Sunteți pe pagina 1din 257

ALEXANDER OF APHRODISIAS

On the Soul Part 1


This page intentionally left blank
ALEXANDER OF
APHRODISIAS
On the Soul
Part 1
Soul as Form of the Body,
Parts of the Soul,
Nourishment, and Perception

Translated with an Introduction


and Commentary by Victor Caston

LON DON   0 ' 9  &' . * +   0 ' 9  YO4 -   SY DN ';


Bloomsbury Academic
An imprint of Bloomsbury Publishing Plc

50 Bedford Square 1385 Broadway


London New York
WC1B 3DP NY 10018
UK USA

www.bloomsbury.com

Bloomsbury is a registered trade mark of Bloomsbury Publishing Plc

First published in 2012


Paperback edition rst published 2014

2012 by Victor Caston

Victor Caston has asserted his right under the Copyright, Designs and Patents Act, 1988,
to be identied as Author of this work.

All rights reserved. No part of this publication may be reproduced or transmitted


in any form or by any means, electronic or mechanical, including photocopying,
recording, or any information storage or retrieval system, without prior permission
in writing from the publishers.

No responsibility for loss caused to any individual or organization acting on


or refraining from action as a result of the material in this publication can
be accepted by Bloomsbury Academic or the author.

British Library Cataloguing-in-Publication Data


A catalogue record for this book is available from the British Library.

ISBN HB: 978-1-7809-3024-4


PB: 978-1-4725-5798-8
ePDF: 978-1-4725-0172-1

Library of Congress Cataloging-in-Publication Data


A catalog record for this book is available from the Library of Congress.

Acknowledgements

The present translations have been made possible by generous and imaginative
funding from the following sources: the National Endowment for the
Humanities, Division of Research Programs, an independent federal agency
of the USA; the Leverhulme Trust; the British Academy; the Jowett Copyright
Trustees; the Royal Society (UK); Centro Internazionale A. Beltrame
di Storia dello Spazio e del Tempo (Padua); Mario Mignucci; Liverpool
University; the Leventis Foundation; the Arts and Humanities Research
Council; Gresham College; the Esme Fairbairn Charitable Trust; the Henry
Brown Trust; Mr and Mrs N. Egon; the Netherlands Organisation for Scientic
Research (NWO/GW); the Ashdown Trust; Dr Victoria Solomonides, the
Cultural Attach of the Greek Embassy in London. The editor wishes to
thank John Thorp, James Wilberding, Michael Grifn, Donald Russell and
William Charlton for their comments, Sebastian Gertz for preparing the
volume for press, and Deborah Blake at Bristol Classical Press, who has been
the publisher responsible for every volume since the rst.

Typeset by Ray Davies


Printed and bound in Great Britain
Contents

Acknowledgements vii
Introduction 1
Note 23
Abbreviations 25
Textual Emendations 27

Translation 31
What is the souls essence? 31
Bodies act and are modified in virtue of non-bodily items 36
The demonstration that the soul is a form 39
In something [is said] in how many ways? 40
The soul is inseparable from the body whose soul it is 44
The soul is not one of the simple bodies 46
The soul itself just on its own is unchangeable 48
The soul is not a harmony 51
The parts of the soul: how many and which ones? 53
That there are not infinitely many powers of the soul 53
There are two kinds of nourishment 59
On the activity of being nourished 61
Heat is the souls tool for nourishing [oneself] 61
Why parts detached from plants live and develop, while those from 63
animals no longer do
On the soul for perceiving 64
On what is intrinsically and extrinsically perceptible 66
How there comes to be light in the transparent 68

Commentary 71
Bibliography 169
English-Greek Glossary 177
Greek-English Index 189
Index of Passages Cited 215
Subject Index 237
In memoriam
Robert W. Sharples
1949-2010
Acknowledgements

As a graduate student in London in 1985-7, I had the good fortune to study


Alexander of Aphrodisias with Bob Sharples. I met with him weekly to
read Alexanders On the Soul and discuss the texts finer points, while
attending the university lectures he offered on Post-Aristotelian philoso-
phy with my supervisor Richard Sorabji and the many seminars they
organised at the old Institute for Classical Studies on Gordon Square. It
was an exciting time to be in the UK working on ancient philosophy. The
turn to Hellenistic philosophy was still relatively new, Long and Sedleys
collection had just been published, and the project to translate all the
commentators on Aristotle was just getting under way. They were wonder-
ful years.
In retrospect, I had no idea just how fortunate I was. No one in the
twentieth century did more to advance the study of the Peripatetics after
Aristotle, from Theophrastus to Alexander. His scholarship was exem-
plary, not merely for its erudition, which was remarkable, or for its range
and sheer quantity, which was impressive, but above all for his critical
judgement. He was unusually alive to the openness of texts to multiple
interpretations and the recalcitrance of counterevidence: short of decisive
refutation, which he took to be rare, one always had to acknowledge the
possibility of alternative readings, and he was ingenious at devising ways
in which they might be defended against even his own favoured view. In
others, these might merely have been methodological scruples. But in Bob
they were also a natural expression of his own characteristic modesty and
his charity towards others. As a teacher, he was unfailingly generous and
supportive. Bob immediately read any draft you gave him and returned it
with copious, thoughtful comments and acute criticisms, often within just
a day or two. I could not have studied Alexander with anyone better.
Like many others, I was greatly saddened to hear of Bobs illness and
untimely demise, and he passed away while this volume was still in
preparation. Both it and the second volume are dedicated to his memory.
I would also like to thank several other people who were crucial in
bringing this volume to term, above all the editor of the series, Richard
Sorabji, who has my deepest gratitude for his detailed and invaluable
comments, his unwavering encouragement, and (not least) his patience
and trust. Thanks are also due to Ian McCready-Flora, who served as my
research assistant in the summer of 2009, and to Michael Griffin, the
viii Acknowledgements
editorial assistant through most of the project. Both made excellent con-
tributions, which I have tried to acknowledge in the notes. Sebastian
Gertz, the editorial assistant who saw the project through to its final
stages, also has my thanks for his yeomans service with all the corrections
and changes. I am grateful as well to the students in my graduate seminar
at the University of Michigan in the winter term of 2009, who worked
through an earlier draft of the entire translation, and to the participants
in a mini-conference on the present volume at the Humboldt-Universitt
in Berlin in March 2011, especially my commentators Jakub Krajczynski,
George Karamanolis, and Miira Tuominen. I learned a great deal from
both groups, who helped save me from quite a few errors. In addition, I
owe a great debt to Ian Crystal for his Herculean efforts to assemble the
index locorum. Such tools are invaluable for making a scholarly work
useful, and they rarely get the thanks they deserve.
Finally, I would like to record my greatest debt of all, to my wife, Ruth,
for her love and support, and for all the joy that she and my two daughters,
Eva and Sarah, have brought me. Without them, I could not have accom-
plished any of the things I have.
Introduction
Alexander of Aphrodisias, easily one of the greatest commentators on
Aristotle later Greek commentators sometimes refer to him as the
Commentator (ho exgts) on Aristotle1 or even as a new or second
Aristotle2 is also one of the foremost Aristotelians in the tradition,
alongside Thomas Aquinas and Ibn Rushd, a philosopher in his own right,
who developed a profound and systematic synthesis of Aristotles views
that was to have lasting influence.

Life and works


We know little of Alexanders life, apart from the fact that he came from
Aphrodisias, a city in Caria (a region in the southwest of modern day
Turkey), and that sometime between 198 and 209 CE he was appointed by
the emperors Septimus Severus and Antoninus Caracalla to a chair in
Aristotelian philosophy in Athens, one of four chairs established for each
of the major schools of philosophy by Marcus Aurelius in 176.3 It has often
been assumed that Alexander is referring to this appointment in the
dedication of his treatise On Fate.4 But recently this has been confirmed
by an inscription from Aphrodisias, which Alexander had dedicated to his
father, who was also named Alexander and was a philosopher as well
(perhaps even the author of On Fevers and the collections of medical
problems attributed to an Alexander of Aphrodisias). The inscription
gives the full name of the younger Alexander (that is, the commentator) as
Titus Aurelius Alexandros.5 We also know that he studied with Sosigenes
and Herminus (who had been one of Galens teachers as well).6 Based on
these facts, we can infer that Alexander was born sometime between 140
and 165.7
Apart from commentaries on Aristotles works, Alexander also wrote
philosophical treatises, articulating and defending in his own voice an
Aristotelian point of view.8 The present work, On the Soul, is one of the few
surviving treatises, along with one on determinism and freedom of choice
(On Fate) and another on material composition (On Mixture). Alexander
also wrote a commentary on Aristotles On the Soul. But the commentary
is now lost and survives only in isolated quotations.9 The present treatise
is sometimes mistakenly referred to as a commentary, because of its
faithfulness to both the sequence of topics and the doctrines put forward
in Aristotles On the Soul. But the purpose of the two types of work is
2 Introduction
nonetheless different: a commentary seeks to elucidate a text; a treatise
seeks to defend the truth. Unlike his commentaries, Alexanders treatise
On the Soul does not focus on Aristotles statements line-by-line or puzzle
about manuscript readings and alternative interpretations. Instead, he
simply presents the view itself, elaborated and argued for in the way he
sees best, referring to Aristotles text only at intervals for added emphasis
and authority.10 That Alexanders treatise agrees so extensively with
Aristotles On the Soul, in both its organisation as well as in doctrine, is
just a sign of how much he thinks Aristotle got right. As Alexander
explains in the opening pages of his On the Soul, he reveres Aristotle
because there is more truth (althesteras) in his views than in others, and
so will have achieved his aim if he elaborates and defends them well
(2,4-9).11 But that obviously permits him divergences, modifications, and
revisions, as well as elaborations and further developments.

The organisation and aims of the treatise


The most obvious similarity between Aristotles On the Soul and Alexan-
ders is the sequence of topics, as just mentioned. But in fact the overlap
is only partial. Alexander only follows the same order as Aristotle from
27,1, where Alexander picks up Aristotles ascent through the powers of
living things, starting with his general remarks about these powers in On
the Soul 2.3; and their discussions remain more or less parallel until the
end of Alexanders treatise, where he adds an appendix on the
hgemonikon or governing part of the soul and its location in the body
(94,7-100,17). Because of this extensive correspondence, much of Alexan-
ders treatise can be consulted as if it were a continuous commentary on
Aristotle On the Soul 2.3-3.13, even though that is not its form or express
purpose.
The first twenty-six pages of Alexanders treatise, on the other hand
a full quarter of the work as a whole do not correspond directly to any
major section of Aristotles On the Soul. In these pages Alexander explores
the Aristotelian notion of form in order to ground and explain Aristotles
central claim that the soul is the form of the body. This section naturally
has many points of contact with the earlier parts of Aristotles treatise. But
they are limited to a few key passages or doctrines. Otherwise, the course
of discussion is very much Alexanders own.
This superficial difference in organisation points to more significant
underlying differences in both method and objectives. Aristotle ap-
proaches the subject dialectically, in his characteristic fashion. He begins
by surveying the reputable views (endoxa) that have been held on the
topic, to see where there are disagreements as well as consensus, and then
teases out various puzzles (aporiai), which set the agenda for his sub-
sequent discussion. His aim is to recover what is right in each position in
a way that also shows how to resolve the difficulties they face, and he later
refers back to these disagreements at various points in the treatise to show
Introduction 3
how his own theory provides a more adequate solution. Aristotle thus actively
engages the earlier history of the subject for the entire first book of On the
Soul. He begins his own positive exposition in earnest only in the second book,
which opens with his definition of the soul as the form of the body.
In contrast, Alexanders approach is not dialectical, but systematic. His
aim is to elaborate the conceptual framework necessary for understanding
Aristotles theory, above all the metaphysical foundations on which the
whole theory rests something which Aristotle does only in the most
cursory way in his On the Soul. If the soul is best understood as the form
of the living body, then we first need to grasp the distinction between
matter and form, and reflect on its implications, before we can appreciate
how it can account for the psychological functions investigated in the
remainder of the treatise. What Alexander offers is a masterful synthesis
of Aristotles views, drawn from across the corpus, to which he adds
extensive supporting arguments and distinctions. It is both insightful and
illuminating. It is also controversial. Although many contemporary philoso-
phers will find it extremely attractive, someone like Paul Moraux (1942),
working in the midst of the neo-Thomist revival, clearly felt it was threaten-
ing and had to be answered in the most forceful terms.12 But there can be no
doubt that it is an Aristotelian vision of real power and profundity.
There is a second respect in which Alexanders approach differs. Under-
standably, he is not concerned so much with Aristotles predecessors as
with his own contemporaries and rivals. For Alexander, unlike Aristotle,
Aristotelianism is already an established position, and Alexanders aim is
to show that it is still very much a live option, indeed the best. He
consciously positions himself midway between the dualism of Middle
Platonists like Atticus (late second century CE) and the materialism of the
Stoics, undoubtedly still the most dominant philosophical force of his day.
It is hard for us not to focus on the sharpened polemical character of
Alexanders arguments and the ways in which he is or is not fair. But we
should not overlook the more important underlying fact that for Alexander
exegesis is not an antiquarian or historical exercise, but a philosophical one.

Naturalism
Alexander makes his opposition to Platonism clear in the opening pages of
the treatise, even though he does not refer to it by name. At 2,17 he uses
a phrase from Platos Phaedo to mark the fundamental difference in
orientation. In the original passage, Socrates diagnoses Simmias doubts
and fears about the fate of the soul: Simmias is at a loss because he cannot
reconcile his belief that (i) the soul is a balance or harmony of bodily
powers, and so something that will dissipate at death, with his conviction
that (ii) the soul is something more divine and more noble than the body
(kai theioteron kai kallion on tou smatos, 91C-D; cf. 94E). Between these
two options, Socrates allegiance to the second is never in doubt, and he
goes on to argue at length against the first view that the soul is in any way
4 Introduction
a function of the body. In contrast, Alexander challenges the underlying
dilemma. On his view, the soul is both something worthy of our greatest
admiration and something that is entirely natural and indeed mortal. We
will cease to find this puzzling, he says, once we appreciate how wondrous
the workings of nature are. By grasping how nature works in general and
how the soul is a part of nature, we will come to find both halves of this
conjunction completely convincing (2,10-25).
Alexander gestures briefly in this passage at biological marvels: he
mentions the structure of our bodies, the regulation of our inner organs,
and the ways in which they are adapted to our external limbs and their
functions. But this is not something he examines in any detail in this
treatise or, for that matter, in any of the extant works. What he focuses on
instead is more basic and metaphysical. On the Aristotelian view, the soul
is the form of the body and so, Alexander argues, the relation between the
two will be essentially the same as in other natural objects: in this respect,
the same basic principles apply to both living and non-living things. Such
things obviously differ from each other in their specific characteristics,
powers, and activities. But this does not mark a fundamental metaphysi-
cal difference. It is due rather to physical differences in their material
composition and organisation. Greater complexity in the powers of things
is to be explained by a corresponding complexity in their underlying
matter (8,517).
For Alexander, then, the distinction between matter and form is only
the beginning. We also need to consider the specific differences between
various forms and the corresponding differences in their underlying mat-
ter. Alexander proceeds to survey the whole of nature, starting with the
four elements and rising through inanimate substances, plants, animals,
and finally humans (8,17-11,13).13 At each stage of this scala naturae,
there is a dramatic increase in the variety of species of natural substances,
their distinctive abilities, and their behaviour. In fact, he argues, the
increase in complexity from one broad echelon to the next is strictly
proportional: the extent to which plants surpass inanimate substances in
their variety and powers is as great as the extent to which animals surpass
plants (10,12-17). Once we have considered the whole of nature in this way
and seen how the powers of natural substances are a function of their
underlying material constitution, he urges, we will no longer be astonished
at the powers of animals and humans: any natural piety they evoke in us
will be the same as what we feel before the equally amazing, though
simpler, powers of natural substances at lower levels, right down to the
elements themselves (11,5-13). There is no fundamental discontinuity
here that would require a different kind of explanation than we use for
other natural substances. What we marvel at here pervades the whole of
nature.
To emphasise the general applicability of these underlying principles,
Alexander repeatedly explains characteristics of the soul by reference to
what holds for forms even in the simplest case, of the elements, such as
Introduction 5
the lightness of fire and the heaviness of earth.14 His point is not that the
elements have a soul or that the soul of living things can be reduced to
combinations of the elements or their features, but rather that many of the
conclusions he draws about the soul follow simply from its being a form,
and so are strictly analogous to characteristics that belong to the forms of
other things, including those which are inanimate or even elemental. The
features that distinguish the animate from the inanimate are, in contrast,
the subject of the more detailed discussions that constitute most of the
subsequent treatise.

Form and matter


The basic framework for Alexanders discussion rests on his under-
standing of form and matter. He introduces the distinction by considering
artefacts like statues, where the components are more evident to us
(2,25-3,13), before proceeding to the more fundamental case of natural
objects (3,13-21). We can draw a further division that will prove pivotal,
though, within the class of natural objects, between simple and compound
bodies (3,22-8). All natural objects are bodies, and all bodies are composed
of matter and form. But some natural bodies namely, compound bodies
are composed from other bodies, whereas simple bodies are not. Evi-
dently the underlying matter in these two cases is quite different. Since
the matter of compound bodies consists of other bodies, they will have a
matter and form of their own; and if they are also compound bodies, then
their matter in turn will be bodies and so composed of matter and form
again; and this will continue so long as the underlying matter consists of
compound bodies. There is obviously potential here for great complexity in
the constitution of bodies. Understanding how the properties of the under-
lying bodies contribute to the properties of the compound will be crucial
for the material account of any natural substance.
Simple bodies like the four elements, in contrast, lack this complexity.
They provide the base case: though themselves composed of matter and
form, their matter is not a body and so not composed of anything further.
Alexander here develops the classic conception of prime matter (3,28-4,20),
or what he calls matter in the fundamental sense (kuris hul). Since
prime matter is not itself a body, it does not contain within itself any
intrinsic form or quality of its own. He defines it, in fact, as what is
amorphous, formless, and shapeless, a formulation used by both earlier
Aristotelians and Stoics, and whose phrasing traces back to Plato and
Aristotle.15 But although prime matter lacks any intrinsic quality of its
own, it nevertheless cannot exist apart from some form or quality. Conse-
quently, it can be separated from form only in thought (6,17-19); it cannot
exist by itself on its own (autn huphestanai kath hautn, 4,8-18.20-1; cf.
18,2-7). Apart from a few brief reports in the doxographic tradition,
Alexanders discussion is the first full elaboration of the idea. It is still
hotly debated whether this is part of Aristotles own conception of matter.16
6 Introduction
At least as important, though, is Alexanders conception of form. As
Alexander points out, the forms of natural bodies are even less inde-
pendent than matter. For while prime matter cannot exist on its own, the
matter of compound bodies can, since they are composed from bodies and
bodies are precisely what can exist on their own. The forms of natural
bodies, in contrast, can never exist on their own, but must always exist in
some matter (4,20-5,2; see also 4,8-9; 5,21-2; 17,11-12). Here again we see
claims that will later become orthodox stated much more fully and directly
than in Aristotle. And while Alexander does not press the point, his
emphasis on the dependence of form has significant ramifications in
context. For if, as he will later make clear, the soul is a form of a natural
body, then the soul cannot exist on its own, but must be inseparable from
the body (17,9-15).
Less well appreciated are Alexanders innovations about forms. In
setting out his notion, he emphasises that forms are something non-bodily
(asmata, 5,18-6,2; 7,9-14). He clearly intends a sharp contrast with the
Stoics, who also hold there are two basic principles or arkhai of things, one
of which is qualityless matter and the other the divine logos that pervades
and shapes all things, giving them their distinctive qualities. On the Stoic
view, however, both principles are bodies.17 But what precisely does Alex-
ander mean when he says that form is non-bodily? He is not claiming that
forms can exist separately from bodies, as we have just seen. Nor is he
making a point about the irreducibility of form to matter, though he surely
is committed to this (17,17-18). Given the fundamental distinction be-
tween form and matter, it is trivially true that forms are immaterial. But
this cannot be his point in the present context, since he also thinks that
prime matter is non-bodily, even though it is surely material. The contrast
again is with the Stoic doctrine of principles. Alexander will later argue
that since forms are not bodies, they are never composite (17,15-18,10),
something that is not true of matter in general, apart from prime matter.
But this does not seem to be the motivation for the view we are now
considering so much as an elaboration or consequence of it.
Alexanders point is more subtle and profound. It is best understood as
turning on the metaphysics of causality. The Stoics characterise their two
principles in explicitly causal terms: one principle acts, the other is acted
upon, and together their interaction results in the constitution of ordinary
objects (see n. 17 above). But on the Stoic view only bodies can act or be
acted upon.18 Hence, the principles of bodies must themselves be bodies
(see n. 17 above). Alexander rejects this model of causality. Only bodies
can affect one another in this way. But the principles of bodies do not
themselves affect anything. Rather, bodies act and are acted upon in virtue
of their forms (7,9-13; cf. 6,21-3; 31,10-25). Forms are thus causally rele-
vant indeed, they are precisely what grounds or explains a bodys ability
to act or be acted upon in a certain way. But they are not themselves the
agents or subjects of change. Alexander is not making a point about
materialism or reduction, but about ontology. A form is not a body itself,
Introduction 7
but something that belongs to a body. Today we take for granted a
distinction between bodies and the properties of bodies; it hardly needs
saying that properties are not themselves bodies. But it was a relatively
new point at the time. Although the basis for it is certainly in Aristotle, he
never states it so fully and explicitly. Alexander is one of the first to
develop the point at length.19 To say that forms are non-bodily, then, is
simply to make a distinction about which ontological category they belong
to. It does not compromise or qualify his materialism in any way.
Alexander uses this conception of form to explain some of Aristotles
more distinctive claims about the soul. The first is that the soul cannot
itself undergo change (akinton): it does not move, grow in size, or undergo
any kind of alteration or modification in its own right.20 The composite
substance to which the soul belongs is what in fact undergoes any changes
ascribed to the soul, not the soul itself. The soul undergoes these changes
only vicariously, in so far as it is an essential part of something that does
properly undergo change as Aristotle and Alexander standardly put it,
the soul undergoes these changes only extrinsically (kata sumbebkos).
Alexanders point, however, is that this is not peculiar to souls, but
something that holds of causal powers, dispositions, and forms quite
generally. None of them undergoes change itself, but rather the things that
bear them do and they do so in virtue of having these dispositions or
powers. The only things that can properly undergo change are things that
exist separately on their own (21,22-22,23).
Alexander further uses this view to explain a second, closely related
doctrine. Because the soul does not undergo any modification itself, it is
not the subject of psychological states properly speaking (23,6-24). As
Aristotle says in what has come to be known as the celebrated Rylean
passage, DA 1.4, 408a34-b18, the soul no more perceives or thinks than it
builds or weaves; to think that it does is, in effect, to make a category
mistake. It is more accurate to say that a living being does these things
with or in virtue of its soul. Alexander emphasises again that this does not
hold merely for the soul, but for all powers and dispositions generally. The
heaviness of earth does not travel downwards by using earth; rather earth
moves downward in virtue of its weight (23,24-24,4).
For this reason, Alexander firmly rejects the suggestion that the soul
might be a homunculus, like the pilot of a ship, a possibility Aristotle
mentions in passing as still unsettled at DA 2.1, 413a8-10. If the soul were
like a pilot, Alexander argues, it would be separable from the living body,
and possibly even a composite itself. But if the soul is a form or disposition
of the living body, and so its completion (entelekheia) and culmination
(teleiots), as Alexander maintains, then the soul is inseparable from it. It
is nothing but the ability of a living thing to function as such, in the ways
characteristic of its kind. It would therefore be better to compare the soul
to the art of piloting, to the principles that govern the operations and
well-functioning of the ship. If the comparison with a pilot is merely a vivid
way of making this point, through a kind of synecdoche, it is harmless. But
8 Introduction
if the comparison essentially involves an inner person who uses and guides
the ship, then it is to be rejected, whether the soul is taken to be incorpo-
real, as the Platonists would maintain, or corporeal, as the Stoics and
other materialists would (15,10-26; 20,26-21,21; cf. 14,4-5; 79,19-20). All
such dualisms face a serious mind-body problem, according to Alexander:
once we have granted their premise that the soul and the body are
separate from each other and different in nature (kekhrismena te allln
onta kai tas phuseis diapheronta), it is no longer clear how the two could
ever constitute a unity (21,16-21).
On the Aristotelian view, in contrast, the soul and the body are not
separable in this way. They are both aspects of the living thing that cannot
exist apart from it or from each other. The way in which the soul is present
in (en) the body, and so the way in which the living thing is constituted
from (ek) them both, cannot be reduced to any other kind of presence or
composition: to that extent, the relation between the soul and the body is
sui generis (11,14-12,7; 14,24-15,5). But since it is just the relation between
form and matter, it is also a commonplace relation, found in all composites,
animate or inanimate, and is not invoked because of any special peculiar-
ity of the soul or mind. The Aristotelian solution to the mind-body problem
is thus not ad hoc, but a principled application of a broader metaphysical
strategy.
Finally, Alexanders conception of form explains a distinctive feature of
his approach. On his view, the forms of natural substances, and hence
souls, are ultimately to be understood as causal powers or dispositions.
Forms are the features of substances that enable them to do or undergo
the activities characteristic of their kind in Aristotelian terminology, a
form just is the ability to perform ones specific function or ergon. Aris-
totelians in the last forty years have taken this approach to be pivotal and
placed great emphasis on it. But the first person to make this central is
Alexander. A striking feature of his treatise, as a quick glance at the
Greek-English Index will show, is how prevalent the word dunamis or
power is much more so, in fact, than energeia or activity, which one
might otherwise have assumed would be more important, given the onto-
logical and definitional priority Aristotle assigns it in the Metaphysics. But
there is no substantive or doctrinal difference here, only one of emphasis.
A power must inevitably be defined as the power to perform or undergo a
certain activity, just as this activity can in turn be understood as its
culmination and fulfilment. Both Aristotle and Alexander stress, more-
over, that the soul and its faculties are to be identified with the first
attainment or completion (entelekheia) of a living body: they are the
features a living thing actually has that enable it to act or behave in certain
characteristic ways. Actually performing these activities represents a
further, second attainment or completion, which need not be taking place
in order for a thing to be alive or endowed with the relevant psychological
faculties.
This approach has even broader ramifications. If the forms of all
Introduction 9
natural substances are first attainments or completions, as Alexander
seems to believe, then the metaphysics of natural substances in general is
best pursued in terms of the causal powers or dispositions they possess, in
virtue of which they perform their distinctive activities. Such an approach
is, in a broad but important sense, a functionalist one: to be a certain kind
of thing is to be capable of performing specific activities, and its form is
just what enables it to do this.21 The basis of this view, again, is clearly in
Aristotle. But Alexanders conception of form as a causal power is the first
fully explicit articulation of it.

