka, Luciana. “Robert Boyle e a química experimental. O Ensaio do Nitro: alguns
relacionados à polêmica com Espinosa.” Cadernos de história e filosofia da ciência 11,
no. 1 (2001): 63–80.
Index
For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on
occasion, appear on only one of those pages.
Abelard, Peter, 150–56, 161, 162 45–47, 95–96, 99–100, 108, 110, 112,
accidental properties. See properties, 126–27, 148–49, 150, 151, 153–54
accidental association of particles, 81, 122,
acids, 31–32, 34, 55–56, 63–64, 65, 137, 163–64
88, 97, 98, 105–6, 107, 123–24, atomism
138–39, 159–60 alchemical, 22
affordance, 81, 103, 128–30 Epicurean, 3–4, 9–10, 20, 21, 22, 25, 35,
aggregate corpuscles, 32, 71–72, 92, 93, 47–51, 60–61, 83, 85
94, 105–6, 107–8, 137, 148, 150. See mechanistic, 3–4, 26, 47–51
also Boyle, Robert: chymical atoms; vitalistic, 22
chymical atomism atoms. See atomism
alchemy and alchemists, 1–2, 3, 8–9, Augustine of Hippo, 20, 20n.49
10–13, 16, 17, 19, 22, 25–27, 30–31,
33–34, 43, 53, 82, 89, 131–32, 161 Basso, Sebastian, 10–11, 24–25, 62, 72–73
Renaissance, 3, 4–5, 6, 8–10, 11–13, Bensaude-Vincent, Bernadette, 26–27, 33,
16–17, 18–19, 22 34, 106
See also Paracelsus and Paracelsians; Boyle, Robert
spagyria anti-reductionism (see reductionism
Alexander, Peter, 113, 115–16, 117, and anti-reductionism)
118, 123–24 chrysopoeia, 1, 89
alkahest, 39, 81, 89, 129–30 chymical atoms, 2–3, 4–6, 26, 58, 88–93,
alkaline salts, 18–19, 25, 160 94, 95–97, 103, 104, 105–8, 109, 113–
analysis, 19, 26–31, 32–34, 43, 57, 63, 64–66, 14, 122, 125, 126, 130, 131–32, 134–36,
67, 71–72, 79–80, 82–83, 87, 88, 91–93, 147, 149–50, 155–63, 164–65, 166–67,
94–95, 106–7, 122, 158 170, 171
Anima Mundi, 8, 18–19, 22–23, 24, 25, corpuscularianism (see
72–73, 75–76, 142 corpuscularianism: Boylean)
Anstey, Peter, 2–3, 85, 102–3, 109–10, emergentism (see emergence of
113–19, 121, 123–24 chemical properties)
aqua fortis, 28–29, 31–32, 34, 63–64, 92–93, essential form, 4–5, 15, 91, 101, 102–4,
105, 114, 123–24, 125–26, 157 105–6, 107–8, 148, 149, 150, 155–56,
aqua regia, 28–29, 41, 92–93, 159, 162–63
123–24, 125–26 extra-essential properties, 90–92, 104
Aquinas, Thomas, 15, 28, 152, 154 ‘lock and key’ analogy, 115–16
archeus, 25, 36–37, 38–39, 40, 84, 160 and Locke, 5, 99–101
Aristotle and Aristotelians, 13–15, 19–20, meaning of ‘mechanical philosophy’,
22, 27–28, 30, 33–34, 35–37, 38, 3–4, 73–78
192 Index
Boyle, Robert (cont.) chemical properties as emergent. See

