Causality and Causal Explanation in

Aristotle Nathanael Stein

Visit to download the full and correct content document:
https://ebookmass.com/product/causality-and-causal-explanation-in-aristotle-nathana
el-stein/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Explanation Beyond Causation: Philosophical

Perspectives on Non-Causal Explanations Alexander
Reutlinger (Editor)

https://ebookmass.com/product/explanation-beyond-causation-
philosophical-perspectives-on-non-causal-explanations-alexander-
reutlinger-editor/

Complexity, Emergence, and Causality in Applied

Linguistics: A Social Realist Viewpoint Jérémie
Bouchard

https://ebookmass.com/product/complexity-emergence-and-causality-
in-applied-linguistics-a-social-realist-viewpoint-jeremie-
bouchard/

Brooklyn Legacies Triss Stein

https://ebookmass.com/product/brooklyn-legacies-triss-stein-2/

Brooklyn Legacies Triss Stein

https://ebookmass.com/product/brooklyn-legacies-triss-stein/
Understanding Ludwig Lachmann's Economics Stein

https://ebookmass.com/product/understanding-ludwig-lachmanns-
economics-stein/

Psychology and Value in Plato, Aristotle, and

Hellenistic Philosophy Fiona Leigh (Editor)

https://ebookmass.com/product/psychology-and-value-in-plato-
aristotle-and-hellenistic-philosophy-fiona-leigh-editor/

Explanation and Integration in Mind and Brain Science

1st Edition David M. Kaplan

https://ebookmass.com/product/explanation-and-integration-in-
mind-and-brain-science-1st-edition-david-m-kaplan/

Reconsidering Causal Powers: Historical and Conceptual

Perspectives Benjamin Hill

https://ebookmass.com/product/reconsidering-causal-powers-
historical-and-conceptual-perspectives-benjamin-hill/

Industrial Craft in Australia: Oral Histories of

Creativity and Survival Jesse Adams Stein

https://ebookmass.com/product/industrial-craft-in-australia-oral-
histories-of-creativity-and-survival-jesse-adams-stein/
Causality and Causal Explanation
in Aristotle
 Causality and Causal
Explanation in Aristotle
NAT HA NA E L ST E I N
 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 2023

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: Stein, Nathanael, 1976– author.
Title: Causality and causal explanation in Aristotle / Nathanael Stein.
Description: New York, NY : Oxford University Press, [2023] |
Includes bibliographical references.
Identifiers: LCCN 2022062180 (print) | LCCN 2022062181 (ebook) |
ISBN 9780197660867 (hardback) | ISBN 9780197660881 (epub) | ISBN 9780197660898
Subjects: LCSH: Aristotle. | Causation. | Explanation.
Classification: LCC B 49 1 .C 3 S 84 2023 (print) | LCC B 4 9 1 . C 3 (ebook) |
DDC 122—dc23/eng/20230314
LC record available at https://lccn.loc.gov/2022062180
LC ebook record available at https://lccn.loc.gov/2022062181

DOI: 10.1093/​oso/​9780197660867.001.0001

Printed by Integrated Books International, United States of America

 Contents

Preface  ix
Abbreviations and Editions Cited of Aristotle’s Works  xiii

Introduction  1
I.1. The Basic Problem  1
I.2. Outline  12
I.3. Themes, Consequences, Comparisons, and Absences  18

PA RT I : C O N C E P T UA L ST RU C T U R E
1. Reading (and Animating) Physics II 3  25
1.1. The Problem of Physics II 3  25
1.2. Physics II 3 in the Context of Book II  29
2. Background 1: Critiques of the Predecessors  37
2.1. Epistemological and Scientific Critiques  37
2.2. Metaphysical Critiques  44
2.3. Summary  53
3. Background 2: Science and Dialectic  55
3.1. Being “in Virtue of Oneself ” and in Other Ways  55
3.2. Physics II 2 and the Autonomy of Natural Science  60
3.3. Posterior Analytics II 11 and the Simple Schema of Causes  62
3.4. Summary  69
4. Physics II 3 in Argumentative Context  71
4.1. Highlights  71
4.2. Implications for Theory and Method  79
4.3. Causal Pluralism  82
4.4. Strengths, Weaknesses, and Open Questions  87

PA RT I I : M E TA P H YSIC S
5. The Realist Challenge  95
5.1. Metaphysical Pluralism  95
5.2. Clarifying the Question  100
vi Contents

6. Causes, Kinds, and Transformations  105

6.1. From Privileged Entities to Intrinsic Causes  105
6.2. Essentialism and Kinds of Natural Change  112
6.3. Pre-​theoretical and Theoretical Accounts of Change  116
6.4. Real Definitions of Transeunt Interactions  123
6.5. Summary  129
7. Causal Kinds and Causal Profiles  131
7.1. Real Definitions and Causal Profiles  131
7.2. The Varieties of Causal Profile  133
7.3. Connections and Correspondences to Other Distinctions  145
7.4. Causal Profiles at Work: Gluttonous Birds  148
7.5. Implications for Puzzles about Aristotelian Causality  152
7.6. Summary  160
8. Discreteness in Agent–​Patient Relations  162
8.1. The Realist Question  162
8.2. “Ways of Being Causes” in Physics II 3  165
8.3. Transeunt-​Causal Change in Physics III 3  174
8.4. Varieties of Discreteness  181
8.5. Discreteness Applied  186
8.6. Aristotle and Modern “Neo-​mechanism”  190
8.7. Summary, Comparisons, and Open Questions  193

PA RT I I I : E P I ST E M O L O G Y
9. Coming to Know Causes  199
9.1. Basic Questions about Grasping Causes  199
9.2. Causes and the Two Images  201
9.3. A Problem of Induction  205
9.4. Stages of Inquiry and Their Associated States or Capacities  212
10. Causality and Epistemic Asymmetries  217
10.1. Manifest Causes and Basic Asymmetries  217
10.2. The Priority of Transeunt-​Causal Efficient Causality  223
10.3. Grasping Transeunt-​Causal Interactions  225
10.4. Understanding Origin-​Dominant Causal Profiles  233
10.5. Summary  235
 Contents vii

11. The Non-​Secret Connexion  237

11.1. Is Some Causality Just as It Appears?  237
11.2. The Special Importance and Epistemic Status of Crafts  240
11.3. The Nature of Blood  244
11.4. Summary: Causal Explanation and Aristotle’s Empiricism  251
Conclusion  255

Bibliography  261
Index Locorum  273
Subject Index  281
 Preface

Philosophical problems get some of their extra bite from the fact that they
are two-​faced: we are typically unsure what to say about the subject matter
itself but also puzzled about the right way to approach it. Questions about
causality and explanation have persistently been a focus of those two sorts of
puzzlement for me—​not knowing what to say about them, and not knowing
how to start looking for answers. Now, Aristotle connects causality directly to
the natural human desire to know, and I think it is, above all, that connection
that is the source of its abiding interest: not just in relation to scientific ex-
planation or metaphysics, but in relation to almost any domain in which we
have an impulse to understand and therefore to explain—​and to feel like we
have the explanation, even at the cost of self-​deception. That impulse drives a
whole range of behavior, from scientific and philosophical inquiry, to histor-
ical explanation, all the way down to arguments about the meaning of elec-
tion results, clickbait, and conspiracy theories. By nature, we desire to know,
especially if there’s a chance that reality is different from the appearances.
If you’re a suspicious type, or just think it’s generally good policy to be
circumspect even around your own ideas, then you may naturally start to
wonder about the difference between good and bad explanations. As a phi-
losopher, you may also start to wonder about good and bad approaches to
explanation, and one way to get a grip on that problem is to cast a wide net,
going back as far as possible. Aristotle is almost as far back as you can go for
a self-​conscious discussion of causality and causal explanation, and a great
many paths in philosophy and its history lead back to his understanding of
causes and their role in explaining the natural world.
But here, I think, we find ourselves in a strange position. On the one hand,
Aristotle says so much about causes, and his thinking about them permeates
so much of his work, that they constitute in a way a main point of entry into
his theoretical and natural philosophy. On the other hand, they are so rarely
a focus of his direct philosophical attention—​so rarely a focus of argument as
opposed to a tool for building an argument, or a framework for guiding an
inquiry about something else—​that it isn’t nearly as clear as one might have
thought what his theory of causality really is. Thus, in ways that I will make
x Preface

more precise in the Introduction, in spite or even because of its centrality to

so much of his work, we cannot really answer some basic questions about
Aristotle’s views on causality in the way that we can for Plato, Hume, Kant,
and others.
This gap is not just unfortunate: given the role Aristotelian causality has
played in the history of philosophy and science, and the role it is supposed to
play in standard accounts of how modern thought and especially modern sci-
ence have broken with the past, it really shouldn’t be allowed. Furthermore,
when we look at contemporary discussions of causation, we find a prolifer-
ation of approaches with very little that counts as common ground or even
common sense, especially as previously dominant approaches have come to
be just one among many. So it would be of interest to get a clear idea of the
theory that is in some ways the point of origin for theorizing about causality
in Western philosophy and science, even for those aiming to flee it. For these
reasons, I want to try to have a philosophical conversation about causality
with Aristotle, a conversation of the kind that has proven so fruitful in other
domains, but which has in some ways been missing here, however much at-
tention has been devoted to the wide array of particular topics on which he
brings his causal notions to bear.
In some ways this book goes all the way back to the beginning for me. The
route that I have just described, from generalized puzzlement about how one
should satisfy one’s desire to know to wanting a better philosophical account
of Aristotle’s theory of causality, is in fact a cleaned-​up version of how I went
from being a confused and dissatisfied undergraduate to whatever it is that
I am now. Along the way I wrote my dissertation on causality and explanation
in Aristotle; some of the ideas in it and from subsequent papers survive here
in some altered form, but it represents a fresh start, and as far as possible I do
not repeat here what I have said in print elsewhere, or cover the same ground.
I aim instead to keep the discussion rather tightly focused on questions about
the kinds of philosophical challenge for which Aristotle thinks that a cor-
rect account of causality is crucial, what he thinks that account is, and how
we should assess it, both as a response to those challenges and as a philo-
sophical theory in general. So while parts of this book will mainly interest
specialists in ancient philosophy, the argument as a whole is meant for an-
yone interested in philosophical theories of causation and explanation and
their history, as well as anyone who has read Aristotle’s thoughts on the topic
of causality and come away wondering what it all really adds up to, and how
we might engage with it.
 Preface xi

Because this project goes so far back, the list of those to whom I’m in-
debted is long. Especially important for me are the debts that have less to
do with the content of the book than with the people along the anything-​
but-​straight path from the beginnings of graduate school to actual employ-
ment who, for whatever reason, saw fit to give me their time, help, or just
some much-​needed encouragement on the basis of what must have been, at
the start at least, very slight evidence of any promise. These include David
Rosenthal, Nick Pappas, Michael Tooley, Bob Pasnau, David Charles, Terry
Irwin, Lindsay Judson, Roger Crisp, Ursula Coope, and Jonathan Lear. The
biggest debts by far, though, accumulated over many years, are owed to Chris
Shields. I wouldn’t know where to start.
As to the content of the book, I have received helpful feedback at various
stages from a variety of audiences, including those at the Ancient Philosophy
Seminar at the University of Oxford, Syracuse University, the University
of Coimbra, the University of Campinas, the Center for the Aristotelian
Tradition at the University of Notre Dame, and in many different contexts
from colleagues at Florida State. I’m grateful for feedback from several
readers, including Andrea Falcon, Devin Henry, Thomas Johansen, Michael
Strevens, and especially Jessica Gelber and the participants at the workshop
she organized at the University of Toronto, as well as Peter Ohlin at OUP.
More materially, it is a pleasure to thank the National Endowment for the
Humanities, a fellowship from which made it possible for me to give this
project my full attention.
Finally, I thank my friends and family, especially my mother, Cecile, and
my wife, Hashi, for support—​again stretching back to a time of great uncer-
tainty and maybe unfathomable decision-​making on my part. I dedicate the
book to my father, Michael, who died right in the middle of the route I have
described, whose love and life remain a constant source of good in my own.
 Abbreviations and Editions Cited
of Aristotle’s Works

[APo.] Posterior Analytics (Ross 1949)