Supervenience and emergentism


The claim that different kinds of substance differ in form, then, is to be
spelled out in terms of the different causal powers each has: these powers
are what makes each the kind of thing it is.
Such differences are not adventitious, moreover: they do not float freely
on top of the underlying matter. A given body cannot take on just any form.
It can only take on those for which it is appropriate as matter it must be
suitable for the kinds of activity characteristic of that form. This functional
constraint on matter, of what sort of matter must be present if a given type
of form is to be realised, is what Aristotle calls conditional necessity.22 The
same constraint applies directly to souls and bodies, and it was the failure
to appreciate this, in both Aristotles and Alexanders opinion, that led
Pythagoreans to their mistaken doctrine of metempsychosis: a human soul
cannot employ the body of a sea slug any more than the principles of
carpentry govern the playing of a musical instrument like an oboe.23 It is
not just that the activities of souls presuppose some bodily activity or other
and so must be embodied (12,7-13,8). Their activities require bodies of a
specific sort, suited to carrying them out.
Aristotle further thinks, more strongly, that each type of body has a
distinct form of its own; and in context it is clear that this principle is
meant to apply directly to bodies and souls (DA 1.3, 407b23-4). Alexander
is even more explicit. Which type of soul a living thing has is a conse-
quence of the kind of body it has (epi ti toutn poiai sustasei
epkolouthsan hai diaphoroi psukhai), so that they covary with one
another (summetaballousin de alllois); more generally, which type of soul
a living thing has follows on the kind of temperament or blend of its body
(ti poiai krasei tou smatos kai h ts psukhs hepetai diaphora, Mant.
104,27-34). In the contemporary technical sense of the term, then, souls
supervene on bodies:24 the same type of body must have the same type of
soul; hence, any difference in the type of soul that living things have
implies a corresponding difference in the type of body.25
Now, it might be objected that the supervenience of souls on bodies is
in one sense trivial: living bodies, eo ipso, necessarily have a soul. It even
seems built into Aristotles definition: the soul, he says, is the form of a
natural body that through its power has life (smatos phusikou dunamei
10 Introduction
zn ekhontos, DA 2.1, 412a27-8), that is, it is the form of something which
can perform the activities distinctive of living things and is thus alive; the
soul simply is that ability or set of abilities in virtue of which the living
thing is alive. A body, therefore, which is incapable of performing these
functions is not the sort of thing a soul can belong to (412b25-6). We might
say that the dead body lying over there is a man, but only through a kind
of courtesy, or as Aristotle says, by homonymy (412b17-413a3). It is not
a man in the primary sense, of something which is living, breathing,
capable of walking, talking, engaging in meaningful interactions, and
experiencing a broad range of phenomena. A corpse is a man only in the
sense that it was a man in this primary sense. But then the soul, as
Alexander rightly points out, is not an accident or extrinsic feature
(sumbebkos) of the relevant kind of body, something that this body might
have or lack, while otherwise remaining the same (14,24-15,5).26 On
Aristotles view, then, the organic body of a living thing is essentially
ensouled: it is by its nature alive, with functioning organs, and so must
have a soul it is animate in the full, etymological sense of the word. It
would be trivial to claim that souls supervene on bodies of this sort.
Wherever there is a living body, there must eo ipso be a soul as well.
Different types of soul, moreover, might also plausibly be thought to
covary with different types of body, when this is understood in terms of the
fully functioning, organic bodies of different types of living thing.
Alexander is willing to go a good deal further than this objection allows,
however. The soul, in his view, is a power and form that supervenes
(epiginomenon) on the particular mixture of underlying bodies, which are
blended together in some specific proportion (25,2-3).27 The underlying
matter here is not the organic body, but the various bodies and materials
that go into its make-up, comparable to the separate ingredients that
make up the famous tetrapharmakon or four-fold drug (24,23-25,2), a
potent cocktail mixed from otherwise ordinary ingredients (according to
LSJ, wax, tallow, pitch and resin). At one point, in fact, he even seems
willing to regard the relevant underlying bodies as extending down to the
level of simple bodies, that is, down to the four elements that ultimately
constitute the body (26,21-2). If we want to explain the vast differences
between different types of souls say, of animals, which can perceive and
move in various ways, and plants, which can only nourish themselves,
grow and reproduce we will need to refer to the number and kinds of
underlying bodies, and how they are blended and joined together (10,14-
19). The material composition of a living substance is thus an origin and
principle (arkh) in the following sense: differences in forms follow from
and supervene on differences in the underlying matter (10,24-6), where
these are understood in terms of ingredients that can exist and be de-
scribed independently of the living substance they come to form.
Supervenience, moreover, is not a phenomenon peculiar to souls and
bodies, but something which holds for all natural bodies. In order to
adequately explain the tremendous diversity of forms and causal powers
Introduction 11
throughout nature, we must refer to the underlying diversity in the
materials that constitute these different bodies (8,13-25). Forms super-
vene on matter quite generally, then: there cannot be a difference in the
former without some corresponding difference in the latter.28 The combi-
nation of underlying bodies is therefore a sufficient condition for the
corresponding form that supervenes upon it.29
Attributing such a position to Alexander sometimes meets with resis-
tance because it is thought to imply a form of reductive materialism or at
the very least a rejection of Aristotles commitment to the explanatory
priority of form, the view that his preferred explanations refer in the first
instance to form rather than matter and so are fundamentally from the
top down. But in fact neither consequence follows. Alexander is clear on
this point. He straightforwardly rejects reduction. The blend (krasis) or
mixture (mixis) of underlying bodies can itself be said to be a kind of
harmony or tuning (harmonia), in line with a tradition reaching back at
least as far as Platos Phaedo, to a view put forward by the Pythagorean
Simmias (85E-86C).30 Whenever the constituents of the body possess a
harmony of the right sort, there will also be a corresponding kind of soul,
which cannot exist apart from it. But, Alexander insists, even if we grant
this much, the soul should not be identified with this harmony; rather, it
is a power that follows from this harmony (24,18-23; 25,3-9; together with
the note on 5,5-6). This distinction might seem overly nuanced. But it is a
direct and principled response to the views of his predecessors. An earlier
Peripatetic, the famous editor of Aristotles works, Andronicus of Rhodes
(first century BCE), had said the soul was either a blend of underlying
bodies or the power that followed upon them; and Galen of Pergamum
(second century CE) criticised Andronicus for not just coming right out and
identifying it with the blend. But from Alexanders point of view, Galen
picked the wrong disjunct.31 Souls and their causal powers do follow from
mixtures of certain sorts. But these powers do not belong to any of the
ingredients in their own right. They are distinctive, new powers that
emerge from this combination of material factors literally, they are born
or begotten upon them (epigennasthai).32
Once again, this is not something peculiar to the soul or mind. As
Alexanders survey of the scala naturae is intended to show (8,17-11,13),
it occurs throughout nature, starting with the elements. Just as the causal
power of being buoyant or light emerges from the elemental qualities of
being hot and dry, from which it is distinct (5,5-6), so new causal powers
arise at each successive level, with the increasing complexity of the
material constitution in compound bodies (8,8-10; 11,11-13). These bodies
are capable of behaviour quite unlike that of the materials that make them
up, which cannot be explained as the result of merely aggregating their
powers or their effects. But distinct causal powers mark off distinct kinds
of things and so a difference in form. Each new level of complexity
therefore involves formal differences that cannot be identified with or
reduced to the levels underlying it as matter. The primary causal expla-
12 Introduction
nation of behaviour distinctive of a certain level will thus always be from
the top, in terms of the higher-level form that emerges, even if the lower
underlying levels determine which form emerges. And even there the role
of form is still paramount. Higher-level forms emerge only in compound
bodies, where the constituent materials are themselves bodies and so have
form and matter of their own;33 and it is these lower-level forms which
contribute something (suntelouss ti) to the higher-level form that stands
in common over them (to epi pasin koinon eidos). It is for this reason that
Alexander describes the higher-level form as in a sense a form of forms
and a culmination of culminations (eidos ps eidn } kai teleiots tis
teleiottn, 8,10-13). All explanation of difference, then, is ultimately in
terms of form: higher-level forms explain the distinctive behaviour of a
substance, while the lower-level forms of the constituent materials are
responsible (aitia) for which higher-level form emerges (10,17-19).
This layered model of reality, as Jaegwon Kim has described it, is a
form of non-reductive materialism which bears a striking resemblance to
nineteenth- and twentieth-century emergentism, both in motivation and
in scope.34 It likewise holds that even though natural substances are
exhaustively constituted from material elements, their behaviour is not
governed solely by the powers and laws that characterise the lower levels.
At each new level, there are new orderly patterns of behaviour, which
must be explained in their own terms by the principles governing that
level. This is evident already at the lowest levels, in the gap that emerges
between the basic properties of the physical elements and the more
complex laws that govern chemical combinations and interactions. On this
view, the phenomena of life and mind are not exceptions to a general rule,
but rather one layer of complexity among many.

Parts of the soul


Since the soul is to be understood functionally, in terms of the causal
powers that set living things apart from inanimate things, it should come
as no surprise that Alexander devotes the remainder of his treatise, much
as Aristotle does, to an examination of the individual powers that distin-
guish souls from other forms.
Alexander follows Aristotle in the division and order of powers, as well
in their underlying rationale. Their criterion is essentially taxonomic:
given how these powers are distributed across living things, it is evident
that some can exist without others and are therefore distinct. In this way,
it can be shown empirically that there are three basic clusters of causal
powers, corresponding to three kingdoms of living things, which together
form a kind of hierarchy. At the bottom are basic vital functions like
nutrition, growth, and reproduction, which are the only vital powers that
plants have (29,1-10); next come the powers of perception and movement,
which distinguish animals (29,10-22); and last the powers of reason and
thought, which only humans have, at least among mortals (29,22-30,6).35
Introduction 13
But in the hierarchy of living things, there is also an important overlap in
the distribution of powers, with each successive echelon incorporating the
powers of the earlier ones: thus, animals have the basic vital powers in
addition to the perceptual ones, and humans have all three kinds. Differ-
ent powers of the soul are typically joined together in a given type of living
thing, then, and cannot be prised apart, even when there are distinct sets
of organs subserving them (37,4-38,11). The distinct powers within a living
thing thus come to possess a certain unity and integrity, even when they
can be found separately in other species (29,26-30,6; cf. 30,6-17).
The serial order of souls and the accumulation of powers within them
has methodological ramifications. The only powers common to every type
of soul are the most rudimentary vital ones. Why not define life and souls
in general in terms of these common powers? If we were to, the resulting
definition would not make clear how or why the soul finds its fullest
expression and culmination in humans, who possess all of the powers
distinctive of souls. On the other hand, were we to use this highest
realisation to define the soul in general, we would exclude each of the lower
kinds of soul, even though they are complete in their kind. The only way
to proceed, then, is through an investigation of the individual powers
which are shared piecemeal between the different kinds of living thing
(16,18-17,1; 28,13-29,1).36
Alexander also makes explicit an underlying premiss of this general
approach, namely, that the soul consists in a finite plurality of powers, and
offers systematic arguments in its favour. He rejects monistic approaches
to the soul he mentions Democritus, but surely the Stoics are in his
crosshairs here37 on the grounds that monists cannot account for psycho-
logical conflict, as Plato argued in Republic IV (27,1-8); more
importantly, such approaches make nonsense of the essential connec-
tion between the organs and structure of a living being and the kinds of
powers it possesses by nature, that is, between its matter and form
(27,8-28,2). At the other extreme, though, we cannot imagine that souls
have an infinite number of powers, since living bodies have a finite
number of organs and so can only engage in a finite number of different
types of activity (28,3-13). Our approach should accordingly take ac-
count of the organs and structure of the body, as well as how the
activities of living things are distributed over time and across species.
The ability to nourish oneself and to reproduce are distinct, not just
conceptually, but in terms of the organs involved and their maturation
at different stages in the development of an individual living thing
(35,9-22). But we also know that there are no species which naturally
have one of these basic abilities and not the other, so that both must
belong to the same general power of soul (29,1-3; 32,15-22). The princi-
ple of individuation here is thus relatively coarse-grained. This makes
it reasonable to think of faculties here rather than just abilities, which
might be more finely individuated.
14 Introduction

Nourishing and reproducing oneself


Alexander offers a very full and valuable discussion of nourishment,
growth, and reproduction, with many important details (31,7-37,3). In the
interests of space, I will simply mention a few of the noteworthy ones. To
nourish itself, a living thing must take in food from the environment and
transform it, assimilating it to the materials that make up its own body,
so that it can grow by accretion. In this respect, Alexanders account
conforms to Aristotles general model of causal interaction: what is in-
itially unlike becomes like through the action of the agent on the patient,
in this case through the action of the living thing on the nourishment
(33,13-28). Alexander attempts to go further, though, by showing how
nourishing oneself differs from other cases where something is enlarged
through the assimilation of contraries. But in the end, he cannot do it
without reference to the soul: nourishment occurs only where the agent of
growth is animate (33,28-34,26).
Alexander further develops Aristotles views of reproduction as well.
Although nourishment and growth are ways of preserving and sustaining
oneself (35,13-16), the production of another thing like oneself is not. But
it bears a certain similarity: it is the closest a mortal being can come to
immortality, namely, by preserving the species to which one belongs
(36,6-9; 32,11-15). The foetus has a soul like its parent, though it does not
completely manifest these powers until the relevant organs for carrying
out their respective activities have developed, apart from the activity of
nourishing itself (36,22-37). Although the foetus engages in this most basic
activity from very early on under its own power (ex hautou, 36,27), it
nevertheless receives its nourishment from the mother. Because of this,
Alexander argues, the foetus can not yet be said to be an animal or alive
simply as such and on [its] own, but is instead a part of the mother (38,4-8).

Perceiving
Alexander devotes almost all of the remaining treatise (38,12-92,14),
nearly half of the treatise as a whole, to the powers of cognition, that is, to
perceiving in general, the individual senses, the common sense, repre-
sentation (phantasia), and understanding (nous), with an interlude on
desire and action (73,14-80,15). The present volume contains only the first
sections of this, concerning perception in general, perceptible objects, and
sight. Again, in the interest of space, I will highlight only some of Alexan-
ders more distinctive contributions here.
Alexander begins his discussion of perception by emphasising what it has
in common with nourishing oneself. Like nourishing, perception also consists
in a causal interaction in which the agent makes the patient like itself
through a kind of alteration or change in quality (38,21-39,6; 39,14-18). But
in this case it is the external, perceptible object that is the agent: the animal
becomes like part of the world, rather than transforming part of the world
Introduction 15
into itself. Alexander explains this teleologically, by appealing to the utility
of perception. In order to survive and thrive, an animal must be able to avoid
threats and pursue opportunities, and it detects these by perception (39,9-10).
What the senses undergo in perceiving is not, however, the same sort of
change or alteration that food undergoes when it is digested. The latter
cases involves a transformation and change of form in the patient: as the
food becomes assimilated to the body, it loses some of its previous causal
powers. The senses, in contrast, do not lose their ability to perceive when
they perceive, as Aristotle makes clear in On the Soul 2.5; on the contrary,
perceiving is just the realisation of this ability. Alexander always marks
this difference with a qualification: perceiving is a kind of (tis) change,
alteration, or modification, or only in a way (ps). But at the same he also
insists that perceiving occurs through (dia) a change, alteration, or
modification of the sense organs (39,1-2) in the primary and literal sense
of these words.38 In terms of recent debates over Aristotles theory of percep-
tion, Alexander rejects spiritualism without embracing full-out literalism:
he insists that perception requires some kind of literal change of the sense
organs, without identifying perceiving with that change.39 In making these
distinctions, interestingly, he does not directly employ Aristotles claim that
in perception the sense receives the form without the matter (aneu ts huls),
though he also describes perception in these terms (39,13-14).
Alexander develops these basic doctrines into a systematic causal the-
ory. On his view, when any of the five peripheral sense organs is
stimulated, it relays (diadidonai) this change to a central organ of per-
ception (39,18-21), which he identifies with the heart, where perception
actually occurs (39,21-40,3). In the cases of the three distance senses
sight, hearing, and smell there is a further link earlier in the chain:
perceptible objects affect our senses only by first affecting an external
medium which relays or transmits this message (diakoneisthai) to the
peripheral organs.40 The basis for this view can be found in various
remarks scattered throughout Aristotles corpus (although his view about
where perceiving occurs is less than clear). But Alexander gives them
added prominence and coherence, by synthesising them into a unified
whole and placing them at the forefront of the theory.

Perceptibles and error


Like Aristotle, Alexander divides perceptible objects into those which are
perceptible intrinsically (kath hauta) and those which are perceptible
extrinsically (kata sumbebkos); the former are divided again into those
which are intrinsically perceptible to one sense exclusively (idia) and those
which are common to or shared by several senses (koina). But Alexander
also corrects or modifies the characterisation of each of these groups,
sometimes significantly.
In the case of extrinsic perceptibles, Alexander states clearly something
some commentators have assumed, but Aristotle never explicitly says,
16 Introduction
namely, that they are not perceptible at bottom (oude aisthta tn arkhn;
41,9), but are perceptible solely in virtue of another quality which is
intrinsically perceptible, to which the perceptible in question is extrinsi-
cally related (41,6-10). On this view, extrinsic perceptibles are not strictly
perceptible at all, but in effect only by homonymy. Aristotle, in contrast,
appears to classify them in his division as genuine perceptibles, even if
they are not perceptible in the primary or fundamental sense (kuris,
418a24-5) that exclusive perceptibles are.41
In the case of common perceptibles, like shape, number, or change,
Alexander claims that they need only be able to affect more than one sense,
rather than all of them, as Aristotle incautiously states.42 This modifica-
tion has the added benefit of making the subdivision between exclusive
and common perceptibles dichotomous. He also remarks that the two
kinds of intrinsic perceptible are interwoven (sumpeplegmena) with one
another (83,17-22) and so jointly affect the peripheral organs (65,11-16);
the effect of the common perceptibles is then relayed to the central organ
(41,4-5). But unlike exclusive perceptibles, the common perceptibles are
not perceived by the power to perceive colours, for example, or odours.
They are perceived only by the common power of perception, which Alex-
ander calls the common sense (65,10-11.16-21).
Perhaps the most striking change, though, is in Alexanders charac-
terisation of intrinsic perceptibles exclusive to a single sense. In On the
Soul 2.6, among other texts, Aristotle claimed that it was impossible for
the senses to be mistaken about the exclusive perceptibles themselves,
though they might be in error about where they were located or which
objects had them.43 But in one passage, Aristotle qualifies this by allowing
that the perception of such perceptibles is either true or involves the least
amount of falsehood (DA 3.3, 428b18-19). Alexander seems to temporise
even further when he maintains only that the perception of exclusive
perceptibles is the most true and even then only when a number of
conditions are satisfied (41,13-15). He holds that such perception is veridi-
cal only in natural or normal conditions, of which he lists five: (i) the
condition of the organ, (ii) the direction in which the object lies, (iii) its
distance from the sense, (iv) the condition of the medium, and (v) the
absence of interference (41,13-42,3). When one or more of these conditions
is not met, he claims, perceptual error is likely to occur.44 This discussion
comes as part of a more general characterisation of perceptual error: it
occurs whenever the change in the sense deviates from the quality in the
object which gives rise to the perception and so fails to correspond to it.45

Seeing, light, and colours


Alexander begins his discussion of sight by considering the types of objects
perceptible exclusively to it, which includes not only colours, which can be
seen in the light, but also phosphorescent objects, which can be seen in the
dark (42,4-7; 43,16-18).46 Thus neither sight nor what can be seen are
Introduction 17
defined in terms of colour: colour is just one of the things that can produce
it. Its role, like that of phosphorescent objects, is causal. Colours are what
can produce the sorts of changes in an illuminated, transparent medium
that in turn can affect an animals eyes in such a way as to produce sight
(42,5-11; 43,11-44,2). Although we do not in general see this effect on the
medium itself, Alexander thinks that in some circumstances we can. He
gives a long list of cases where the air surrounding an object is as
though tinged (hsper khrnnumena) with its colour or that of a facing
object (42,11-19).
To see a colour, the intervening medium of air or water must not only
be transparent, it must also be illuminated. Like Aristotle, Alexander
holds that light is not a body that travels through the medium. But in
Alexanders view, it is also not a condition or modification of the medium
which the medium takes on in the way that matter receives a form or
quality (42,19-22). The medium does not itself undergo any alteration or
modification in its qualities when it is illuminated, but rather a relational
change what some have called a mere Cambridge change due to the
presence of fire or something similar in the medium (42,22-43,4; 43,8-11;
45,2-5).47 Unlike other changes, illumination occurs instantaneously with-
out duration (akhrons), even across vast distances (43,11): it occurs
immediately, as soon as fire or some other illuminant is present in the
medium and ceases as soon as it is removed (42,22-43,1).
When the medium is illuminated in this way, it realises fully its nature
as something transparent something, literally, through which an object
appears (diaphanes).48 It is this power which allows it to relay the colours
of objects to sight and so function as a medium. But interestingly this is
not what makes them see-through (diopton). On the Aristotelian view, all
bodies have some degree of transparency in them (44,20-1), even if they
are not see-through at all; and in bodies with definite boundaries or
surfaces of their own, the transparency in them is responsible for their
colour (44,18-23; 46,12-14), different colours being due to the admixture of
opaque materials like earth (44,5-45,10). It is when a transparent material
lacks a definite boundary of its own, like air or the sea, and is shaped
instead by the bodies around it, that the material lacks a colour of its own
and is transparent in the fullest sense of being see-through (44,23-44,5;
46,7-11).49 Light is like a colour of these transparent bodies (44,16; 45,1;
though compare 45,17-19). But in the end, light no more constitutes their
own proper colour than does their being tinged by surrounding bodies.

In the remainder of the treatise, which can be found in the next volume,
Alexander continues his examination of the individual senses hearing,
smell, taste, and touch and the common sense. He then goes on to
consider other abilities animals possess, such as representation, desire,
and striving, as part of the soul capable of perception, before turning
finally to the understanding, which in the sublunary realm can only be
found in humans.
18 Introduction

Notes
1. Although this claim is generally accepted (Zeller 1923, 817-18 esp. 818 n. 1;
Thillet 1984, liii esp. n. 2; Sharples 1987, 1179 esp. n. 23), it has more recently been
contested by Barnes et al., who argue that this and similar phrases are not
honorific titles they are ordinary referring expressions, much as we might use
the author at the end of a book review, without meaning The Author par
excellence (1991, 4 n. 28). The objection is clever and has some merit. But it is not
borne out by the evidence as a whole.
There are a few cases in philosophical commentaries where ho exgts is used
anaphorically to pick up a prior mention of Alexander or some other commentator
in context. This is most obvious where a contrast is being drawn between the
commentator and the author commented on: thus Simplicius contrasts what
Aristotle says with what his commentator Alexander says (in Phys. 1170,1-2.13)
and Proclus contrasts Plato and Porphyry in the same way (in Tim. 1.322.4-7);
similarly, Olympiodorus contrasts the commentator Alexander with the great
philosopher Ammonius (in Meteor. 263,20-1). There are also a few cases of anaph-
ora without any contrast being made: thus, Simpl. in Phys. 313,30 appears to be
referring back to Alexander at 312,1 and 1176,32 refers back to him at 1175,13;
similarly, 213,34 refers back either to Alexander at 213,23 or Syrianus at 213,24.
The greatest number by far occur in the Platonic commentaries of Damascius: he
sometimes refers to Syrianus simply as the commentator (in Phaed. I 222, 235,
236, 243) and even more frequently to Proclus (in Parm. 312, 394, 421; in Phileb.
133, 200, 242; in Phaed. I 297, 405; II 14, 25, 88, 89, 100, 110, 128, 129, 132,
145, 147; on Damascius various forms of reference for these authors, see e.g.
Westerink 1959, xvii). Philoponus similarly refers to both Taurus and, more
commonly, Proclus as the commentator on Plato (de Aet. Mundi 311,18; 482,22;
520,25; 521,3; 524,14; 553,9).
But this sort of case is the exception rather than the rule. In virtually all of the
passages in the commentaries on Aristotle where the singular ho exgts is used
to refer to a specific individual, it is as part of an epithet of Alexander specifically
(e.g. Alexander the commentator, Alexander the commentator on Aristotle, the
Aphrodisian commentator): Them. in DA 120,18; Syr. in Metaph. 53,12; 100,6-7;
122,11-12; 195,12 (Syrianus uses ho hupomnmatists instead of exgts in all but
the second passage); Ascl. in Metaph. 93,18; 245,26-7; Ammon. in de Int. 13,19;
39,14; 100,20; 202,4; Simpl. in DC 121,12; 700,9; in Phys. 707,33-4; in DA, 52,28;
Philop. in An. Pr. 126,20-1; in GC 62,17-18; 82,12-13; 214,23; 226,17-18; de Aet.
Mundi 48,10; 191,19-20; Michael, in EN V 584,3. (I am grateful to Michael Griffin
for discussion on some of these passages.) In all of these cases, the phrase the
commentator is a mark of distinction: it is not used in the commentaries on
Aristotle for any other commentator, except once, for an unnamed commentator
distinguished from Alexander at Olymp. in Meteor. 273,27-8. This honorific use is
confirmed by other epithets used for Alexander. Simplicius refers to him as the
most authoritative commentator on Aristotle (tou gnsitatou tn Aristotelous
exgtn, in Phys. 258,16-17; ton gnsiteron, 80,15-16) as well as the most
industrious of them (ho philopontatos, in Phys. 795,34-5). Syrianus calls him the
most industrious as well (in Metaph. 54,12-13).
Someone might conceivably object that the epithet the commentator is only
used to distinguish him from other Alexanders, such as Alexander the Great (ho
megas), as he is in a late discussion of homonymy (Michael in EN IX-X 4,17). But
there is no reason to think that there is any danger of confusing these or any other
Alexanders in these contexts generally; and even if there were, it is always possible
Introduction 19
(and indeed common) to use place names to distinguish such homonyms. It is
noteworthy, then, that Alexander of Aphrodisias is virtually the only one who
receives the epithet the commentator.
2. Syrianus in Metaph. 100,6; David [Elias], in Categ. 128,12-13. The first
passage praises a commentator on Aristotle as the new or more recent Aristotle
(ho neteros Aristotels), but does not identify him further by name, though he
mentions the Aphrodisian commentator at 100,4 and 10. (On the use of the
comparative, see Thillet 1984, xxix, cf. xxvii n. 1.) The second passage speaks
explicitly of the instructor Alexander, who by right is himself like a second
Aristotle (edei gar auton hoion deuteron onta Aristoteln). Zeller took the reference
in both passages to be to Alexander of Aphrodisias (1923, 805-8 n. 2 at 807). This
was contested by Moraux 1967, 177-82, although he is forced to emend both texts
in crucial ways. For a convincing and detailed response, see Thillet 1984, xix-xxxi.
3. Philostratus VS 566; Lucian, Eunuch 2-3; cf. Dio Cassius 72.31. For more
detail, see esp. Oliver 1970, 80-91; Lynch 1972, 170; Todd 1976, 6-7; Oliver 1977,
166; Glucker 1978, 146-53; Oliver 1981, 216-17, 221-5 (including his critique of
Dillon [1977] 1996, 233); Hahn 1989, 126-8; and most recently Watts 2006, 33-5.
Oliver 1981, 224 emphasises that we do not know the subsequent fate of the chairs
after Marcus Aurelius death; cf. also Lynch 1972, 177, though contrast Watts
2006, 34-5. On the mention of two Peripatetics in Lucian, see Lynch 1972, 171;
Glucker 1978, 148-50; Hahn 1989, 126-7 n. 30. Thillet speculates (1984, lxxviii esp.
n. 2) that On Fate was dedicated in response to a promotion (164,14) in the
Peripatetic section of the faculty at the University of Athens. But it need only be
a reference to Alexanders initial appointment to the imperial chair of Aristotelian
philosophy, which does not presuppose a university or a faculty, as the work of
Lynch and Glucker has persuasively shown.
4. For the dedication, see Fat. 1, 164,1-3.13-15, although see Sharples commen-
tary (1983, 125 ad loc.) on Caracallas antipathy to Aristotelians. Alexanders
failure to mention Gaeta in his dedication suggests that the treatise must have
been composed before the latters becoming Augustus in 209: see Todd 1976 1 n. 3;
Thillet 1984, vii-viii n. 3, cf. lxxv; and Sharples 2005a, 47, though Barnes et al.
(1991, 1 n. 1) regard this as reasonable, but not perhaps absolutely mandatory.
5. For the inscription, see Chaniotis 2004a, 388-9 (inscription no. 4, with photo)
and 2004b. For an extensive discussion of its implications, see Sharples 2005a
(including the identification of the title diadokhos with the imperial chair, at 52-3;
and the possible attribution of works to Alexanders father, 53-6).
6. For the evidence on his teachers, see Sharples 1990, 86-9; 2005a, 49-50; also
Thillet 1984, viii-xxxii. On Sosigenes himself, see Moraux 1984, 335-60 and on
Herminus, 361-98. Morauxs further conjecture (1967) that Aristotle of Mytilene
was also one of Alexanders teachers is, despite its influence, highly questionable:
against it, see Thillet 1984, xi-xxi and Opsomer and Sharples 2000 passim (with
further references on the question at 256 n. 30).
7. For extensive, critical discussion, see Sharples 2005a, 50-2.
8. For a comprehensive survey of Alexanders works and the secondary litera-
ture on them, see Sharples invaluable survey (1987), 1182-99.
9. Alexander refers to his commentary (hupomnma) on On the Soul at Quaest.
1.11a, 21,18; 1.11b, 22,26; Philoponus expressly speaks of Alexanders own On the
Soul as a distinct treatise (idion biblion) at in DA 159,18-19. For further discus-
sion, with references, see Sharples 1987, 1186, paragraph (g); Moraux 2001,
317-53. Moraux argues that the treatise makes use of formulations from the
commentary, which it expands on and elaborates, and is therefore later than the
commentary (2001, 338). For a collection of the fragments on the understanding
(nous), see Moraux 1942, 203-21.
20 Introduction
10. A similar difference can be observed in Aquinas treatises and commentar-
ies: even though his commentaries clearly reflect his overall outlook, their
immediate purpose is to explain Aristotles text and this constrains what he is
willing to mention or devote his energies to, in contrast with his own systematic
works.
11. All references to Alexanders On the Soul are to the Berlin Academy edition
of Alexanders Scripta minora in the Supplementum Aristotelicum, vol. 2 part 1,
edited by Ivo Bruns (1887).
12. Morauxs sympathies are plain, despite his protests against those who
consider Alexanders solutions only in comparison with Thomas, since he still
takes Thomism to be the ultimate criterion (1942, xxii-xxiii). His frequent denun-
ciations of Alexanders materialism of how it is fundamentally anti-Aristotelian,
a view which could not be more foreign to Aristotles theories (31, 33, 39, 42, 44,
48) and harbours heresies (172, 173), with a monstrous and genuine contradic-
tion at its core (11, 34, 44, 48, 167, 168-9, 173) reveal just how much is at stake
for someone with Morauxs allegiances.
13. Alexander explicitly restricts his scope in this treatise to the sublunary
realm, where substances come to be and perish, and so excludes from consideration
the souls of the divine movers of the heavenly spheres and the Prime Mover
(28,26-8; cf. 1,2).
14. 9,14-25; 11,5-13; 20,19-22; 22,4-12.15-23; 23,24-24,4; cf. 5,4-18; 13,22-14,3.
15. For detailed references, see the note on 3,28-4,4.
16. For references, again see the note on 3,28-4,4.
17. D.L. 7.134 = SVF 2.299 = LS 44B 3, retaining the MSS reading see LS
1.274, 2.266; this is confirmed by Alex. Mixt. 225,3-5 = SVF 2.310.
18. For references, see the end of the note on 7,9-13.
19. For references, see the note on 7,9-13.
20. In one sense, the soul also does not come to be or perish either, since
Aristotle denies that there is coming to be or perishing of forms in general (Metaph.
7.8, 1033b5-8). But in a broader sense, he allows it, since the form or essence is
what comes to be in another thing (ho en alli gignetai, 1033b7); that is, there is
a moment when there is something with a given form, where there had not been
before, and there is another later moment when there is not any longer (7.15,
1039b23-7; 8.3, 1043b14-16; cf. 6.3, 1027a29-30; 8.5, 1044b21-2; 12.3, 1070a15-17).
For a good discussion of this nuance, see Frede and Patzig 1988, 2.136-7. Aristotle
is even willing in fact to speak about the form supervening on the matter
(epigignesthai) in these cases, although not in the technical sense of the term that
we will see in Alexander (1035a12; 1036a31, b6; cf. 1035a5). Alexander in any case
speaks of the soul both as coming to be in this broader sense (see e.g. DA 24,3-4
below) and as supervening on matter (10,25; 24,19; 25,2; 26,22), while also being
openly committed to the orthodox denial of forms coming to be in the narrower
sense (ap. Simpl. in de Cael. 578,32-579,2). For more extended discussion, see
Accattino 1995, 198-200.
21. Functionalism in this broad sense, concerning the nature and power of
substances, does not imply functionalism in the narrower sense used in contempo-
rary philosophy of mind, which defines mental states in terms of their inputs,
outputs, and causal role with respect to other mental states, although the two
senses are obviously compatible with one another. There are good reasons, how-
ever, to think that Aristotle is a functionalist only in the broad sense outlined
above, and not in the narrow sense: see my 2006, 320-2.
22. PA 1.1, 639b21-30, 642a1-13; Phys. 2.9 passim.
23. Aristotle, DA 1.3, 407b13-27; cf. 2.2, 414a22-5; Alexander DA 27,15-28,2.
Introduction 21
For more on metempsychosis, including Alexanders critique, see note to 27,15-
28,2 below.
24. For more on this technical notion, with references to the contemporary
literature, see my 1997, 313-14. For discussion of the authenticity of Mant. 104,
see the end of n. 92 on 10,24-6.
25. The converse, it is important to note, need not hold: for all that has been said
here, it might be possible for there to be a difference in bodies without a corre-
sponding difference in souls, in which case the same soul might be found in
different types of body, so long as each was suitable, a possibility contemporary
philosophers refer to as multiple realisability. But this just shows that the present
claim about supervenience is distinct from the earlier one, which is meant to rule
out metempsychosis.
26. For further references, see the note on 14,24-15,1 below. In spite of these
clear denials, Moraux 1942 claims that Alexander is nonetheless committed to this
view and so in contradiction both with himself and Aristotle (xix, 42-4; cf. 9-10).
27. For Alexanders use of supervene (epiginesthai) and further references, see
the note on 25,2 below.
28. Quite generally, but not universally: even leaving aside questions about
forms that exist apart from matter, it is clear that the most basic elemental forms
the four elemental qualities of hot, cold, wet, and dry do not supervene on prime
matter, since each of these different forms can belong to the same underlying
substratum. But this is not surprising either, since prime matter has no qualities
or differences of its own and so no form essential to it. Apart from this base case,
though, all other natural forms the forms, that is, of compound bodies will
supervene on their underlying matter. See note on 10,17-19 below.
29. For further discussion, see the note on 10,24-6 below.
30. For more on this conception of the soul and the ensuing debate, see my 1997.
31. For references, see the note on 25,2-4 below.
32. For Alexanders use of emerge (epigennasthai) and further references, see
the note on 5,5-6 below.
33. Emergence does not occur in simple bodies, because their underlying matter
is prime matter. Since prime matter is compatible with all of the elemental
qualities, none of them supervene on it: being completely without form, it does not
determine which elemental qualities will be instantiated. See n. 28 above.
34. Kim 1993, 337-9; McLaughlin 1992 passim.
35. Alexander explicitly excludes divine beings from consideration in this
treatise, on the grounds that if they can be said to have a soul at all, it is only by
homonymy: see 28,25-9 below with corresponding note.
36. Alexander would likely fault a disjunctive characterisation of the soul as
that which, e.g., enables a living thing to nourish itself or perceive or think for
equally failing to capture the sense in which the highest soul constitutes the fullest
and most complete expression of soul powers.
37. See note on 27,6-7 below.
38. For discussion and references, see (3) in the note on 38,21-39,2 below.
39. For a brief summary of this debate, see my 2006, 328-30; and at greater
length in my 2005.
40. For the terminology of relaying and transmitting see the notes on 39,18-21
and 41,1-3 respectively.
41. Alexander may not always take such a hard line: for discussion of a possible
alternative in Quaest. 3.8, see the final paragraph of the note on 41,8-10 below.
42. DA 2.6, 418a10-11, a18-19, though he seems to recognise the difficulty at
Sens. 4, 442b4-7.
43. For the many references, see the note on 41,13-42,3 below.
22 Introduction
44. Similar lists are reported by various sceptics and are sometimes associated
with the Stoics epistemology. But whether they are its ultimate source or, as I
think more likely, the Academic Carneades, or possibly even Aristotelian sources,
cannot be determined from our evidence. For references, see the last paragraph of
the note on 41,13-42,3 below.
45. For further discussion of perceptual error and the correspondence principle
presupposed, see the note on 41,10-13.
46. We might be tempted to regard the appearance of phosphorescent objects
equally as colours. But Alexanders point is that their appearance in the dark is
different from their actual colour: an objects own colour is the way it looks in the
light.
47. On mere Cambridge changes, see (2) in the note on 38,21-39,2.
48. For this etymology, see 43,5-8 below, along with the note for further
references.
49. Alexander discusses transparent solids like glass, horn, and mica, which
have definite boundaries of their own, in other treatises: for references and
discussion, see the note on 42,7-8 below. For Aristotles view, see Sorabji 2004b,
131.
Note