mechanicism (see mechanicism: properties: emergent
Boylean) Chemical Revolution, 53, 94–96
microstructure, 4–6, 28–29, 34, 83, 87, chrysopoeia, 1, 89, 90, 101–2n.58
90–99, 101, 103, 104–5, 106, 107–8, See also alchemy
109, 111–12, 113–14, 117–18, 122, chymical atomism, 2–3, 4–6, 26, 50, 58,
123, 126–27, 136, 140–41, 147, 158, 88–93, 94, 95–97, 103, 104, 105–8,
160–61, 166–67, 170 109, 113–14, 122, 125, 126, 130,
pneumatics, x, 76–77, 84, 113, 126, 131–32, 134–36, 147, 149–50, 155–
128–29, 144–45 63, 164–65, 166–67, 170, 171
scientific research programme, 4–5, see also Boyle, Robert: chymical atoms
73–78, 96 chymistry, 9, 9n.3, 10–11, 13, 16, 17, 18,
subordinate and intermediate causes, 19–20, 21–22, 24–25, 27, 30, 32–33,
92–93, 97, 98–99 37, 39, 41, 43–44, 56, 57, 58–59, 60,
texture, 4–5, 64, 65, 66–67, 90, 91, 69–71, 72, 79–80, 81, 82, 89, 99
93–94, 98, 103–4, 107, 108, 109, Clericuzio, Antonio, 1–2, 11, 72–73, 122
111–12, 114, 117–18, 119–20, Cordemoy, Géraud de, 56–57, 58
121, 122–23, 124–27, 132–33, 134, corpuscular concretions. See Boyle,
135–37, 140–41, 144, 148–49, 150, Robert: chymical atoms
156–57, 158, 159–60, 161–62, 163, corpuscular philosophy. See
166–67, 170 corpuscularianism
Bruno, Giordano, 8–9, 22–23 corpuscularianism, 3–5, 6–7, 9–11, 13,
60–62, 67, 69–70, 80
Cartesian mechanism. See mechanicism: Boylean, 79–89, 93, 94, 96, 98, 111–12,
Cartesian 116, 137–38, 139–40, 142–43, 144–
Cecon, Kleber, 92–93, 94, 96 45, 147, 152, 162–63, 166–67, 170
Chalmers, Alan, 53, 54–55, 76–77, 96–97 Cartesian, 47–48, 49, 54–55, 169
Charleton, Walter, 29, 47–48, 50 mechanistic, 28, 33, 42, 45–46, 47–48,
chemical alterations, 35, 37–38, 42, 50, 64, 49, 50–51, 56, 60–70, 72, 73, 75–77,
85, 90, 91–92, 106, 107, 115–16, 126, 79–89, 93, 94, 96, 98, 111–12, 116,
132, 161 137–38, 139–40, 142–43, 144–45,
in Boyle, 64, 85, 90, 91–92, 106, 107, 147, 152, 162–63, 166–67, 169
115–16, 126, 132, 161 vitalistic, 19–25, 28, 34–35, 40–41, 42,
chemical bond, 41, 107, 158 47–48, 75, 91
chemical composition, 29, 31, 32–33, corpuscularism. See corpuscularianism
50, 58, 59, 77, 86, 88, 89, 90, 93–95,
104–5, 107, 130–31, 147, 148, 149– Democritus, 22, 89
50, 156–57, 160, 163, 167. See also Descartes, René and Cartesianism, 2,
molecules 3–4, 9, 16, 43–44, 45–60, 62, 66,
in Boyle (see Robert Boyle: 68–71, 74–75, 85, 86, 97, 100, 120,
chymical atoms) 134, 159–60
chemical elementarity. See Boyle, downward causation, 112,
Robert: chymical atoms; Gassendi, 130–31n.47, 138–39
Pierre; negative-empirical concept;
Sennert, Daniel effluvia, 5, 139–40, 142, 163
chemical properties as dispositional. See elementarity, 96, 106 (see also mereology;
properties: dispositional parts and wholes)
Index 193
operational, 5–6, 26–27, 57 (see also Helmont, Jan Baptista van, 3, 10–11,
Boyle, Robert: chymical atoms; 18–19, 21, 23, 25, 34–44, 47–48, 70,
Lavoisier, Antoine Laurent; 72–73, 81, 91
negative-empirical concept; Sennert, Henry, Desmond Paul, 150–51, 153–55,
Daniel 156, 161–62
elements Henry, John, 143–45
chemical, 7, 27, 41, 50, 57–58, 79–80, heterogeneity. See mixts
81, 88, 90, 94, 95–96, 129–30, 170 heterogeneous substances. See mixts
theory of, 13–14, 16–17, 28, 38, 81, 132 heuristic
emergentism. See properties: emergent negative, 3–4, 73–78, 99, 142
Epicurus and Epicureanism, 3–4, 9–10, 20, positive, 3–4, 73–78, 99
21, 22, 25, 35, 47–51, 60–61, Hirai, Hiro, 1–2, 20n.48
70–71, 75–76n.91, 83, 85–86, 120, Hohenheim, Theophrastus von. See
134, 159–60, 165–66 Paracelsus and Paracelsians
essential form. See Boyle, Robert; essential homogeneity. See homogeneous
form; microstructure; texture substances
essential parts. See parts, essential homogeneous substances, 24, 26, 28, 29–
essential properties. See properties: 30, 31, 50, 71–73, 87, 88, 91–92, 94,
essential 105, 106–7, 151–52, 170
essential structure. See Robert Boyle: humoral theory, 16, 21, 36
essential form Hunter, Michael, 2, 3
hydrostatics, 4–5, 54, 76–77, 84,
ferments, 10, 11, 24–25, 34–35, 36, 37, 38– 99, 144–45
44, 71, 72–73, 98 hylomorphism, 10–11, 13–14, 26–34, 36,
Ficino, Marsilio, 8–9, 11, 19–20 49–50, 67, 77–78, 85, 148, 153–54
final causes. See teleology See also Aristotle and Aristotelian;
fixedness. See fixity Sennert, Daniel; substantial form
fixity, 63–67, 96, 111, 122, 126–27, 129–30, hypostatical principles. See tria prima
133, 135, 160, 165
Fontenelle, Bernard le Bovier de, 69, 70 integral parts. See parts, integral
Fracastoro, Girolamo, 21, 139–40 integral wholes. See wholes, integral
Galen, 16, 21 Joly, Bernard, 53–54, 65–66, 69, 72