[APr.] Prior Analytics (Ross 1949)
[Cat.] Categories (OCT)
[DA] De Anima (OCT)
[DC] De Caelo (Bekker 1831)
[De Juv.] On Youth, Old Age, Life and Death, and Respiration (Bekker)
[De Motu] Motion of Animals (Nussbaum 1978)
[De Somno] On Sleep and Waking (Bekker)
[GA] Generation of Animals (Peck 1943)
[GC] De Generatione et Corruptione (Joachim 1922; Rashed 2005)
[HA] History of Animals (Bekker)
[Met.] Metaphysics (Ross 1924)
[Mete.] Meteorology (Bekker, Fobes 1919)
[EN] Nicomachean Ethics (OCT)
[PA] Parts of Animals (Bekker, Peck 1937)
[Phys.] Physics (Ross 1936)
[Pol.] Politics (OCT)
[Top.] Topics (OCT)
 Introduction

I.1. The Basic Problem

Causality is at the heart of Aristotle’s philosophical project, and at the same

time it so permeates his work that it can come to seem indistinguishable
from that project as a whole, or whatever part of it one happens to be looking
at. It is at the heart of his project because, in his view, it is at the core of un-
derstanding in the most ambitious sense: our highest cognitive achievement
as rational beings, whether identified as wisdom (sophia) or scientific knowl-
edge (epistêmê), is held to consist in coming to grasp the causes of things.1 At
the same time, when he articulates, in Metaphysics I, how his own thought
has advanced and improved upon that of his predecessors, he famously does
so by casting the history of attempts at a rational understanding of the nat-
ural world in terms of the progressive discovery of the four causes (aitiai),
the progress being nonetheless so limited that, he says, “in a certain way they
have all been described before, while in another way, not at all” (Metaphysics
I 10, 993a14–​15). He thus thinks that causality is both at the heart of any the-
oretical undertaking and that the correct understanding of it is central to his
own philosophical achievement.
For all that, and for all the influence his claims about causality have had
over the centuries, it is surprisingly difficult to say what the theory is, or
whether it is even a theory at all. By this I mean that it is unclear, from what
Aristotle says and what his commentators have said about it, (a) how he
answers the main philosophical questions about causes to which he thinks
his predecessors’ answers are flawed, and (b) how his answers bear on the
main questions we confront in thinking about causality in general, such that
they could be usefully critiqued, defended, developed, and compared with
others.

1 ‘Wisdom’ is used in Metaphysics [Met.] I, ‘scientific knowledge’ or ‘understanding’ (sometimes

a better translation of epistêmê) is more often the term used elsewhere, as in the Posterior Analytics
[APo.]. Regardless of the term, Aristotle consistently identifies grasping causes with a higher grade of
cognitive achievement, close to if not identical with our primary aim as potential knowers.

Causality and Causal Explanation in Aristotle. Nathanael Stein, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197660867.003.0001
2 Causality and Causal Explanation in Aristotle

Here are some of the questions I have in mind. Granted that Aristotle is
some kind of pluralist, in that he recognizes multiple different types or modes
of causality, what is the basis of those distinctions, and why should we accept
them? Are these causes subject to further distinctions or further unification?
What are the criteria by which a certain relationship is to be considered a
causal one, rather than, say, a correlation or a merely conditional relation-
ship? What, precisely, are the metaphysical relationships between causes and
their effects? How do we, on the basis of experience or observation, come
to understand causal relationships, so that we are justified in thinking that
certain elements of nature are genuinely responsible for others, rather than
simply correlated with one another? These are not just questions we wish to
answer in our own, modern discussions of causality. As we shall see, they
overlap considerably with the questions Aristotle puts to his predecessors,
and they underlie the criticisms he makes of their views.
Now, it is easy enough to list “the four causes” by the names tradition-
ally applied to them by later commentators—​the efficient, formal, material,
and final causes—​and to analyze some examples of artifacts like statues and
houses in terms of them.2 Indeed, a typical account of Aristotle’s theory of
causality will do just that, and then show how Aristotle intends those modes
of causality to apply to far less obvious relationships, such as the constitu-
tion of a natural organism and its relationship to its own body and bodily
activities.
Such a presentation is fine as far as it goes, but it mainly serves to sharpen
the difficulties we face in understanding Aristotle’s views about causality and
natural science. A reader may well find the initial examples quite clear, the
later ones quite unclear, and then wonder at the end of it what has actually
been presented—​a theory? a doctrine? a schema? It has been called all of
these, suggesting that commentators themselves are uncertain about it, and
indeed there is good reason to be so without a sense of either an argumenta-
tive background or some other motivating concern—​and Aristotle does not
give one when he presents his distinctions. The standard presentation does
not provide answers to questions of the sort I have just raised.3

2 As is often noted, these labels are not Aristotle’s, though he has standard ways of referring to

them—​e.g., the efficient cause is often referred to as “the origin of the change” (hê archê tês kinêseôs)
or “the principle from which the change originates” (hothen hê archê tês kinêseôs). The labels appear
to trace back at least to the second century CE: Ptolemy refers to four “well-​known causes: material,
acting, formal, and final” (“tessara eisi polluthrullêta aitia: hulikon poiêtikon eidikon telikon,” Musica,
§ 26 ln. 2).
3 Certainly, there is no simple way to cast the standard presentation as a theory, though

commentators sometimes do this. Aristotle does not claim in these chapters that citing all these four
 Introduction 3

It is a much more straightforward matter, by contrast, to reconstruct the

core philosophical commitments of Platonic, Humean, or Kantian accounts
of causality, and to identify contemporary thinkers whose views constitute,
deliberately or not, developments of those positions, than it is for Aristotle’s.
There are, of course, many problems and disputes that arise in reconstructing
or evaluating their views—​I do not mean to suggest that the content of Kant’s
theory of causality, for example, is especially obvious. I mean, rather, that
the task of presenting their theories in such a way as to make reconstruc-
tion and evaluation possible is easier for these thinkers, in part because they
raise and address philosophical questions about causality more directly than
Aristotle does. When Plato has Socrates introduce the theory of Forms as
causes in the Phaedo, for example, Socrates gives explicit statements about
the types of questions he was investigating, why he was dissatisfied with his
predecessors’ answers, what he thought a good answer should do, the log-
ical status of his own answer, and how (and to what extent) he thinks that
answer solves the problems he raises.4 That is, even if the content of the view
needs clarification, the core claims are explicit and the philosophical context
is well-​marked. Further, when Plato returns to the general topic of the causal
explanation of coming-​to-​be in nature in the Timaeus, he has Timaeus in-
troduce several distinctions among types of causes; here, too, there is a clear

causes is either necessary or sufficient for understanding any natural thing, or for understanding a
special kind of natural thing (e.g., substance), nor does he list four causes for any one particular kind
of thing, even an artifact. The closest he gets to such a claim in Physics II is in c­ hapter 7, 198b5–​9,
where he states that “we render the Why in all ways,” sketching all four types of answer, but this is not
itself a claim that all or some particulars have all four types of cause, nor does it present reasons for
thinking these four modes are necessary for causal explanations. Rather, it is supposed to follow from
the preceding discussion, none of which involves a claim that they are jointly necessary or sufficient
for explaining a particular entity or type of entity. In one of the rare passages where he does list four
causes for a single entity (a human being), Met. VIII 4, 1044a32–​b1, Aristotle only says that, when
seeking something’s cause, since causes are spoken of in many ways, one must state all possible (kinds
of) causes. He then immediately gives examples of cases in which there are fewer than four causes to
be named.

4 Phaedo, 96a5–​107b10, on which see, for example, Vlastos 1969; Fine 1986; Devereux 1994; Sedley

1998; Silverman 2002, ch. 3; and Dancy 2004, ch. 13. There is, of course, no shortage of treatments
of Hume’s arguments in the Treatise and the Enquiry; recent treatments include Kail 2007. For a
full-​scale account of Kant’s theory of causality, see Watkins 2004. Modern treatments of a broadly
Platonic character include, prominently those of Dretske 1977; Tooley 1977, 1987; and Armstrong
2016, ch. 6, who all account for causation in terms of nomological relationships between universals,
despite Armstrong’s scruples about the nature of universals themselves. An explicitly neo-​Kantian
approach is framed by Price 2007 (see also the introduction to the volume in which it is included).
Neo-​Humean approaches are, of course, widespread, including not only the various types of counter-
factual analysis but also more reductive regularity approaches inspired by Mackie 1974.
4 Causality and Causal Explanation in Aristotle

attempt to spell out the reasons for introducing these distinctions and what
work they are meant to accomplish.5
Philosophical theories are also partially self-​reflective, however: they con-
tain both claims about the topic at hand as well as a view, implicit or explicit,
about how we should think or argue about that topic itself. Second-​order
questions of this sort include, for example: Why do we need a theory of cau-
sality at all? What problems or disputes is a theory of causality meant to re-
solve? What are the standards we might use to evaluate such a theory, and
what types of evidence have the most authority?
Here again, at least the broad outlines of this second-​order thinking are
more explicit for philosophers like Plato, Hume, and Kant. Plato has Socrates,
in the Phaedo, present his dissatisfaction with “materialist” accounts of nat-
ural phenomena, especially insofar as their explanations of generation and
change display the kind of inadequacy he thinks generally affects perceptible
reality—​the kind of instability by which something alleged to be a source of
beauty in one context will be a source of ugliness in another, making both
claims suspect.6 Thus, according to the Phaedo, an adequate theory of cau-
sality must yield the right sort of stable relationship without which purported
explanations crumble far too easily. Further, in the Phaedo, the Forms are ex-
plicitly presented as a hypothesis that would allow us to render the properties
and changes of particulars intelligible, in the absence of the kind of teleolog-
ical account that Socrates was hoping to get from Anaxagoras—​that is, one
that appeals to goodness or beauty (100a2–​b9; cf. 101d1–​e3). In the Timaeus,
a teleological account of the right sort is given, and it is in this context that
Timaeus distinguishes different types of causality: one intelligent cause that
aims at what is best, and another, “necessity,” that does not.7 In this way, Plato
gives us the theoretical status and the theoretical context of his main claims
about causality, as well as some relatively clear adequacy conditions for a
good account of the causes of coming-​to-​be.
Hume has a very different approach, of course, and his presentation
enables us to see the resulting differences in second-​order thinking quite

5
See Strange 1985 and Johansen 2004, ch. 5.
6
See Phaedo 96e6–​97b7 for the point that the types of cause to which these earlier thinkers appeal
are subject to this kind of instability, though its precise nature is controversial. For the inadequacy of
explanatory principles that do not appeal to Forms generally, see Rep. V, 478e7–​479e5.
7 See especially Tim. 46c7–​47c4 and 47e3–​48b3. Necessity includes what is described as the “wan-

dering” cause (hê planômenê aitia), as well as the notion of an auxiliary cause (sunaitia); on these and
their relation to necessity, see Johansen 2004, ch. 5. Cf. also the general claim at 28a4–​6, that every-
thing that comes to be (pan to gignomenon) does so necessarily by virtue of some cause.
 Introduction 5

starkly: he thinks that we must come to terms with causation against the
background of certain starting points concerning how anything at all comes
to be present to the mind, which include a sharp cleavage between relations
of ideas and matters of fact.8 Many post-​Humeans, in turn, present their
views against the background of the challenges he raises for the notion of
a necessary connection, and especially the problem of avoiding the kind of
skepticism they threaten, including, most famously, Kant.9 In Aristotle’s case,
however, the second-​order thinking is far less clear, despite the fact that he
evidently thinks that his predecessors were not just incorrect about causes
but were thinking about them in profoundly wrong ways—​and the standard
presentation does not really help here either.
I submit, then, that despite the abundance of texts and commentary we
have about causes in Aristotle, we do not really have a good account of his
theory of causality, in the sense I have described. In part I think this is because
of the way causality permeates so much of Aristotle’s work, as mentioned at
the outset—​so often the frame, so rarely the focus—​but there are other, more
concrete reasons as well.
Aristotle’s own presentation in those canonical chapters of Physics [Phys.]
II is surely partly to blame. Even though these chapters are the ones most ex-
plicitly concerned with general theses about causality in Aristotle’s surviving
works, they are not a philosophical inquiry into causes as a distinctive or
disputed subject in its own right, the way, say, Phys. III 1–​3 is an examination
of change (kinêsis), Met. VII an examination of substance (ousia), and Met. IX
of actuality (energeia). Phys. II as a whole is concerned with nature (phusis)
and the problems that arise in understanding it, especially as an object of sci-
entific inquiry. In Phys. II 3, without preamble or obvious philosophical mo-
tivation, he simply says that, having discussed the notion of nature and how
the natural scientist differs from the mathematician—​an odd-​seeming ques-
tion to which I return later (§ 3.2)—​we should discuss the kinds of cause and
how many there are (194b16–​17). He presents a list with some examples—​
some obviously appropriate, some unclear—​draws a variety of distinctions