Text and editions. The text used for this translation is from Ivo Bruns
Berlin Academy edition of Alexander of Aphrodisias Scripta minora, in
the Supplementum Aristotelicum, vol. II, part 1 (Berlin: G. Reimer, 1897).
Any changes are flagged by an asterisk in the translation and noted
separately in the Textual Emendations.
The division of Alexanders On the Soul into two parts does not corre-
spond to a division in our manuscripts or editions, but was made purely
because of the exigencies of publication.
All references to Alexanders other works, as well as that of the other
Greek commentators on Aristotle, are to the Berlin Academys Commen-
taria in Aristotelem Graeca (1882-1909).
A key to the abbreviations for other ancient texts can be found in the
Index of Passages at the end of the volume, along with the name of the
editor, where more than one edition is commonly in use.

Translation. Translating Alexander into readable English presents cer-


tain challenges. He is, if you will, one of the first authors to write in
academic German: he is given to frequent Schachtelstze and long series
of subordinate clauses (in which he occasionally gets lost himself). Worse,
he standardly conditionalises his formal arguments into a single sentence
that includes not only the many premisses, but also their individual
elaborations and justifications. Preserving these syntactical features of his
treatise would serve little purpose. In many cases, it is possible to break
his sentences into smaller units without affecting their underlying logical
relations.
To make the structure of Alexanders inferences even clearer, though, I
have taken the liberty of using the modern typographical convention of
nested paragraphs with identifying numbers or letters for the different
premisses or senses of a term being distinguished. This is not in fact a
great departure from Alexander, since he sometimes uses numbers him-
self to distinguish different conditions and claims (e.g. 22,27-23,3;
41,15-42,2). I have also sought to be as consistent as possible in rendering
his terminology, so that it is easier to see how much more natural his
inferences and assumptions might seem as originally framed. (Variations
in the translation of each term are recorded in the Greek-English Index at
the end of each volume.) In quite a few cases, I have also avoided custom-
24 Note
ary translations, where another rendering would bring out etymological
connections to other terms or invigorate an otherwise arid and abstract
construction. Instead of translating teleisis as perfection and entelekheia
as actuality, for example, I have rendered them as culmination and
completion respectively, to bring out their connections with the underly-
ing notion of a telos or end. Similarly, instead of translating aisthtik
dunamis as the perceptive faculty or potentiality, I have used the power
for perceiving to bring out the emphasis Aristotle and Alexander both
place on the corresponding activity. The unfamiliarity of these transla-
tions (and in some cases inelegance) will undoubtedly rub some readers
the wrong way. But my hope is that they will make Alexanders positions
and arguments seem less arcane or obscure.

Commentary. In line with the principles of the series, my notes have two
primary aims: (i) to make philosophical sense of the arguments in the main
text and (ii) to provide the necessary background to understand them by
supplying the references to Alexanders predecessors and contemporaries
to Aristotle above all, but also various Stoics, Platonists and Peripatetics
and cross-references to Alexanders own works. Unfortunately, I have
had to limit myself severely with regard to the secondary literature: to
have included them more fully would have swollen the size of the volumes
unduly. The two main exceptions I have allowed myself are recent com-
mentaries on the treatise, a superb one in Italian by Accattino and Donini
(= AD) as well as a French one by Bergeron and Dufour (= BD), as they are
directly on the treatise itself and unlikely to be widely known or available
to English readers. Although I cite them most often to mark our disagree-
ments, I have learned immensely from them both.
Abbreviations

Abbreviations of ancient works can be found in the Index of Passages


Cited. The abbreviations below are for modern reference works.

AD = Paolo Accattino and Pier Luigi Donini. 1996. Alessandro di Afrodisia:


Lanima. Traduzione, introduzione e commento. (= Biblioteca Univer-
sale Laterza, v. 447.) Rome: Laterza.
BD = Martin Bergeron and Richard Dufour. 2008. Alexandre dAphrodise:
De lme. Texte grec introduit, traduit et annot. Paris: Librairie Phi-
losophique J. Vrin.
DK = Hermann Diels and Walther Kranz (Hrsg.). [1903] 1951. Die Frag-
mente der Vorsokratiker. Griechisch und Deutsch. 6. Auflage. 3 Bnde.
Berlin: Weidmann, 1951.
Dox. Gr. = Diels, Hermann. 1879. Doxographi Graeci. Collegit, recensuit,
prolegomenis indicibusque instruxit. Berlin: G. Reimer.
LS = A.A. Long and D.N. Sedley (eds). The Hellenistic Philosophers. 2 vols.
Cambridge: Cambridge University Press, 1987.
SVF = von Arnim, Hans Friedrich August. 1903-24. Stoicorum Veterum
Fragmenta. 4 vols. Leipzig: B.G. Teubner.
This page intentionally left blank
Textual Emendations

The following list contains the emendations to Bruns text which are
followed in the translation (marked there by an *).

5,7 Retaining the MSSs autou instead of Bruns hautou.


6,7 Retaining the MSSs autous instead of Bruns hautous.
6,11 Accepting ADs correction of the MSSs te to ge.
6,23 Supplying <pur> a second time, just after pur, following
the Hebrew translation.
7,20 Retaining the MSSs autou instead of Bruns hautou.
14,1 Accepting Bruns suggestion <en> toi loipi.
14,8 Correcting the MSSs kekhristo to ekekhristo.
14,24 Reading diapheron instead of the MSSs diaphorn or
diaphern (or Bruns emendation, diaphoron).
15,6 Keeping the MSSs autou instead of Bruns hautou.
16,1 Omitting hn kai teleiotta kai entelekheian ethos
Aristotelei legein with Bruns.
16,13 Accepting Bruns suggestion to correct the MSSs
dunamenou to dunamenon.
18,1 Accepting Bruns emendation, to einai, instead of the
MSSs to eidos.
18,2 Retaining auts with the MSS instead of Bruns hauts.
18,10 Reading tou smatos with the Hebrew translation instead
of tn tou smatos mern with the MSS.
19,5 Accepting Useners conjecture ti, with einai understood,
instead of the MSSs to.
20,9 Accepting Bruns emendation of hautou and hauti (twice)
for autou and auti (twice).
20,22-3 Accepting ADs emendation tn psukhn hs eidos instead
of Bruns tn hs psukhn eidos.
21,1 Emending the text to read all hols rather than the
MSSs all ei hols or Bruns more drastic all oud hs.
22,18 Accepting Bruns emendation hautn for the MSSs autn.
22,27 Omitting kai, following the Hebrew translation.
27,9 Omitting Bruns insertion of zia.
28,5 Accepting Bruns suggestion that we read esesthai, with
the Hebrew translation, rather than the MSSs legesthai.
28 Textual Emendations
30,2 Retaining the MSSs dunamen instead of Bruns
dunamenn.
31,14 Accepting Bruns hauti instead of the MSSs auti.
33,14 Accepting Bruns deletion of troph legetai.
34,4 Inserting huph before hn, with AD.
34,5 Retaining the MSSs tauta, instead of Useners emendation
tanantia, printed by Bruns.
34,21 Accepting Bruns tode instead of the MSSs to de, and
inserting < tode> afterwards with AD.
35,4 Reading khris toutou, following ADs suggestion, instead
of the MSSs khris tauts.
35,24 Accepting Victorius correction of the MSSs phusiks to
phutiks (printed by Bruns).
36,5 Retaining the MSSs autou instead of Bruns hautou.
36,17 Accepting Bruns emendation of hauti for auti, and his
suggestion that we read homoion poiousa for the MSSs
homoiousa.
37,14 Accepting Bruns suggestion that we read oude instead of
the MSSs oute.
38,5 Accepting Bruns emendation of pherouss for trephouss,
in line with the Hebrew translation.
39,5 Omitting the first kai with AD.
39,21 Reading ep with AD rather than the MSSs eit (acute) or
Bruns eit (circumflex).
40,5 Omitting Bruns addition of phutiks before ts psukhs.
40,19 Accepting Bruns correction of the MSSs homois ei to
homoisei, in line with the Hebrew translation.
40,21 Reading the neuter idia, rather than the plural idiai (MS
V, Bruns) or the feminine idia (MS a). Cf. 41,3.
41,6 Inserting ha following the first aisthta, with AD, instead
of Bruns suggested insertion <ta kath hauta ouk aisthta,
ha> at the same point, following the Hebrew translation.
41,10 Accepting Bruns emendation of hi for the MSSs .
44,8 Accepting Bruns emendation of hupo, following the
Hebrew translation, instead of the MSSs auto.
45,3 Accepting Bruns suggestion that we read the singular
auto, rather than the plural auta in the MSS.
45,23 Accepting Bruns emendation of monon for the MSS
monn.
Accepting Bruns emendation of smata for the MSS
khrmata.
46,11 Retaining the MSS horiston instead of Bruns emendation
of ahoriston (following the Hebrew translation).
Alexander of Aphrodisias
On the Soul

Part I: Soul as Form of the Body,


Parts of the Soul, Nourishment,
and Perception

Translation
This page intentionally left blank
Alexander of Aphrodisias, On the Soul 1,1

My aim is to speak about the soul that belongs to a body subject to


coming-to-be and perishing:1 to say what its essence (ousia)2 is and
what its powers are, how many powers there are, and what their
difference is from one another.3 For given that
1. a person ought, before all else, to obey what the gods com- 5
mand;
2. Know Thyself is commanded and decreed by the Pythian
god,4
and so not merely by a god, but one who is believed to foreknow the
future5 and who proclaims and declares that
3. each person, through knowledge of himself,6 will achieve the
life in accordance with nature;7
but
4. self-knowledge consists in the knowledge of that in virtue of 2,1
which [each person] is himself;
5. a human is a human in virtue of the soul;
it follows necessarily that
6. any person who intends to obey the god and live the life in
accordance with his nature must first analyse the soul and
know just what it is.8
Just as I have great reverence for Aristotles works on other subjects,
since I find more truth in the views passed down from him than in 5
what others have claimed, I regard what he states in his theory of the
soul in the same way as well. I will therefore have fulfilled my aim if
I can set out his claims about the soul as clearly as possible and offer
suitable arguments to show how each of them is well formulated.9

What is the souls essence?10 10


Anyone who is going to examine in detail what is said about the soul
and reach agreement on characterisations of its essence must first
appreciate the extraordinary magnificence of nature, before all else.
For once we have learned just what sort of thing nature is and have
been persuaded that its workings are more incredible than any
marvel produced by art,11 we will more easily be convinced by what
32 Translation
15 is going to be said about the soul.12 For the view that its powers and
activities cannot easily be harmonised with what is said about it,
because they are more divine and greater than any bodily power,13
is no reason to get mired in puzzles about the soul. Hence, those who
do not wish to go against sensible claims about the soul ought first to
20 look at the construction of the body that has the soul and to investi-
gate the regulation of its inner parts and then afterwards the grace-
ful agreement of the external parts with them.14 For after studying
these, it will no longer seem so incredible that the soul, which has
within it so many origins and principles of change (arkhai),15 is
25 something that also belongs to this body, which is constructed so
incredibly and magnificently.
Now those who have made it their business to analyse such things
find the following, more than anything else, to be true and evident:
that every bodily, perceptible substance (ousia) is a compound of both
3,1 (i) some underlying thing, which we call matter, and (ii) a nature
that shapes and gives definition to this matter, to which we give the
name form.16 Anyone who looks at the products of various arts will
see clearly that this is how things are. Whichever artefact one
5 chooses, one finds something in it which is underlying matter and
something which is form: bronze, stone, wood, wax, or whatever is
shaped by art is an underlying subject, while that in it which is
produced by the artisan is form. For the statues matter is bronze or
stone (if the statue happens to be made from one of these), while its
form is a particular kind of shape and particular kind of contour,17
10 which is what the art produces and indeed is art. For anything
artfully produced in the underlying matter by someone who pos-
sesses the art is art.18 The body that is naturally suited to receive the
art, but which does not [already] have what is produced in it by art
as part of its own definition, is the arts matter.19
Just as each body constituted and formed by art exhibits this
duality, the same duality can be found even before that in natural
15 bodies, that is, bodies which have been formed naturally, since art
imitates nature, not nature art.20 In each natural body as well, then,
there is something that is underlying matter and something in [the
matter] that is form. Thus, in the case of bronze, either water or the
vaporous emission is matter (on the assumption that all metals that
can be melted have moisture as their matter);21 while its form is a
20 certain type of consistency and solidity.22 So just as art is the form
produced in something by art, nature is likewise the form produced
in things formed naturally.23
There is a difference between natural bodies, however. For some
of them are simple, while others are compound.24 The underlying
matter of compound bodies is also a natural body, composed out of
25 matter and form, since every natural body is a compound of these.25
Simple bodies, in contrast, do not have a further compound underly-
Translation 33
ing them, since in that case [a simple body] would be compound too.
But if what underlies them is not a compound, then it is not a body
either, given that every body is a compound of matter and form. What
underlies simple bodies and serves as their matter, then, is going to
be a kind of nature that is simple and separate from form, and which 4,1
by its own definition is amorphous, formless, and shapeless. Because
it is formless and said to be so, that which rids it of the aforemen-
tioned lack, by having come to be in it, is given the name form. The
former sort of nature is what one would call matter in the fundamen-
tal sense.26 For the underlying subject in compound bodies that is, 5
their matter is not matter simply as such,27 because it is conjoined
with a form and there is something further which in turn is its
matter. But in bodies where the underlying subject is simple, their
matter is matter in the fundamental sense and simply as such, which
cannot exist on its own28 (because anything that does [have matter
and exists on its own] is a compound body, as was stated earlier).29
As it is always found together with some form, it is separated from it 10
in thought alone, being related to the forms of simple bodies in the
way that wax is related to shapes.30 For any [piece of] wax one
conceives of is necessarily conjoined with a shape, since it is impossi-
ble to conceive of wax which is shapeless. Yet to be wax is not to be
conjoined with some particular shape, for if a determinate shape
were included in its essence, then [wax] that was not conjoined with 15
this shape, but some other one, would no longer be wax. Matter in the
fundamental sense is related to the forms of the simple natural
bodies in just the same way. For it cannot exist separate from any of
them, and yet to be [matter of this kind] does not consist in being
conjoined with any of them either. For then it would not be simple
and non-bodily, nor could it be the matter of some other simple body,
since it would perish with the loss of its own form.31 20
It is not only matter of this sort that cannot exist by itself just on
its own, but the form that comes to be in it too. In fact, this is true of
form even more than it is true of matter.32 For in the case of matter,
even though matter in the fundamental sense cannot exist on its own,
the matter in compound bodies, conceived as proximate [matter],33 is
not matter simply as such, but the matter of some specific thing (tode
ti),34 and it does exist on its own. The matter in artefacts is certainly 25
of this sort. In contrast, none of the forms that come to be in matter
is ever able to exist on its own, whether it belongs to a simple body or
to a compound one. Thus the forms produced by the arts in various
underlying matters are inseparable from them, even though these 5,1
[matters] are capable of existing separate from the [forms].35
Now, a form produced by art is not a substance (ousia),36 since art
is not [a substance] either; for this sort of form is also art, as was said
earlier.37 But a form produced naturally is a substance, because
nature is in fact a substance and this sort of form is a nature.38 In the
34 Translation
5 case of fire, which is a simple, natural body, heat and dryness, as well
as39 the lightness that emerges from them and above them,40 are its
form, while what underlies these is its matter, which by its nature*
is not any of these, but is receptive of them and their contraries
equally.41 (It is because of this nature that the transformations of the
simple bodies into one another occur.42) While neither of these [sc.
either its form or its matter] is a body, what is composed out of them
10 is by that very fact a body and fire, for it possesses a principle of
upwards motion namely, lightness from its nature and substance
(taken as form). This lightness, as fires form and nature, does not
itself undergo change. For how could lightness ever undergo change
itself just on its own, when it belongs to something else and is not just
on its own? For lightness is a power of the body that has it; but no
15 power is separable from what it empowers.43 For this reason, even
though it does not undergo change on its own, it is responsible44 in
the body which has it for the change that arises from this body, and
it is in virtue of this that the body which has it is said to change,
because45 undergoing change belongs to the body on account of and
in virtue of lightness, which is its power and nature.
Neither of the simple bodys basic components, then, is a body. For
20 the matter [of a simple body] is not a body either,46 since every body
is tangible47 and conjoined with a contrariety of some sort, whereas
the matter [of a simple body] lies beyond every tangible quality and
contrariety;48 and a body is existent, whereas matter needs form in
order to be.49 For the same reason, form is not a body either, because
6,1 it is incapable of being on its own, since it is inseparable from the
matter, as has been shown.50
Each of these, on the other hand, is a substance. For a natural form
is a substance, just as matter is,51 since the parts of a substance are
substances; or better still, it is because each of them is a substance
that what is composed out of both is also a substance and a single
5 nature of some kind.52 This is unlike artefacts, which are substances
in virtue of the underlying matter and qualities in virtue of their
forms.53
Those who are puzzled as to how it is possible for a body to come
from non-bodily items54 would not be puzzled if they knew them* and
so were to appreciate that anything that comes to be a specific thing55
comes to be from something that is not that thing, as hot comes from
not hot, but cold does not come from cold or white from something of
10 the same sort.56 For even if something else hot is responsible for the
particular thing that becomes hot becoming so, nevertheless the very
thing* that becomes hot becomes hot from [having been] not hot. For
something that is already [hot] does not become [hot]. Consequently,
even if someone were to accept bodys coming into being, he would
have to say that it comes to be from what is not body.57 But given that,
as we claim,
Translation 35

1. body does not come to be simply as such58


since there is always body, if the universe is in fact eternal and
incapable of either coming to be or perishing;59 and 15
2. the coming into being of some particular body, that is, of any
body, is from something which is not that particular body,
it follows that
3. the coming into being of some particular body is not from
what is not a body simply as such after all.
For just as we separate matter from form in thought and in account,
even though it is not actually separable for though it is separable
from this particular form, it is inseparable from form simply as such,
just as wax is inseparable from shape60 in the same way the coming 20
into being of body as body is something that likewise occurs only in
thought and in account.61
Although fire and each of the other [simple] bodies is in possession
of just what it is from both the underlying matter and the form in it,
[each of them] is just what it is in virtue of form.62 For fire is fire* in
virtue of form, and earth is earth in virtue of the same thing [viz.,
form]; and the same argument applies to water and air as well. For 25
each of them is just what it is said to be in virtue of form, since
1. each of them is a specific thing in virtue of its difference from
the rest
2. each has this difference from the rest in virtue of its own form
since the underlying matter in all of them is the same.63 Therefore,
3. each of these is just what it is in virtue of form.
And if
4. that in virtue of which each thing is just what it is, is its 30
culmination64
since the culmination of something that comes to be is the stage at
which it, having come to be, ceases coming to be, as it has already 7,1
come to be and developed completely, and it ceases coming to be
whenever it has come to be fully in its form,65 then it would make
sense to say
5. each bodys culmination is in its form.66
Yet if
1. each thing formed by art and, even prior to that, each thing 5
formed naturally is just what it is in virtue of its form
2. [the form] is the culmination of each,
36 Translation
then it is true, by conversion,67 that
3. that in virtue of which every compound of matter and form
has its being is its form and culmination.