Galilei, Galileo, 110–11, 113
gas, 37, 38, 63–64 Kim, Mi Giyung, 12, 16
Gassendi, Pierre, 3–4, 21, 47–51, 61, 85, Kuhn, Thomas, 68, 149–50, 168
86, 87–88
Geber, 27, 34–35, 42 Lakatos, Imre, 3–4, 73–78
generation, 21, 35, 36, 38, 40, 87, 127–28, See also scientific research programme
133, 148–49 Lavoisier, Antoine Laurent, 9–10n.3, 26,
gold, 1, 23–24, 31, 32, 89, 123, 125–26, 33, 51–52, 57–58, 59–60, 94–96
129–30, 135, 151–52, 157, 160 Lémery, Nicolas, 58
analysis of, 31, 32, 39n.119, 41, 93–94, Leucippus, 89
105–6, 114, 123–24, 125–26, 157, 160 Llored, Jean-Pierre Noël, 5–6, 147, 155–56
Locke, John, 4–5, 99–101
Hall, Marie Boas, 26 logoi spermatikoi. See semina rerum
Harré, Rom, 5–6, 128–29, 147, 155–56 Lucretius, 5–6, 20, 22, 163–67
194 Index
manifest qualities. See qualities, manifest See also Boyle, Robert: chymical
mechanical affections, 45–46, 52, 59, 64, atoms; elementarity: operational;
65–66, 68–69, 70–71, 83–84, 86, 90– Sennert, Daniel
91, 94, 96–97, 98, 102–3, 109, 111–12, negative heuristic. See heuristic: negative
114, 116–18, 119–21, 126, 132–33, Neoplatonism, 8–9, 11, 18–19, 20–21,
134, 135–37, 142, 144, 163, 164, 169 22, 24, 35
mechanical philosophy. See mechanicism Newman, William, 4–5, 26–27, 33–35,
mechanicism, 1–2, 3–4, 7, 9n.3, 9–11, 16, 41, 65, 89, 94, 105–6, 136–38,
22, 23, 24, 26, 29, 30–31, 168, 170–71
33–34, 35, 37–38, 40–42, 43–44, 45, niter
47–51, 60–69 aerial, 10, 18–19 (see also anima mundi;
Boylean, 3–7, 34, 69–78, 79, 83–89, potassium nitrate)
90–108, 109, 110–20, 121–45, 146, fixed and volatile, 63–64, 65, 66–67
148, 149–50, 152, 154–55, 156–57, redintegration of niter (see potassium
159–60, 162–63, 167, 169, 170, 171 nitrate: redintegration of)
Cartesian, 2, 3–4, 45–47, 51–60, 62, 66, spirit of, 63–64, 65, 66–45
68–70, 85 nitric acid. See aqua fortis
mechanism. See mechanicism nominalism, 100, 101, 154–55
mechanistic qualities. See qualities, non-mechanistic qualities. See qualities,
mechanistic non-mechanistic
Meinel, Christoph, 60–61, 62, 104–5
menstruum, 89, 98 O’Toole, Frederic, 119–20, 124,
mercury 125–26, 157
element, 62, 93–94 occult qualities. See qualities, occult
principle (see tria prima) Oldenburg, Henry, 63, 66–67
mereology, 2–3, 5–6, 13–14, 15, 46–47,
49, 96–97, 103, 107, 109, 133, 145, Paracelsian principles. See tria prima
146, 147–67 Paracelsus and Paracelsianism, 3, 4–5, 10–
See also parts and wholes 11, 12, 16–21, 23, 24–25, 26, 33–35,
Mersenne, Marin, 37, 54–55 36, 38, 42, 43, 47–48, 56, 58, 63, 77,
microstructure. See Boyle, Robert; 79, 81, 82, 121, 129–30, 131–32, 139–
essential form; microstructure; 40, 142, 160, 161
texture See also alchemy: Renaissance; spagyria
minima naturalia, 3, 19–25, 28, 34–35, parts and wholes
37–38, 40, 41–42, 86–87, 90–91, contiguous wholes, 151–53, 154–55,
98, 107, 111–12, 114, 122, 126, 133, 158, 162–63
135–36, 148, 158, 160 continuous wholes, 151–53, 155–56,
mixts, 5, 26, 27–31, 32, 34, 79–80, 81, 158, 162–63
87–89, 91–92, 94, 119, 129–30 essential parts, 153–54, 160–61
mixtures, 37, 40, 41–42, 43, 63–64, 66, integral parts, 153–54, 159, 160–61
87–89, 91–92, 97, 121, 123–24, 126, integral wholes, 151–53, 154–56,
128, 132, 134–35, 136, 159–60, 163 158–59, 162–63
molecules, 24, 49–50, 87–88, 90, 95, 109, Pasnau, Robert, 15, 45–46, 101, 148,
147, 150, 170 149, 150
Peterschmitt, Luc, 57, 59
natural kinds, 7, 29, 46–47, 77–78, 92–93, positive heuristic. See heuristic: positive
97, 99–104, 105, 107–8, 157 potassium nitrate, 10, 18, 159–60, 161
negative-empirical concept, 26–34, 58, redintegration of, 3–4, 19, 33, 62–64, 65,
59–60, 92, 106 66–67, 91–92, 97, 161
Index 195
precipitation, 31–32, 33–34, 41, 93–94, redintegration of niter. See potassium