8 See, canonically, Treatise I.III.II and Enquiry § 4.I.

9 More recently, many analytic philosophers have approached the question following the broad
pattern applied to other general notions, with a focus on ordinary usage, the “method of cases,”
and so on; see Paul and Hall 2013 for a sustained treatment. Others, approaching from the philos-
ophy of science, build theories of causal explanation while eschewing the task of giving an analysis
of causation itself, such as Woodward 2003, who rightly laments the fractured nature of contem-
porary discussions of causality across different sub-​disciplines (3–​4). Nevertheless, Hume’s critique
continues to cast a shadow even over these other approaches.
6 Causality and Causal Explanation in Aristotle

and sub-​classifications, and then moves on to a discussion of chance and luck

(chs. 4–​6), before returning in II 7 to give a brief account of the way in which
a natural scientist should approach causal explanation. In short, the standard
presentation and its limits are no accident, since they mirror Aristotle’s
own. Phys. II seems somehow both highly informative and philosophically
silent at the same time, giving us a wealth of distinctions, examples, and
specifications, but no arguments, no claims to insight, and no sense that what
is being revealed here is the key to what everyone else missed.10
Another reason Aristotle’s theory is unclear is no doubt the pluralism that
lies at its core—​indeed, it is pluralistic in ways that extend beyond the ob-
vious fourfold distinction, as I shall argue. That pluralism has, along with
modern assumptions about causation, made it possible to treat the modes
separately, and so to assume that if Aristotle does aim to answer questions
about causality, those answers lie in his claims about efficient or material
causes. It also makes it more natural for scholars to examine topics in a piece-
meal way—​to look at his account of the causes of substance (ousia), or tel-
eological explanations in biological contexts, for example—​especially since
for these topics Aristotle does indeed make his motivating concerns and his
argumentation more evident.
Finally, there is the persistent question of the relationship between
Aristotle’s mature thinking about causality and natural science in the Physics
and Metaphysics [Met.], on the one hand, and his account of scientific un-
derstanding in the Posterior Analytics [APo.], on the other.11 The latter work
presents a theory of scientific demonstration with causality at its core: the

10 Aristotle does, of course, extensively discuss his predecessors’ views and point out their

shortcomings relative to his own. In particular, Metaphysics I contains a rich source of under-
standing for Aristotle’s views about the defects of his predecessors’ theories; Phys. I presents a case
for recognizing matter and form as principles of change, against the background especially of earlier
accounts (and rejections) of change; and Phys. II 8 defends the idea that nature acts for the sake of
something—​the thesis of so-​called natural teleology. But these piecemeal, and in some cases notori-
ously unsatisfying, discussions do not give an account motivating or defending the fourfold distinc-
tion itself either, nor do they tell us exactly how Aristotle’s conceptions of these causes are distinctive
enough that it makes sense for him to say that in a way they have not even been discussed at all before
him. In the Metaphysics the causes function as a lens for clarifying the nature of wisdom: his discus-
sion of his predecessors confirms for us that, since no other modes of causality have been identified
other than the ones described in the Physics, we can be satisfied with using them as organizing prin-
ciples for the search for wisdom. He explicitly refers us to the Physics for a full discussion of the
causes at Met. I 3, 983a34. The discussion of principles in Phys. I likewise tends to emphasize the
similarities between Aristotle’s account and the “principles” of his predecessors (e.g., that they all
recognize contraries as principles), rather than the differences. Finally, the enormously controversial
arguments for natural teleology in Phys. II 8, on most readings, make use of a prior understanding of
teleology which, he argues, extends to nature and natural things (see § 1.2).
11 The Posterior Analytics is universally regarded as an early work, and earlier than the Physics;

Barnes 1994, xiv supposes it was written while Aristotle was still at Plato’s Academy.
 Introduction 7

middle terms of demonstrations are supposed to pick out causes of what

they demonstrate, and essences are among the first principles of demonstra-
tion. These accounts of demonstration and essence have much to say about
causes and causal explanation, but they do not present a theory of causality
as such—​for the most part the nature of causality appears to be taken for
granted. Nor is it clear, notoriously, how exhaustive this account is meant to
be as an account of causal explanation (i.e., of the ways in which we appeal
to causes to render phenomena intelligible), or how much of it is meant to
apply to Aristotle’s natural scientific work. Even if we do not claim, as some
commentators have, that Aristotle abandons the Analytics model when he
turns to natural science, we nevertheless cannot assume that the APo.’s dis-
cussion of demonstration is informed by Aristotle’s mature thinking about
causality—​as we shall see, there are good reasons to think that it is not. Thus,
there is a great deal of difficulty in linking Aristotle’s most sophisticated
discussions of causality with his most developed account of scientific expla-
nation, despite the prominence of causal notions in the latter.
There are, then, good textual and philosophical explanations for the fact
that it is difficult to discern a clear and unified theory of causality in Aristotle’s
general remarks about causes, and for why scholars and philosophers have
not reconstructed one. Yet he is explicit that his predecessors only had a par-
tial grasp of whatever causal principles they did identify, and that his own
achievement does not lie simply in having collected the whole set. Indeed,
for each mode of causality, Aristotle is explicit that it has not just been imper-
fectly grasped, but ignored or outright rejected by some of his predecessors.12
Such an insistence makes little sense without some general reflection as to
what makes for an adequate understanding of causality and causal explana-
tion. Thus, we also have good reason to suppose that Aristotle does have a
general theory, at least to the extent that what he says about causes and causal
explanation is guided by a reasonably determinate sense of how his own
views about causes are superior to those of his predecessors. Tracing this line
of thought is the main goal of this book.
Commentators have raised a number of interlocking questions, however,
about whether, broadly speaking, it is appropriate to take Aristotle’s concerns
with aitiai to match with our concerns about causes, which would imply that
tracing this line of thought would not in fact yield a theory that is usefully
compared with modern theories of causality. It is common, for example,

12 I discuss these passages in Chapter 2.

8 Causality and Causal Explanation in Aristotle

to note semantic worries about whether ‘aitia’ really means or should be

translated as ‘cause’, rather than ‘because’ or ‘explanation’. These worries are
sometimes coupled or confounded with conceptual worries about whether
Aristotle’s claims about aitiai (or Plato’s for that matter) should be interpreted
as claims about causes (i.e., whether their concept of aitia, if they have one,
is a match for our concept of cause, if we have one). There are, further, more
theoretical worries about whether certain modern assumptions about causes
are incompatible with Aristotelian or Greek assumptions about aitiai, which
would suggest that the modern theoretical enterprise is in some important
way different from the ancient one.13
The semantic claim by itself is not compelling as long as we do not in-
sist that ‘cause’ must mean what recent Anglophone philosophers have taken
causes to be, and so I will follow the tradition of translating ‘aitia’ and ‘aition’
as ‘cause’ except where noted.14 The more substantive worries about concep-
tual and theoretical overlap are not borne out either, as I hope will be clear
from what follows. Even at this stage, however, there are good reasons to

13 For example, whether Aristotle’s efficient causes are “things that do things to other things,” as

M. Frede 1980 contests, or whether final causes can genuinely be responsible for things that are tem-
porally prior to them, or whether we must assume causes are events.
14 Occasionally ‘explanation’ is indeed better, but not because there is anything misleading about

‘cause’ in most contexts. For what it is worth, the Latin ‘causa’ apparently first entered several
Romance languages as meaning “thing” (and is the ancestor of ‘chose’ in French, ‘cosa’ in Spanish);
later, it re-​entered those languages, and English, as ‘cause’, with the technical sense from its use in
medieval philosophical discussions, having been used to translate ‘aitia’ into Latin from the classical
period onward (OED). There was thus no decision to use an antecedent English word ‘cause’ to trans-
late Aristotle’s ‘aitia’; rather, ‘cause’ owes its usefulness as a translation of ‘aitia’ to its intimate history
with the term. Discomfort with the usage seems to have begun only in the late nineteenth and early
twentieth century, despite criticisms of Aristotle’s own distinctions stretching back to Theophrastus.
Grote 1872 uses ‘cause’ throughout his book on Aristotle, without noting any particular difficulty or
awkwardness of translation, and Mill 1872, in his Logic, is comfortable using ‘cause’ in his discussion
of Aristotelian as well as “Scotch” causes. But Heath 1876 criticizes Mill for this usage, and Wicksteed
and Cornford’s 1929 translation of the Physics warns us that “it does violence to the English idiom
to call the material out of which a thing is made, or the distinctive attributes which define it, its
‘causes’. . . . Paraphrase or barbarism offer the only escape from using English words in a sense that
they cannot really bear” (126–​27). Among recent commentators, Ross 2004, Ackrill 2001, 111,
M. Frede 1980, Annas 1982, Barnes 1994, 89, Charlton 1992, 98, Hankinson 1998, 132, Lear 1988,
28, Moravcsik 1974, 1975, Santayana 1925, 183–​84, Sorabji 1980, 40, and Vlastos 1969, 294f. all note
that the translation requires qualification in one way or another, but their claims range from minor
caveats (Ross, Ackrill) to sharp warnings that the translation is deeply misleading (Vlastos, Frede,
Annas). On the other hand, some commentators have remained unmoved: see, e.g., Gotthelf 1987a,
Freeland 1991, and Furley 1996. Interestingly, the seventeenth-​century authors of the so-​called Port-​
Royal logic (Arnauld and Nicole 1981) make the opposite complaint about scholastic definitions of
cause, which they claim are imprecise and do not seem to capture all of the kinds of cause, so that it is
better to leave ‘cause’ undefined—​“l’idee que nous en avons étant aussi claire que les définitions qu’on
en donne” (238). There is, of course, no point in arguing that ‘cause’ and ‘aitia’ mean the same thing,
but nor do we make progress on the more substantive questions by suggesting that, for semantic
or conceptual reasons, Aristotle’s claims about aitiai are really claims about something other than
causes. For a clear survey of the history and evolution of the term ‘aitia’ itself, see Natali 2013.
 Introduction 9

suppose that the basic problems underlying Aristotle’s inquiries are essen-
tially the same as the ones underlying modern ones.
There are many reasons for wanting to understand the nature of causality,
but the presentation in Met. I expresses two of the central ones. First, cau-
sality is at the heart of explanation and the pursuit of knowledge or under-
standing, especially in natural science. A good account of causality, Aristotle
clearly thinks, is required if we are to understand and justify a broad swath
of our knowledge-​seeking practices; to say what kinds of explanation are
good or bad, in what contexts, or for which topics; to distinguish genuine
from pseudo-​explanations; to know what kinds of things can be causally
explained, and what cannot; to know what can be explained at all and what
must be understood some other way, or can only be described.
Second, as Aristotle’s exposition makes clear, one’s approach to causality
bears on fundamental questions about appearance and reality, and how they
relate to one another. Imagine a debate between three parties: one tells us that
to understand the appearances we confront in everyday life and in observing
the natural world, we need to seek the material stuff that underlies those
appearances and controls them from below; another tells us that we need
to find the abstract patterns and ultimate realities that govern the world of
passing appearances, which can only be grasped by the mind and not the
senses. A third interjects, and tells us that the other two are both wrong: we can
grasp and explain fundamental reality without “descending” or “ascending”
to a different ontological plane, and yet still arrive at a profound type of un-
derstanding, as “scientific” or theoretical as we might wish, rather than a
merely detailed description of the shifting appearances we call the natural
world. This is a philosophically dramatic clash of viewpoints, and it is more
or less the story Aristotle tells us about himself and his predecessors.15 He is
explicit that a proper understanding of causality itself is what allows us to see
this third way, yet we do not have his argument, really, as to how his own ac-
count of causality helps ground his contention. Like him, however, we tend
to assume that ultimate reality and genuine causality go together, and it is
common to argue that some phenomenon is mere appearance, or of ontolog-
ically lesser status, or epiphenomenal, because it is not causally efficacious or
arises incidentally from different causal processes that are genuine.