Bodies act and are modified in virtue of


non-bodily items
Not only does what each thing is come from its form, as do its
10 differences from other beings, but the differences between bodies
with regard to acting and being modified also come about in virtue of
forms. For when bodies act or are modified in so far as they are
bodies, they act or are modified in a particular way due to the form,
since eah of these also is in virtue of the difference in forms. So it
makes sense to claim that bodies act and are modified in virtue of
non-bodily items.68
15 In natural bodies where the underlying subject is simple that is,
[where the underlying subject is] matter in the fundamental sense,
which underlies the simple, basic bodies that in turn are elements of
other bodies both the form and nature of these [simple bodies]
will be simple as well. Because of this, their nature will be respon-
sible for just a single simple change. For if nature is an origin and
principle of change, then a simple nature will be the origin and
principle of a simple motion, and [the motion] will be simple and
20 single in the fundamental sense. Hence, some of these will be
moved exclusively upwards or exclusively downwards in virtue of
their* nature.69
With other [bodies], in contrast, where the proximate subject is no
longer simple, but already a kind of body and compound for every
body is compound in this way70 the form in these cases will be more
complex and more advanced as well,71 as will their nature. This, too,
makes sense. For the form in the underlying matter also makes a
8,1 contribution to the form of [the compound body],72 and all the more
so if there is not just one body underlying it [but many], as is
necessarily the case with natural compound bodies. For the underly-
ing subject is not a single simple body in any of them, since the
natural transformation of a single simple body is into another simple
5 body.73 Consequently, if there is to be a compound natural body, over
and above the simple bodies, then it must have several simple bodies
as its underlying subjects, where the number is one corresponding to
the difference between the forms in them. This is the reason why
things of this sort are compound bodies. The [body], then, which has
many differing forms conjoined with matter as its underlying sub-
10 jects, has a nature and form that is of necessity more complex and
more advanced, since each nature in the bodies underlying it makes
a contribution to the form that stands over them all and is common
Translation 37
to them.74 For this sort of form is in a way a form of forms and a kind
of culmination of culminations.75
For this reason, we should also not be surprised by the difference
amongst forms in natural bodies, since we plainly find what is
responsible for their diversity in the underlying subjects. For it 15
makes sense that the number of forms in the bodies underlying them,
together with their distinctive mixture,76 should bear responsibility
for such great diversity.77 For even when there is just a single
underlying subject that does not contribute to any difference in
things composed out of it for the matter [in question] is of this sort78
the pairing of dryness or moisture with the heat or coldness in it is 20
responsible for a very great difference in the things composed out of
it, such that one pair of them comes to be fire, another air, another
earth, and another water, and one comes to be heavy and another
light, and others possess each of these in a derivative sense. Given all
this, how could it fail to make sense that bodies produced from a
certain sort of mixture and blend of all of them will differ enormously
from the others, in both their forms and their powers for change?79 25
Consider, then, how immensely trees and plants, which are all
natural compound bodies, differ from simple bodies in their powers 9,1
for change. The [simple bodies] have an origin and principle within
themselves for just a single, simple motion, while each [tree and
plant] has within itself the origin and principle of being nourished
and of motion in all directions by growing; in addition, each of them
also has a power of producing something like itself.80 Although plants 5
have all these features in common, they further differ thoroughly
from each other in form because they are different in matter and in
the mixture and blend of the matter underlying them.81 For trees are
the most distinct from each other, much more than even vegetables,
which are plants too, as well as natural, compound bodies. But among
[vegetables], too, we see yet again the greatest difference from one 10
another, and nothing else is responsible for this other than what was
just mentioned. This sort of form and this sort of culmination is
already the first power of the soul. For if every living thing is alive in
virtue of a soul,82 and plants are alive since to nourish oneself and
grow is to be alive then plants will have a soul and their form will
be a soul.83 For just as with simple bodies, the form corresponds not 15
to the activity that issues from the power, but to the power from
which the activity issues. For the nature of earth does not consist in
being moved downwards, since it is no less earth when it remains
stationary. Rather, [the nature of earth] is the power from which its
being moved downwards issues, namely, heaviness. Each [of these]
is a culmination too: both heaviness, since it is a form, and the
activity that occurs in virtue of it, since the activities that issue from 20
dispositions and powers are their culminations. In cases where both
occur, the culmination that corresponds to the disposition and power
38 Translation
is first and exists prior in time to the activity that occurs in virtue of
it.84 The same holds for compound bodies too. So if forms are not the
activities issuing from powers, but the powers instead, which, be-
25 cause they exist prior to the activities issuing from them, are first and
come before them, the first culmination of plants will also be a form
and a soul.
10,1 Now, once this has been agreed to and granted, I do not think
anyone will dispute that an animal is likewise a natural, compound
body, composed of soul and body. If, as a body, it gets its being from
them, and every body is a compound of matter and form,85 then
clearly one of them will be the animals matter and the other its form.
5 But from what has been said about plants, it is clear that the soul is
the form, although this was not intelligible from what had been
demonstrated at the beginning of our discussion. For it was granted
that that in virtue of which each thing has its being is its form and
culmination.86 What it is to be an animal belongs to an animal in
virtue of a particular kind of soul, namely, the one for perceiving,
10 since it is due to this that an animal is an animal and differs from
everything else that is not an animal.87 In an animal, therefore, the soul
is a form of the body, since the forms in all bodies are forms of the
underlying matter in them, where its soul, as a form, is more advanced
than the [soul] in plants to the same extent that the soul in them is
[more advanced] than the forms in the simple, primary bodies.88
15 The differences between souls for perceiving preserve the same
proportion again in relation to each other that the forms of simple
bodies have to the soul of plants and that the soul of plants in turn
has to the soul for perceiving.89 What is responsible for the propor-
tional difference between them is the difference in the bodies
underlying them, as regards their number and the particular kind of
blend, mixture, and formation.90 And that makes sense. For if the
20 beginning (arkh) is not just half, but the largest part of the whole,91
then it makes sense that a difference in the origins (kata tas arkhas),
even if small, can be responsible for a great difference. This is obvious
from the forks in roads too. For a small disparity between them is
often responsible for travel to places a great distance from each other.
But the proximate matter of each thing that comes to be is also an
origin. So it makes sense that the difference in forms that supervene
25 on differences in the proximate matter follows from them. For not all
matter can receive the same culmination.92
So in things at the ground level, which do not yet have a body
underlying them, the form is simpler too; whereas in things in which
a compound body with distinctive parts useful for different activities
11,1 underlies them, the form is conjoined with many powers, since it is a
form and culmination of a complex and organic body.93 For just this
reason, then, the soul is a form of this sort of body.94 The soul that
has a simpler organic body is simpler, while the soul comprising
Translation 39
many powers is more advanced and its underlying body, of which this
particular sort of soul is a form, is more complex and more organic. 5
If someone on first hearing thinks it incredible that the soul is this
sort of bodys form and nature,95 just as lightness is fires [form and
nature],96 but then looks at the distinction between the bodies that
have these forms and sees how unspeakably greater the construction
of animate bodies is than that of simple bodies, he would no longer 10
think those who speak in this way are saying anything incredible, since
it is not incredible, but makes the most sense of all, to say that the
proportion that the bodies underlying the forms have to one another is
the same as the one that their forms preserve to each other.97

The demonstration that the soul is a form


That the soul is a form can be shown as follows. It is agreed that
1. an animal98 is composed out of soul and body. 15
But
2. something is said to be composed out of things99 either
a. in the way that [something] is composed out of parts that
are preserved in the whole and the sort of things into which
it is also divided,
as when we say that a house is composed out of bricks;100 or
b. in the way that something blended is composed out of
things,
like honey water,101 for we say that this is [composed] out of water
and honey, but that honey water is no longer something that is
divided into these, due to the fact that neither of them is preserved
or still retains its nature after blending;102 or 20
c. in the way that [something] is composed out of matter and
form,
as we say that a statue is composed out of bronze and a shape.103 But
if being composed out of things is said in these ways, then
3. an animal should be composed out of soul and body in one of 12,1
these ways.
But
4. it cannot be in the first way (a),
since that would then be a juxtaposition of soul and body, so that the 5
body would no longer be animate throughout;104 and if so, then the
soul would contribute not to the character of the animal, but rather
to its size.105 But
40 Translation

5. it cannot be in the second way (b) either,


since it is not the case that the animal comes to be when the soul and
the body perish, as happens when things come to be from a blend of
things.106 The only option left, then, is to say that
6. an animal is composed of soul and body (c) as matter and
form.
One might also learn that the soul is a form of the body, and not a
substance itself just on its own (aut kath hautn),107 from its activi-
10 ties. For no soul activity can occur apart from a bodily change,108 just
as there could not be motion from natural tendencies either,109 since
the body engages in these activities in virtue of the power within it.
For being nourished, growing, and generating another like oneself
occur in virtue of a soul power, but the changes belong to the body.
Moreover, an animal perceives through the perceptual organs, which
15 are bodies, and desires themselves plainly occur when a certain body
undergoes change.110 For appetites, passions, and rage occur in ani-
mals in this way; in fear as well there is an alteration and contraction
of the body. Representation111 also occurs through a body, if at any
rate it depends on perception in activity, as will be shown.112 No one
20 would deny that in striving,113 too, what undergoes change is a body.
Even thinking (phronein), if it does not occur without representation,
would itself be something that occurs through a body too.114 If it is not
possible to conceive of any activity of soul apart from a bodily change,
then clearly [the soul] is something that belongs to the body and is
inseparable from it. For if it were separate it would be superfluous,
since it would not be able to engage in any of its own activities on its
own.115 It is evident, then, that [the soul] is the form of the body and
25 that the changes in virtue of it come about in virtue of the bodys
13,1 aptitudes.116 For on some occasions, when there are strong, external
factors responsible, which are capable of setting in motion anger,
fear, appetite, or some other affect like this, we are either not affected
in any way at all or only superficially and to a small extent; whereas
on other occasions when [external factors] are minor or incidental, we
5 overflow with affects, whenever the body is excited and primed and
predisposed towards the affect, either due to a shortage or abundance
or combination of certain humours. For we become angry more easily
when there is an excess of bile, and often we are afraid or upset in
circumstances because our body is thus disposed in a certain way.117

In something [is said] in how many ways?


One could show that the soul is a form of the body in the following
10 way as well. Everyone agrees that the soul is in the animals body.
But even though being in something is said in many ways,118 the
Translation 41
statement the soul is in a certain body cannot be meant in any other
way. For something can be said to be in something either
a. in the way that a species (eidos) is in a genus
as, for example, human is in animal; or
b. in the way that a genus is in [its] species
as animal is in winged and footed and aquatic, since the being of the
genus that is divided into species is in the species that come from the
genus and are contradistinguished from one another. Or 15
c. in the way that a part is in a whole
as a hand is in a body; or
d. in the way that the whole is in its parts,
since the whole body of a human being is said to be in the head, neck,
trunk, and other parts from which the whole body is made up. One
thing is also in another
e. in the way that something is in a vessel or in a place,
as we say, for example, that Dion is in the marketplace or wine is in 20
the jug. Further,
f. characteristics which belong to things extrinsically (ta
sumbebkota) are also said to be in them,
as white is said to be in the white body, and
g. things that are blended are said to be in the blend made
from them,
just as honey and wine are in the honeyed wine.119 But in addition to
all of the ways just mentioned, something is also said to be in
something
h. in the way that a form (eidos) is in matter.
For that in virtue of which something that is composed from others
itself has being is [in] the rest* in the way that a form is in matter. 14,1
This is how a particular kind of shape is in a statue, which is
composed from bronze and shape, and how heaviness is in earth,
which is composed from matter and [heaviness].120
Now, while in something is said in these ways since a pilot will
be in the ship in one of the ways just mentioned121 the soul is not in 5
the body (c) in the way that a part of [the body] is in it. For in that
case it would be a body, since part of a body is a body, and it would
contribute to the animals size, not its character.122 Moreover, not all
of the body would be animate, but each of them would be separated*
from the other, unless a body can pass and extend through a body
42 Translation
10 (something not at all easy to conceive).123 Nor is the soul (d) a kind of
whole, where the parts of the body are parts of the soul, so that the
soul would be [in them] in the way that the whole is in its parts.124
Nor is the soul in the body (a) in the way that a species is in a
genus, since the body in which the soul is present is not the genus of
the soul. For each of these is one in number, whereas genera and
species are common and universal; and further the soul would be a
15 body.125 But the soul is also not in the body (b) in the way that a genus
is in its species, for the reason mentioned above126 and because if the
soul were said to be the genus of the body, every body would be a soul,
just as every human is an animal.127
Nor is the body (e) a place for the soul either. For in that case the
soul would be a body, since anything that on its own is in a place is a
body,128 and further the body will be a place, since the body will thus
20 be either an empty extension or the boundary of the container (as
these are the views concerning place).129 Moreover, what is in a place
contributes nothing towards the places being a particular sort of
thing, but the animate body gets its differences with respect to other
bodies from the soul.130 Nor is [the body] like a vessel. For a vessel is
also a place for that which is in it, differing* from place only in that
it is carried along with the things that are in it.131
25 Nor is the soul in the body (f ) in the way that an extrinsic
15,1 characteristic is in an underlying subject. For the soul is a substance
and can receive contraries. But no extrinsic characteristic is a sub-
stance.132 Moreover, what is in some underlying subject is not
responsible for the being of what underlies it, since what underlies it
can also exist separately from it, whereas the organic body in which
5 the soul is present gets what it is to be organic from the soul.133
Nor again is the soul in the body (g) in the way that the ingredients
from which something is blended are in the blend. For none of the
things that are in something in this way still retains its nature, as
neither wine nor honey are preserved in honeyed wine;134 whereas
the soul and the body both persist in the animal. Moreover, every
blend is [a blend] of bodies, so that the soul in that case would be a
body too.135
10 Nor can the soul be in the body in the way that a pilot is in a ship.136
For if one conceived of the pilot (1) as the art of piloting, then the soul
would be in the body in the way that a disposition and form is in
matter, since dispositions are in the things of which they are dispo-
sitions in this way, since they are both non-bodily and inseparable
from the things which they are in.137 If, on the other hand, one
conceived [of the pilot] (2) as the pilot himself that possesses this
15 disposition, then the soul will be a body (since the pilot is endowed
with a body),138 and it will be in some circumscribed part of the body
in the way that [something is] in a place, and not all of the body will
be responsive to itself or have an accompanying awareness, since not
Translation 43
all of it will have soul.139 In addition, the soul in that case would move
the body by force and so, being this sort of thing once it had become
separated from the body, it would be able to enter into it again, with
the result that the same body would sometimes be animate and 20
sometimes not. If it were such a thing, what would be responsible for
the souls entering the body and how would it enter? What would be
responsible for its staying?140 And even before that, what would its
essence and nature be? If the soul is like a pilot, rather than the art
of piloting, it will also be composed itself from matter and some form.
[But then] this form would be the soul, since what it is to be a soul comes 25
from the same thing that what it is to be a pilot comes from. For each
thing has what it is to be just the [sort of] thing it is in virtue of its form,
and the pilot gets what it is to be a pilot from the form, not the matter.141
If the soul cannot be in the body in any of the ways just mentioned,
the only option left is that the [soul] is in the [body] in the way that
a form is, since each things form is that in virtue of which it is just
what it is, and it is in virtue of the soul that the animal is an
animal.142 Now, given that 16,1
1. the soul* is a form
as has been shown;143 and
2. [it is] an enmattered form144
since it is [the form] of a body, and a natural body at that for it is
not artificial, like the body of a statue is a natural body which is not
simple, like the body of fire, but compound and organic;145 and it has
further been shown that
3. the form is a culmination of that of which it is a form;146 5
and it is characteristic for Aristotle to say that
4. the culmination is a completion (entelekheia),147
since it is responsible for the fact that the object to which it belongs
is complete (en ti telei), it makes sense, then, that he offered this sort
of characterisation:
5. [the soul] is a first completion.148
For there are two kinds of completion, (a) one which is a disposition
and power and (b) another which is the activity issuing from that
power, and of these the power is first and the culmination corre- 10
sponding to it is the form.149 [The soul] is a completion, then, that is,
6. it is the first [completion] of a natural, organic body.150
For an organic body is one that has many different parts able to
subserve the soul powers.151 Hence, [Aristotle] says that the organic
body has life by its power,152 using the phrase has life by its power
44 Translation
for is able to actively live.* For something that already possesses an
15 advanced soul can do and undergo many things in virtue of it: [it can]
be nourished, grow, reproduce, perceive, desire, think, speak, act, be
healthy, be ill. For things are no less animate even when they are not
in activity, but still capable of it. The qualification as a power makes
clear that [the soul] is the first completion.153
This is in fact the essence (ousia) of the soul.154 For it comprehends
under a single characterisation many souls that are not the same in
form as each other and yet also stand in a certain order with respect
20 to one another, such that one of them is less advanced and first, while
the next is more advanced than it, because it has some other power
in addition to these; and then a third [type of soul], which in addition
17,1 to these powers has certain other powers yet again.155 When things
are related to each other in this way, their general characterisation
(koinos logos) cannot be perspicuous for the following reason. On the
one hand, their definition has to be attributed to all of them. On the
other hand, none of the more advanced ones among them can be fully
5 expressed in such a characterisation, since then the less advanced
ones would no longer be comprehended under the definition. Hence,
the general characterisation of the soul is simpler and more general
[than a definition], since it does not make reference to any of the
[powers] thought above all to be soul powers. For this sort of charac-
terisation is based on something that belongs to all these things in
common (koins).156

The soul is inseparable from the body whose soul it is


If the soul is a form, as has been shown,157 then necessarily
10 i. it is inseparable from the body whose soul it is
ii. it is non-bodily
iii. it is incapable of undergoing change on its own158
since every form is of this sort. For a body is a composite that exists
on its own,159 but a form is [a form] of something else since this is
the sort of thing a completion and a culmination are and cannot be
without that of which it is [the form], just as a limit cannot be apart
from that of which it is the limit. The soul, therefore, cannot be
15 separated and exist on its own.160 But then it is not a body either.161
A form, furthermore, cannot be a body along the lines of those who
claim, like the Stoics, that every body is either matter or composed
from matter.162 For form is not matter, since the latter is qualityless,
while the former is a kind of quality.163 But it is not composed from
matter either. For if the form were composed from matter and
form,164 then in the first place, what is different from the composite
will be the same as it;165 and in the second place, this will continue to
20 infinity, since this [second] form will be composed from matter and
Translation 45
form, and that form in turn will require matter and form again. For
if someone were to claim that a form is composed in this way from
matter and form, but not from some further form, because it has [its] 18,1
being* together with matter,166 then matter on their view would not
be qualityless, as its definition states, since it requires some form or
quality in order to be. But if they claim that matter is qualityless,
even though it is incapable of existing separate from a quality, on the
grounds that the quality is not included in its own nature, then by 5
the same argument the form and quality ought to be separate from
matter as well, even though their existence is conjoined with matter,
since the matter is not contained in their own nature.167 Further-
more, how can absurdity be avoided if someone claims that the
matter, when it receives the form and the quality, also receives some
matter in addition? For that is what they must say if they claim that
the form and the quality are an enmattered body.168 10
It is a mark of confusion to maintain that since
1. the parts of a body are necessarily bodies*
just as the parts of a surface, a line, or a time are surfaces, lines, and
times, respectively; and
2. the form and the matter are parts of an animal
therefore
3. [each is] also a body.169
For [form and matter] are not parts of the body in such a way that 15
the body can be carved into them. For the parts of the body into which
it is carved contribute to the bodys size and continue to exist after
they are carved.170 Form and matter are not parts of the body in this
way, but in the way that bronze and its contours are parts of the
statue. Dividing up the statue does not result in them, as it does into
a head, trunk, and legs. But the composite is composed from them as 20
parts, even though not in the same way.171 For the shape of the statue
is a part, though not in a way that contributes something to its size
it contributes to its character instead and not as something that
can persist in separation from the matter.172 For the division of the
body and other continuous objects mentioned above173 results in parts
that have a size and persist after the division. For this reason, [the 25
shape and the bronze] are not parts of body simply as such, but of the
particular sort of body of which they are a part, and they are that in
virtue of which [the body] is of that particular sort, even though they
are not a body [themselves].174
The following argument does not show anything either. It main-
tains that
1. anything whose part is a body is itself a body
46 Translation
2. the sense that is part of the soul is a body
so therefore
3. the soul is a body.175
19,1 For if one conceives of the sense as the perceptual organ, what is
conceived is a body, but not a part of the soul; whereas if one
conceives of it as the power for perceiving, then it is a part of the soul,
but will not be what is conceived as a body.176 For if the soul is a body,
and a body not in the way that matter is, it will be composed from
5 matter and form, given that every body other than matter* is such
on their view.177 But if so, then the soul will in fact be the form in that
body. For if what underlies [that body] is a body and so a specific
thing capable of existing on its own since it is either breath
(pneuma) or something else which is not animate then that which,
when added to this persisting body, makes it animate from some-
10 thing inanimate would be the soul on their view.178 For one cannot
claim that breath comes to be a soul by adding a quality to the breath.
For if adding the quality to breath produced an essential transforma-
tion in the breath, one could claim that breath is transformed into
soul. But if it remains breath after adding the quality, then the
15 breath would not be a soul nor would the quality added to it make it
a soul, since the breaths essence is not transformed; rather [the
quality] is just an extrinsic characteristic of breath. For no natural
body can convert from one nature to some other without having been
transformed essentially. The nature of soul and breath are differ-
ent.179 One surely cannot claim that breath is a genus of soul, since it
20 has existence on its own. For no genus is such as to exist on its own.180

The soul is not one of the simple bodies


Those who claim181 that the soul is a form of the body, but make it one
of the underlying bodies in the animal, such as fire or air or some
other body, fail to realise that they are making part of the matter a
form of the remaining matter. But this is absurd. For the nature of
25 form and matter is different, as has been shown,182 since one is an
underlying subject, while the other is in it; and one is shaped, while
the other is that in virtue of which it is shaped. As a consequence,
those who make such claims will have to say that fire or air is a
culmination of the other three bodies and that it comes into being
from their mixture, since that is the sort of thing a culmination is.183
But then it would no longer be a simple body, since whatever comes
30 into being from the mixture of several bodies is not simple, and there
would not be four simple bodies. One of the four bodies cannot come
to be from the mixture of the others either, given that it is necessary
for something constituted from a mixture to exhibit all the powers of
Translation 47
the ingredients from which it has been mixed. For mixture differs
from perishing in that mixed ingredients do not perish entirely. But
it is impossible to maintain that moisture and coldness are present 35
in fire or that dryness and coldness are in air.184 (Of the four [simple
bodies], moreover, which one specifically would be a form of the other 20,1
three, rather than another?)185 The only powers in the animal, fur-
thermore, would be those that belonged to the body which had been
produced from the blend of the others: if what had been produced was
fire, the powers would be heat and dryness, and if it was some other
[body], they would be its powers, since this is what had been produced
from them. If, on the other hand, the powers of the other bodies were 5
to persist in the animal no less than its powers,186 then it could not
have been produced from them and so would not be a form. But if not,
then it is not a soul either.187
If, furthermore, the soul is one of the [simple bodies], then either
not all of the animal will be animate or, alternatively, a body will
extend through a body.188 But what view could be more absurd than
that? For how can a body, which is filled with itself and does not have
any empty space inside it, receive another body into itself which is
similarly filled with itself?* For if something which is full of itself is 10
to receive something else, which again is likewise full of itself, then
nothing will prevent it from receiving another yet again and still
another: since it was no less full on its own than it is now having
received something else into itself, it will not be prevented from
receiving a second thing, and thus nothing will prevent the largest
body from coming to be in the smallest. Yet how is it not absurd to 15
say that a place that has been filled up by a certain body is able to
receive another body while staying the same, since the body that
filled it remains in it and has not changed place? We have, however,
spoken about the absurdity of this theory elsewhere at length.189
Those who speak in this way, moreover, do not preserve the
analogy between the form in an animal and the forms of simple 20
bodies and the forms of artefacts. For heaviness is not this sort of
form, nor is lightness or shape.190
If one were to say, moreover, that the soul, understood as a form,*
is one of the simple bodies, on the grounds that it is the form, and not
the composite, which informs the matter, then it should be the form
of this body namely, lightness, or the heat and dryness conjoined 25
with it, or a form of one of the simple bodies which is a soul on the
view of those who speak in this way.191
Those who say that the animals soul is a form, but a form in the
way a substance is, which is separable by itself just on its own as the
pilot of the ship is, on the grounds that the pilot is also a form and
culmination of the ship, do not speak correctly either. For the pilot is
not the ships form or its culmination, since [a ship] is still a ship even 21,1
without the pilot.192 Or rather* the pilot might be said to be a form of
48 Translation
the activity of the ship, since the activity, as [the activity] of a ship,
comes about in the ship in virtue of the particular shape which is its
form and the pilot contributes something to the activity. But the soul
is a form of the animal in so far as it is an animal, and it is not
animate only when it engages in activity. It is just because of this that
5 the soul is the first completion [of the body].193 That the soul is not
[the form] of an animal in the way that the pilot is a form of a ship is
clear from the fact that the ship persists, even when the pilot is gone,
whereas the animal is no longer an animal once the soul is gone.
Hence, the soul bears more of an analogy to the shape of the ship
instead. For the ship has its being in virtue of this, just as the animal
10 does in virtue of its soul.194 The pilot, moreover, is not present
throughout the entire ship, but animals entire body is animate.195 In
addition, if the soul were in the body in this way, it would also be
possible for it to re-enter the body again after having been separated.
For what will there be to keep the soul within the body, if it is this
sort of thing?196 The body, furthermore, will not be moved in virtue of
15 its own nature; hence it will be moved contrary to nature. But every
forced change contrary to nature in fact occurs due to meeting bodily
resistance, so that by this argument the soul would again be a body.197
In general, it would also make sense to ask those who say that the
soul is a form of the body in this way, just what joins and holds the
two of them together, given that they are separate from one another
and differ in their natures, so that what is composed from them is
20 and remains a single thing? For it is difficult to find something
responsible for the unity of such things, both when they first come
together and after they have come together.198

The soul itself just on its own is unchangeable


Since the soul is a form of the body of the sort discussed above,199 and
such a form is inseparable from the body, it will also perish along
with the body (as much of it, at any rate, as is a form of a perishable
body).200 But in that case the soul will also be incapable of undergoing
25 change on its own. For not everything that is responsible for a change
in something necessarily undergoes change itself.201
Anything which is detached and separated from the things it
changes and which changes them through bodily contact necessarily
undergoes change itself when it effects change. For things that push,
22,1 pull, spin, or carry something are in fact responsible for the change
in the things changed by them because they themselves undergo
change in this way.202
In contrast, where the powers and dispositions of what undergoes
change are responsible for change of some sort in the things that
possess them, not only do they not need to undergo change them-
selves in order to produce change, they cannot. For it is impossible
Translation 49
for something to undergo change on its own if it is not separable on 5
its own. But dispositions, powers, and the forms of bodies quite
generally (which, it has been shown, include the soul) are all like
that.203 For heaviness is responsible for earths downward motion and
can produce change in this direction in it without undergoing change
on its own. For how could heaviness be changed on its own, if it is a
form and a nature of the body which posseses it? In the same way,
the soul of animals has responsibility for all the changes that belong 10
to [the animal] as an animal, since it is in virtue of the soul that an
animal has the power to be changed in this way. It changes the body
in this way without undergoing change itself just on its own.204
That having been said, the soul does undergo change jointly with
the body undergoing change and is changed extrinsically, something
that necessarily holds for any form that resides in and accompanies
the body of which it is the form, because it is inseparable from it.205 15
In fact, whenever the form is properly responsible for the bodys
change, although the form primarily changes the body on that occa-
sion, it also changes itself extrinsically, since it undergoes jointly
with the body the change it produces in it. For heaviness changes
itself extrinsically,* whenever it is responsible for the downward
motion of the body that has it. The soul too, then, will change itself 20
extrinsically whenever the animal that possesses it is changed in
virtue of it. For although it will also undergo change extrinsically
when the animal changes place due to some other cause, it would be
changed by itself only in the case where the change comes about in
virtue of itself.206
Further, if [others] say that the soul undergoes change precisely
because it produces change,207 then it should also be said to be at rest 25
whenever it is responsible for the bodys rest. For what kind of
arbitrariness would it be if the soul produces change by undergoing
change itself, yet does not also bring the body to rest by being at rest?
But if it did, then change would no longer belong in the essence of the
soul.208 For if* (i) its proper function is exhibited more during rest 23,1
than in change, then (ii) being at rest will more properly and more
naturally belong to the soul. But the first, (i), is in fact the case, since
the most proper function of the soul, understanding, occurs more
through the souls being at rest than by its undergoing change.209 For
this reason, understanding occurs more in old age than in youth, and
more in the sober than in the drunk, and more when the body is at 5
rest than in motion.210 Being at rest, therefore, naturally belongs to
[the soul] more than undergoing change does.
Accordingly, it is more truthful to say, not that the soul undergoes
change or that it is at rest in its own right, but that the animal does
each of these things in virtue of the soul. For just as we do not say
that the soul walks or sees or hears, but that the human being does
in virtue of the soul,211 so too with all the other activities and changes 10
50 Translation
in which one engages as something animate and as a human, it is not
the soul which engages in activity and undergoes change,212 even
though we are often tempted to [say this] because the body that
undergoes change in activities of this sort is not as discernible to us
as it is [in other cases].213 But even in these cases, an animal and a
human engage in activity in virtue of the soul, in virtue of which [a
15 human] possesses what it is to be human. For a human feels pleasure
and feels pain, becomes angry and is afraid, loves and hates, learns
and understands, recollects and remembers in virtue of the soul,
since [a human] is able to do these things because it has this kind of
form and this kind of culmination.214 For just as a wrestler wrestles
in virtue of the disposition for wrestling, but the disposition for
20 wrestling does not itself wrestle; and the oboist plays oboe in virtue
of the disposition for playing the oboe, but the disposition for playing
oboe does not play the oboe;215 and the weaver weaves but the
disposition for weaving does not weave; so one should suppose that
the same holds likewise for those activities which things possessing
a soul engage in in virtue of being animate.216 For in none of these
cases does the soul in its own right engage in a vital activity; rather
that which possesses the soul does so in virtue of it.
25 The view that these particular activities belong to the soul because
it uses the body as an organ is simply not true. For this case is like
that of other powers and dispositions. No power or disposition en-
gages in activity by using that of which it is a disposition. Rather, it
is the other way around: the things that possess the powers and
dispositions engage in activity in virtue of these powers and disposi-
tions.217 For heaviness is not carried downwards by making use of
30 earth, of which it is a power. Rather, earth is carried downwards in
24,1 virtue of heaviness, which is its power and form, its culmination and
completion. The same holds for the soul as well, since it is likewise a
power and form and completion of the body that has it, as it comes
into being from a certain mixture and blend of the primary bodies, as
has been shown.218
5 What engages in activity related to the soul is that in which the
governing part of the soul is found,219 since that is what is initially
animate in its own right;220 and it uses the organic parts of the body
for the activities related to the soul.221 For the organs animals use for
activities related to the soul are congenital parts of the body, not
10 separate parts as in the case of the arts.222 For sinews, hands, feet,
and perceptual organs are body parts of this kind. A sense organ, for
example, is a part of the organic body for awareness of a perceptible.
The soul is an origin and principle for animate things and some-
thing that is responsible in so far as it is productive of the vital
changes in them. But it is also responsible in the way that a form is,
as has been shown,223 and in addition as a that for which, that is, in
15 the way that an end is [responsible], given that for things that have
Translation 51
a soul [the soul] is an end and a culmination, while the things that
lead to the end are for the end.224 For everything produced naturally
is produced for something, namely, the end in it.225

The soul is not a harmony


One should not assume that people who claim that
1. the soul is a form that supervenes on a particular sort of
mixture and blend of the bodies underlying it 20
[also] claim that
2. the soul is a harmony.226
For suppose that the soul cannot be separate from this sort of blend
and mixture; it does not thereby follow that it is the same as the soul.
For the soul is not a particular kind of blend of bodies which is what
a harmony is but a power that emerges above a particular kind of
blend,227 analogous to the powers of medicinal drugs, which are
assembled from a blend of many [ingredients]. For in their case too,
the mixture, composition, and proportion of drugs such that one of 25
them, it might turn out, is 2:1, another 1:2, and another 3:2 bear
some analogy to a harmony. The power, however, which emerges
from the blend of drugs exhibiting this harmony and proportion is not
likewise a harmony too. For while the harmony is the proportion and
composition of the mixed ingredients, the ointments power is not the 25,1
proportion by which the ingredients are mixed.228
The soul is also of this sort. For the soul is the power and form that
supervenes229 on the blend of bodies in a particular proportion, not
the proportion or composition of the blend.230 For it would make more
sense for someone to say that health is a harmony than to say that 5
the soul is, since the former comes closer to a harmony than the soul
does. For health is a balance of various things, where this balance is
just a composition and mixture of certain things in a certain propor-
tion.231 The soul, in contrast, is not a balance, but the power [that
supervenes] on the balance: it cannot be without this balance, but is
not [the same as] it.232
Harmony is said in two ways.233 It is said in one way of the
composition of bodies. For when [bodies] are composed so as not to 10
admit anything similar between them, they are said to be in har-
mony and a composition of this kind is said to be a harmony. It is
said in another way of a proportion of mixed ingredients, such as when
something has been mixed in a proportion of 2:1 or 4:3. For it seems that
something which has been mixed in this way where, say, one ingredi-
ent in the mixture is double, the other half has been mixed 15
harmoniously and that this particular kind of mixture is a harmony.
With the first kind of harmony, the harmony is not the same as
52 Translation
what is produced from the things brought into harmony. For when
the wood in a bench is put together harmoniously it does not follow
that the bench is therefore a harmony; and when stones are arranged
harmoniously it does not follow that the temple that has been pro-
duced from their harmony is thereby a harmony as well. With this
20 kind of harmony, the harmony is one thing and what has been
produced from things composed harmoniously is another. The same
also holds for harmony when taken as a proportion. For supposing
that certain things have been mixed in a certain proportion and
harmoniously, it does not thereby follow that what has been produced
from this sort of mixture is [itself] a harmony and a proportion.234
26,1 A citharas being in harmony, moreover, is not the same as its
making a certain sort of sound. For it is no less in harmony when it
makes no sound, whereas the sound that is produced out of and from
this sort of harmony is something other than the harmony from
which it is produced.235
Someone might deny that a proportion is a harmony whenever
5 things are mixed in a certain proportion, since the proportion of 2:1
is not a harmony when it occurs in wine and honey. For there is
harmony in a specific kind of composition of melodies and rhythms,
not in just any random thing or in any random proportion.236 For
every harmony is constituted by a determinate proportion. But the
mixture of hot, cold, dry, and moist things that is, the mixture of
elements that the soul is a form of is not constituted by a determi-
10 nate proportion. For the same soul persists even when the powers in
a mixture are strengthened or weakened to a certain extent. For if
someone were to say that in general the mixture of just any thing in
some proportion is a harmony, this would commit one to saying that
everything composed is in harmony, since some proportion is to be
found in every composed thing.237
In contrast, the soul would be a harmony, or a harmonious compo-
sition of certain bodies, rather for those who make the soul out of a
15 specific kind of mixture and composition of things.238 This group
includes the Stoics, who claim that the soul is breath composed in a
certain way from fire and air.239 It also includes the Epicureans, since
on their view the soul is composed from many different bodies.240
Even on Platos view, the substance of the soul is constituted from a
20 composition of things and composed in a certain proportion, as he
says in the Timaeus.241 As I was just saying, the soul would be a
harmony for someone who speaks like this rather than for someone
who claims that the soul is a disposition, power, and form that
supervenes on a certain kind of blend and mixture of simple bodies.242
For on the view of those who think that the soul is in a way the very
things that are composed, the composite gets what it is to be a soul
25 from a certain kind of composition; so if this composition is a har-
mony, then the composite would get what it is to be a soul from the
Translation 53
harmony. But for someone who thinks that the soul is not the things
composed simply as such, but instead a power that emerges above a
certain sort of blend and mixture of the primary bodies, the blend of
matter will have a proportion, but the soul will have its being not in
virtue of the harmony and blend (as they are committed to saying),
but in virtue of the power that emerges above it.243 30

The parts of the soul: how many and which ones?