98, 105–6 nitrate: redintegration of
prima mixta, 28, 88–89 reduction to the pristine state, 4–5, 26–27,
prima naturalia, 87, 91, 107–8, 28–29, 30–31, 32–33, 34, 62–63,
134, 136–37 85, 104–8
prime matter, 17, 28, 29, 48, 82, 82n.9 reductionism and anti-reductionism,
Principe, Lawrence M., 1–2, 41 6, 10, 43–44, 49–51, 70, 71, 72–73,
properties 96–97, 107–8, 109, 113–20, 121–45,
dispositional, 2–3, 5, 6, 39, 53–55, 150–51, 156–57, 158, 162–63, 168,
77–78, 96–97, 108, 109–10, 112, 169, 171
113–20, 121–30, 131, 135–36, relational properties. See properties:
137–38, 139–45, 146, 167, 168, 171 relational
emergent, 2–3, 5–6, 77–78, 84, 97, re-synthesis. See redintegration
108, 109, 114, 120, 125–26, 130–38, Rohault, Jacques, 56, 57–58
144–45, 146, 147–48, 150, 166, 168, Rossi, Paolo, 11–12
169–70, 171 Royal Society of London, 50–51, 127
essential, 15, 28, 30, 77–78, 83, 87, 88,
90, 91–93, 101–2, 103, 104, 105, 106, salts
107–8, 149, 159 chemical, 18, 25, 27–28, 31, 39, 41, 63,
higher-level and lower-level, 29, 64, 93, 121, 126, 128, 129–30, 157,
49–51, 52, 55–56, 73, 97, 108, 131, 158, 159–60
135, 136–37, 138, 139–40, 144–45 salt-petre or saltpeter. See potassium
novel, 5–6, 66, 68–69, 108, 109, 113–14, nitrate
125, 126, 131, 133–35, 146, 152– scientific research programme, 3–4,
53, 154–55, 156–57, 158, 159–61, 72–78, 96–97
169, 171 secondary qualities. See qualities,
relational, 2–3, 5, 6, 23, 77–78, 96–97, secondary
109–10, 111–12, 113–20, 121–30, seminal principles. See semina rerum
131, 138, 139–45, 146, 147, 156, 168, semina rerum, 3, 10–11, 19–25, 34–43, 71,
169, 171 87, 140
See also qualities Sennert, Daniel, 3–5, 10–11, 21, 23, 25,
Pyle, Andrew, 75–76 26–34, 42, 43–44, 47–48, 49–50,
60–61, 62–63, 70, 72–73, 82, 83, 90,
qualities, 2–3, 5, 23–24, 28, 30, 31, 35, 91, 95–96, 104–5
45–46, 49, 50, 52, 64, 85, 90, 96–97, sensible qualities. See qualities, sensible
100–1, 102, 109, 110–20, 121–38, Severinus, Petrus, 35, 47–48
156–57, 158, 163–64, 166 Shapin, Steven, 127–28
chemical, 10, 34, 53, 55–5 6, 60, spagyria, 9, 10–11, 16–20, 39, 75, 76, 79,
82, 84, 108, 109, 112, 113–2 0, 80, 81–82, 84, 93, 129–30, 131–32,
121–3 8, 160 136, 142
cosmical, 77, 97, 98, 139–45 See also alchemy: Renaissance;
mechanistic, 111 Paracelsus and Paracelsians
non-mechanistic, 111–12, 113–20, Spinoza, Baruch, 3–4, 33, 43–44, 50–51,
121–38, 139–45, 160 62–63, 66–68, 169
occult, 111–12 spirit of niter. See aqua fortis
primary, 110, 111–12 spirit
secondary, 96, 110, 111–12, 121–30 chemical, 18–19, 24–25, 34–44, 63,
sensible, 113–20, 121–38, 139–45 64–65, 66–67, 72–73, 126, 128,
See also properties 129–30, 159–60
196 Index
spirit (cont.) 130–38, 145, 146, 147–48, 150, 158,