15 Indeed, some of the puzzles in Met. II, especially the sixth (998a20–​b14) and eleventh (1001a4–​

b25), set up an opposition between the Platonist and reductive materialist approaches in just this way,
such that the dissolution of the puzzle will require seeing how each is mistaken.
10 Causality and Causal Explanation in Aristotle

Thus, while one might raise questions about the extent to which Aristotle’s
accounts of aitiai fit with or diverge from contemporary accounts of causation,
the motivations underlying his reflections on aitiai are of clear philosophical
concern to us, and they connect very closely to basic questions about causality.16
In view of his pluralism and the broad but still limited scope he places on causal
explanation, his view is also of interest to recent debates about the nature and
scope of non-​causal explanation.
There are two further motivations for trying to formulate Aristotle’s theory
of causality in the unified way I have proposed. First, there are a number of
persistent but seemingly more circumscribed puzzles and controversies in
Aristotle’s work which, I think, are best understood and answered in light of
a general theory of causality rather than independently of it. These include
questions about the scope of his commitments to teleological explanation in
nature, his understanding of causal powers and causal action, the nature of his
essentialism, and his commitment (or lack thereof) to the so-​called Principle
of Causal Synonymy (the principle that only something F can be a genuine
cause of something else’s being F), as well as his frequent but sometimes ob-
scure claims that formal and final causes are “the same” or “one.” It is valuable to
pursue piecemeal analyses of various types of causal relationship in Aristotle’s
work, but problems like these also call for a synthetic understanding that treats
them in relation to a theoretical whole, absent any reason to think this cannot or
should not be done.
Second, Aristotle’s four-​causal framework is often identified as at the
core of the pre-​modern, pre-​scientific view of nature that was abandoned
or at least thoroughly revised by the scientific revolution. Even though
it has been well-​established that the most radical versions of this way of
reading the revolution are caricatures, there is an enduring supposition,
even among philosophers and historians of philosophy, that somehow the
Aristotelian conception of causality, especially in light of its commitments
to hylomorphism and teleological causality, cannot really be separated from

16 These reservations about aitiai and causes should be further distinguished from the one

considered in a moment, namely, whether Aristotle’s views about causes are so tightly connected to a
pre-​modern understanding of nature that they do not apply in a modern scientific context. This ques-
tion is important and substantive, but it is sometimes miscast as a question about whether Aristotle’s
notion of aitia is really a notion of cause, or whether his word ‘aitia’ is appropriately translated by
‘cause’ in English. Given the philosophical motivations I have just described, the idea that for con-
ceptual reasons, in his investigations of ‘aitiai’, Aristotle is engaged in a deeply different project than
the one in which Hume, Kant, Mackie, Hart and Honoré 2002, Armstrong, Lewis 2000, or Woodward
2003 are engaged when discussing causation, simply is not true in any worthwhile sense.
 Introduction 11

a pre-​scientific understanding of nature and its metaphysics.17 At the same

time, when we read these chapters of the Physics or Metaphysics I, many of
their main statements seem plainly reasonable. Indeed, for all the worries
about their suitability for a modern scientific worldview, all of Aristotle’s
four causes continue to be deployed in causal explanations in both scien-
tific and philosophical contexts, in various guises, whether separately or
together: material and efficient-​causal explanations often stand alongside
teleological and functional explanations, various forms of essentialism are
developed and defended, and many examples of what are now often called
“grounding” explanations are recognizably types or descendants of what
Aristotelians would have called material-​or formal-​causal explanation.18 But
the problems that made Aristotelian views of nature seem inconsistent with
modern science do not really turn on whether or not all of these relationships
deserve the label of ‘cause’, so it remains unclear how and how much our own
explanatory theories and practices are really different from Aristotle’s. Part of
the difficulty in assessing those differences lies again, I think, in the fact that
presentations of Aristotle’s account of causality tend to be so closely bound
up with particular applications or cases, especially that of substance, that it is
simply not clear how much paradigmatically Aristotelian metaphysics is re-
ally embedded in Aristotle’s claims about causes.19
An account of Aristotle’s theory of causality therefore holds the promise of
clarifying more local questions about his views, as well as answering broader
questions about the continuities and discontinuities between ancient, medi-
eval, and modern scientific approaches to the natural world and our under-
standing of it.

17 For an account of the development of early modern science which is sensitive to the ways

in which it was and has been taken to reject various strands of Aristotelian natural science, see
Shapin 1996. For an account of some key developments between Scholasticism and early modern
conceptions of causality, see Schmaltz 2008, 2014. For a survey and analysis of different historical
claims about “the Scientific Revolution” and its relation to Aristotelianism, see Cohen 1994, ch. 4.
18 The nature and scope of “grounding explanations” are controversial and the subject of an ever

expanding literature. See the contributions in Raven 2020, many of which note the connections to
Aristotle and other ancient authors.
19 To cast some doubt on this tendency, it is enough to note that Aristotle only mentions substance

(ousia) once in Phys. II 3–​9, and only to make the point that a teleological explanation should explain
something in virtue of its being better not unqualifiedly, but in relation to a substance. This is the
very last line of c­ hapter 7 (198b9). There are only a handful of other uses in book II, all of which are in
­chapter 1, which is a dialectical study of the notion of nature—​a context in which Aristotle often uses
the term for substance in describing the views of his predecessors.
12 Causality and Causal Explanation in Aristotle

I.2. Outline

This study thus has two main motivations: to understand Aristotle’s theory
of causality in the philosophical context of his critiques of his predecessors’
views about causation and causal explanation, and to understand it in such a
way that it can fruitfully be considered as a philosophical approach alongside
broadly Platonic, Kantian, and Humean approaches. These aims are interde-
pendent, of course.
Because of the various ways in which causality permeates Aristotle’s work
so thoroughly, however, it can seem as though such a project would require
nothing short of a full reckoning of his theoretical philosophy, and so would
in the end accomplish little. Nor can we dismiss the skeptical hypothesis that
Aristotle does not really have a full-​blown theory, as opposed to a reasonable
and flexible framework that he applies and extends in a piecemeal way—​not
randomly, but not because of any overarching general account of causality,
explicitly or implicitly held.
I think, however, that we can see a general theory at work that answers
to these main motivations, and that we can isolate the basic elements of his
answers to these main philosophical questions we tend to raise about cau-
sality, especially in light of the ways Aristotle himself raises and recognizes
them. The only way to decide the question is to see what results from the
attempt.
My strategy is first to animate the key chapters about causality, namely
Phys. II 2–​9, but especially 2–​7.20 As I have noted, these chapters present
Aristotle’s distinctions without philosophical motivation or defense, but we
can see what they accomplish, in theoretical terms, by reading them against
their philosophical background, particularly the criticisms Aristotle makes
elsewhere of his predecessors that connect most directly to questions about
causality, as well as the general constraints concerning causation and ex-
planation that he describes in other works, such as On Generation and
Corruption, the Metaphysics, and the Posterior Analytics. Against this back-
ground, several features of those key chapters stand out, which allow us to
delineate some important conceptual claims Aristotle makes about causes,
and which govern the four-​causal framework as a whole. This is the task of
Part I.

20 While Phys. II 3 contains the canonical presentation, II 2 has already, in responding to the de-

marcation challenge in relation to mathematics (discussed below), introduced some key points about
all four modes of causality.
 Introduction 13

In each of Parts II and III, I then frame a primary question about cau-
sality which is also at the heart of Aristotle’s critique of his predecessors, and
argue for an interpretation of the way the views in Phys. II and III express
commitments that respond to those questions and problems. Part II raises a
metaphysical question, while Part III focuses on an epistemological one.
In more detail, the three parts of the book run as follows:

I.2.1. Part I

After briefly discussing the canonical chapter of Phys. II 3 and the problems
it raises, both by itself and in the context of the whole line of discussion
in Phys. II (Chapter 1), I turn to an examination of Aristotle’s criticisms
of his predecessors, focusing especially on criticisms that relate to cau-
sality (Chapter 2). Those criticisms mainly divide into two sorts. Some are
motivated by worries about how the natural world can be a genuine object
of scientific understanding (e.g., in light of worries about whether some-
thing fundamentally characterized by change and instability, as nature is,
can be fully understood). Others relate to metaphysical worries about the
relationships between certain types of proposed causal principle and the
phenomena of which they are claimed to be causes. Next, I turn to a dif-
ferent set of background philosophical considerations pertinent to Phys. II 3,
having to do with general constraints on explanation and explanatory prac-
tice (Chapter 3).
I then return to Phys. II 2–​7 to highlight the philosophically signifi-
cant claims of these chapters against this background, and I assess their
implications for my main questions (Chapter 4).
Three key interpretive claims emerge from this part, which in different
ways reflect Aristotle’s critiques of his predecessors’ views:

(1) Aristotle, I think, diagnoses many of the problems with his

predecessors’ accounts of nature as stemming their commitment to
a kind of reductionist approach, requiring what I call a special meta-
physics for causal explanation. This approach typically connects causal
explanation with a special ontology and/​or privileged relationships to
such an ontology. This reductionist approach, as I argue, is pursued in
different ways by Platonists and materialists alike, though its precise
14 Causality and Causal Explanation in Aristotle

nature, and the problems Aristotle sees with it, must be distinguished
from other types of reductionism.
(2) The schema presented in the Physics represents an extension of a
simpler schema given in Posterior Analytics II 11—​simpler in that
it appeals to fewer and more basic concepts. That simple schema
presents four ways of accounting for a situation in which something
answering to a predicate is said to hold of some subject—​that is, in
some appropriate sense of the term, a fact. Phys. II 2–​7 take these
distinctions and extend them to the task of explaining things and the
changes they undergo rather than true predications, in particular to
the more controversial task of stating the causes that are intrinsically
(kath’hauto) connected to stable kinds of change or thing, whether
natural or artificial—​that is, to things that are not co-​incidences or
otherwise unique.21
(3) While Aristotle’s four-​causal schema is often loosely described as being
comprised of four distinct metaphysical roles or relationships, this is
inaccurate and misleading in important ways. They are rather what
I shall call second-​order roles, since they each cover what Aristotle
thinks of as clusters of importantly different and more determinate
metaphysical roles and relationships—​and the differences matter
a great deal for a variety of reasons that have not been appreciated.
Thus, the schema in Phys. II represents an ambitious extension of a
logically more basic (and possibly earlier) schema for explaining facts
of any sort, to a more metaphysically committed task of explaining

21 Here and throughout the book I use ‘intrinsic’ as a placeholder translation of ‘kath’hauto’, fol-

lowing Irwin 1988, Freeland 1991, and Meyer 1992, among others—​and only in this way except
where noted. ‘Kath’hauto’ (“with respect to itself ”; often translated ‘per se’ or ‘in its own right’) is
usually contrasted by Aristotle with ‘kata sumbebêkos’, literally something like “with respect to being
placed or occurring together,” often translated by ‘per accidens’, or ‘accidental’, ‘incidental’, or ‘co-​
incidental’. It is also contrasted, more rarely, with ‘kath’heteron’, for example, at De An. I 3, 406a4—​
“with respect to another,” or ‘per aliud’. I discuss some of these contrasts later in § 3.1. These are all
quasi-​technical terms, and Aristotle’s use of them is sometimes flexible or unclear; all translations
have their own drawbacks. ‘Intrinsic’ is similarly quasi-​technical, and usually not bad, but it should
above all not be taken to imply that something is internal. In Aristotle’s translated usage, then, a
doctor is normally an intrinsic and external cause of health in a patient, and so Aristotle’s claim in
Phys. II 1 is that to have a nature is to have an intrinsic origin of change which is also internal in some
important sense. There is a modern usage of ‘intrinsic’ in relation to causation which is different,
though, I think, adjacent to this one; for that sense see, e.g., Menzies 1999, Hall 2004b. I will use
‘co-​incidental’ for ‘kata sumbebêkos’, since ‘intrinsic’ requires a contrasting English term, and I am
reserving ‘extrinsic’ for broader work (see again § 3.1); here again the term is something of a place-
holder, since not everything that ‘co-​incides’ with something else in this usage is a mere accidental
concomitant.
 Introduction 15

regularly occurring phenomena in terms of their intrinsic causes. It is

nevertheless far more metaphysically neutral than it may appear and
is often presented.