We have made clear what the essence of the soul is and what follows 27,1
from it, to the extent it was possible to offer a general overview of it
in a short space. The next thing to speak about are its parts, how
many there are and which ones, after we have shown that there are
several parts, though a finite number [of them].244
To show that the soul has several powers and not just one power 5
which seems to be several as Democritus and various others
thought,245 because of the changes and activities [it undergoes] at
different times, in relation to different things, and by different means
it is sufficient to point to the conflict of powers with one another,
which occurs both in people who control themselves and in those who
do not.246
But it could be shown most clearly by the fact that (1) nature does
not seem to do anything superfluous247 and (2) there are things* that
have a power of soul, but cannot engage in other activities beyond it, 10
even when there are objects present which the activities correspond-
ing to these powers concern.248 For if someone were to claim that
these activities do not occur in them [solely] due to a lack of organs,
he will undermine the claim that nature does nothing superfluous.
For the souls being capable of engaging in many activities will be
superfluous in them, since it will be prevented by nature itself from 15
engaging in these activities due to a lack of organs.249
How could it not be absurd, moreover, to claim that a human does
not differ from frogs or any other random animal with respect to the
power of the soul, but with respect to some organic body? This view
will not differ from the belief in metempsychosis that some maintain,
if the difference between animals and between their activities does 20
not come from the powers of the soul, but rather from organs. For
even though the same soul will be in all animate things, they will engage 28,1
in distinctive activities in virtue of the differences in their bodies, which
is just what [the proponents of metempsychosis] maintain.250

That there are not infinitely many powers of the soul


It is evident from what has been said that there are several powers
of the soul. But it is also evident that there are not infinitely many
powers,251 from the fact that 5
54 Translation

1. nature does nothing superfluous


2. an infinite number of powers would be* superfluous
since an animal cannot engage in activity corresponding to infinitely
many powers.252
Furthermore, since the parts of the body that has a soul are
limited, it makes sense that the parts of the soul in it are limited
too.253
Furthermore,
1. nature does everything for something254
2. in cases where one thing is first and another second, the first
is always directed towards the second.
10 But
3. the soul is a nature and by nature there is a certain order
among the powers in it.
Therefore,
4. there is a power that is highest among them, and the ones
that precede it are for its sake.
For if the sequence of the souls powers went on infinitely, it would
undermine the principle that nature does everything for some-
thing.255

***

Since the powers of the soul are finite in number, we must investigate
how many there are and which ones. Consider the following argu-
15 ment. As has already been stated,256
1. in cases where one thing is first and another second, where
one is less advanced and another more advanced, it is not
possible to formulate the essence accurately and in detail with
a general characterisation.
For while the nature of the object is most fully expressed in the more
advanced cases, the general characterisation cannot make them
20 explicit, since such a characterisation would no longer fit the less
advanced cases and so would no longer be general.257 But
2. the soul is this sort of thing.
For among the powers of the soul, some are first and simpler, and so
because of this are also less advanced, while others come after them,
and still others yet again above them, so that all of them are related
to each other in such a way that the earlier powers can be separated
25 from the subsequent ones, but the subsequent ones cannot occur
Translation 55
without the earlier ones.258 (As I said earlier, our account concerns
the soul of things that are subject to coming-to-be and perishing. For
the definition stated above is in fact [a definition] of this [type of] soul.
For the soul of the gods if this even ought to be called a soul is
said to be a soul homonymously with this.)259 But then
3. it is necessary for anyone speaking about the soul to give an
account of each of its powers individually.260 29,1
The first [power] of the soul, then, in animate things which are
subject to coming-to-be and perishing is the [power] for nourishing
[oneself], to which the [powers] for growing and for reproducing are
both linked.261 For the animate is separated from the inanimate by
this first power. Thus we attribute being alive, which is due to souls 5
presence, to things that have at least this power; and being alive
belongs to all animate things, not only plants, on account of the power
for nourishing [oneself], just as perceivers perceive on account of the
power for perceiving. For of necessity when living things lose the
former power, they die, whereas everything that has a share in it is
alive. Even living things that possess additional powers are alive
because of the first powers presence. It is for this reason that we 10
define being alive in terms of nourishing oneself and growing.262
The [power] for perceiving and striving, as well as the [power] for
desiring and representing constitute the second [type of] soul. These
[powers] cannot themselves occur in anything apart from the [power]
for nourishing [oneself], despite the fact that the [power] for nourish-
ing [oneself] also exists apart from them (since the [nutritive power]
in plants is like that).263 Things that have the [power] for perceiving
that is, animals, since an animal is defined by this power264 15
necessarily have the [power] for nourishing [oneself] too. But the
[power] for perceiving is linked to whatever the [power] for desiring
and representing is linked to; conversely, the [power] for desiring and
representing is also linked to the [power] for perceiving.265 And both
the [power] for representing and the [power] for perceiving are linked
to whatever the [power] for striving and changing place is linked to,
since anything moved spatially in place in virtue of the soul present 20
in it is, of necessity, capable of representing, perceiving, and nourish-
ing [itself] as well. It is not the case, on the other hand, that
everything capable of representing and perceiving is thereby capable
of moving spatially as well.266
The last type of soul power is the one generally called rational, yet
contains within it the powers for deliberating, believing, knowing,
and understanding.267 These are the souls most advanced powers, 25
both because they possess something of superior value in themselves
and because a soul that has these specific powers also has all the 30,1
other powers of soul (although not everything that has those powers
also has a share of these*).268 For this reason, even though there are
56 Translation
many soul powers in [living] things in which the rational power is
present, the soul constituted from them all is one, because no sub-
sequent power can occur without the power that comes before it.
5 Rather, they all belong to it as parts, where successive [powers] are
joined to the preceding ones and the preceding ones are because of
this expanded and developed.269
In virtue of the difference in souls and soul powers just mentioned,
there is also a quite general division of animate things, in virtue of
which we say that some things that have a soul are plants and others
10 animals, and some animals are rational and others non-rational.
Those which have all of the powers just mentioned are called ra-
tional and advanced in reference to the highest power in them.270
Humans are the only animal of this sort. All those, on the other
hand, that do not have a share in reason, but have perception, are
non-rational animals, both those which have the power of chang-
ing place in addition to perception and those that have perception
apart from this. They are referred to in this way [viz., as non-ra-
15 tional] due to an absence and lack of the highest power. Plants are
[animate things] governed solely by the power for nourishing [one-
self] and reproducing.271
Since, then, the soul is like this and for this reason cannot be
revealed through a general characterisation intelligibly, it is necessary
for those who speak about the soul to develop their account of each of its
20 powers individually. For this is how the kind of essence and nature it
has will be discerned.272 Our account of them will continue only until the
nature of each [soul] and the differences between [their] powers have
been indicated, since the puzzles and queries that can be raised about
each of them would be inappropriate to the present treatise, as they
25 would require much more leisure to pursue. For the endless review of
claims and difficulties which arise for each power would get in the way
of setting out the nature of the soul intelligibly.273
Though we speak of various parts of the soul, it does not follow that
we thereby make the soul into something extended. For not every-
thing which has parts is necessarily an extended thing. For number
is not extended either, even though it is composed from parts. But in
addition to not being divisible into parts in the way an extended thing
31,1 is, the soul is also not divisible into parts in the way a number is. For
we do not divide the soul as though it were composed from the parts
into which we divide it as separate things.274 Rather, we divide the
soul by ennumerating the powers it has and by ascertaining the
differences between them, just as if one were to divide an apple into
5 its fragrance, lustre, shape, and flavour. For dividing an apple in this
way is not like dividing a body, even though the apple is certainly a
body, nor is it like dividing a number.275

***
Translation 57
Now, given that
1. life belongs to each living thing through the presence of a
soul276
and
2. plants are alive
it should follow that
3. plants have a soul
and
4. their soul is the power in virtue of which they are alive.
But
5. plants are alive in so far as they are nourished and grow.277
Therefore,
6. their soul and form is the power in them for nourishing 10
[themselves] and growing.278
For every compound is said both to be and to engage in activity in
virtue of each [component] in the compound: it both is what it is and
engages in its own activities, first, in virtue of its disposition and
form, and second, in virtue of the matter in it that underlies it.279 The
second [component] in virtue of which it is a given thing is what has
received [something] within itself;* while the first is what it has 15
received. For example, the healthy animal, which is composed from
health and a body, is said to be healthy due to health and also due
to the body, where health is the disposition and form the body has
received, and the body, in so far as it is healthy in a secondary
sense, is the underlying matter that has received health.280 If, 20
then, in the case of being alive, living things are alive due to the
soul first and the body second, since they are composed from a soul
and a body and they are said to be alive due to both, that is, due
both to the soul and to the body, then it can also be shown from
this that the soul is a form of the body and what this particular
kind of body has received, while the body is the souls matter and
what has received [it].281
The soul for nourishing [oneself] is therefore a form of the body 25
that has it. It is clear that this form is the soul, and not simply a
nature, from the fact that none of the other things are alive that are
held to have a nature, but not a soul. For none of the simple bodies is
said to be alive, since something is said to be alive in virtue of a souls
presence. Certainly everything that has a soul is alive, and nothing 32,1
is alive which does not have a soul.282
Furthermore, if
58 Translation

1. the powers in virtue of which seeds are active are similar in


kind to those which belong to the things the seeds are produced
from,
then
2. the powers in virtue of which the seeds of animate things are
active will themselves be soul powers.283
But
5 3. all the seeds produced from animate things are first active in
virtue of the power for nourishing [oneself].284
Therefore,
4. the power for nourishing [oneself] is a soul power.
Given, then, that the power for nourishing [oneself] is both a soul and
first among the souls powers, we should speak about it first and show
just what its essence and nature is.285 The account of the soul for
reproducing is linked to the account of the soul for nourishing [one-
self].286 For just as being nourished and growing are activities that
10 belong properly to the soul for nourishing [oneself], so too does
reproducing something similar to what sows seed. For making an-
other thing like oneself is a function that belongs to living things
which are not maimed (as eunuchs are) or generated spontaneously,
but fully developed, so that they all can share, in the way that they
can, in existing forever and in a likeness to the divine. For everything
15 formed naturally longs for this and does everything natural for its
sake.287 But
1. the soul is what is responsible for something animate doing
what is in accordance with its own nature
and
2. generating something like itself is one of the things an
animate thing naturally brings about,
since plants and animals that have not been maimed are able to
produce things like themselves in virtue of their own nature. There-
fore,
20 3. the power for reproducing is also a soul power.
But
4. plants, which only have the soul for nourishing [oneself], are
able to reproduce in addition to being nourished and growing.
Therefore,
Translation 59
5. the power to reproduce is part of the power and soul for
nourishing [oneself] too.288
The account for both, then, will be a shared one.
Since in all subjects one comes to know what is less clear by means
of things which are more clear,289 one ought to approach the present
case in this way as well. In particular, since the activities correspond- 25
ing to the souls powers are clearer than the powers from which they
arise into activity perceiving, for example, is clearer than percep-
tion and being nourished clearer than the power for nourishing
[oneself]290 we should begin with being nourished. For in addition
to being more intelligible, being nourished is also prior with respect
to the definition of the power from which it arises. For those things,
which [other] things include in their own definition, are prior in
definition to those which are not contained in their definition.291 30
Being nourished is related in this way to the power for nourishing 33,1
[oneself]. For being nourished takes place in what is nourished due
to the power for nourishing [oneself], whereas the power for nourish-
ing [oneself] does not require its activity in order to be. For it is
impossible for activities arising from dispositions to be without the
dispositions, whereas the dispositions can also be found without the 5
activities.292
But in fact anyone who speaks about being nourished ought to
speak about nourishment first. For being nourished cannot be made
intelligible if nourishment that by which something nourished is
nourished remains unknown. The same order of inquiry will hold for
the souls other powers too. For in their case as well the activities are
more intelligible than the dispositions. But the account of the objects
that the activities are directed towards is linked to the discussion of the 10
activities.293 Anyone, therefore, who intends to show what the soul for
nourishing [oneself] is ought first to speak about nourishment.

There are two kinds of nourishment


Nourishment seems to be of two kinds, since (i) what we ingest from
outside, though still uncooked, is called nourishment* as well as (ii)
what has already been digested and incorporated.294 Because of this 15
duality in nourishment, it is sometimes said that contrary is nour-
ished by contrary, while at other times that like is nourished by
like.295 For nourishment that is still uncooked and undigested (that
is, the first kind of nourishment) is contrary to what is nourished,296
since a transformation has to take place in it for it to be assimilated
to what is nourished for cooking is just this sort of thing and a
transformation is in every case [a transformation] into a contrary.
For there is no transformation of something white, in so far as it is 20
white, into anything except black or one of the intermediate colours,
60 Translation
which likewise comes about in virtue of a transformation into a
contrary; nor is there a [transformation] of something hot into any-
thing except cold or, conversely, of cold into [anything except] hot.
What is sweet, in so far as it is sweet, is not transformed into white
25 or black, or cold or hot, but into a contrary flavour.297 So, too, in the
case of nourishment that is still uncooked: if it is transformed by
cooking and comes to be like what is nourished, then prior to this
cooking and transformation it must be contrary to it. For this reason,
all nourishment of this kind must be capable of being cooked and of
being transformed into what is nourished by it.298
34,1 In this way, then, a contrary is nourishment for a contrary. But
not every [contrary]. For it is not the case that everything which is
transformed into its contrary is thereby capable of nourishing it.
What is high-pitched, for example, is transformed into something
low-pitched and vice versa; what is shapeless into something shaped;
and what is white into something black. But it is not the case that
they are thereby nourished because of this and made to grow by* the
5 things which have undergone a transformation to become the same.*
The contraries that serve as nourishment, rather, are those which,
by transforming into their contraries, can cause the contraries into
which their transformation has resulted to grow. But bodies are
capable of causing growth. Nourishment, therefore, must be a body
which is contrary to and distinct from what is nourished.299 But even
though nourishment must be the sort of thing which can enlarge
[something], it does not follow from this that anything which en-
larges something enlarges it in the way that nourishment does. For
10 in fact the simple bodies, which we call elements, are enlarged by
those [bodies] which are transformed into them, but not in the way
that [something is enlarged] by nourishment. For water is said to be
nourishment for fire [only] in a broader sense, since the only things
that are nourished in the fundamental sense are those which have a
soul for nourishing [oneself], since that is what nourishing belongs
to. Therefore, nothing inanimate is nourished, even if it grows.300
So nourishment is a contrary, since the first [kind of] nourishment
15 is so. But in so far as what has already been incorporated also counts
as nourishment, and like is incorporated into like, what is alike is
nourishing too. For nourishment, as they say, is the part of nourish-
ment that nourishes,301 and what has been incorporated is
nourishing, and what has already become like what is nourished is
of this kind.302
What is nourished is something that has a soul for nourishing
[itself]; nothing else is nourished.303 This is clear from the following.
20 Although it can be said that what nourishes and what is nourished
each becomes larger from the compounding of likes since one could
not become larger instead of [the other]* in the compound, as each is
alike it cannot be said that what has been transformed from
Translation 61
nourishment has [itself] been nourished and grown by means of
nourishment. Rather, that to which it has been added [has been
nourished and has grown], because the latter possesses a soul that
nourishes and remains untransformed, while the nourishment, 25
which has been transformed by its power, is incorporated into the
animate thing that persists.304

On the activity of being nourished


This is the sort of thing nourishment is, then. The activity of being
nourished, in contrast, comes about when the soul for nourishing
[oneself] transforms nourishment ingested from outside by cooking 35,1
[it], thereby assimilating and incorporating it into the body whose
form and power it is.305

Heat is the souls tool for nourishing [oneself]


In activities and transformations of this sort, heat is [the souls]
tool.306 Because of this, some even thought that this is what nour-
ishes, transferring the power of the source and principle that is acting
to that by means of which it acts, since they saw that nothing can be 5
nourished apart from it*.307 Now while heat is also capable of produc-
ing change, there is no determinate or orderly change that primarily
has heat as its source and principle. When something is being nour-
ished, on the other hand, it develops in a certain sequence and order
and has a limit, which are characteristics of soul, not fire.308

***

Things that are nourished also grow by means of what they are
nourished by, as has already been said.309 To grow and to be nour-
ished are not the same, however: they do not occur at the same times, 10
nor does each of them have the same function. For an animal is
always being nourished for as long as it exists it is for just this
reason the most continuous of soul activities.310 But things which are
nourished are not always growing, since they are nourished as long
as they exist, but some also shrink as they age. The function of what
nourishes is to preserve what is nourished, but [the function] of what
causes growth is to add to the magnitude of what is nourished. So 15
among powers there is accordingly one that is able to preserve the
being and essence [of a living thing] since the power for nourishing
[oneself] is of this kind while the [power] for growing is primarily
able to produce an increase in size.311
After the activities just mentioned, of growing and being nour-
ished, next comes reproducing and leaving something behind like
oneself, since it is not the case that whatever has each of the other
62 Translation
20 [powers] has it too, but rather all those which are not maimed or
undeveloped, as was stated earlier.312 Of the powers that belong to
the soul for nourishing [oneself], this is the last to develop. For the
activity corresponding to it, since it naturally belongs to [living
things] that are already developed, is developed at that stage too.313
For to reproduce something like itself, as was already said before, is
each natural things end.314
These, then, are the activities of the soul for vegetation* or first
25 soul:315 nourishment, growth, and reproduction. But then the powers
from which these activities arise should also be apparent.316 The
power for nourishing [oneself], which is souls first power,317 can
preserve what possesses it through its own activity, which occurs in
36,1 response to the presence of nourishment.318 The souls power for
causing growth can cause the body that possesses it to grow by means
of nourishment through its own activity. Finally, the souls power for
reproducing can produce something similar to that which possesses
it.319 It uses nourishment itself in a way too, since the seed by means
of which reproduction occurs is the end product of the final stage of
nourishment, as has been shown elsewhere,320 since [the power for
5 reproducing] is a more developed [power] than those which precede
it. For the activity of the power for reproducing does not contribute
to [the living thing] itself, its* preservation, or its development.
Rather, in [living] things which are already developed it is responsi-
ble for the generation of another thing like [themselves],321 through
a longing for a kind of reinvigorated immortality. Through this
power, nature provides what is mortal with a share in immortality,
in the way that is possible for it.322
10 There are three things involved in being nourished: that which
nourishes, that which is nourished, and that with which it is nour-
ished. The soul for nourishing [oneself], or first soul, is that which
nourishes; that which is nourished is the body whose form is the
power just mentioned; and that with which it is nourished is the
nourishment.323 The same account holds for growing and reproducing
as well.324
Now, given that
1. it is proper to name all things after the end [telos] and the
most advanced [teleiotatou] part of their makeup,325
15 thus, for example, we say that a human is rational, even though [a
human] is also able to perceive; and
2. the first souls end, as has been shown, is to produce some-
thing similar to oneself326
it would make sense, then, that
3. the first soul is called [the soul] for reproducing,
Translation 63
since it produces something similar to what possesses it, even though
in addition to this it is most of all [a soul] for nourishing [oneself] and
for growing.
The soul and power for nourishing [oneself] is responsible for the
initial formation of the animals body as well as for its being, increase, 20
and growth.327 For its function is to enable whatever it happens to be
in to nourish itself and to grow. Since this soul is [already] present in
the seed when it is sown,328 once it receives matter suitable for it,329
it forms the [matter] in such a way that what is formed from it comes 25
to be like what emitted it through nourishment and growth and so
becomes responsible for nourishing itself and growing when nourish-
ment is present.330 What is inside the womb engages in activity on its
own solely in virtue of this soul power, since even though it possesses
the sources and principles for the other powers and suitable condi- 37,1
tions for the dispositions that its parent likewise possessed, it does
not yet have these in activity, since in a way it does not yet possess
the parts through which the activities of those powers [are exer-
cised].331

Why parts detached from plants live and develop,


while those from animals no longer do
Although some parts detached from plants can live and develop, 5
those detached from animals no longer can.332 We should regard the
following as responsible for this.
Plants only have the soul for nourishing [oneself], which is present
throughout the entire body. Because of this, plants are simpler and
do not need as many organs. Any of their parts that can keep the
power for nourishing [oneself] in it after being detached by means of
abundant nourishment is preserved and becomes like the whole
[plant]. For these parts engage in activity in virtue of the same power 10
that enables things generated from a seed to engage in activity, since
this is also what their culmination consists in.333
In the case of animals, though, [the powers] are not distributed
throughout the entire [body] in this way and so not in the detached
part either. For the soul of animals is not homogeneous in the way
that the soul of plants is, nor is there the power of some of these
[souls] present in them as it is in seeds. This is because, first, these 15
parts cannot receive some of these [souls] themselves,334 and second,
they cannot draw nourishment in just any way, but only through
certain specific organs, which cannot be added to things that have not
been formed from a seed principle, as roots can to plants.335 For roots
are formed easily in the presence of nourishment and grow without 38,1
needing much management. But no animal can be nourished without
the specific organs for nourishing [itself], and it cannot form them
without the seed that possesses their powers.336
64 Translation
5 For this reason, even though animals are nourished in the womb,
it is still as parts of the mother bearing them.* For while they are
nourished in virtue of the power in them, they receive nourishment
in so far as they are parts [of the mother]. Hence, animals still inside
the womb are not yet said to be an animal or alive simply as such and
on their own.337
The parts of animals often do live for a little while, though, after
they have been divided, in so far as (with regard to soul powers) the
10 parts can retain traces of the activities and powers they possessed
when they were unified and connected to the whole [animal].338

On the soul for perceiving


This, then, is what the souls first power is like. Next after the soul
for life and vegetation339 the soul which has been shown by our
arguments about it to be a form of the body that possesses soul and
whose powers were just said to be the powers for nourishing [oneself],
15 growth, and reproduction340 after the account of this soul, then, the
next to be discussed would be the soul for perception, since this soul
is second after the previous one. For anything that has a further soul
power necessarily also has the power for perceiving, just as it has the
power for nourishing [oneself].341 And it is this power in virtue of
which living things, in addition to being alive, are also thereby
animals as well.342
20 Perception is a large topic and deserves a treatise of its own.343 But
in a nutshell, one could say in general as regards the soul for
39,1 perceiving that it is a power of soul in virtue of which whatever
possesses it is able to have cognition of the perceptibles it receives
with an activity related to them, by becoming like them through a
kind of alteration.344 For just as the power for nourishing [oneself]
requires nourishment to be active, since its activity concerns this, so
the power for perceiving likewise requires perceptibles, since its*
5 activity concerns them:345 it is for being aware of them and having
cognition of them.346 Because of this, when they are not present it is
not active, but is rather perception in power,347 analogous to those
who have a kind of knowledge, but are not engaged in the activity
corresponding to it. For in a similar way they are knowers through
their power and inactive perception is perception through its
power.348
Perceptibles are external because perception is directed towards
10 something useful, since it is for obtaining or avoiding something, and
these things are external.349 The perceptible and the sense are unlike
[each other] before the activity [of perception], but in activity they
come to be alike, since perceiving occurs in virtue of a sort of likening
that comes about through alteration. For perception in activity is the
perceptibles form coming to be in what can perceive apart from the
Translation 65
matter.350 But if the likening occurs through alteration, clearly they 15
are different and unlike before the alteration, as was also shown to
be the case with nourishment. It makes sense, then, to say that the
sense is modified by what is like in one way, and in another way by
what is unlike, since it is alike [now] because it was modified.351
This modification comes about in the primary body that has the
soul for perceiving through certain organs, which are of a different
nature than the perceptibles which they serve, because they have a 20
power to be modified by [perceptibles]; and whenever one is modified
in these ways, it relays the modification to* [the primary body].352
[The primary body] is situated in the neighbourhood of the heart,
where the governing part of the soul is also entirely present, as our
discussion will show as it proceeds. For it makes sense that the
highest form is located where what it is to be an animal is present 40,1
most. But what it is to be an animal is present most in what is hot
and moist, and the region around the heart is like this. For it is an
origin and well-spring of the blood with which we are nourished and
also of breath, and these are moist and hot.353
Such, then, is the general characterisation of perception. Now
given that
1. there are five parts of perception;354 5
2. they are related to each other in a similar way to the parts of
the soul,* where one is first and another subsequent;355
3. standardly, when things are related in this way, a general
characterisation does not ever express the nature of the subject
matter in question intelligibly; rather, if one intends to speak
about these things appropriately and in detail, one must always
develop an account of each of them individually;356
it then follows that
4. anyone who speaks about perception ought to speak about 10
each part individually.357
The power for perceivings parts are touch, taste, smell, hearing, and
seeing or rather sight. Of these, touch is separated from the rest
(though taste is in a way also a [form of] touch), as there are in fact
some animals who only have this sense. But it is impossible for any
of the other senses to be found without touch in any animal. For it is
not possible for there even to be an animal in the first place without
this sense.358 15
But those who intend to speak about the powers for perceiving and
the activities corresponding to them must, since the activities of each
sense concerns a perceptible of its own, analyse perceptibles first
briefly, if they are to make these [activities] clear, since perceiving
occurs by likening* to them.359
66 Translation