vital, 8, 10, 11, 16–17, 18–19, 24–25, 161–62, 165–67, 168, 169, 170–71
34–44, 60, 62, 71, 72–73, 76–77, 142 synthesis, 19, 27–28, 29–30, 33, 43,
Stahl, Georg Ernst, 51–52, 89 64–65, 82–83
Stengers, Isabelle, 26–27, 33, 34, 106, 145
Stoics, 20–21, 35 taxonomical classification, 4–5, 7,
structure. See structuralism: chemical 97, 99–104
structural explanation. See structuralism: teleology
chemical immanent vs. extrinsic, 75–76,
structuralism 75–76n.91
chemical, 2–3, 4–6, 23, 26–34, 54–55, texture. See Boyle, Robert; essential form;
64, 66–67, 70–71, 77–78, 166–67 microstructure; texture
in Boyle (see Boyle, Robert; essential three principles. See tria prima
form; microstructure; texture) transmutation. See alchemy; chrysopoeia
substantial form, 3–5, 11–12, 13–16, transposition of particles, 91–92,
22–23, 28–31, 33–34, 43–44 104, 133
Boylean rejection of, 67, 68, 79, 83, 84, tria prima, 3, 4–5, 10–11, 16–20, 38–39,
91, 92–93, 99–101, 102, 103–5, 123, 50, 77, 79–81, 82, 83, 84, 132, 140
148, 149, 152–53, 154, 155–56, 159,
160–61, 162–63 vacuum, 62, 86, 143
Cartesian rejection of, 45–47, 49 Venel, Gabriel François, 69–70
sulfur and sulfuric compounds, 16, vitalism, 1–2, 3, 8–13, 16–17, 19–25, 33–
17–18, 39, 50, 97, 105–6, 107, 126, 35, 37, 40–41, 42, 43–44, 47–48, 52,
128, 159–60 69–70, 71, 91, 104–5, 139–40
sulfur principle. See tria prima volatility, 1, 64, 82, 111, 133, 135, 159–60
supervenience of chemical properties,
2–3, 5, 6, 7, 108, 109, 113–14, 126, Webster, John, 21, 23–24