I.2.2. Part II

The metaphysical neutrality of the schema in one sense makes it flexible but
also raises worries about explanatory power. If there is no special metaphysics
for causality, what account can we give of the relations between causes and
what they cause? These worries derive from a more general set of tensions
surrounding the metaphysics of causation, to which Hume’s famous rejection
of the notion of “necessary connexion” is one kind of response. The tensions
are broadly due to the fact that causes and effects are somehow supposed to be
connected tightly enough to ensure that their relationship is not one of mere
co-​incidence or regular co-​occurrence, but not so tightly connected that the
phenomenon collapses into a kind of self-​realization of an effect.
Finding an acceptable compromise between these various pulls is, I sug-
gest, one of the main problems for giving a good account of the relationship
at the core of the metaphysics of causality: that between causes and what they
cause. Aristotle himself, in various guises, raises some of these same concerns
for his predecessors. How, then, does he think causes in nature relate to what
they cause, and how can his account avoid the metaphysical problems he
raises for his predecessors that pertain to the tension I have described?
Since Aristotle, on my account, rejects the appeal to a special ontology
for causal explanation, and since he is also committed to a kind of met-
aphysical pluralism within each mode, he does not think that there is one
universal kind of connection between, say, material causes and what they
cause. Rather, I argue, his position involves extending the features com-
monly associated with essentialism about natural kinds to the whole range
of regular phenomena that have intrinsic (i.e., kath’hauto, or per se) causes
(Chapter 6). These include not just substances and the processes by which
they are generated and destroyed but also those changes, activities, and states
of affairs which always or usually occur in the same way, by the same causes.
For these phenomena, too, the natural scientist should give a so-​called real
definition, that is, one which specifies what they are by stating why they are—​
by stating their intrinsic causes. I argue that Aristotle does just that for dif-
ferent kinds of change, giving a brief and cryptic statement of how this is to
16 Causality and Causal Explanation in Aristotle

be done in Phys. III 3, and following precisely the suggested model in certain
key cases, especially clearly in his so-​called chemical work, Meteorology IV.
Clarifying this point requires finding a path through a thicket of philosoph-
ical and interpretive difficulties about essences and the different types of def-
inition and definienda Aristotle recognizes.
Aristotle further recognizes, I argue, different types of what I call “causal
profile,” corresponding to different types of regularly occurring phenomena
with different arrays and types of causes (Chapter 7). As on many standard
presentations of the four causes, some natural phenomena do indeed have all
of their causes intrinsically, the way substances do. Many, however, do not,
and even among those that do, there are important differences in the ways
the causes of a causally defined kind relate to each other, and to the phenom-
enon they explain.
I argue that Aristotle observes an especially important distinction be-
tween what I call “origin-​dominant” and “end-​dominant” causal profiles.
This distinction is expressed briefly but relatively clearly in Met. VII 17,
though Aristotle does not go into detail. When we elaborate on this distinc-
tion, however, we can offer satisfying interpretations of some persistently dif-
ficult questions and passages, including the nature of his commitment to the
so-​called Principle of Causal Synonymy, and the various passages in which
he seems to identify the formal and the final causes (on which see the final
section of Chapter 7).
I then return (Chapter 8) to the challenge of accounting for the various
features of causal relations that seem to be in tension. Focusing especially
on the question of discreteness between efficient causes and what they
bring about, I articulate the components of a transeunt-​causal change (i.e.,
a change brought about in one thing by something else) and how they re-
late to the change itself. Several distinctions then need to be drawn between
different ways in which things may be considered discrete or not discrete,
and these in turn need to be applied to the different components Aristotle
recognizes in a causal interaction, as well as the distinction between actu-
ality and potentiality. The resulting response to the question of discreteness
is quite complex, but that complexity is an inevitable function of several key
features of Aristotle’s view. Rather than admitting a simple answer, which
would be inappropriate given the complexity of the phenomena, the ques-
tion of discreteness and the distinctions drawn allow us to see how Aristotle
negotiates the different demands we face in giving a good account of causal
interactions.
 Introduction 17

I.2.3. Part III

Finally, I turn to questions about the epistemology of causality. These issues

are, I think, as important as any of the others, but perhaps the least developed
in Aristotle’s work, as well as that of his commentators.22
The problem, in effect, is one of appearance and reality, or the relationship
between the so-​called manifest and scientific images of nature. As I argue
(Chapter 9), in Aristotle’s terms, this is the problem of moving from what
is “more knowable to us” (gnôrimôteron hêmin) to what is “more knowable
by nature” (gnôrimôteron phusei), or “more knowable without qualification”
(gnôrimôteron haplôs), and how causality figures in this transition. That
Aristotle takes this question very seriously is clear, but the brief texts in which
he confronts it directly are some of the most cryptic in the whole corpus, and
the commentary tradition, understandably enough, largely follows the texts.
Describing his account is therefore necessarily more of a reconstructive ef-
fort than that of the previous sections. Each point I make has a firm basis in
the texts, however, while building on what I take to have been established in
Parts I and II, and the view that results is a recognizably Aristotelian one that
fits with his main philosophical commitments.
I focus especially on aspects of what is “more knowable to us”—​the mani-
fest image—​and how Aristotle thinks we progress from it to a grasp of causes,
in such a way that he would have reason to claim that his views make a scien-
tific grasp of nature possible in ways that his predecessors’ do not. The main
elements of my account are that (1) there are epistemic asymmetries among
different modes of causality and with respect to different causal profiles
(Chapter 10); and (2) at least some causality is such that it can be grasped
with our more basic cognitive powers of observation, that is, without infer-
ence and without going beyond the conceptual resources available to ordi-
nary perceptual judgment (Chapter 11). Some causality is, we might say, just
as it appears, at least to careful observation, and so for these phenomena there
is no necessary gap between appearance and reality. To illustrate the view as a
whole and use it to clarify some of Aristotle’s claims about particular natural
phenomena, I examine Parts of Animals II 2–​3, which is concerned with how
we should approach a scientific understanding of the nature of blood, and, fi-
nally, connect that account back to some of the problems raised by Aristotle’s
remarks on induction.

22 Claims about causal inference are, of course, at the heart of Hume’s critique as well.
18 Causality and Causal Explanation in Aristotle

Thus, in contrast to what I take to be the implication of most readings

of Aristotle as well as a standard contemporary picture, understanding the
causal structure of the natural world is not a matter of moving from an image
that lacks causal content or causal concepts to one that does. That picture
of epistemic progress fits more naturally with the kind of appeal to a spe-
cial metaphysics for causal explanation that Aristotle rejects. Nor are all
four causes epistemically on par with one another. Rather, I argue, the path
from the most basic to the most sophisticated kinds of causal knowledge is
anchored in non-​inferential knowledge of certain kinds of cause and cau-
sality, and it is marked by different and more sophisticated causal concepts
as it progresses.

I.3. Themes, Consequences, Comparisons, and Absences

As the summary in the previous section makes clear, pluralism is a recur-

ring theme. Aristotle is not just a pluralist about the modes of causality, but
about the different types of relationship within the various modes between
causes and what they cause, and about the different types of epistemological
task we confront in coming to know them. This is thoroughly in character
for a philosopher who generally seeks to observe salient differences among
general groupings, and to privilege the more determinate specifications of
lower-​level kinds over highly general and, he thinks, often deceptively unin-
formative accounts. These pluralistic commitments are important for many
reasons, but a central one I think is this: Aristotle thinks that if we follow
the approaches of his predecessors, the best we can achieve is a kind of “co-​
incidental” knowledge of natural things—​that is, we can only know them in-
directly, and therefore only in a qualified way. If natural phenomena are to be
objects of scientific understanding in their own right, however, we must be
able to know them unqualifiedly—​hence, not just in relation to something
else—​and this, he thinks, requires a more articulated approach to causation
that can allow for different types of explanandum, the knowledge of which
places correspondingly different cognitive demands on us.
Another theme is that Aristotle’s views about causality are often studied
and assimilated too closely with his views about especially important cases of
causally explicable phenomenon, most prominently substance. Indeed, part
of what motivates this book is the thought that Aristotle’s claims about cau-
sality have suffered not just from their presentation, but from a subsequent
 Introduction 19

failure to disentangle them from these more prominent applications of

causal analysis. Substance is indeed basic on Aristotle’s view in many ways.
For this reason, however, it is not only difficult to understand and a source
of persistent controversies, but it also lies at the extreme end of the spectrum
of natural beings in a way that makes Aristotle’s causal concepts harder to
clarify. We can do better in that regard by attending to the full range of phe-
nomena Aristotle thinks admit of genuine causal explanation, and indeed,
since Aristotle ultimately aims to understand substance in light of causality,
we can gain a better understanding of his views about substance as well if we
can tease the two apart.23
I have taken my primary task to be arguing for the account of Aristotle’s
theory itself, rather than about whether it succeeds or how it compares with
other accounts, and so this introduction has largely followed that line of
argument. Along the way, however, I do draw connections, raise and con-
sider objections, and make some assessments of the view. One theme of
those sections is that, if Aristotle’s theory of causality has modern relatives
or at least natural friends, they are in the family of views in recent philos-
ophy of science known as “neo-​mechanism.”24 That connection is rarely
noted in the scholarly or philosophical literature, but Aristotle and the neo-​
mechanists share similar motivations—​to avoid reductionism and to frame
causal explanations that respect the contours of the phenomena they are
addressing—​as well as the view that natural phenomena are often complex
wholes made up of simpler, causally interacting processes which always or
usually occur in the same way. I discuss the similarities and differences espe-
cially in § 8.6. Another main “exoteric” theme is that comparing Aristotle’s
view to modern ones is often very fruitful, but rarely direct. When we un-
derstand what he claims and what he does not claim about causation, I think
two aspects of his view stand out especially clearly in contrast with a pair of
related assumptions common to a great deal of contemporary work. One is

23 For similar reasons I have also paid less attention to the difficult topic of the inter-​transformation

of the elements, especially since it is bound up with the question of prime matter. For a recent study of
that question, see Lewis 2008. Likewise, I have tried not to use artifacts as primary illustrations as far
as possible, since their role as models for substance and for causality more generally is easily tangled.
Causality in general, I submit, is seen most clearly through examples in between these extremes of
substance and prime matter, which lie at the limits of natural science, as Aristotle conceives it.
24 Perhaps the most prominent contribution is Machamer, Darden, and Craver 2000, but see also

Glennan 2016 and Craver and Tabery 2019. For a useful survey, see Andersen 2014a, b. These views
and their relationship to Aristotle’s should be sharply distinguished from views known under the
more standard “mechanist” label. Aristotle’s relationship to mechanism in that sense is discussed in
Berryman 2009.
20 Causality and Causal Explanation in Aristotle

the assumption that giving a genuine, scientific account of nature involves

moving from the “ordinary” ontology with which we are presented in eve-
ryday life to a special ontology consisting only of objects which meet cer-
tain more stringent criteria for existence or causal power. A version of this
assumption is held, he thinks, by the materialist and Platonist reductionists
whose approaches he rejects, but, of course, the manner in which he rejects
the assumption depends on how he construes their views and his own. The
other is the common and related assumption that causal relations somehow
supervene on non-​causal facts, events, or states of affairs, such that giving an
account of causation amounts to analyzing the relationship between these
basic, non-​causal facts and the supervenient causal ones.25 That assump-
tion has clear and profound implications for discussions of the metaphysics
of causation, and less clear but equally profound implications for questions
about the epistemology of causation. I think that Aristotle rejects it, and has
decisive reasons to do so.
Since I have attempted to trace Aristotle’s theory of causality in argu-
mentative context, there are a number of other issues one might expect to
be discussed in a book about Aristotelian causality which are not. These
include the role that notions of necessity play in Aristotle’s account of cau-
sality, his theory of motion (including projectile motion, with its notorious
problems), and theology, all of which have been given attention in their own
right. Further, for the reasons given earlier, I have discussed substance and
attendant questions about hylomorphism only as much as necessary for the
argument.26
The position as I have sketched it in this introduction in some ways nec-
essarily seems more complex than its underlying philosophical motivations.