On what is intrinsically and extrinsically


perceptible
20 Now, to speak generally, some perceptibles are (A) intrinsically
perceptible (kath hauta), while others are (B) extrinsically percepti-
ble (kata sumbebkos).360
Of those which are intrinsically perceptible, some are (A1) percep-
tible to one sense exclusively (idia):* some, like colours, are
perceptibles of sight exclusively; others, like sounds, are perceptibles
of hearing exclusively; others of smell, as odours are; others of taste,
as flavours are; and still others of touch, as the tangible contrarieties
25 are, including [degrees of] heat and cold, dryness and moisture,
roughness and smoothness. For these are perceptible to the corre-
41,1 sponding sense, but no other sense can perceive them intrinsically.361
Other things, however, which are intrinsically perceptible are (A2)
common (koina): these are the kind of thing that can be grasped by
perception, but there are several senses which transmit them.362 For
the perceptible exclusive to each sense occurs together with features
the perceptual organs include when they relay them to the sense of
5 these things.363 Change, rest, number, shape, and extension are this
sort of thing.364
(B) Those things are extrinsically (kata sumbebkos) perceptible
which,* because they happen to belong (sumbebkenai) to some
perceptible, are themselves called perceptible too. For example, if
someone were to say that foam is perceptible, it would be because
foam happens to be white, which is perceptible.365 So things said to
be perceptible in this way are not perceptible at bottom, because the
10 sense is not modified in any way by them in so far as* they are such
things.366
Of the things that are intrinsically perceptible, the senses make
mistakes with regard to common perceptibles, since the senses are
not modified [by them] in such a way as to be like the corresponding
objects. For perceptual error consists in just this: due to certain
circumstances, the modification that occurs in the sense is of a
different sort than and unlike that from which it arose.367 In contrast,
[the senses] are most true with regard to exclusive perceptibles, as
15 long as they preserve the conditions in which they have the capacity
to be aware of these perceptibles. These are, first, that the perceptual
organs are healthy and in their natural state; second, the position of
the perceptible (for sight cannot have awareness of what is located
behind oneself); and third, the commensurateness of the distance,
since an awareness of perceptibles does not occur at just any distance
from the perceptual organs. Beyond these conditions, the medium
through which there is awareness of perceptibles must also be in a
20 suitable condition for transmission to the perceptual organs. For it is
42,1 not possible to see if the transparent [material in the medium] is not
Translation 67
illuminated. Finally, [the medium] must not be disturbed by any-
thing. For one cannot hear what one wishes when loud sounds create
a disturbance.368

***

Colours369 and things which are not seen in the light, but are seen in 5
the dark, are both perceptibles exclusively of sight.370 The latter
cannot be colours, though, since they do not fall under the definition
of colour. For colour is what is capable of producing change in that
which is actively transparent.371 Air, water, and any solid which does
not have a colour of its own are transparent,372 and when they are
changed in a certain way by colours, they are able to transmit to sight
so that there is awareness of [colours]. They become actively trans- 10
parent whenever they are illuminated, for colours can produce
change in them if they are in this condition.373 For it is clear that light
and transparent [materials] that have been illuminated are changed
in a certain way by colours, from the fact that in many cases when
colours are seen through the light, one sees the [transparent mate-
rial] come to be the same colour and carry the colour along with it.
For it itself appears golden from the presence of gold, purplish from 15
murex dye, and greenish from foliage. Often one can see facing walls
or the ground to be this sort of colour, as though they were tinged with
the colour of these things, or even people, if they happen to be
standing nearby, because what is illuminated relays this particular
type of colour from one set of things to the other by being modified.374
For the colour comes to be present in what is illuminated and in light 20
in the same way that light comes to be present in what is transparent,
though what is transparent does not receive light or light colour in
virtue of an effluence or in the way matter [receives something].375
In fact, when the things that produce these [effects] have gone
away, the colour immediately leaves the light as well (in the case
where the things that tinge it go away) and light leaves the 43,1
transparent (in the case where what illuminates it is not present).
The sort of change that arises from both sources occurs in what
receives them in virtue of a presence (parousia) and a particular
sort of relation (skhesis), much as [the reflections] in mirrors come
to be present in them.376
Now illuminated things, as I said, are actively transparent.377 For 5
when something can actively appear through (diaphanesthai) them,
they are actively transparent (diaphan) in the fundamental sense,
receiving their culmination and their own form (in so far as they are
transparent) from light.378 For light is the activity and culmination of
transparent [material] in so far as it is such.379
68 Translation

How there comes to be light in the


transparent380
10 This light comes to be by the presence of fire or the divine body in the
transparent. For light comes about in virtue of the illuminants
relation to things whose nature it is to be illuminated.381 For light is
not a body and so comes to be instantaneously.382
If colour is visible in light, then it is also able to change it. For, as
has been shown in the inquiries into how we see, the perception and
cognition of colours occurs because (i) what is actively transparent
that is, what is illuminated is first modified by the colour, since
colour is able to change it, and then (ii) the eye is modified by this [sc.
15 what is transparent], since the eye itself is also transparent.383
(Things seen in the dark, therefore, cannot be colours. How these are
seen and in what circumstances has also been addressed in the
writings on how we see.)384 For light, which has been tinged anew by
20 each visible thing along a straight line to the eyes aligned with it,
44,1 relays the exclusive modification, since it was itself modified due to
them;385 and [the eyes] in turn are also able to receive a reflected
image themselves because they are both smooth and transparent.386
So given that
1. seeing occurs because the perceptual organ received the
colour and in some way becomes likened to it;387
2. what can receive colour is something that lacks a colour of its
own or can be seen only with difficulty;388
3. what has this characteristic most is transparent [material]
5 that is moist and without a definite shape;389
4. both air and water are this sort of thing,390
[it follows that]
5. the perceptual organ through which there is awareness of
colours is composed of these [materials].391
For the interior of the eye is composed of water,392 for the following
reason. Of the transparent [materials] that lack a definite shape,
water straight off can retain the modification produced in it by
colour,* because of its density and consistency; for air is not like
10 this.393 Seeing, then, comes about through the activity concerning
visible things, which in turn comes about through the eyes power for
seeing, in virtue of a likening to [visible things]. The power for seeing,
in contrast, is that in virtue of which its possessor can, when altered
by visible things, become aware of them and have cognition of them,
through the activity concerning them.394
As regards light something that is visible in the highest degree
and responsible for other visible things being seen395 it has already
15 been stated just what it is, namely, an activity of transparent [mate-
Translation 69
rial] that lacks a definite shape, in so far as it is transparent.396 But
light is also said to be the transparent [material]s colour, in so far as
it is transparent.397 For in the same way that colour is in things that
have colour, light comes to be in transparent [materials] and is
responsible in the way just mentioned. Colour, which can change
what is transparent in this way, is itself in transparent [material]
too, but in something with a definite shape.398 For every body, be- 20
cause of its ability to receive colour, has a larger or a smaller share
of the nature we call transparency,399 since this is the proximate
matter of colours. For if there is a colour, there must also be the
nature that can receive it.400
Transparent [material] without a definite shape something
which is and is said to be transparent (diaphanes) in the highest
degree because it receives daylight (phaos), that is, light (phs), and
is responsible for other things appearing (phainesthai) is coloured 25
by something external [to it].401 For light is in a way a colour of what 45,1
is transparent in the form stated above402 and occurs in it due to the
presence of something whose nature it is to illuminate it. This is what
fire and the divine body are like.403 For light comes about in virtue of
their having a certain relation to it [sc. the transparent material].*
Hence, it is illuminated when they are present, but when they leave,
the light departs together with them.404 5
The transparency in solids, in contrast, has the colour inside,
co-mingled and compresent.405 Hence, it is not present in it at one
time and not another, as happens with light in transparent [materi-
als] without a definite shape.406 Among [materials] which are
transparent in this way, some are more and others less transparent,
because of a certain amount of mixing in of a minimally transparent
body, such as earth. For it impedes the transparency of whatever it 10
has been mixed into.407 It is also responsible for the difference be-
tween colours. For the colour in any transparent [material] is due to
the presence of something bright (leukon) for fire is in fact just this
sort of thing whether as something external to it or mixed in with
it. But the difference in degrees of transparency and the mixture of
various amounts of something whose nature it is to illuminate it
produces the differences between colours. For [the colour] is bright 15
and luminous in things in which the transparent is purer and in
which there is more colourant, but it is less so and dull, already
tending towards dark (melan), in things where the transparent is
present [only] to a certain extent. When there is something mixed in
with the transparent [materials], however, that makes them trans-
parent in actuality, they have a colour of their own compresent with
them, which is called colour in the fundamental sense (since light is
not like this).408
Colour is also the boundary of transparent [materials] with a 20
definite shape in so far as they are transparent.409 For in so far as
70 Translation
they are a body, their boundary is a surface; but in so far as they are
transparent, it is a colour. The colour is thus compresent with the
surface.410 For colours make their appearance (phantasia) on the
surface, since only the boundary of bodies is apparent (phaneron),*
and the colour is both apparent and something that appears (phaino-
25 menon).411 For that in virtue of which bodies are coloured is also that
in virtue of which they are visible;* but they are visible in virtue of
46,1 the boundary. So it makes sense that colour is said to be the boundary
of the transparent in a [body] with a definite shape in so far as it is
transparent.412 The power present in fire and other things which are
able to produce light makes them visible in the highest degree and is
responsible for other colours being seen as well, since these things
can produce light, by which things seen are seen. We should regard
this nature as responsible, when mixed in larger or smaller quanti-
5 ties in transparent [materials] with a definite shape, for the
production of colours and the differences between them.413
That colour is the boundary of the transparent, in so far as it is
transparent, is indicated by the fact that transparent [materials]
which do not have a boundary of their own those without a definite
shape are of this sort do not have a colour of their own either, since
10 the boundary of a body is compresent with its boundary as something
transparent. For [these materials] are tinged by something from
outside in the same way as they are bounded and shaped by it.414 For
anything whose boundary is definite* as a body is also [definite] as
something transparent; and anything transparent whose boundary
is definite has a colour of its own, since the boundary and the colour
coincide.415 Yet although they coincide, they are not the same. For a
15 single body has a single surface; but it is not the case that if there is
a single surface, there is also a single colour, since it is possible for
the same surface to have even contrary colours in it.416 Transparent
[materials] with a definite shape have a boundary, whether they are
illuminated or not, but they only have a colour when they are
illuminated as well.417
Commentary