Marina Paola Banchetti-Robino - The Chemical Philosophy of Robert Boyle - Mechanicism, Chymical Atoms, and Emergence-Oxford University Press (2020)

Uploaded by

Document Information

Original Title

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

Marina Paola Banchetti-Robino - The Chemical Philosophy of Robert Boyle - Mechanicism, Chymical Atoms, and Emergence-Oxford University Press (2020)

Uploaded by

Copyright:

Available Formats

The Chemical Philosophy of Robert Boyle

Published in the United States of America by Oxford University Press

© Oxford University Press 2020

All rights reserved. No part of this publication may be reproduced, stored in

You must not circulate this work in any other form

Library of Congress Cataloging-in-Publication Data

4. Boyle’s View of Chemical Properties as Dispositional, Relational,

for Contemporary Philosophy of Chemistry,” in Jean-​ Pierre Llored (ed.),

1 Principe, The Aspiring Adept, 5.

2 Hunter, Robert Boyle Reconsidered, 5.

first, to demonstrate that Boyle regarded chemical properties as properties that

a detailed account of the relationship between Cartesian mechanicism and

Although the study of such an important historical figure as Boyle is valu-

1 Bloor, “Durkheim and Mauss Revisited,” 78–​80.

corpuscularian theories of matter imply a mechanistic hypothesis. More will be

the rejection of which helped to define Boyle’s commitment to the mechanical

1.1 The Vitalistic Character of Renaissance Alchemy

4 Clericuzio, Elements, Principles, and Corpuscles, 37.

the 16th century, chemistry “had no organized structure whatsoever, no theories

6 Rossi, The Birth of Modern Science, 137.

Given these considerations regarding the scientific status of early modern

1.2 The Scholastic Theory of Substantial Form

Mereology and the Status of Form,” 715–​736.

14 Aristotle, The Complete Works of Aristotle, Z.17, 1041b32–​33.

Substance, and Mechanism.”

it is one of Aristotle’s most cherished ideas that material and efficient

We find that it is this increasingly naturalistic and wholly concrete conception

1.3 Paracelsian Spagyria and the Tria Prima

Theophrastus von Hohenheim, a.k.a. Paracelsus (ca. 1493–​1531), is a sem-

microcosm and macrocosm and, in this affirmation, he is once again inspired

a powerful stream of speculation wondered whether saltpeter contained the sal

40 Debus, “The Paracelsian Aerial Niter,” 47.

1.4 Semina Rerum, Minima Naturalia, and

physicalistic connotations, albeit still within the context of a generally vitalistic