25 This view and its status are shown very clearly by Paul and Hall 2013, who present their general

survey as assuming “a broadly reductionist outlook, according to which facts about which events
cause which other events are fixed, somehow, by (i) the facts about what happens, together with (ii)
the facts about the fundamental laws that govern what happens. Minimally, that’s a supervenience
thesis: no two possible worlds differ with respect to what causes what without differing with respect
to what happens or with respect to what the fundamental laws are that govern what happens. But, as
will become apparent, the most important philosophical approaches to causation aim for something
arguably stronger, namely, an account of causation that lays bare the causal ontology in terms of how
causal facts are grounded in, reduce to or depend upon these more basic facts” (7–​8).
26 Likewise, I do not discuss some topics that are prominent in recent discussions of causation,

since they are not of central importance to clarifying Aristotle’s theory. These include questions about
whether causation is transitive, whether absences can be causes, whether causes must be prior in time
to their effects, and how causal relations should be “modeled.” In some cases, I think relatively clear
answers do result from the theory as I have reconstructed it (as, for example, for the question of si-
multaneity, which should be clear by the end of Part II), while in others the question may simply lose
its relevance, since some modern questions only result from adopting a certain framework.
 Introduction 21

Because of his pluralism and his tendency to prefer causally determinate

levels of specification to general ones, Aristotle’s views often resist easy sum-
mary. But the philosophical import of the position and the motivations be-
hind its pluralism are reasonably straightforward. Against the persistent
tendency to look for explanations behind, outside, above, or below the
appearances, which ultimately treat them as somehow only of diminished
status, Aristotle presents a picture according to which the things that we con-
front in the appearances are themselves primary realities, even if they are
usually not ultimately just as they appear. His response to the various kinds
of explanatory reductionism is a version of Butler’s dictum that “Everything
is what it is and not another thing,” and one of his aims is to show that this re-
sponse still leaves room for a scientific understanding of those realities that is
profound, and does not consist merely in redescribing the way things appear.
Some parts of the picture I present are heterodox, some are not, but the
picture as a whole is perhaps neither heterodox nor orthodox, since, for the
reasons mentioned earlier, I do not think there is a clear orthodoxy. There
has not been a place at the table for an account of the whole, and there really
should be.
 PART I
C ONC E PT UA L STRU CT U R E
 1
Reading (and Animating) Physics II 3

1.1. The Problem of Physics II 3

In formulating the most basic questions about Aristotle’s four causes, both as
constituting a theory and as a response to certain problems, we are at a disad-
vantage. The chapter in which the four causes are presented, Phys. II 3, does not
argue for the distinctions, nor are they presented in any explicit context of dis-
pute or inquiry into the nature of causes. As Ross (1936) says, “We do not know
how Aristotle arrived at the doctrine of the four causes; where we find the doc-
trine in him, we find it not argued for but presented as self-​evident” (37). Our
textual evidence is a series of statements of varying clarity, the philosophical
aims and motivations of which are not apparent.
After introducing his topic in Phys. II 1 as the study of nature, clarifying what
he takes this to imply, and then distinguishing the natural scientist from the
mathematician in II 2, Aristotle simply says:

Having distinguished these things we must examine causes, both what sorts
there are and how many they are in number. For since this inquiry is aimed
at knowing, and we do not think we know each thing until we grasp the Why
for each thing (and this is to grasp its primary cause), it is clear that we must
also do this concerning generation and destruction, and every natural trans-
formation (metabolê),1 so that knowing their principles, we may attempt to
refer each of the things being investigated back to them.
So then, one way in which ‘cause’ is said is as that out of which some-
thing comes to be and which is inherent2 in it; for example, the bronze of

1 Translating Aristotle’s terms related to change is difficult: he recognizes three canonical forms

of ‘kinêsis’—​increase/​decrease (auxêsis/​phthisis), alteration (alloiôsis), and locomotion (phora);

these are included under the even more general heading of metabolê, which also includes genera-
tion and destruction (see Phys. V 1 for the canonical distinctions). He does not always seem to ob-
serve the differences strictly. I use ‘change’ for ‘kinêsis’ throughout and choose ‘transformation’ as
the least awkward for ‘metabolê’, but where the terminology matters I will provide Aristotle’s term in
parentheses.
2 ‘Enuparchontos’ also presents some difficulties of translation, which in the end are difficult to

disentangle from philosophical questions about the nature of matter in a hylomorphic compound.

Causality and Causal Explanation in Aristotle. Nathanael Stein, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197660867.003.0002
26 Causality and Causal Explanation in Aristotle

the statue and the silver of the cup, and the genera of these things. Another
way is the form, i.e. the paradigm, and this is the account of the essence, and
its genera (for example of the octave, the ratio two-​to-​one, and in general
number), as well as the parts in the account.
Further, the primary principle from which the transformation or rest
originates: for example, the person who has deliberated is a cause, and the
father of the son, and in general what makes (to poioun) of what gets made
and what transforms (to metaballon) of what gets transformed. Further, as the
end: and this is that for the sake of which, for example, health of walking about.
For when asked, “Why does he walk about?” we say, “in order to be healthy,”
and speaking in this way we think we have given the cause. And however many
things come to be leading up to the end, when something else is bringing
about the change, for example, of health, slimming or purging or medicines or
instruments—​all these things are for the sake of the end, but they differ from
one another by being some of them functions and some of them instruments.
So then, the causes are spoken of roughly in this many ways, and it
follows that, as the causes are spoken of in many ways, there are also many
causes of the same thing, not co-​incidentally; for example, both sculpting
craft and the bronze [are causes] of the statue, not in accordance with any-
thing else [ou kath’heteron ti] except insofar as it is a statue, though not in
the same way, but rather the one as matter and the other as the origin of the
change. (194b16–​195a8)

This text provides the most detailed description in the corpus of the four
causes as a group.3 The rest of the chapter presents:

The example of metal suggests not just something present in what comes to be, but that the material is
somehow continuously the same sort of thing from beginning to end. This seems incorrect for many
cases such as biological generation, making translations such as ‘component’ (Reeve 2018) or ‘per-
sistent’ (Hardie and Gaye in the Revised Oxford Translation) appear too strong (see Henry 2019, chs.
2–​3). In any case, given Aristotle’s examples later in the chapter and the philosophical questions he
works through about matter here and elsewhere, we should not expect the term itself to be especially
committal.

3 It is not immediately clear whether the fourfold distinction implies that ‘aitia’ is a multivocal

term, i.e., ‘pollachôs legomenon’, or whether Aristotle might have another kind of multiplicity in
mind, such as genus/​species relations. Here he mainly speaks of “ways” (tropoi) of being causes, but
he does also say that they are “spoken of in this many ways” (195a3); at 195a29 he says that ‘aitia’ is
pollachôs legomenon, but where this is not to indicate the main fourfold distinction. Elsewhere he
uses the language of multivocity more explicitly; see, e.g., DA II 4, 415b8–​9, Met. I 3, 983a26–​27,
and Met. VIII 4, 1044a33. There is some variation in his expressions, however: we find “forms” (eidê)
of causes at Met. II 2, 994b28, and “kinds” (genê) at Met. III 2, 996a19–​20. I discuss the nature of
this pluralist commitment in Stein 2011, though I revisit the issue here from a somewhat different
 Reading (and Animating) Physics II 3 27

(1) Some entailments of drawing these distinctions among the causes

(195a3–​14);
(2) Confirmation that all the things that are “now” or “in fact” called
causes fall under these four headings (195a15–​26);
(3) A further set of distinctions among the ways in which causes and
“effects” are specified, resulting in what he claims is an array of six
main possibilities.4 Causes may be specified as follows: particular or
general; particular co-​incidental or general co-​incidental; simple or
complex. Each of these may be further specified as either actual or po-
tential (195a26–​b21);5
(4) Some further claims about how causes should be pursued or specified
(195b21–​30).

This presentation is thus in some ways very detailed and informative, but
philosophically inanimate, in that Aristotle does not argue for these claims
and distinctions, nor does he provide a sense of what motivates or justifies
them either in general or in context, so that even the seemingly informative
parts turn out to be somewhat obscure. (Why, e.g., is it important to say that
one way (tropos) of being a cause is as a co-​incident, and that Polycleites is in
this sense a co-​incidental cause of a statue?)
Indeed, Ross’s statement actually goes slightly beyond what we can say
about the four causes, given Aristotle’s presentation. Though it is common
to call it a doctrine, as Ross does, if we do not have a clear sense of why we
are meant to accept these claims or what their precise import is—​nor who
‘we’ refers to when he says that “we do not think we know each thing until we

perspective in § 4.3; the variations in expression may indicate genuine ambivalence, which would fit
with my account. See also Gourinat 2013, § 1.

4 For convenience I will use ‘effect’ to refer to what causes are causes of, regardless of category or

kind; Aristotle himself does not generally use a single word for what causes cause, but occasionally
uses an equivalent periphrasis (e.g., ‘those things of which the causes are causes,’ as at Phys. II 3,
195b7), and we do find ‘what is caused’ (to aitiaton) at APo. II 16, 98a36 and 98b3.
5 Particularity and generality are initially referred to as ‘prior’ (proterôs) and ‘posterior’ (husterôs),

but re-​glossed immediately below at 195b13–​14 as particular (kath’hekaston) and general (hôs to
genos); co-​incidental causes may also be “farther” (porrôteron) or “nearer” (egguteron), which may
be a matter of whether they are “directly” co-​incidental to the cause as such or co-​incidental to a
co-​incident (see Ross 1936, 512–​13). ‘Co-​incidental’ brings in an implicit contrast with what is in-
trinsic (kath’hauto), subsequently made explicit by referring to intrinsic causes as causes “properly
speaking” (ta oikeiôs legomena). The final distinctions are being spoken as potential (hôs dunamena)
and as active (hôs energounta). Some of these terms are ones Aristotle uses consistently for talking
about causes, but others are rare, and the language varies even within the passage. It is not entirely
clear how to take some of them individually, nor what the aim is of the set of distinctions as whole;
I discuss the latter question in § 8.2.
28 Causality and Causal Explanation in Aristotle

grasp the Why for each thing”—​then we also lack a clear sense of their overall
status. Do they constitute a metaphysical theory, a reflection of common
beliefs, a survey of ordinary Greek usage, a “model,” claims resulting from a
long examination of nature and especially natural substance, or something
else? All these have been suggested, though some are less plausible than
others.6
While the distinctions among causes are presented without argument
or justification here, they are not supposed to be merely obvious to anyone
without the right philosophical background—​they are not “self-​evident” in
the strong sense. Indeed, Aristotle is at pains to point out, especially in Met.
I and Phys. I, that they take a fair amount of thought to grasp clearly, and
the confusions he attributes to his predecessors about them are not minor.
Further, his claim that in one way all four causes have been described before,
but in another way they have not been described at all (oudamôs), makes
little sense if they are supposed to be self-​evident.
Nor do they appear to be the results of analyzing ordinary talk.7 Some of
the examples Aristotle gives are simple and ordinary enough, such as per-
haps the examples of efficient and final causation—​though the example of
the deliberating person is less obvious. However, this does not apply at all
to the example of essence (the ratio of the octave) or in general the distinc-
tion between form and matter, which Aristotle has only just worked through
in such a dialectically complicated way in Phys. I. Nor are the intellectual
defects of his predecessors described in Met. I attributed to a failure to mark
ordinary modes of speech.
Ross’s tentative suggestion, that the doctrine is the result of considered re-
flection on natural and technical processes, aided by his grasp of the work of
his predecessors, is no doubt more likely, but still vague. It may also under-
state the extent to which the context is Academic: as both he and Charlton

6 See, e.g.: Ross 1936 (reflections on nature and artistic production, with some help from his

predecessors); Graham 1987, chs. 3 and 6 (a “model”); Irwin 1988, § 51 (common beliefs); Charlton
1992, 99 (ordinary Greek usage); Hennig 2008 (metaphysical theory). Exacerbating the difficulty is
probably the fact that Phys. II 3 and its distinctions are so familiar to readers of Aristotle, and so per-
vasive, that it can be hard to avoid reading uses and problems from elsewhere in the corpus back into
the presentation here.
7 Pace Charlton 1992, who claims that “it is obvious . . . that the doctrine with which we are

presented here is the immediate result of a survey of how we ordinarily speak” (99), thus suggesting
that the ‘we’ is either “we Greeks” or “we humans who speak about such things,” and that the
distinctions are justified to the extent that they capture distinctions observed by ordinary speakers.
(He cites Wieland 1962 as well as Phys. 194b24, b34–​35, and 195a3–​4, 15; but these passages are not
compelling and only involve formulae that Aristotle uses in both technical and non-​technical ways.)
 Reading (and Animating) Physics II 3 29

point out, Aristotle’s exposition contains a rare use of ‘paradeigma’ to indi-

cate the formal cause, suggesting an audience familiar with Platonism.8
This discussion, as I have suggested, hardly answers the questions we
would like to have answered: Why should we believe that there are four
types of cause, and that they are these four? What philosophical problem or
problems do these distinctions allow us to solve, which previous thinkers
were unable to solve? What did they get wrong that Aristotle has gotten
right? What, specifically, does this discussion accomplish? Yet this chapter,
along with Phys. II 7, is clearly the canonical statement of the causes and their
role in natural science (as the reference to his discussion in the works on na-
ture at Met. I 3, 983a34 suggests).
I think that we can illuminate the status and philosophical import of these
claims by viewing the chapter in three contexts: (1) the whole of Phys. II;
(2) the criticisms, in other works, of his predecessors’ accounts of causes that
reflect his views about their more substantive mistakes; and (3) background
debates and discussions about explanation and causation in general, and
in natural science in particular. After all, despite its lack of argumentation
or philosophical stage-​setting, it is a reasonable working assumption that
Aristotle’s canonical presentation of the causes in Phys. II would show traces,
in some fashion, of the results of his reflections about causes in general and
the improvements he thinks must be made to the work of his predecessors.
By placing the chapter in these three contexts, I think it will become clear
that the common way of presenting the four-​causal schema, as primarily
grounded in considerations of the causes of things such as substances and
artifacts, though to some extent prompted by Aristotle’s own presentation,
has it conceptually backward. Applying the four-​causal schema to things and
the changes they undergo represents an extension of a conceptually more
basic theory about the ways in which, and the means by which, predicates
apply to their subjects—​namely, in virtue of the subjects themselves, or in
virtue of something else.