1. [1,2] By restricting his discussion to the souls of bodies that come to be and
perish, Alex. effectively brackets any questions regarding divine souls from consid-
eration and how his account might apply to them (cf. 28,25-29,2; Mant. 104,10; also
in Metaph. 373,5-8). This group would include not only the First Cause, which is a
non-bodily understanding or nous (89,9-15; cf. Quaest. 1.25, 40,8-9) if it can even
be said to have (or be) a soul but also, importantly, the soul of the divine and
eternal body which constitutes the celestial spheres (Quaest. 1.25, 40,9-30). This
restriction fits with Arist.s own view that we should not generalise from the
perceptible region around us (ho peri hmas tou aisthtou topos), where bodies
come to be and perish, to the universe as a whole, since the bodies of the celestial
spheres are eternal (Metaph. 4.5, 1010a29-30; cf. Alex. in Metaph. 310,23-33;
77,24-6). In Alex.s view, bodies that come to be and perish the subject of what he
calls Third Philosophy (in Metaph. 251,36-7) are to be found exclusively in the
region beneath the moon, which is the domain of contingency, fate, and providence
(Mant. 181,21-7; 182,32; Quaest. 1.25, 41,10-11; cf. 2.19, 63,15-26; in Meteor. 6,5-6).
2. [1,3] 16,18. The Greek ousia, although often rendered as substance, is
derived from the participle for the verb to be. It can thus signify both (a)
something that really is, i.e. a being or a substance, as it often does below
Alexander says that Aristotle always calls beings (ta onta) substances (ousias, in
Metaph. 117,25) but also (b) what something really is, its essential nature,
especially when it is used (as here) in locutions of the form the ousia of X.
Understood in this second way, it is not limited to the essence or form of composite
individuals, but can be used for the essence of other things as well, including souls.
On this broader sense, see Alex. in Metaph. 374,37-375,9 (cf. Arist. Metaph. 5.8,
1017b21-2). Alex. develops the question of the souls ousia somewhat differently in
the opening of the Mant. 101,3-18.
3. [1,2-4] Alex. follows this programme, but only in broad outline. He does claim
explicitly at 16,18 to have stated the essence of the soul and he focuses directly on
the number of powers at 27,3-28,13 and the powers themselves at 28,13-31,6. But
his remark at 27,1-2 suggests that the whole treatise up to that point has been a
discussion of the essence of the soul (2,10-26,30), just as the differences between
the powers will in some sense be the subject for the remainder of the treatise
(31,7-92,11), as is confirmed by his formal conclusion at 92,12-15 (before what are,
in effect, two appendices, 92,15-94,6 and 94,7-100,17).
The questions Alex. sets here are modelled on those with which Arist. begins
his DA (1.1, 402a7-8). But they are more specific: Arist. says that in addition to the
nature and essence of the soul, we should investigate its attributes (hosa sumbe-
bke) and whether any belong to the soul alone. Alex. does not raise the latter
question here, which is potentially much broader in scope, and focuses exclusively
on the powers of the soul instead.
4. [1,6] That is, Apollo: cf. Fat. 31, 202,9; 203,1.8.16.20.24. The epithet refers to
Apollos slaying of the dragon Python, in establishing his own shrine at Delphi. For
the different versions of the myth, see Fontenrose 1959. For a recent introduction
72 Notes to page 31
to Apollos role in prophesy, see Graf 2009, ch. 3, and for the Delphic oracle,
Johnston 2008, 22-60.
Someone might question the sincerity of Alex.s appeal to divine prophecy here
in light of his own reservations about divine foreknowledge: at Fat. 30, he argues
that not even the gods can know that contingent events will happen, only that they
might or might not (see esp. 200,24-201,3 and 201,13-21). But Alex. still believes
that prophecy and petitionary prayer have a useful function, since on his view,
unlike the Stoics, we can avert fate, and through prophecy the gods advise us to
resist the consequences of a baser nature (enhistasthai ti ts phuses kheironos
akolouthiai, Mant. 185,33-186,4 at 185,36; cf. Fat. 6, 171,7-11; 17, 188,11-17; and
31 passim). In these passages, Alex. is concerned not only with the nature of the
species, as here, but with the nature of an individual as well (17, 185,11-33; cf. 6,
170,9-171,7).
5. [1,6-8] In contrast with the tradition that ascribes this saying to the Seven
Sages (see Wilkins 1917, 1-2, 6, 8), Alex. speaks of the command as divine in origin,
as other later authors do (Cic. Leg. 1.58-9, cf. Tusc. 1.52; Juv. Sat. 11.27; Jul. Or.
6, 183A, 188B), though Alex. goes further by appealing to Apollos special authority
in virtue of his foreknowledge. Arist. may have been gesturing in this direction in
his lost dialogue On Philosophy, when he claimed the command was older than the
sages, originating instead from the Pythian oracle (Clem. Strom. 1.14.60.3 = Arist.
fr. 3 Rose3).
6. [1,8] I have retained the gender of the masculine pronouns in the Greek here,
as less disruptive and distorting than more neutral alternatives we prefer today.
It is clear that Alex. has in mind every human (anthrpos, 2,1), not just men.
7. [1,9] Elsewhere, Alex. maintains that one of the universally shared intuitions
(koinai ennoiai) about fulfilment (eudaimonia) is that it consists in living in
accordance with nature and the life in accordance with nature (to zn kata phusin
kai ton kata phusin bion, Mant. 162,32-163,1). He argues that virtue alone is not
sufficient to secure it: if we are deprived of either the goods of the body or of the
soul, we are not living in accordance with nature (163,1-18). The phrase in
accordance with nature, he claims, should be understood to mean in accordance
with natures intention (kata to boulma ts phuses akouetai, 163,13-18, cf.
163,21-3. 28-31).
The most striking parallel for this conception of fulfilment, as well as for the
whole opening argument (see next note), comes from an Academic, Antiochus of
Ascalon (first century BCE), who similarly uses the phrase living in accordance
with nature (secundum naturam vivere) to define the end or aim of life (Cic. Fin.
5.26; 2.34 = Antiochus fr. 9a Mette; cf. 5.44, Acad. 1.19). Antiochus disavows any
originality here: he repeatedly represents his views as shared by the Old Academy,
the Peripatetics, and (with qualifications) the Stoics. The view seems a natural one
for Peripatetics to hold; but the formula living in accordance with nature is not
attested for Peripatetics prior to Alexander. The evidence for the Old Academy, in
contrast, is unequivocal. The definition is attributed several times to the fourth
head of the Academy, Polemon (fourth to third century BCE), who wrote a treatise
with that title (Cic. Fin. 2.34 = fr. 127 Gigante; 4.14 = fr. 129; for the title, Clem.
Strom. 7.32.9 = frs. 97, 112); and the general view is also credited to the third head
of the Academy, Xenocrates (fourth century BCE) at Cic. Fin. 4.15-18 (= fr. 79
Heinze), as well as to Aristotle (ibid.). It may conceivably be behind the definition
attributed to the second head of the Academy, Speusippus, too (Clem. Strom.
2.133.4 = fr. 101 Isnardi Parente).
The Stoics also use this phrase. Zeno, who studied with Polemon (SVF 1.10-11,
Notes to page 31 73
13), wrote a treatise with that title as well (D.L. 7.4 = SVF 1.41) and is said to have
followed Xenocrates and Polemon in their view that the elements of fulfilment are
nature and what is in accordance with nature (Plut. Comm. Not. 1069F = SVF
1.183; Cic. Fin. 4.61). It is also listed among the Stoics various definitions of the
end (Stob. Ecl. 2.77.16-19 Wachsmuth = SVF 3.16 = LS 63A 1; cf. 2.78.1-6; Mich.
in EN 598,30-2 = SVF 3.17). But the Stoics use of this notion differs in two key
ways. (1) In characterising the end, the Stoics emphasise our being in agreement
with the nature of the cosmos as a whole (D.L. 7.89 = SVF 1.555), and not just with
the specific nature of human beings, as Alex. and Antiochus claim. (2) The Stoics
seem to distinguish a life lived in agreement with nature (homologoumens ti
phusei zn), a more canonical definition of the end, from one which is lived merely
in accordance with nature (Fin. 4.25-6, 4.43). The so-called natural goods are not
goods, strictly speaking, on their view: while things that contribute to the life in
accordance with nature (khreian sumballomenn pros ton kata phusin bion) are
preferred (progmena) and so have value (axia), they are morally indifferent
(adiaphoron, D.L. 7.105 = SVF 3.126; Stob. Ecl. 2.80.3-4 Wachsmuth = SVF 3.136).
We are to choose them because, and only in so far as, they fit in with the plan and
order of nature as a whole, that is, because they agree with nature as a whole, and
not because they are good in themselves. On this difference, see N. White 1979,
148-9, 151-3, 157-8; 1985, 61, 66-8; Striker [1983] 1996, 289; [1991] 1996, 224. Alex.
criticises the Stoics on precisely this point, arguing that they will have trouble
assigning the life in accordance with nature to any of their categories of value
(Mant. 167,17-168,1 = SVF 3.45).
8. [1,4-2,4] The overall argument here is surprising if one associates the
Delphic command with knowing ones own personal traits and limitations, a
subject on which Aristotelian psychology has little to say. Alex. takes the command
instead to concern our common nature as humans and more generally as beings
who come to be and perish (1,2), that is, as mortals (an interpretation which had
some currency in antiquity: see Wilkins, 1917, ch. 7) and so is arguing, in effect,
that the recognition of mortality is essential to a life lived in accordance with
nature and thus to human fulfilment. Alex.s controversial views on the mortality
of the human soul at the end of the treatise are thus already foreshadowed here in
its opening argument. Interestingly, he does not invoke the notion of reflexive
self-understanding, so central to his own theory of the understanding (nous), which
would figure prominently in interpretations of the Delphic command by later
Platonists (Wilkins 1917, ch. 8 offers a preliminary survey). For a thorough history
of the maxim, see Courcelle 1974-75, t. 1.
The closest antecedents we have for Alex.s argument are again Academic (see
preceding note). An extended version of the argument is attributed to Antiochus of
Ascalon by Cicero: see Fin. 5.44; cf. 5.26-7, 34. 41; 4.25-6; also Aug. CD 19.3. Earlier
evidence for antecedents is thinner. Arist. mentions the Delphic command in his
extant writings only in passing (Rhet. 2.21, 1395a20-6; though cf. MM 2.15,
1213a13-26); he appears to have discussed it in his lost dialogue On Philosophy
(frr. 1-3 Rose3), but we do not know how extensively or to what end. The Stoic
evidence is somewhat better: according to the emperor Julian, the Stoics made the
Delphic command the core of their philosophy (kephalaion tithentai philosophias)
and wrote many treatises on it (Or. 6, 185D-186A). There is perhaps some basis for
this assessment in Epictetus (1.18.17; 3.1.25; cf. 1.6.25; 2.6.14) and in Seneca
(Cons. ad Marc. 11.1-5; NQ 3, praef. 15; cf. Ep. 94.28; 121.3), but there is little
before them. The best parallel is an argument that it is impossible to fulfill the
Delphic command without a knowledge of the natural world generally (Cic. Fin.
74 Notes to pages 31-32
3.73 = SVF 3.282; cf. Leg. 1.61 = SVF 3.339; Epict. 1.1.10; I am grateful to Brad
Inwood for discussions on the Stoic evidence). The closest parallels before Antio-
chus are in Plato: Socrates describes his efforts to obey the Delphic command in
the Phaedrus as an attempt to find out what sort of being he is, a monster like
Typhon or one that has a share in the divine (229E-230A); and in the Alcibiades I
it is argued that to know ourselves is to know our souls (130E), though it seems
there to be more a matter of knowing the individual self rather than the nature of
the species. (There are other contrasts as well, of course: Plato is concerned
primarily with our immortal and rational self, rather than the mortal nature we
share with animals. I am grateful to Richard Sorabji for emphasising this.)
9. [2,4-9] Notice that Alex. does not assign some kind of ultimate authority to
Arist.s views, dismissing any who disagree. His wording suggests rather that
others do attain the truth in some measure, but that Arist. does to a much greater
extent and so deserves our full efforts at understanding him. Accattino and Donini
1996 (= AD) aptly cite Arist. Metaph. 2.1, 993a30-b7, which argues that while
everyone manages to get at some part of the truth, some do more than others.
10. [2,10 title] The section titles are found in the principal manuscript V
(Venetus Marcianus gr. 258) and in the Aldine edition of 1534, but are printed by
Bruns only in the apparatus. I have printed them here because, even though they
might not stem from Alexander, they are nonetheless a part of the tradition and
generally helpful.
11. [2,13] Art here renders the Greek tekhn, which should not be understood
as restricted to fine arts, but applies to any kind of productive and practical
knowledge, of knowing how to make or do things, including the art of housebuild-
ing, the art of medicine, and even the art of governing. As such, it implies a grasp
not only of the general regularities in nature, but also the causes responsible for
producing various effects (see esp. Arist. Metaph. 1.1, 981a12-b9; cf. An. Post. 2.19,
100a6-9). The analogy between nature and art that Alex. exploits here is central
for Arist. as well: see esp. Phys. 2.8, 199a9-b33; PA 1.1, 639b14-23; see also G.E.R.
Lloyd 1966, 287-90.
12. [2,12-15] 3,2-21. The rhetorical comparison with the products of art, as well
as the language of marvels and wonders (cf. 11,5-13), is reminiscent of the famous
passage where Arist. exhorts his readers to the study of the parts and materials
from which animals are composed (PA 1.5, 644b23-645a26, esp. 645a10-17).
Alex.s confidence about his readers being more easily convinced (pisteusomen,
2,14) contrasts strikingly with Arist.s more guarded remark at the opening of his
On the Soul, that this is a subject about which it is the most difficult, in each and
every way, to secure conviction. (1.1, 402a10-11)
13. [2,17-18] 11,5-13. Alex. is alluding to Platos Phaedo (91C-D), where Socra-
tes addresses the doubts and fears Simmias has because he regards the soul both
as (i) something more divine and more noble than the body (kai theioteron kai
kallion on tou smatos; cf. 94E) and (ii) a tuning (harmonia) or disposition of the
body, a theory that has relevance for Alexanders own treatment of the soul (note
his use of epharmozein here at 2,17). The preceding section of the Phaedo (89C-
91C) also stresses the difficulty of arriving at settled views in these matters and
the consequent risk of misology. It may also be the case, as both AD (105) and
Bergeron and Dufour 2008 (= BD), 234 suggest, that Alex. is responding to
contemporary Platonic views here, citing a fragment of Atticus that excoriates
Arist. for rejecting the incorporeal nature of the soul (Euseb. PE 15.9.7-8 and 11,
2.370.4-9 and 371.1-6 Mras = fr. 7.34-43, 57-63 des Places).
14. [2,18-22] Arist. DA 1.3, 407b13-26; 2.2, 414a22-7. As AD point out (106),
Notes to page 32 75
Alex. does not carry out such a detailed examination here (see also Donini 1971,
68). At most he speaks in broad, schematic terms about the increasing complexity
of bodies, from simple bodies to plants and then animals (see 8,13-11,13).
The English regulation renders oikonomia; cf. 38,2. In an extended simile,
Arist. describes nature, in using different material to form the parts of the body in
the embryo, as a good house-manager (oikonomos), who lets nothing go to waste
(GA 2.6, 744b16-21). The Stoics also speak of the regulation or management
(dioiksis) of the universe: Alex. Fat. 22, 192,12.28 = SVF 2.945 = LS 55N; Plut.
Stoic. Rep. 34, 1050A = SVF 2.937.
15. [2,23] I have used the phrase origin and principle throughout to render the
important term arkh, apart from a few cases where only one of these senses could
be in play. In general an arkhe is a source, origin, or basis of a thing or process,
that in some sense determines or characterises what issues from it or (in another
of its senses) governs or rules it.
16. [2,25-3,13] Mant. 101,22-6. The distinction between matter and form, which
Alex. introduces here simply as the opinion of the experts, is one of the most central
and characteristic Aristotelian doctrines. For the view that all perceptible sub-
stances are composed of matter and form and more generally for the distinction
betwen matter, form, and the composite formed from them, see esp. Cael. 1.9,
277b30-278a10; DA 2.1, 412a6-12; Metaph. 7.8, 1033b12-19; 8.1, 1042a24-31; 8.2,
1043a14-28; 12.3, esp. 1070a9-13.
17. [3,7-9] Mant. 102,1-7. Arist. uses statues frequently as an example to
illustrate various points about the four causes, but the one closest to the present
context is GA 1.18, 724a23-6, where Arist. speaks of the material being shaped
(skhmatisthentos), as Alex. does at 3,6 above. Cf. also Metaph. 5.2, 1013b6-9; Phys.
1.5, 188b19-20; 1.7, 190b5-6; 2.3, 195a32-5. Arist. also uses other artefacts to make
key points about the soul: he uses the example of a house to show how psychological
states should be defined (DA 1.1, 403b3-9) and an axe to explain the way in which
the soul is a form of the body (2.1, 412b10-17).
18. [3,9-11] Alex. here relies on the analogy with nature he makes fully explicit
just below (3,19-21, with note): the form or design produced by art is itself said to
be art, just as the form of a natural object is its nature. Arist. says that art is the
form of the product at Metaph. 7.9, 1034a24; GA 2.1, 734b36-735a3; 2.4, 740b25-9;
cf. PA 1.1, 640a31-2.
19. [3,11-13] Elsewhere Alex. generalises this description to all matter, as an
underlying subject suitable for receiving the form in question, at Mant. 148,25-7;
in Sens. 44,25-7.
20. [3,15-16] in Meteor. 197,3; Eth. Prob. 25, 148,15; cf. Mant. 101,28-102,1.
Both parts of this antithesis occur nearly verbatim in a passage believed to derive
from Arist.s lost dialogue, the Protrepticus (B13 Dring = Iambl. Protr. 49.28-50.1;
cf. B14 = Iambl. Protr. 50.2-14). But the positive claim, that art imitates nature
can be found in the extant works too: see Meteor. 4.3, 381b3-9 (cf. a9-12); Phys. 2.2,
194a21-2; 2.8, 199a15-17. See G.E.R. Lloyd 1966, 287-90. Ian Flora has pointed out
to me that we also find versions of the positive claim in e.g. Philo (Ebr. 90,
2.176.13-14 Cohn-Wendland = SVF 3.301) and Galen (UP 1.408.5-13 Helmreich).
On the theme more generally in antiquity, see Close 1969, 469-72.
21. [3,17-19] in Meteor. 177,20-178,10. Much of Arist.s theory in the Meteorol-
ogy is based on his distinction between two types of emission (or, as it is often
rendered, exhalations), one vaporous, the other fume-like or smoky (Meteor. 1.4,
341b6-12; 3.6, 378a19-20). Metals are the result of the vaporous emissions being
compressed and solidifying (3.6, 378a26-31). For the specific claim that bronze, like
76 Notes to pages 32-33
other metals that can be melted, is made of water, see Arist. Meteor. 4.10, 389a7-9;
cf. Metaph. 5.4, 1015a9-10; 5.24, 1023a28-9 (cf. Alex. in Metaph. 421,33-4; 430,8-9).
On Arist.s account of metals, see Eichholz 1949.
22. [3,19] in Meteor. 212,24-33. Alex. suggests at the very end of this commen-
tary that the form of natural bodies can be understood quite generally as a certain
kind of solidity or consistency (227,11-14; cf. 226,5-6). On solidification in general,
see Arist. Meteor. 4.6-7.
23. [3,19-21] Arist. Phys. 2.1, 193a31-b8. See n. 18 on 3,9-11 above.
24. [3,22] Mant. 103,32-3; Arist. Cael. 1.2, 268b26-7; 1.5, 271b17-18; PA 2.1,
646a12-24. AD are right to note (107) that compound here cannot mean composed
of matter and form, as it generally does (e.g. 2,27), since for Alex. all bodies,
including simple bodies, are compounds of matter and form (see 3,27 below). What
distinguishes compound bodies is that their matter is also a body and so itself a
compound of matter and form. This analysis can be reiterated until we finally
reach simple bodies, which, though compounds of matter and form themselves,
have matter that is not a compound of matter and form and so not a body either.
25. [3,24-5] This last generalisation extends to celestial bodies: see Alex.s
discussion of their underlying matter in Quaest. 1.10 and 15.
26. [3,28-4,4] Alex. here articulates a traditional conception of prime matter
(materia prima), an ultimate material substrate underlying all bodies, which has
no form as part of its own nature beyond its mere receptivity to forms (for the
latter, see Quaest. 2.7, 52,28-30; cf. 1.15, 27,22-3). It is inseparable from the bodies
it underlies and so from form, but in itself is nothing in actuality. For Alex. on
matter in the fundamental sense (kuris), see in Metaph. 159,11-17 (cf. 158,9-14;
159,3); 430,3-6; Quaest. 2.24, 75,5-10.16-19, which explicitly identify it with prime
matter (prt hul). For Alex.s views on prime matter more generally, see also in
Metaph. 213,2-4; 357,13-18; 359,10-11; 421,34-6; in Meteor. 224,14-19; Quaest.
1.17, 30,2-6. For the description of it as formless, see Mixt. 11, 226,15; Mant.
104,19; shapeless, Quaest. 2.7, 52,20-53,30 (cf. 2.3, 49,30-1); and also unshaped
(arrhuthmistos), 1.10, 21,1-2.
Arist. uses all of these phrases himself to describe matter: formless and
amorphous (Cael. 3.8, 306b17), amorphous (Phys. 1.7, 191a10), shapeless (Phys.
1.7, 191a2; cf. 1.5, 188b20; 1.7, 190b15), and unshaped (Phys. 2.1, 193a11). He also
speaks of matter in the most fundamental sense (GC 1.4, 320a2-5) and even of
first or prime matter (GC 2.1, 329a23, a29; Phys. 2.1, 193a29; Metaph. 5.4,
1015a7-10; 9.7, 1049a24-7; GA 1.20, 729a32; cf. Metaph. 5.6, 1017a5; 8.4, 1044a23;
GA 2.1, 733b26). Arist. explicitly notes, though, that this last phrase can be used
for either end of the hierarchy, for the last or proximate matter of a substance as
well as for the first or ultimate one (Metaph. 5.4, 1015a7-10), and in a number of
these cases he uses quite ordinary materials as examples, such as the element
water. The best reasons for thinking Arist. is committed to prime matter are his
views about the transformation of the elements into one another (e.g. GC 2.1,
329a24-35 (cf. a8-11); 2.7, 334a15-25; Cael. 3.8, 306b16-21), together with the
requirement that some matter must persist through change that can equally
possess or lack even these elemental qualities (see esp. Phys. 4.9, 217a21-33; Cael.
4.5, 312a30-3; GC 1.4, 319b2-4; 1.5, 320b12-14; 2.5, 332a17-20, a34-b1). The closest
he comes to describing such a qualityless substratum is in a thought experiment
in Metaph. 7.3, 1029a7-30, for the admittedly negative purpose of showing that it
cannot be substance. But the interpretation of these passages remains controver-
sial. For classic examinations of the evidence for prime matter, see: contra,
Charlton [1970] 1983, 129-45 (also Rashed 2005, xcii-ciii); and pro, C.J.F. Williams
Notes to page 33 77
1982, 211-9. For more recent interpretations, with references to the literature:
contra, M.L. Gill 1989, ch. 2 and Appendix; pro, F. Lewis 2008.
Many of the key features of prime matter, though, can be found uncon-
troversially in Platos characterisation of space as the mother and receptacle of
all perceptible qualities (Tim. 49A-52B): he says that it is amorphous (amorphon,
50D7, 51A7, cf. 50C1), without any other nature (50E2, cf. B7-C2), invisible (51A7),
and completely receptive (51A7; cf. 50B8, D4-E1). There can be little doubt of the
profound influence this passage has exercised over the development of the notion
(Charlton [1970] 1983, 141-5), since Arist. understood this conception of space to
play the same role that he assigned to matter (Phys. 4.2, 209b11-16).
Alex.s precise formulation, though, has two nearly verbatim parallels in the
tradition, which suggest that by his time it was already something of a philosophi-
cal commonplace.
(1) A report deriving from the doxographer Aetius (c. 100 CE) attributes the
view to both Aristotle and Plato (or just Plato, in Stobaeus version) that matter is
bodily, amorphous, formless, shapeless, and qualityless in its own nature (sma-
toeid amorphon aneideon askmatiston apoion), but comes to be a receptacle of
forms, like a nurse, wax tablet, and mother (1.9.4-5, Dox. Gr. 308.4-9); a very
similar version occurs again in Alcinous summary of Platos doctrines from per-
haps the mid-second century CE (Didask. 8, 162.32-39, 163.4-10 Whittaker). The
same themes, expressed somewhat differently, had been already attributed two
centuries earlier by Antiochus of Ascalon to both Academics and Peripatetics (Cic.
Acad. 1.27). For further parallels for the individual terms, see Whittaker 1990, 96 n.
139. It should be noted (against AD, 107) that the parallels for the Peripatetic Boethus
(first century BCE) are merely verbal, since the pre-existing matter he describes as
amorphous and formless is expressly distinguished from the underlying substratum:
ap. Them. in Phys. 26,20-4; Simpl. in Phys. 211,15-18 (passages which may stem from
Alex.s lost commentary on the Physics see Moraux 1973, 170).
(2) One of our testimonies for the Stoics provides another verbatim parallel.
Calcidius reports that Zeno of Citium (fourth to third century BCE), the founder
of the Stoa, believed that just like wax and the many shapes it can take on see
Alex.s use of this comparison below at 4,9-20 in the same way there is no form
or shape or any other quality that belongs intrinsically to the matter underlying
all things [sic neque formam neque figuram nec ullam omnino qualitatem propriam
} fundamenti rerum omnium silvae], though it is always inseparably conjoined
with some quality. (Calc. 292, 295.3-6 Waszink = SVF 1.88 = LS 44D 3; cf. 311,
311.7-8 Waszink; Origen Orat. 27.8, 368.2-10 Koetschau = SVF 2.318). Elsewhere
we find Zeno and Chrysippus, the third head of the school (third century BCE),
defining substance (ousia) as the prime matter of all things and the Middle Stoic,
Posidonius (second to first century BCE), describing it as qualityless and amor-
phous in itself, though it always has some given quality and shape: Stob. Ecl.
1.132.27-133.11, 18-23 Wachsmuth = Arius Didymus fr. 20, Dox. Gr. 457.25-458.11
= LS 28q and in parts, SVF 1.87, 2.317 and Posidonius fr. 92 Kidd; cf. Calc. 290,
294.8-9 Waszink = SVF 1.86, cf. 293, 295.17-18 Waszink; D.L. 7.150 = SVF 2.316;
Ps.-Galen Qual. incorp. 19, 273-4 Giusta = SVF 2.323a; Simpl. in Phys. 227,23-6 =
SVF 2.326. As one of the Stoics two fundamental principles, matter is described
as qualityless and shapeless substance (apoios ousia, D.L. 7.134 = SVF 2.300 =
LS 44B 2; askhmatistos, Sext. Emp. Adv. Math. 9.75 = SVF 2.311 = LS 44C 1, cf.
9.11 = SVF 2.301). See also 18,2-3 below.
These two parallels may not be independent. If the older Stoics actually used
the word matter, it suggests an Aristotelian influence, as was long thought (see
78 Notes to page 33
esp. Hahm 1977, ch. 2). But even those who are more sceptical of Aristotelian
influence, like Sandbach (1985, 35-6), still link Stoic views with the Academic
tradition, esp. Xenocrates. There is thus no reason to think that Alex. is simply
appropriating Stoic views here. And there is in any case a critical difference
between their views. For Alex., although prime matter is a constituent of every
body, it is not itself a body, since it is not a compound of matter and form (3,26-7).
For the Stoics, in contrast, prime matter is a body, since it is capable of being acted
upon (D.L. 7.134 = SVF 2.299 = LS 44B 3, retaining the MSS reading see LS
1.274, 2.266; this is confirmed by Alex. Mixt. 11, 225,3-5 = SVF 2.310).
27. [4,5] The phrase simply as such here renders the Aristotelian technical
term hapls, which indicates that a term is to be understood without further
qualification or restriction. See e.g. Arist. Top. 2.11, 115b29-35 (cf. GC 1.3, 317b5-
7); Alex. in Top. 215,22-216,10; also Arist. GC 1.3, 317b5-7.
28. [4,8] Mant. 120,2; Quaest. 1.8, 17,10; 1.17, 30,3-6; Mixt. 13, 228,22; Arist.
GC 1.5, 320b16-17; cf. 2.1, 329a24-7.
29. [4,8-9] In fact, Alex. does not explicitly say earlier that the matter of
compound bodies can exist on its own: when he introduces the distinction at 3,23-4,
he says only that their underlying matter is a natural body. He assumes that all
natural bodies can exist on their own, apart from any further mixture or compound
they may enter into or from any form they may take on. This is confirmed below at
4,23-5 and 17,11-12.
30. [4,9-15] cf. 6,20. The example of wax is used to make a similar point at
Quaest. 1,8, 18,32-4, 19,3-6; 2.7, 53,25-30. The last passage notes that while
qualityless in its own nature, matter is not qualityless in its very being, since it
always has some quality or other; Mant. 121,32-4 adds that it is only separable in
thought (epinoiai). Alex. may have in mind Arist. Cael. 3.7, 305b29-30 and Metaph.
5.26, 1024a4-5 which notes that the same wax can take on different shapes; cf. also
GC 1.10, 327b14-15, which argues that shape is not something mixed with wax.
Arist. also uses wax prominently as an example in DA 2.1, 412b6-8, but he is
concerned there with the unity of matter and form and not Alex.s point that
although having a shape is essential to wax, no particular shape is.
Compare Descartes use of a similar argument based on the variable qualities
of wax in the Second Meditation (AT 7.30-2) to show that we can only perceive the
essence of wax and bodies in general by the intellect alone.
31. [4,18-20] If having a particular form were part of the nature of prime
matter, then it would be a body, compounded of matter and form itself. But then
anything which had prime matter as its matter would be a compound body,
contrary to the hypothesis that we were considering the matter of a simple body.
In addition, if the form that was essential to prime matter were the form of one of
the elements, such as fire, then prime matter could not serve as the matter of
another simple body like earth, since on the present hypothesis it would cease to
exist if it were to lose the form of fire.
32. [4,20-2] Mixt. 13, 228,20-1; Quaest. 1.17, 30,5-6. Alex.s claim that forms in
general or at any rate those forms that come to be in something (he remains silent
on eternal substances) cannot exist apart from matter, even less than matter can
exist apart from form, is essential to his overall position concerning the soul in this
treatise. Despite the explanatory and metaphysical primacy of form (including the
soul), it is nevertheless existentially dependent on body.
33. [4,23] cf. 10,24; 44,21. Alex. uses the term proximate matter (prosekh
hul) to refer to the matter that directly underlies a compound body, in contrast
with the various matters beneath it that constitute the proximate matter (or
Notes to pages 33-34 79
constitute the matter that constitutes the proximate matter, and so on). In contrast
with prime matter, it is the last or highest matter in the hierarchy rather than
the first or ultimate one: see esp. in Metaph. 358,36-359,4 (cf. 359,10-11); also
212,35-213,4; cf. 357,16-18; 430,3-6. (In the case of the elements, first and last
trivially coincide: Quaest. 2.24, 75,5-10.) Although the phrase proximate matter
does not occur in Arist., the notion does: see esp. Metaph. 5.4, 1015a7-10.
34. [4,24] The phrase tode ti, literally a this-something, is a technical expres-
sion Arist. uses to speak in the abstract of something that belongs to a specific kind,
a particular instance or token of a certain type. This is usually an individual
substance (Metaph. 7.3, 1029a27-30; cf. Alex. Mant. 101,18-21), but not always:
see, e.g., 6,8; 7,12 below. For a discussion of the expressions semantic range in
Arist., see M.L. Gill 1989, 31-4. For Alex.s use of the phrase, see Sharples 1999.
35. [5,1] Following the Hebrew translations construal of ekeinn and toutn.
36. [5,1-2] Mant. 103,29; Arist. Metaph. 8.3, 1043b21-3; cf. 7.17, 1041b29-30;
12.3, 1070a13-20. For a recent treatment of the Aristotelian texts, see Katayama
1999. Alex.s claim here is limited: he denies that the form of an artefact is an
ousia, something that is in the most fundamental sense of is (see n. 1 above, on
1,2). As for the artefacts themselves, he believes they are substances only in virtue
of the natural substances underlying them: Mant. 121,17-19; cf. Arist. Phys. 2.1,
192b16-23. See n. 51 on 6,2-3 below.
37. [5,2-3] 3,9-11.
38. [5,1-4] Quaest. 1.21, 35,6. The Aldine edition reverses the natural order of
each of these inferences, using therefore (ara) instead of because (hoti).
39. [5,5] A substantial question of interpretation turns on how we read the
second occurrence of the Greek kai in 5,5. If we take it (A) in its most common
sense, as in my translation above, to mean and or as well as, then heat and
dryness together with the lightness that emerges above them (see n. 40 on 5,5-6
below) are the form of fire. Alternatively, we might take it (B) as epexegetic or even
corrective, as AD seem to, to mean or rather (anzi), with the result that it is not
hot and dry, properly understood, which are the form of fire, but just lightness (AD
xii, 110; cf. Accattino 1995, 187-8). In favour of (A), one might point to the fact that
matter is said to underlie them, using the plural pronoun toutois (5,7), which
suggests that lightness is either part of the form of fire or just a form of fire (as 5,11
might also be understood). On the other hand, one might find some evidence for
(B) in the argument at 7,14-21 that the form and nature of simple bodies must be
simple and responsible for a single, simple change, such as upwards motion
(Accattino 1995, 190).
A serious difficulty for (B) is that lightness, like the two elemental qualities,
belongs to elements other than fire: just as heat also belongs to air and dryness
also belongs to earth, so lightness belongs to air as well as fire (Mant. 126,25-8;
Mixt. 4, 218,2-5; in Metaph. 259,17; cf. DA 8,22). But then lightness cannot by itself
be the form of fire. It is only the combination of heat and dryness, and thereby
lightness, which is distinctive of fire. (It may thus be significant that Alex. is
non-committal about the precise identity of the form of fire at 20,25-6.) To adopt
(A), though, we must take Alex.s argument at 7,14-21 as claiming either (i) that
the nature of fire is capable of a single change in the narrow sense that it is capable
of only a single motion in space, or (ii) that each form or part of a form is simple
and responsible for a single change, so that lightness would be responsible for
motion upwards, heat for heating, and dryness for drying.
It should be noted that Arist. himself distinguishes lightness and heaviness
from the two fundamental contrarieties of hot and cold and of moist and dry, which
80 Notes to page 34
constitute the forms and principles of bodies and thus their differentiae as bodies:
GC 2.2, 329b6-17; 2.3, 330a30-b7; DA 2.11, 423b27-9; cf. Meteor. 1.2, 339a13-14;
PA 2.2, 648b9-10. He regards lightness and heaviness, in contrast, as conse-
quences of the elemental qualities (akolothousin, PA 2.1, 646a16-20), from which
they derive or trace back to (anagontai, GC 2.2, 330a24-6; cf. 329b32-4). On both
of these points, Alex. agrees: for hot, cold, wet, and dry as elemental qualities, see
Mixt. 13, 229,30-230,5; Mant. 127,3-5.12-17; in Meteor. 4,25-5,1; for lightness and
heaviness as consequences of these, see DA 58,26-59,1 in addition to 5,5-6 with
accompanying notes; also cf. in Meteor. 14.18-19.
Arist. himself does not, however, draw an analogy between the elements form
and the soul, as Alex. does here. Gottschalk (1986, 253-4; cf. 1987, 1113-14) argues
that the Peripatetic Andronicus of Rhodes (first century BCE) set a precedent here,
but Accattino (1995, 187 n. 13) is rightly more circumspect.
40. [5,5-6] Literally, is generated from and above them (ek } kai epi }
gennmen); cf. 24,23-4.28; 26,27-9. This pregnant use of the preposition above
(epi) recurs at key points throughout the treatise: see the Greek-English index,
under gennasthai epi and also epigignesthai (on which, see n. 229 on 25,2 below).
I have chosen the language of emergence here, not only because it fits well with
the spatial imagery, but because Alex. uses it, as nineteenth- and twentieth-cen-
tury emergentists do, to advance a non-reductivist position (against Accattino
1995, 187, 199-200; AD 110). Although certain higher-level properties, such as
lightness, depend upon and covary with lower-level properties present beneath it,
higher-level properties are nevertheless distinct, new causal powers. This is espe-
cially clear in the case of lightness, the power to move upwards, which is not
identical to the power to heat or to dry or to a mere combination of the two, even
though it is always found together with heat. This nuanced combination of
distinct new causal powers, which nonetheless depend on and covary with lower
powers beneath them is crucial to his characterisation of the soul as something
which emerges from the specific blend and composition of the body, but is not
identical to it, in virtue of having distinct powers. In general, the higher-level form
does not belong to the underlying bodies from which the compound arises (Quaest.
2.24, 75,24-6). See n. 90 on 10,17-19, n. 92 on 10,24-6, and n. 229 on 25,2 below.
Several centuries later, John Philoponus will draw an even more fine-grained
distinction between those characteristics that merely supervene (epiginesthai) on
other characteristics underlying them and those which follow (hepesthai) or are
the result (apotelesma) of them. He takes these last two relations to be stronger
than mere supervenience: what follows from something is explained by it or is even
reducible to it. Philoponus accepts what he calls supervenience, while rejecting
these stronger notions: see Sorabji 2003, 156-8, with updates in notes in 2004a,
1.193-5 and 199-204 and in the new introduction to his [1987] 2010, 33-4; see also
Berryman 2002. For translations of the key texts from Philoponus, see Sorabji
2004a, 1.199-203. But Philoponus uses supervene differently than Alexander, in
so far as he does not require covariance: Philoponus argument at in DA 51,13-29,
for example, assumes that it is possible to have different psychological states,
while holding the total bodily state fixed. As both Sorabji and Berryman acknow-
ledge, the supervening qualities are not a direct function of the changes in the
underlying mixture for Philoponus but are underdetermined by it.
Alex., however, does not distinguish between supervenes and follows in the
way Philoponus does (pace Sorabji), but frequently uses these terms as equiva-
lents. In claiming that higher-level forms supervene on the mixture of underlying
materials, Alex. accepts covariation and so supervenience in the contemporary
Notes to page 34 81
technical sense of the term, as I have been using it, unlike Philoponus. He is fully
explicit on this point with regard to the soul: the different types of soul covary with
different types of body (summetaballousin), the type of soul being a consequence
(epkolouthsan) of the underlying constitution (sustasei) and blend (krasei), from
which it follows (hepetai, Mant. 104,27-34; for further discussion of this text, incl.
its authenticity and relation to Alex.s DA, see n. 92 on 10,24-6 below). But Alex.
also rejects reduction, maintaining the explanatory primacy of the supervenient
form: see n. 90 on 10,17-19 below; cf. n. 92 on 10,24-6. He thus accepts a stronger
relation between body and soul than Philoponus does, while also rejecting Galens
still stronger, reductionist view.
Alex.s ultimate position about the soul and the body, though, is just an instance
of a more general view about the relation between matter and form. As he states
below, differences between forms follow (hepesthai) quite generally from differ-
ences in the underlying matter on which they supervene (tn ginomenn ep autais,
10,24-6); and in his commentary on the Topics, he discusses a specific case at
length. The white colour of snow, which supervenes (epiginomenon, in Top. 51,3)
on underlying material conditions, which he elaborates in great detail (50,18-29),
follows necessarily from them (ex ananks parakolouthein, 50,20; ex ananks
hepetai, 50,23; epakolouthoun, 50,30; cf. epakolouthounta, 50,10). But while he