In addition to the concept of semina rerum, the notion of minima naturalia

Paracelsians in the 17th century, however, continue to consider vitalism to be

63 Webster, Metallographia, or an History of Metals, as cited in Clericuzio, “Alchimie, philosophie

corpusculaire, et minéralogie dans la Metallographia de John Webster,” 299–​300.

with an internal principle of organization, life and sensibility, namely, the

66 Clericuzio, “The Internal Laboratory,” 59.

1.5 Daniel Sennert’s Structural Hylomorphism

Newtonian matter theory, and it was later resurrected in Bernadette Bensaude-​

Negative-​empirical principles that defined the temporary limits of analysis were,

73 Bensaude-​Vincent and Stengers, A History of Chemistry.

returned to their pristine state when the original microstructural configuration

83 Newman, Atoms and Alchemy, 138–​139.

87 Newman, Atoms and Alchemy, 122.

91 Newman, Atoms and Alchemy, 128.

macrolevel effect.”97 It must certainly be recognized that Sennert is a man of his

1.6 Jan Baptista van Helmont and the Chemical

97 Newman, Atoms and Alchemy, 136.

99 Clericuzio, Elements, Principles and Corpuscles, 56.

development. In a 1631 letter to Marin Mersenne, he accuses Aristotle of having

Helmont appeals to the distinction between merely physical superficial

“mutation of essence.” Such “mutation of essence” or transmutation of water into

116 Newman, “The Corpuscular Transmutational Theory of Eirenaeus Philalethes,” 65–​66.

solve every substance, including gold.

121 van Helmont, Imago fermenti, in Ortus medicinae, 69.

124 Newman and Principe, Alchemy Tried in the Fire, 66.

129 van Helmont, Progymnasma meterori, 72.

to a strict mechanicism entailed sacrificing the status of chemistry as systematic

“Mechanicism,” or the mechanical philosophy, is the view “according to which

2.1 The Cartesian Rejection of Substantial Forms

for Contemporary Philosophy of Chemistry,” in Jean- Pierre Llored (ed.),

1 Bloor, “Durkheim and Mauss Revisited,” 78–80.

Mereology and the Status of Form,” 715–736.

14 Aristotle, The Complete Works of Aristotle, Z.17, 1041b32–33.

Theophrastus von Hohenheim, a.k.a. Paracelsus (ca. 1493–1531), is a sem-

corpusculaire, et minéralogie dans la Metallographia de John Webster,” 299–300.

Newtonian matter theory, and it was later resurrected in Bernadette Bensaude-

Negative-empirical principles that defined the temporary limits of analysis were,

73 Bensaude-Vincent and Stengers, A History of Chemistry.

83 Newman, Atoms and Alchemy, 138–139.

116 Newman, “The Corpuscular Transmutational Theory of Eirenaeus Philalethes,” 65–66.

unity, nor does he wish to re-conceptualize the notion of form in mechanistic

Pierre Gassendi (1592–1655) was arguably the strongest proponent of mech-

10 See Pinet, “La philosophie de la matière de Galilée à Newton,” 67–82.

nature of matter. Nonetheless, since Rohault favors a hypothetico-deductive

Ultimately, in the mid-18th century, Lavoisier would circumvent these questions

4KNO3 + 3C → CO2↑ + 2NO2↑ + N2↑ + 2K2CO3 (K2CO3 — alkaline

2NO2 + H2O → HNO3 + HNO2 (HNO3 —acidic and volatile)

3) Redintegration (or re-synthesis) of potassium nitrate:

63 Oldenburg, The Correspondence of Henry Oldenburg, 449–459.

Louis Lémery et Etienne-François Geoffroy,” 3.

part il identifie le corpuscularianisme au mécanisme cartésien, c’est-à-dire à l’idée selon laquelle

By the mid-17th century, it is evident to natural philosophers that most chemical

20 Boyle, The Origin of Forms and Qualities, 325–326.