1.2. Physics II 3 in the Context of Book II

Placing Phys. II 3 in its local context will help illuminate some of the con-
tent of its claims as well as their status, since I think one of the sources of

8 Cf. Natali 2013, 51.

30 Causality and Causal Explanation in Aristotle

confusion is that it is often unclear whether Aristotle is making a general

claim about causation or a more restricted one that depends on further met-
aphysical commitments.
Though the transition from Phys. II 2 to II 3 is not philosophically straight-
forward, it is not completely abrupt either. As whole, book II presents a case
for a specific understanding of natural science as the study of a relatively ex-
pansive array of things with “natures,” conceived of as enmattered forms—​
a nature being an internal source of various sorts of regular change and
transformation—​and which are intrinsically rather than co-​incidentally re-
lated to the results they produce.9 A discussion of causes, then, is necessary in
order to formulate claims about the phenomena with which the natural sci-
entist is concerned, and the ways in which they are scientifically intelligible.
Book II orients its initial questions about nature with respect to the
possibilities yielded by book I, which has established that the “principles”
(archai) of natural science are form, subject, and privation, in the sense that
these are the basic concepts that make change, or changeable being, a pos-
sible object of inquiry.10 In laying down the principles of something called
phusikê, and, even more boldly, the epistêmê of nature (Phys. I 1, 184a15)
or phusikê epistêmê—​the scientific understanding of nature—​it should be
kept in mind that Aristotle is describing something whose status is either
contested or at least uncertain: Aristotle thinks that on principles like those
embraced by a monist like Parmenides or a materialist like Empedocles,
the very idea of natural science is unintelligible.11 More explicitly, Plato has

9 For recent discussions of the definition of nature in Phys. II 1, see Kelsey 2003, 2015;

Stavrianeas 2015.
10 On the possible relationships between books I and II and the rest of the Physics, see Ross 1936;

Menn 2019.
11 Aristotle’s terms ‘cause’ and ‘principle’ are sometimes extensionally equivalent, but they interact

in complex ways, especially in connection with questions like this one about what we might call the
“conceptual foundations” of natural science. As he says in Met. V 1 (1013a16–​17), all causes are prin-
ciples, since what is common to all the main uses of ‘principle’ is that they are the “first things from
which [something] is or comes to be or is known” (18–​19); however, not all principles are causes (in
that sense). ‘Principle’ is the conceptually broader term, which is why Aristotle can elucidate the
notion of nature as a cause, and indeed the role of efficient causes in general, by telling us what kind
of principle they are. He also uses the term to indicate conceptually fundamental notions or theses,
such as the ultimate starting points for demonstration (see APo. I 2, 72a7f., and I 10 for discussions
of demonstrative principles). Since Aristotle, in the Phys., is also investigating the foundations of
natural science, and he thinks that recognizing certain types of cause and certain theses about cau-
sality are essential to that project (e.g., what I refer to later as the NRC principle), causes turn out to
be principles in more than one sense. Very often, as we shall see later, he will use ‘principle’ to refer to
what, for his predecessors, plays a role equivalent to one or more of his causes, in a way which I think
evokes both the conceptual and generative senses of starting points. That phusikê is supposed to be an
Aristotelian science is clear from many references in the physical works, including the Physics itself,
right from the start. The phrase ‘phusikê epistêmê’ itself is used elsewhere, including PA. I 1, 641a36,
Met. VI 1, 1025b19, XI 4, 1061b28.
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Angola and the
River Congo, vol. 2
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.

Title: Angola and the River Congo, vol. 2

Author: Joachim John Monteiro

Illustrator: Edward Fielding

Rose Monteiro

Release date: May 26, 2022 [eBook #68176]

Language: English

Original publication: United Kingdom: Macmillan and Co, 1875

Credits: Peter Becker and the Online Distributed Proofreading Team

at https://www.pgdp.net (This file was produced from
images generously made available by The Internet Archive)

*** START OF THE PROJECT GUTENBERG EBOOK ANGOLA

AND THE RIVER CONGO, VOL. 2 ***
 ANGOLA
AND
THE RIVER CONGO.
BY
JOACHIM JOHN MONTEIRO,
ASSOCIATE OF THE ROYAL SCHOOL OF MINES, AND CORRESPONDING
MEMBER OF THE ZOOLOGICAL SOCIETY.

IN TWO VOLUMES.
Vol. II.

WITH MAP AND ILLUSTRATIONS.

London:
MACMILLAN AND CO.
1875.
All Rights Reserved.
 LONDON:
PRINTED BY WILLIAM CLOWES AND SONS,
STAMFORD STREET AND CHARING CROSS.
 CONTENTS.

PAGE
CHAPTER I.
Country From Ambriz To Loanda—Mossulo—
Libongo—Bitumen—River Dande—River Bengo 1
—Quifandongo
CHAPTER II.
City of Loanda—Natives—Slavery—Convicts—
20
Theatre and Morals
CHAPTER III.
Division of Angola—Wretched Pay of Officials
—Abuses by Authorities—Evils of High Import
Duties—Silver Mines of Cambambe—Journey 50
to Cambambe—Exploration—Volcanic Rocks—
Hornbill—The Plantain-eater—Hyenas
CHAPTER IV.
Province of Cazengo—Golungo Alto—Gold—
Wild Coffee—Iron Smelting—Former
84
Missionaries—Customs—Natives—
Productions
CHAPTER V.
River Quanza—Calumbo—Bruto—Muxima—
Massangano—Dondo—Falls of Cambambe—
112
Dances—Musical Instruments—Quissama—
Libollo
CHAPTER VI.
Country South of the River Quanza— 151
Cassanza—Novo Redondo—Celis—Cannibals
—Lions—Hot Springs—Bees—Egito—
Scorpions—River Anha—Catumbella
CHAPTER VII.
Town of Benguella—Slave-trade—Mundombes
—Customs—Copper—Hyenas—Monkeys—
180
Copper Deposit—Gypsum—Hornbills—Birds—
Fish—Lions
CHAPTER VIII.
Country between Benguella and Mossamedes
—Mossamedes—Curious Deposits of Water— 212
Hyenas—Welwitschia mirabilis—Mirage
CHAPTER IX.
Climate—Cookery—Drunkenness—Fever—
Native Treatment—Ulcers—Smoking Wild- 233
hemp—Native Remedies
CHAPTER X.
Customs—Burial—White Ant—Wasps—Fruits
—Scents—Spitting-snake—Scarabæus— 268
Lemur
CHAPTER XI.
Conclusion 307
Appendix 315
 LIST OF ILLUSTRATIONS.

To face
View of the City of St. Paul de Loanda
page 20
Bellows—Marimba—Native smiths—Rat-trap ” 93
Maxilla and Barber’s shop—Carrying corpse for
burial—Quissama Women, and manner of pounding ” 147
and sifting meal in Angola
Mundombes and Huts ” 185
Native-smelted Copper—Powder-flask—Mundombe
Axe—Manner of securing Fish for drying—Hunters’
fetish (Benguella)—Manner of carrying in the hand ” 190
(native jug)—Gourd-pipe for smoking Diamba—
Wooden dish—Double-handled hoe
Welwitschias growing in a plain near Mossamedes ” 229
Pelopœus spirifex and nest—Devil of the Road—
Dasylus sp.—Caterpillars’ nests—Mantis and Nest ” 277
—Manis multiscutatum and Ants’ nests
ANGOLA AND THE RIVER
CONGO.
 CHAPTER I.
COUNTRY FROM AMBRIZ TO LOANDA—
MOSSULO—LIBONGO—BITUMEN—RIVER
DANDE—RIVER BENGO—QUIFANDONGO.

The distance from Ambriz to Loanda is about sixty miles, and the
greater part of the country is called Mossulo, from being inhabited by
a tribe of that name. These natives have not yet been reduced to
obedience by the Portuguese, not from any warlike or valorous
opposition on their part, but entirely from the miserable want of
energy of the latter in not taking the few wretched towns on the road.
Incredible as it may appear, it is nevertheless a fact that to the
present day the Mossulos will not allow a white man to pass
overland from Libongo (about half-way from Loanda) to Ambriz,
although this last place was occupied in 1855, and several
expeditions have since been sent to and from Ambriz to Bembe and
San Salvador. Nothing could have been easier than for one of these
to have passed through the Mossulo country and to have occupied it,
at once doing away with the reproach of allowing a mean tribe to bar
a few miles of road almost at the gates of Loanda, the capital of
Angola.
One of these expeditions, on its return from chastising the natives
of a town on the road to Bembe for robbery, was actually sent to
Loanda by road. The Governor-General (Amaral) was then at
Ambriz, and being unacquainted with the negro character, and
having mistaken humanitarian ideas, gave strict orders that the
natives of Mossulo, who had committed several acts of violence,
should not be punished, but that speeches should be made to them
warning them of future retribution if they continued to misconduct
themselves. Their towns and property were not touched, nor were
hostages or other security exacted for their future good conduct.
 The natural consequence was that this clemency was ascribed by
the natives to weakness, and that the Portuguese were afraid of their
power, as not a hut had been burnt, a root touched, or a fowl killed,
and they consequently, in order to give the white men an idea of their
power and invincibility, attacked some American and English
factories at Mossulo Bay, the white men there having the greatest
difficulty to save their lives and property; a Portuguese man-of-war
landed some men, and so enabled the traders to get their goods
shipped, but the factories were burnt to the ground. This was in
September 1859.
I was at Ambriz when the expedition started, so I determined to
join it, and examine the country to Loanda.
The expedition consisted of 150 Portuguese and black soldiers,
and as many armed “Libertos,” or slaves, who are freemen after
having served the Government for seven years; these “Libertos”
dragged a light six-pounder gun. The commander was Major (now
General) Gamboa, an officer who had seen upwards of twenty years’
service in Moçambique and Angola, and to whom I was indebted for
great friendship during the whole time I was in the country. The
major and two officers rode horses; two others and myself were
carried in hammocks. We started one afternoon and halted at a
small village consisting of only a few huts, at about six miles south of
Ambriz. There we supped and slept, and started next morning at
daybreak. The start did not occupy much time, as the Portuguese
troops and officers in Angola do not make use of tents when on the
march, and their not doing so is undoubtedly the cause of a good
deal of the sickness and discomfort they suffer. In the evening we
arrived at the Bay of Mossulo, where we were hospitably entertained
by the English and American traders there established.
The country we passed through on our march was of that strange
character that I have described as occurring in the littoral region of
Ambriz. In the thickets dotted over the country a jasmine (Corissa
sp.) is a principal plant. It grows as a large bush covered with long
rigid spines, and bears bunches of rather small white flowers having
the scent of the usual jasmine. Also growing in these thickets, and
very often over this species, are two creeping jasmines—the
“Jasminum auriculatum” (J. tettensis? Kl.) and “Jasminum
multipartitum?”
Various kinds of birds abounded, principally doves and the
beautiful purple starlings, and on the ground small flocks (from two to
four or five) of the bustards Otis ruficrista and Otis picturata were not
uncommon, appearing in the distance like snakes, their heads alone
being visible over the tops of the short rough grass as they ran
along. A small hare is found in abundance, and also several species
of ducks in some small marshes near Great Mossulo. Of larger game
only some small kinds of antelope are found.
I had gone on some distance ahead of the troops, and on
approaching one large town, about a dozen natives armed with
muskets stopped my hammock, and told me I must return to Ambriz,
as no white man could be allowed to pass. I told them that the
soldiers were close behind, and that resistance would be useless, as
their town would be taken and burnt if they attempted any; they,
however, still persisted in not letting me go forward, so I had to wait
for a few minutes till they saw Major Gamboa and the two officers
approaching on horseback, when they scampered off into the bush
without even saying good-bye, and on our entering the town we
found it deserted save by the king and a few other old men, who
were all humility, and protested that they would never more insult or
ill treat white men.
Major Gamboa was perfectly convinced of the uselessness of only
talking to blacks, his intimate knowledge of them telling him that the
only safe plan would have been to have burnt the towns on the road
and taken the king and old men to Loanda as hostages, but he had
to obey his instructions, and the result was that they attacked the
factories and killed a number of natives. The Portuguese, however,
instead of punishing this outrage, tamely pocketed the affront, and
left the Mossulos in undisputed possession of the road.
In these towns were the largest “fetish” houses I have seen in
Angola. One was a large hut built of mud, the walls plastered with
white, and painted all over inside and out with grotesque drawings, in
black and red, of men and animals. Inside were three life-size figures
very roughly modelled in clay, and of the most indecent description.
Behind this hut was a long court the width of the length of the hut,
enclosed with walls about six feet high. A number of figures similar in
character to those in the hut were standing in this court, which was
kept quite clean and bare of grass. What, if any, were the uses to
which these “fetish” houses were applied I could not exactly
ascertain. I do not believe that they are used for any ceremonials,
but that the “fetishes” or spirits are supposed to live in them in the
same manner as in the “fetish” houses in the towns in the Ambriz
and Bembe country. At one of the towns we saw a number of the
natives running away into the bush in the distance, carrying on their
backs several of the dead dry bodies of their relatives. I hunted in all
the huts to find a dry corpse to take away as a specimen, but without
success; they had all been removed.
Next day we continued our journey, and bivouacked on the sea-
shore, not very far from Libongo, and near the large town of
Quiembe.
On the beach we found the dead trunk of a large tree that had
evidently been cast ashore by the waves, and had been considered
a “fetish;” for what reason, in this case, I know not, as trees stranded
in this way are common. It was hung all over with strips of cloth and
rags of all kinds, shells, &c. As it was dry, it was quickly chopped up
for firewood by the soldiers and blacks.
The following morning our road lay along the beach till we reached
the dry mouth of the River Lifune, a small stream that only runs
during the rainy season. We then struck due inland for about three
miles to reach the Portuguese post of Libongo, consisting of a small
force commanded by a lieutenant. This officer (Loforte) I had known
at Bembe, and he gave us a cordial welcome.
The “Residencia,” or residence of the “Chefe,” as the
commandants are called, was a large, rambling old house of only
one floor, and it contained the greatest number of rats that I have
ever seen in any one place.
One large room was assigned to the use of Major Gamboa, two
officers, and myself, a bed being made in each corner of the room.
We had taken the precaution of leaving the candle burning on the
floor in the middle of the room, but we had scarcely lain down when
we began to hear lively squeaks and rustlings that seemed to come
from walls, roof, and floor. In a few minutes the rats issued boldly
from all parts, running down the walls and dropping in numbers from
the roof on to the beds, and attacking the candle. We shouted, and
threw our boots, sticks, and everything else that was available at
them, but it was of no use, and we could hardly save the candle. It
was useless to think of sleep under these circumstances, for we
considered that if the rats were so bold with a light in the room, they
would no doubt eat us up alive in the dark, so we dressed ourselves,
and pitched our hammocks in the open air, under some magnificent
tamarind-trees, and there slept in comfort.
Libongo is celebrated for its mineral pitch, which was formerly
much used at Loanda for tarring ships and boats. The inhabitants of
the district used to pay their dues or taxes to government in this
pitch. It is not collected at the present time, but I do not know the
reason why.
I was curious to see the locality in which it was found, as it had not
been visited before by a white man, so Lieutenant Loforte supplied
me with an old man as guide, and Major Gamboa and myself started
one morning at daybreak.
We had been told that we might reach the place and return in
good time for dinner in the evening, and consequently only provided
a small basket of provisions for breakfast and lunch; we travelled
about six miles, and reached a place where we found half-a-dozen
huts of blacks belonging to Libongo, engaged in their mandioca
plantations. These tried hard to dissuade us from proceeding farther,
saying that we should only reach the pitch springs next morning. I, of
course, decided to proceed, but Major Gamboa, who did not take the
interest in the exploration that I did, determined to return to breakfast
at Libongo at once, leaving me the provisions for my supposed two
days’ journey.
After a short rest I started off again, and about mid-day arrived at
the place I was in search of. It was the head of a small valley or
gully, worn by the waters from the plain on their way to the sea,
which was not far off, as although it could not be seen from where I
stood, the roll of the surf on the beach could just be heard. It must
have been close inland to the place where we had bivouacked a few
nights before, and had burnt the “fetish” tree for firewood.
The rock was a friable fine sandstone, so impregnated with the
bitumen or pitch, that it oozed out from the sides of the horizontal
beds and formed little cakes on the steps or ledges, from an ounce
or two in weight to masses of a couple of pounds or more.
Although it was very interesting to see a rock so impregnated with
pitch as to melt out with the heat of the sun, I was disappointed, as
from the reports of the natives I had been led to believe that it was a
regular spring or lake. My guide was most anxious that I should
return, and as I was preparing to shoot a bird, begged me not to fire
my gun and attract the attention of the natives of the town of
Quiengue, close by, whom we could hear beating drums and firing
off muskets. Next day we knew at Libongo that these demonstrations
had been for the purpose of calling together the natives, to attack the
factories at Mossulo Bay.
There was great talk at Loanda about sending an expedition to
punish these natives, but, as usual, it ended in smoke, and no white
man has since been allowed to pass through the Mossulo country.
Several years after, the King of Mossulo sent an embassy to me at
Ambriz, begging me to open a factory at Mossulo. On condition that
I, or any white man in my employ, should be free to pass backwards
and forwards from Loanda to Ambriz, I promised to do so, and was
taken to the king’s town at Mossulo, where it was all arranged. I did
not believe them, of course, but I gave a few fathoms of cloth and
other goods that they might build me a hut on the cliff at Mossulo
Bay, which they did, and I then declared myself ready to send a clerk
with goods to commence trading, as soon as they should send me
hammock-boys to carry me to Loanda. As I expected, they never
sent them, and for several years, whilst the hut on the cliff lasted, it
served as a capital landmark to the steamers and ships on the coast.
The Governor-General at Loanda, to prevent traders from
establishing factories at Mossulo Grande, warned us at Ambriz that if
we did so we must take all risks, that he would not only not protect
us, but that all goods for trading at Mossulo would have to be
entered and cleared at the Loanda custom-house. Far from such
disgraceful pusillanimity being censured at Loanda, it was, with few
exceptions, considered by the Portuguese there as a very
praiseworthy measure.
The rock of the country at Libongo is a black shale; also strongly
impregnated with bitumen. A Portuguese at Loanda, believing that
this circumstance indicated coal in depth, sunk a shaft some few
fathoms in this shale, and I visited the spot to see if any organic
remains were to be found in the rock extracted, but could not
discover any. About half way from Libongo to the place where I saw
the bituminous sandstone formation, I observed a well-defined rocky
ridge of quartz running about east and west, which appeared to have
been irrupted through the shale.
The ground about Libongo is evidently very fertile, the mandioca
and other plantations being most luxuriant, and I particularly noticed
some very fine sugar-cane. Some of the tamarind-trees were
extremely fine, and on the stem of a very large one a couple of the
“engonguis,” or double bells, were nailed, which had belonged to the
former native town there, and as they are considered “fetish,” no
black would steal or touch them.
A few hours’ journey (or about fifteen miles) to the south of
Libongo is the River Dande, navigable only by large barges, and
draining a fertile country.
It is only within the last two years that the value of this river, for
trading or produce, has attracted attention at Loanda, and I am glad
to say that it was owing to two foreign houses that trading was
commenced there on anything like a respectable scale. The interior
is rich in coffee, gum-copal, ground-nuts, and india-rubber, and this
country promises an important future; cattle thrive here, and Loanda
is now supplied with a small quantity of excellent butter and cream
cheese from some herds in the vicinity of this river near the bar.
 Limestone is also burnt into lime, which is sold at a good price at
Loanda; and were the Portuguese and natives more enterprising and
industrious, the banks of the river would be covered with valuable
gardens and plantations; but apathy reigns supreme, and the
authorities at Loanda prevent any attempt to get out of this state by
the obstructions of all kinds of petty and harassing imposts, rules,
and regulations, having no possible aim but the collection of a
despicable amount of fees to keep alive and in idleness a few
miserable officials.
The country is comparatively level, and calls for no particular
description, till about eighteen miles southward the high and bold cliff
of Point Lagostas (Point Lobsters) marks the bay into which runs the
beautiful little river Bengo, or Zenza, as it is called farther inland.
This is even a smaller river than the Dande, though more
important from its near proximity to Loanda, and the remarks as to
the wonderful indifference and hindrance to the development of the
River Dande, apply with still greater force to the Bengo, a very mine
of wealth at the doors of Loanda! It is hardly possible to restrain
within reasonable limits the expression of surprise at the fact that
Loanda, with its thousands of inhabitants, should be still destitute of
a good supply of drinking water, when there is a river of splendid
water only nine miles off, whence it receives an insignificant and
totally inadequate supply brought in casks only, carried by a few
rotten barges and canoes that are often prevented from leaving or
entering the river for days together, on account of the surf at the bar.
A small cask of Bengo water, holding about six gallons, costs from
twopence to fourpence! All kinds of fruit and vegetables grow
luxuriantly on the banks of the Bengo, and yet Loanda, where
nothing can grow from its sandy and arid soil, is almost unprovided
with either—a few heads of salad or cabbage, or a few turnips and
carrots being there considered a fine present.
At Point Lagostas a good deal of gypsum is found, and also
specimens of native sulphur.
Both the River Bengo and the River Dande are greatly infested by
alligators, and a curious idea prevails amongst all the natives of
Angola, that the liver of the alligator is a deadly poison, and that it is
employed as such by the “feiticeiros” or “fetish”-men.
The Manatee is also not uncommon in these rivers;—this curious
mammal is called by the Portuguese “Peixe mulher” or woman fish,
from its breasts being said to resemble those of a woman. Near the
mouth of the Dande this animal is sometimes captured by enclosing
a space, during the high tides, with a strong rope-net made of
baobab fibre, so that when the tide falls it is stranded and easily
killed. I was never so fortunate as to see one of these animals, and
am therefore unable to describe it from personal observation, but it is
said to be most like a gigantic seal. I once saw a quantity of the flesh
in a canoe on the River Quanza, and was told that the greater part
had been already sold, and I had given me a couple of strips of the
hide of one that had been shot in the River Loge at Ambriz. These
strips are about seven feet long and half an inch thick, of a yellowish
colour, and semi-transparent. They are used as whips, being smooth
and exceedingly tough. The flesh is good eating, though of no
particular flavour, and is greatly liked by the natives. The marshes
and lagoons about the River Bengo are full of wild duck and other
water-fowl, and are favourite sporting places of the officers of the
English men-of-war when at Loanda. The Portuguese, not having the
love of sport greatly developed, seldom make excursions to them.
The country from the Bengo to Loanda rises suddenly, and the
coast line is high and bold, but the soil is very arid and sandy, the
rocks being arenaceous, evidently of recent formation, and full of
casts of shells.
There is much admixture of oxide of iron, and some of the sandy
cliffs and dunes close to Loanda are of a beautiful red from it. The
vegetation is, as might be expected, of a sterile character, being
principally coarse grass, the Sanseviera Angolensis, a few shrubs,
euphorbias, and a great number of giant baobabs. Though the
vegetation is comparatively scarce, birds of several species are
common; different kinds of doves are especially abundant, as are
several of the splendidly coloured starlings; kingfishers are very
common, and remarkable for their habit of choosing a high and bare
branch of a tree to settle on, from whence, in the hottest part of the
day, they incessantly utter their loud and plaintive whistle, and, after
darting down on the grasshoppers and other insect prey, return
again to the same branch.
The exquisitely coloured roller (Coracias caudata) is also very
common in the arid country surrounding Loanda.
The pretty runners (Cursorius Senegalensis, and C. bisignatus, n.
sp.) are also seen in little flocks on the sandy plains, and are most
elegant in their carriage as they swiftly run along the ground. Two or
three species of bustards are also common.
The great road from the interior skirts the River Bengo for some
miles to the bar, where it turns south to Loanda; and the last resting
or sleeping place for the natives carrying produce is at a place called
Quifandongo, consisting of a row of grog-shops and huts on either
side of the road.
It is a curious sight to see hundreds of carriers from the interior
lying down on the ground in the open air, each asleep with his load
by his side. A march of two hours brings them to a slope leading
down to the bay, at the end of which Loanda is built.