regards the underlying material conditions as part of the essence of snow, its white
colour is only an extrinsic feature (50,29-51,4; cf. 50,9-11). So in this case it is clear
that while the snows colour is a necessary consequence of underlying conditions
and inseparable from them, it is nevertheless a distinct and irreducible feature.
One of Alex.s main points in the passage above is that this non-reductive,
layered metaphysical conception is not peculiar to life and consciousness, but
something present throughout nature, right down to the chemical and physical
properties of the elements, much as modern emergentists later claimed. See my
1997, esp. 309-19 and 347-54; cf. also Accattino 1995.
41. [5,8] Todd notes (1974, 212 esp. n. 20) that while the term rendered here as
receptive (epidektikos) is not itself found in Arist., it is plausibly based on GC 1.4,
320a2-5 (which does in fact use dektikos).
42. [5,8-9] Arist. holds that the four elementary bodies can all be transformed
into one another, if not directly, then through a cycle of changes: see esp. GC 2.4-5;
Cael. 3.6.
43. [5,11-15] Alex. here sets up a crucial analogy for his view of the soul: that
although it is a power for change, it does not undergo change itself because it is
inseparable from the body (21,22-24,17). He explicitly appeals to an analogy with
heaviness at 22,7-10 and 18-20, though he takes this characteristic to belong to
forms in general (22,14-15; cf. Phys. 5.1, 224b5 and 11-12).
44. [5,16] Although commonly rendered cause, the Greek aitia and its cognate
aition are closer to the English because: an aitia is something cited in an
explanation as responsible for a certain state of affairs, its ground or explanans. In
many cases, this is a cause; but not all explanations are causal explanations. I have
chosen to stay closer to the verb from which the Greek term is derived, aitiasthai,
to hold responsible or accountable, even when causes are involved, to show that
this broader explanatory notion is at issue.
45. [5,17] The Aldine edition again reverses the natural order of this inference,
using therefore (ara) instead of because or for the following reason (hoti).
46. [5,19] 3,26-7.
47. [5,20] Arist. GC 2.2, 329b6-10; DA 2.11, 423b27-9.
48. [5,20-1] 3,28-4,4.
82 Notes to pages 34-35
49. [5,21-2] 4,7-9.23-5. Cf. 17,11-12.
50. [5,22-6,2] 4,20-5,1.
51. [6,2-3] See esp. Arist. DA 2.1, 412a6-12; 2.2, 414a14-16; Metaph. 7.10,
1035a1-4; 7.13, 1038b2-3; 8.1, 1042a26-31; 8.2, 1043a26-8; 12.3, 1070a9-13. By
restricting this doctrine to natural substances here, Alex. goes somewhat beyond
Arist. (though 412a11-12 might suggest it). Alex. unambiguously denies that
either artefacts or the forms of artefacts are substances in their own right, but only
in so far as natural substances underlie them: cf. 5,1-2 and Mant. 121,17-19.
52. [6,3-4] Mant. 121,23-5; 122,4-15. Note the strong explanatory priority Alex.
assigns to matter and form. It is not simply that parts of a compound substance
must be substances themselves; the latter fact is what explains the former. On the
principle that the parts of substances are substances in Alex., see Rashed 2007a,
ch. 2. Arist. has reservations about this principle, at least if stated without further
qualification: Metaph. 7.13, 1039a3-8 (though compare a14-19); 7.16, 1040b5-16,
1041a4-5. There are nonetheless traces of it in the corpus: cf. An. Pr. 1.32, 47a24-8;
Phys. 1.6, 189a32-3. On the last two passages and Alex.s comments on them, see
Rashed 2007a, 51-2.
53. [6,4-6] Elsewhere Alex. only claims that the forms of artefacts are qualities
and not substances (5,2; Quaest. 1.21, 35,6); here artefacts themselves are said to
be qualities. Having explicitly restricted substantial forms to natural objects, he
cannot regard the artefacts as substances in their own right (5,2), but at most in
an extended sense in virtue of their materials (Mant. 120,5-7, 121,17-19).
54. [6,6-7] The worry Alex. considers here is triggered by his conclusion at 5,18-19.
He need not have the Stoics exclusively in mind here, though they certainly fall within
the target zone, since they take the two principles of bodies, matter and divine reason,
to be bodies themselves (D.L. 7.134 = SVF 2.299 = LS 44B 3, retaining the MSS reading
see LS 1.274, 2.266). See also Alex.s criticism of the Stoic mereological principle that
every part of a body is a body at 18,10-27 below.
55. [6,8] Here tode ti does not designate an individual, but just a determinate
kind of thing in any category. See n. 34 above on 4,24.
56. [6,8-10] Arist. Phys. 1.5, 188b21-6; 1.8, 191b13-16; GC 2.4, 331a14-16.
57. [6,12-13] Mant. 121,29-30.
58. [6,13-14] Arist. GC 1.3, 317a32-b5; Cael. 3.2, 302a3-7.
59. [6,14-15] Quaest. 2.19, 63,8-28; cf. in Metaph. 103,8; 158,7-13; 159,10-26;
213,2-17; in Meteor. 6,15-16; 78,11-12. Also Arist. Cael. 1.10, passim; 2.1, 283b26-
284a13; cf. 2.14, 296b33. The Stoics accept that there is always body, since god is
eternal and a body (see n. 54 on 6,6-7), even though they believe the world-order
(diakosmsis) is perishable: Philo Aet. 9, 75.14-76.3 Cohn-Wendland = SVF 2.620;
cf. 131, 113.5 = SVF 1.106.
60. [6,19] 4,9-15.
61. [6,6-20] See the closely parallel argument at Mant. 121,31-122,4. Alex.s
argument has the structure of a constructive dilemma: either there was a time
when bodies first came about or there have always been bodies; but either way, he
argues, bodies come from non-bodily items in some sense. But Alex. is not indiffer-
ent between the two. Alex. does not himself accept the first horn, but the second:
there have always been bodies. But he introduces a key distinction: since each body
comes from another body, it does not come from what is not a body tout court, but
only from what is not the particular sort of body it is. In this way, he can accept
the general principle, while avoiding the absurdity of the first horn, of non-bodily
items existing separately on their own. Bare matter and form are only abstractions
and never exist apart.
Notes to pages 35-36 83
As 5,18-19 shows, however, Alex. is still committed to non-bodily items, even if
they do not exist separately. His counterargument above relies on a diachronic
model throughout: Alex. focuses exclusively on the pre-existing items bodies are
produced from (ek). But the initial objection can equally be raised on a synchronic
model: someone might worry whether bodies can be composed from (ek) non-bodily
items like form and matter, and ultimately prime matter, at a single instant. Alex.
believes that they can and are, but does not defend it here.
62. [6,21-2] in Metaph. 20.5-6. This claim about the central role of form is
extended below to the form of compound bodies (7,4-5), and eventually to the souls
of living things. BD nicely observe (238) that this general principle underlies the
claim made in the opening argument that the soul is that in virtue of which each
human is a human (2,1) and so provides the key to self-knowledge.
63. [6,28] Mixt. 13, 229,30-230,2.
64. [6,30] Culmination (teleiots) is one of Alex.s key technical terms in On the
Soul. Arist. himself uses it only twice (Phys. 3.6, 207a21; 8.7, 261a36), though the
idea is completely orthodox. It derives from telos, the Greek word for end or goal,
and thus signifies the full realisation of some given kind of thing, closely related
to entelekheia: see esp. 16,5-7. The alternative rendering teleiots as completion
captures its etymology, but understates the normative role of the end that is so
crucial to Aristotelian teleology, while rendering it as perfection overshoots the
mark and introduces irrelevant theological overtones. On Alex.s use of this term,
see Todd 1974, 213-14.
65. [6,30-7,2] in Metaph. 349,14-16.
66. [7,2] Arist. Metaph. 9.8, 1050a15-16; cf. 5.4, 1015a10-11; 5.24, 1023a33,
together with Alex.s brief comments, in Metaph. 359,30-2 (cf. 21-2) and 422,16-17.
67. [7,6] in An. Pr. 46,2-16; 29,1-30,6.
68. [7,9-13] Mant. 116,18-31. Arius Didymus (first century BCE), in his sum-
mary of Aristotelian physics, speaks similarly of bodies making and being modified
by non-bodily powers (tais asmatois dunamesi, fr. 5, Dox. Gr. 449.12-14 = Stob.
Ecl. 1.141.19-22), though Kupreeva 2003 has questioned the dates for this evidence
(309 n. 40). And while Arist. himself does not contrast bodies and their non-bodily
items or powers in just so many words, he does come reasonably close to it at Sens.
4, 441b12-15: nothing is naturally such as to act or be modified in so far as it is fire
or earth, nor does anything else in fact; everything acts or is modified just in so far
as there is a contrariety in it. In his commentary, Alex. construes this as a claim
about bodies in general (in Sens. 73,10) and portrays the Aristotelian view as
positioned midway between the Stoics and the Platonists: For Arist. does not think
that bodies act or are modifed, as the Stoics maintain, or on the other hand that
non-bodily items do, as the Platonists thought, but rather [bodies do so] in virtue
of the non-bodily contrarieties in them (73,18-21). Bodies thus act in virtue of their
features or powers, which are not themselves bodies, much as someone today
might claim that bodies act in virtue of their properties. Both of the competing
views are thus in a sense right and in a sense wrong. When Alex. speaks of Platonists,
he might have in mind Plato himself, Phaedo 100A-105C (cf. 95E-96A), or still
more likely Arist.s critique of Plato at Metaph. 1.9, 991b3-9; GC 2.9, 335b7-24. The
Stoic claim that only bodies can act or be acted on is central to their ontology: Cic.
Acad. 1.39 = SVF 1.90 = LS 45A; Aet. 1.11.5, Dox. Gr. 310.6-7 = SVF 2.340 = LS
55G; Plut. Comm. Not. 30, 1073E = SVF 2.525 (cf. 40, 1080F); Sext. Emp. Adv.
Math. 9.211 = SVF 2.341 = LS 55B; Stob. 1.138.14-139.4 = SVF 1.89 & 2.336 = LS
55A; Nemes. Nat. Hom. 2, 20.14-17, 21.6-9 Morani = SVF 1.518 = LS 45C; Euseb.
PE 15.14.1, 378.19-379.2 = SVF 1.98 = LS 45G; cf. Sext. Emp. Adv. Math. 8.263 =
84 Notes to pages 36-37
SVF 2.363. On the dispute between Alex. and the Stoics on this point, see
Kupreeva 2003, 307-15.
69. [7,18-20] Arist. Cael. 1.2, 268b26-269a1, a25-7; 3.3, 302b6-8; 3.4, 303b4-5.
On different possible construals of the argument at 7,14-20 see n. 39 on 5,5 above.
70. [7,22] 3,24-7.
71. [7,23] 8,9-10. The comparative teleioteros, here rendered by the phrase
more advanced, can be used to indicate not merely whether something is fully
developed, and so superior in comparison with members of its own kind, but also
whether it is superior across kinds, as here. Similar remarks apply to the semantic
range of the superlative and the various privative forms: see the Greek-English
Index under teleios and atels. For the normative force of teleological notions, see
n. 64 on culmination (teleiots) at 6,30.
72. [7,24-8,1] Note that while the higher-level forms of compound bodies in
some sense depend on matter, they do so via form: the lower-level forms of the
underlying bodies, which serve as proximate matter, contribute (suntelei, 7,24;
8,11) to the higher-level form and so help to explain its complexity and sophistica-
tion (8,13-15).
73. [8,3-5] A single simple body that is, one of the four elements can serve
as the (pre-existing) matter for a different body, by being transformed into it. But
such transformations only result in a different element and thus another simple
body, because they only involve an exchange of elemental qualities (GC 2.4-5,
passim; cf. 2.7, 334a16-26). To get a compound body, therefore, more than one type
of body is needed. Arist. actually goes further, arguing that every compound body
ultimately includes all four elements: GC 2.8, 334b31-335a9.
74. [8,10-12] Alex. begins to lay out the ways in which the nature of a higher-
order form depends upon the underlying matter, in particular on the forms of the
underlying bodies: see esp. 8,15-17; 10,17-19 and 24-6; 11,11-13; and 23,29-24,4,
with their corresponding notes.
75. [8,12-13] Arist. uses this phrase, but to describe the understanding (nous),
which is capable of receiving all forms, and so is a form of forms (in the way that
the hand is a tool of tools) at DA 3.8, 432a1-3. Alex.s point is quite different: as
the form of proximate matter, the higher-level form is in a sense the form of the
lower-level forms so combined and organised.
76. [8,16] This is the first use of the notion of a distinctive mixture (mixis) or
blend (krasis) that will become central to Alex.s explanation in the pages that
follow. See n. 90 on 10,17-19. Elsewhere, Alex. distinguishes between mixtures and
blends, the latter being a species of the former: blends are those mixtures where
the ingredients are not preserved separate from one another, but come to have a
single underlying substratum (Mixt. 13, 228,27-36; cf. Arist. Top. 4.2, 122b30-1);
see Joachim 1903, 73-4. In the DA, Alex. typically uses mixture in conjunction
with blend to signifiy the species.
77. [8,15-17] The different character of higher-level forms depends upon differ-
ences in the underlying matter, specifically their forms. See 10,17-19 and 24-6;
11,11-13; and 23,29-24,4, with their corresponding notes.
78. [8,17-18] Alex. clearly has in mind the matter of a simple body or element
here, that is, prime matter, since a compound body will have not one, but several
bodies as its matter, which would contribute to its difference (8,5-13).
79. [8,17-25] Alex.s argument is that in the base case of elemental bodies there
is already a striking amount of diversity, which is due solely to the different
pairings of elemental qualities, since the matter in this case is entirely without
forms of its own and so makes no distinctive contribution. In the subsequent cases
Notes to pages 37-38 85
of compound bodies, the matter consists of several types of body, which will also
contribute forms of their own and so will only add further to the diversity of the
bodies that result. Arist. speaks of couplings (suzeuxeis) of elemental qualities in
elemental bodies at GC 2.3, 330a30-b7.
80. [9,1-4] Mant. 105,10-12 (cf. 103,36-8); Arist. DA 2.2, 413a25-31. The sim-
plest living things, plants, differ from simple bodies in their causal powers. The
three cited here the power to eat and be nourished, to grow, and to reproduce
are the distinctive powers of the soul for nourishment (threptik psukh). See
below at 9,11-14 and then at greater length at 29,1-10 and 31,7-38,11.
81. [9,4-7] The different forms of plants are explained by the different blend of
materials underlying them. This strong explanatory claim presupposes a covaria-
tion between plant forms and their matter: there is no difference in the type of
plant without some corresponding difference in the type of material blend. The
former types supervene on the latter.
It is surprising that throughout this discussion Alex. does not mention differ-
ences in structure and in anhomoeomerous or non-homogenous parts, which are
mentioned only at the end of the treatise (98,4-5), but are prominent elsewhere
(e.g., in Meteor. 224,24-225,16 and 227,15-22, where he signals the subsequent
discussion of this topic in PA 2). Taking structure into account would not, I think,
fundamentally alter Alex.s emergentist strategy (see n. 40 on 5,5-6), but it would
complicate the detail considerably, and he may have left it out of his presentation
for conveniences sake. (His remark about parts at 10,28-9 is non-specific.)
82. [9,12] For both Alex. and Arist. this claim represents the core of the common
concept of soul or psukh: it is the principle of life (arkh tn zin, Arist. DA 1.1,
402a6-7), that in virtue of which living things are alive and perform the activities
distinctive of living things (2.2, 414a12-13), whatever that in fact turns out to be.
Different theories differ over just what that is and its precise nature. As BD point
out (240), this is the first mention of soul since the beginning of the treatise.
83. [9,12-14] 78,5-6; 92,18-20; Arist. DA 2.2, 414a12-14. AD rightly stress (117)
how with these assumptions, Alex. moves from speaking merely about the forms
of complex bodies like plants to souls proper.
84. [9,14-22] Here again Alex. argues that the principles Arist. applies to the
soul are just instances of a broader pattern, which is already at work at the most
basic level of elemental bodies. The study of the soul is to this extent continuous
with the rest of natural science, a point he makes again later in his treatment of
the soul using the example of the heaviness of the element earth (23,29-24,3; on
heaviness as a form, cf. n. 39 on 5,5 above). He appears to have made a similar
point in his lost commentary on the de Caelo: see Simpl. in Cael. 380,29-381,2.
In the present case, he is concerned with the distinction between potentiality
and actuality, that is, between a power (dunamis) and its exercise or activity
(energeia). Alex.s reading is notable for its unambiguous, emphatic identification
of form with the power, something which is at best implicit in Arist.s view. In this
way, Alex. gives explanatory primacy to powers, esp. causal powers, even though
powers are defined by reference to their exercise (contrast Arist. Metaph. 9.8,
1050b2-6). Alex. develops the point here as a consequence of Arist.s view that
while a power and its exercise can both be viewed as stages of completion or
entelekheia, the soul is to be identified as the first completion (DA 2.1, 412a21-8;
cf. Alex. Mant. 103,10-20). Alex. frames it here as a metaphysical principle which
holds of forms and powers quite generally, substituting his own favored term,
culmination (teleiots), for completion. He links the two terms explicitly below at
16,4-10.
86 Notes to page 38
85. [10,3] 2,26-3,2; 3,23-5.
86. [10,7-8] 6,24-7,8.
87. [10,8-10] The claim that perception distinguishes animals from other living
things and so is a part of their definition is fundamental in both Arist. and Alex.
For Alex., see 29,14-15; 40,14-15; 78,6-7; 92,21-3; Mant. 105,12-13; Quaest. 2.27,
77,14-15; in Sens. 3,19-20; 9,1-2; in Top. 17,9-11; 160,20-1; 376,11-15; 381,17;
394,6-7; 407,23-4; as well as the formulaic definition of animal as perceptive
animate substance (ousia empsukhos aisthtik): Quaest. 1.11, 21,25-6 (= 23,27-9);
22,9-12 (= 24,11-14); in Metaph. 161,8-9; in Top. 93,1-3; 124,1-2; 355,21-4; 405,2-5
(commenting on Arist. Top. 135a16); 407,23-4; 432,1; 463,11-15; 493,3-4. Cf. also
DA 30,13; 66,5; 67,2; 93,17-18; in Sens. 5,21-4; 8,2-3; 9,1; in Top. 16,23-4; 89,16-17;
391,6-7; in Metaph. 198,29-30; 315,33-4; 316,18-19.
For Arist., see DA 2.2, 413b1-4; 3.12, 434a27-31; Sens. 1, 436b10-12; Somn. 1,
454b25; Iuv. 1, 467b24-5; 3, 469a18-20; 4, 469b4; PA 2.5, 651b3-5; 2.8, 653b22-3;
3.4, 666a34; GA 1.23, 731a30-b4; 2.1, 732a12-13; 2.2, 736a29-30 (cf. b1); 2.5,
741a9-10; 5.1, 778b32-3; Metaph. 1.1, 980a27; EN 9.9, 1170a16; cf. Top. 5.2,
129b26; DA 3.2, 427a14-16; 3.13, 435b5-7, b16-17; HA 1.3, 489a17-18; 4.8, 533a17-
18; PA 2.1, 647a21; 2.8, 653b22-3; 2.10, 656a3-4; 3.4, 666a34; 4.5, 678b2-4; GA 3.7,
757b15; 5.1, 778b32-3; Metaph. 7.10, 1035b14-18; EN 6.2, 1139a20; though com-
pare the caveats at HA 8.1, 588b17-18 and PA 4.5, 681a19-20.
88. [10,12-14] Throughout On the Soul, Alex. frequently appeals to analogy
(see under analogia and analogon in the Greek-English Index). But in this pas-
sage, he also intends it to carry the quantitative sense of a proportion, where the
central terms are identical: in a series of three terms, A : B :: B : C, as regards how
advanced (teleios) each is. The golden ratio is an example of this kind of proportion.
Underlying this proportion is a deeper analogy, though, concerning the relation
that holds between the higher-level forms at each stage and the lower-level ones
directly beneath them: each new level emerges from and is in some sense deter-
mined by the previous one (see n. 90 on 10,17-19 below).
89. [10,14-17] Alex., much like Arist., thinks that the relation between the souls
of different animals replicates the larger relation that holds between animals and
plants: but where Arist. is only concerned with a kind of independence (whether
one set of powers can be found instantiated without the others), Alex. makes a
much stronger, quantitative claim about the proportionality that holds between
them. It is not clear exactly which differences Alex. has in mind here, however, and
so just how extensively this proportionalilty is supposed to hold: some animals
have only the sense of touch, others have more than one sense, and some have all;
some animals can move, while others cannot; some have memory, while others lack
it.
90. [10,17-19] The nature of these differences is in some sense to be accounted
for (aitia) by differences in the underlying bodies, specifically their number,
mixture, blend, and formation (kata te plthos kai kata poian krasin te kai mixin
kai sustasin): see 8,15-17.22-5. This claim presupposes that the higher forms
covary with these underlying factors: given a certain mixture or blend of bodies, a
certain higher-level form follows or is a consequence, something stated explicitly
below at 10,24-6. If so, these higher-level forms supervene, in the technical sense,
on the mixture of underlying bodies: which higher-level forms occur is a function
of the blend, mixture, and formation of the underlying bodies. Cf. 8,10-12; 11,11-
13; 23,29-24,4; 25,2; 26,22.
Someone might worry whether this dependence on matter conflicts with the
explanatory primacy Alex. assigns to form and that the bottom-up account he
Notes to page 38 87
introduces here clashes with his top-down approach elsewhere (e.g., 7,9-14). But
there is in fact no contradiction. The matter which is responsible for higher-level
forms is always a body of some sort and so has a form of its own. All bodies differ
from one another in virtue of their forms, moreover, and Alex.s claim is simply that
such differences at lower levels have consequences at higher levels, since higher-
level forms covary with certain kinds of underlying bodies and hence with their
forms and thus supervene on them in the technical sense (Mant. 104,27-34 and also
discussion in n. 92 on 10,24-6 below). In so far as these bodies underlie the
higher-level compound, they are material factors of it. But it is still their forms that
play the key role. Contrast this sort of case with prime matter, which is without
any determination or form of its own. It is precisely because prime matter can
equally occur with any given form or without it that no form covaries with prime
matter and so none supervenes on it in the technical sense. Even the simplest case
of supervenience presupposes lower-level forms that determine the matter, as
when lightness emerges from the heat and dryness of fire (5,4-6); neither heat nor
dryness, in contrast, emerges from or supervenes on prime matter. The only
matter that can be responsible for higher-level differences is matter that has a
form of its own and is therefore already a body.
91. [10,19-20] Though proverbial (Corpus paroem. gr. Diog. 2.97; cf. App. prov.
1.41 Leutsch-Schneidewin), this saying occurs prominently in the methodological
remarks that conclude Arist.s function argument at EN 1.7, 1098b7; cf. Polit. 5.4,
1303b29-31. But the closest parallel may be SE 34, 183b22-6, which develops the
point most fully and uses the superlative megiston, rather than the comparative.
Ian Flora points out to me that the superlative can also be found in Plato Rep. 2,
377A (cf. Laws 6, 753E6-754A2), which Arist. himself may have in mind.
92. [10,24-6] It is because matter is an arkh, an origin and principle, of the
difference in forms that it is responsible (aitia) for these differences, as Alex.
claims above at 10,17 (see accompanying note). The difference in higher-level
forms is here clearly said to be a consequence (hepesthai) of differences in the
matter. These need not be causal consequences of the matter, but simply a
question of covariation: necessarily, whenever there is matter of a certain kind,
there is a higher-level form of a certain kind there cannot be any difference in
the forms without some difference in the underlying matter. In light of this, AD
are wrong to claim (115) that there is absolutely no passage in On the Soul where
matter is said to be a sufficient condition for organic form (and likewise BD 33-4),
against Moraux (1942, 41). Both sides wrongly assume that such sufficient condi-
tions would entail a very strong form of reductionist materialism. But an
emergentist strategy like Alex.s can accept sufficient conditions while rejecting
reductionism. See n. 40 on 5,5-6.
That Alex. does regard the underlying matter as a sufficient condition in the
present passage is confirmed by a parallel passage, Mant. 104,27-34: the bodys
blend (krasis) is responsible (aitia) for the souls coming into being in the first place
(104,28); the difference between souls is a consequence (epkolouthsan) of the
specific formation (sustasei) of the body and one covaries with the other (sum-
metaballousin alllois, 30-1); and the difference in souls is a consequence (hepetai)
of the specific blend of the body (33). For further discussion, see my 1997, 348-9.
Some have worried (e.g., Sorabji 2003, 156 and 2004a, 193, 194-5, following
Sharples) that Mant. 104,28-34 diverges from the view in Alex.s DA and might not
even be written by Alex. But the concern is unfounded, as the two texts do not differ
on these key points at all. Alex. states equally in the DA that higher-level forms,
including the soul, follow (hepesthai, 10,25; 26,29) from the underlying blend and
88 Notes to pages 38-39
mixture, which is therefore said to be in some sense responsible (aitia, 10,17) for
the higher-level form and an origin of it (arkh, 10,24).
93. [11,1] I have used organic here for the Greek organikos. One might worry
that this rendering is a false friend: the Greek organon, from which it derives, is
used not only for physiological organs, but for implements, tools, instruments, and
anything quite generally that serves as the means for some specific function or end.
There is a dispute, moreover, about its use in Arist.s definition of the soul in DA
2.1 as the form of a natural body (412a28-b6), where it has variously been taken to
signify that the body either (a) has organs or (b) is itself organ-like or tool-like. (For
discussion, see Menn 2002, 108-11.) Alex. clearly favours the first reading, (a). He
characterises the organikon body as one that has many different parts able to
subserve the soul powers (16,11-12; Quaest. 2.8, 54,9-10) and explicitly rejects the
view that the soul uses the body as an organ or tool (hs organi) at 23,24-8. He
takes Arist.s use of the word in the definition of soul as an elaboration of the
earlier phrase capable of having life (Quaest. 2.8, 54,6-9.11), thus bringing it even
closer to ordinary connotations of the English organic. I have accordingly trans-
lated organon as organ, which Alex. uses almost exclusively for physiological
organs in On the Soul. (For the rare exceptions, cf. 35,2 and 96,8.)
94. [11,2] Alex. is gesturing broadly at elements of Arist.s definition of the soul
at DA 2.1, 412a19-b6.
95. [11,5-6] Alex. returns to the strategy he advanced at 2,10-25: that once we
understand how marvellous nature is, we will no longer be puzzled as to how
bodies can be endowed with the powers characteristic of living things. See n. 97
below on 11,11-13.
96. [11,7] 5,11-15.
97. [11,11-13] Alex. extends the proportion between different kinds of forms to
the matter underlying each of them, because of their dependence on this matter:
if the difference between forms is a function of the difference in the underlying
matters composition and make-up (7,21-8,25; 9,4-7; 10,17-19.24-6), then the mat-
ter of more advanced forms, such as the souls of animals, will itself be more
complex and advanced than the matter of less advanced forms, such as the souls
of plants. Instead of positing a non-bodily soul separate from all matter, whose
powers are more divine and greater than those of any body (cf. 2,17-18), Alex.s
solution to the earlier aporia is to appeal to a layered conception of nature, where
increasingly complex forms depend in some way on the increasingly complex
bodies that serve as their matter. What is important about these underlying bodies
are their forms and powers, which depend, in turn, on the simpler bodies underly-
ing them, whose forms and powers depend on the bodies underlying them, and so
on again until we ultimately reach the elemental forms and prime matter. At each
level, new forms emerge as a result of the complexity and organisation of the forms
of the bodies underlying them.
98. [11,14] AD suggest (122-3) that the reason Alex. does not make this claim
about living things (ta znta) generally, but restricts himself here to animals (ta
zia), is that he is arguing dialectically from what has been agreed upon (homolo-
geitai), since the Stoics deny that plants have a soul: Mant. 105,6-12; 118,12-14 =
SVF 2.711 (cf. 118,19-21); cf. DA 31,25-32,1 below. See also Galen Intr. 14.726.7-11
Khn = SVF 2.716 = LS 47N; Philo Leg. Alleg. 2.22-3, 1.95.7-20 Cohn-Wendland =
SVF 2.458 = LS 47P.
99. [11,15] Throughout this treatise, I have rendered the Greek construction
esti ek tinos literally, is from something or is out of something as is composed
out of . Arist. distinguishes different senses of from or out of something in a
Notes to page 39 89
number of passages (Metaph. 5.24, passim; cf. also 2.2, 994a22-b3; 8.4, 1044a23-5).
But for the most part his discussion does not line up with Alex.s here, which
exclusively concerns different kinds of composition. (Arist. Metaph. 14.5, 1092a21-
35 is closer in this regard, but not much.) Alex.s discussion has more to do with
the distinction between different types of mixture or combination that Arist.
discusses in GC 1.10, esp. his distinction between composition (sunthesis) and
mixture (mixis) at 328a5-15, corresponding to (a) and (b) above. Alex. refers to the
first case as juxtaposition below (parathesis, 12.2). He explicitly links juxtaposi-
tion and composition at Mixt. 13, 228,28-34.
But Alex. is not canvassing different types of composition here just for the sake
of completeness. His strategy is more pointed. The three kinds of composition
directly correspond to three theories of the souls relation to the body: atomists like
the Epicureans think that soul atoms are juxtaposed with other atoms in the body
(a); the Stoics hold that there is a total blending of the two (b); and the Aris-
totelians make the soul the form of the body (c). Afterwards, Alex. will make an
additional argument (12,7-13,8), directed against substance dualism and so effec-
tively against the Platonists, to round out his critical survey.
100. [11,15-17] This sort of mixture, where the ingredients are preserved intact
in their original form (like rice and beans), is what Alex. himself calls juxtaposi-
tion (parathesis) below at 12,2; see also his use of this term to characterise one of
the Stoic forms of mixture at Mixt. 3, 216,17-22. For Epicureans and atomists
generally, all mixture is of this type, including the body and soul: Epic. Ep. ad Hdt.
63-6 = LS 14A 1-5; Lucr. 3.425-58 = LS 14F 2-3.
101. [11,17-18] 15,5-8; Mant. 116,7; in Metaph. 334,31-4; cf. also DA 13,22 and
Mant. 116,12, which use the example of honeyed wine; also cf. Arist. Metaph. 8.2,
1042b15-17. Alex.s use of these examples and the terminology suggest the Stoics
view, esp. in the Mant. passage: the whole soul pervades the whole body and is
blended with it (hs holon di holou dikein kai kekrasthai ti smati, 116,5-6); for
the attribution to Chrysippus, see Mixt. 4, 217,32-218,1 (= SVF 2.473, p. 155, ll.
24-30 = LS 48C 10) together with 3, 216,4-14. But the view Alex. goes on to
elaborate in the present passage differs in key respects: see n. 102 on 11,18-20.
Both of these views, it should be noted, hold that the animal is a blend of soul and
body, which is quite distinct from the view that the soul itself is a blend of the
bodies underlying it, as Galen claims in Quod animi mores. Regarding the latter
view, see n. 226 on 24,18-20 below and Alex.s discussion of harmonia theories of
the soul more generally at 24,18-26,30.
102. [11,18-20] Alex.s characterisation here differs from that of the Stoics.
According to Chrysippus, the ingredients are preserved in total blending and can
be extracted from it again in fact, the possibility of extraction is a distinctive
characteristic of blending (Mixt. 3, 216,25-217,2 = SVF 2.473, p. 154, ll. 19-28 = LS
48C 4; cf. Mant. 6, 220,20-2). The case Alex. describes above, in contrast, is what
the Stoics call fusion throughout (sunkhusei di holn), in which the ingredients
perish and some third thing is created (Mixt. 3, 216,22-5 = SVF 2.473, p. 154, ll.
15-19 = LS 48C 3; cf. Philo Conf. Ling. 184-7, 2.264.23-265.15 Cohn-Wendland =
SVF 2.472). For a quick overview of the Stoic theory of blending, see LS 1.292-4
and M. J. White 2003, 146-9. For more detailed studies, see Todd 1976, 21-88;
Sorabji 1988, ch. 6; E. Lewis 1988.
Alex., however, is not trying to report Stoic doctrine faithfully here. As at Mant.
116,5-13, he blends his presentation of their view with his own criticism, drawing
out what he thinks are the real options and consequences, since he finds the whole
notion of total blending absurd: he argues that it collapses either into juxtaposition
90 Notes to page 39
or, if there is mixture throughout as here, fusion (Mixt. 7, 220,23-221,25; cf. 1,
213,2-13). In the latter case, the ingredients will not be preserved, as the Stoics
claim, but perish (7, 221,11-15.20-5). This accords with his own view, according to
which the ingredients in a blend are not preserved in actuality that would be
juxtaposition and so could in a sense be said to perish when the blend comes into
being: see 12,5-6 below, but also Mixt. 13, 228,34-6; 15, 231,15-19.25-6. (Alex.
would embrace this designation even more strikingly if we were to accept, with
Todd, Montanaris emendation phthora at 230,15 and 19, in place of the MSSs
mixis; on which, see Todd 1976, 178.) But it is critical to note that even for Alex.
this is not complete annihilation (out } pantapasin phtharei), since the ingredi-
ents are in a way preserved (szomenn): see Mixt. 14, 230,17-19 especially, but
also 231,10-12; 15, 232,13-18.28-31. Alex. adopts the orthodox Aristotelian position
that the ingredients still exist in one sense (namely, potentially) though not in
another (actually): GC 1.10, 327b23-31, 328b17-19.
103. [11,20-1] This is, of course, the Aristotelian alternative Alex. favours. See
above, 2,25-3,13 (incl. the example of the statue at 3,7-9).
104. [12,2-3] 15,15-17; 20,6-7; 21,10-11; Mant. 104,21-2; cf. 115,30-4. Alex.
assumes here that the body of a living thing is animate throughout, a point on
which certain Stoics agree (Mixt. 4, 217,36 = SVF 2.473, p. 155, ll. 25-8 = LS 48C
10; Hierocles Elem. Eth. 4.6-8 Long). Alex.s position might be thought to be a
straightforward consequence of Arist.s hylomorphism: since the soul is the form
of the whole living body (holou zntos smatos, DA 2.1, 412b17-23), one might
think that no part of the body could be without at least the most basic vital powers
of the soul. (The soul for perception is a different matter, since not all parts of the
body are sensitive: see 74,29-75,2 and 97,4-6 below, with notes.) Arist.s views
about homonymy also suggest this, as when he claims that there will be no part
[of the body] without a share of soul, except in name (GA 2.1, 735a6-8, cf. 734b23-7;
also 1.19, 726b22-4). But Arist. does not always regard this as a consequence of his
views. At the end of MA 10, he says that there need not be soul in each part [of the
body] (703a36-7); rather the other parts live by being naturally attached to the
bodys central principle, in which the soul, like a ruler, resides (703a29-b2).
Nussbaum suggests (1978, 153) that Arist. is merely denying that each part is a
separable animal with a soul of its own. But on the most straightforward reading
of the Greek, Arist. is making the stronger claim that there need not be any soul
present in certain parts of the body at all. Tracy more plausibly suggests (1983,
336-7) that the claim is meant to apply to the soul only in so far as it is an efficient
cause of change, which is the main concern of the argument in that context, and so
is compatible with his view elsewhere that the soul is the form of the entire body
while the heart is the principle or origin of the activities that distinguish living
beings as such; on this reading, Arist. is simply rejecting multiple principles (cf.
PA 3.4, 665b27-9).
105. [12,3-4] Alex. appears to be drawing a categorial distinction between the
character (poion) of an animal and its size (poson). But presumably he has in mind
the broader sense of character here, in which one can speak of the genus and
differentia of a substance as a kind of characteristic, rather than just those which
more strictly fall into the category of quality (poiots): Arist. Categ. 5, 3b13-21;
Metaph. 5.14, 1020a33-b2, b14-15; 5.28, 1024b4-6; Top. 4.2, 122b16-17 (though
compare Metaph. 7.12, 1038a19-20); Alex. in Metaph. 399,1-6; 400,28-31; 429,12-
14.17-19; in Top. 314,22-4. If so, then the argument is not merely dialectical, as AD
123 suggest, in order to win Stoic concessions, but something Alex. can and should
accept on his own terms (as one would expect if the present argument is to count
Notes to page 40 91
as a demonstration): the soul contributes not merely to the size or quality of a living
thing, but to its essence and substance as well.
106. [12,5-6] The argument here is clearly meant to run as follows: (1) if the
animal is a blend of soul and body, then the soul and body perish when the animal
comes to be; but (2) when the soul and the body perish the animal does not come
to be, but rather dies; therefore, (3) the animal is not a blend of body and soul. The
Stoics would deny (1), since on their view the ingredients are preserved in a true
blend. But Alex. rejects this account of blending and so does not concede it here
even for the sake of argument. On the differing views of blending and the dialectic,
see n. 102 on 11,18-20 above.
107. [12,8] Alex. uses the phrase itself just on its own (aut kath hautn) to
indicate that the soul, though an ousia in the way a form is (6,2-4), is not an
independent substance that can exist itself just on its own. This way of putting it
gives added emphasis to his rejection of substance dualism, since the phrase is
used first by Plato to distinguish the Form of F from the many things which are F:
the Form is just F itself on its own, taken by itself without anything else to which
it attaches. Alex. uses it here as elsewhere in the treatise to speak more generally
of whether a given thing as such can exist independently: see the Greek-English
Index, under auto kath hauto. Platonists are clearly a target here, as AD rightly
suggest (124), though Alex. need not have any specific ones in mind: they cite
Atticus fr. 7 Des Places and Alcinous Didask. 25, but neither uses the phrase aut
kath hautn. Like Arist., Alex. clearly intends the argument that follows to be a
broadside attack against substance dualism in general.
108. [12,9-13,8] Alex. expands and develops the programmatic argument from
Arist. DA 1.1, 403a3-25. The differences are instructive. Arist. starts from a puzzle
about whether there are any states or modifications (path) that belong to the soul
exclusively or whether they are all shared by the body and the soul (a2-5), and
suggests guardedly that most seem to fall into the latter category: he lists several
emotions as well as perception in general. Understanding (noein) is the one he says
is most likely to belong to the soul alone, but even that will not be separable if it
requires representation (a8-10). He then lays down the condition that the soul will
be separable if, and only if, some modification is exclusive to soul (a10-12), but
then argues that the body undergoes something concurrent with all modifications
of the soul (a16-19), concluding that these modifications are all enmattered logoi
(a24-5), using examples again drawn from the emotions.
Alex. is more forthright, comprehensive, and systematic. He begins by unambi-
guously endorsing the general thesis that all activities of the soul involve the body
necessarily and without exception, and immediately sets this within the context of
his broader naturalistic strategy, by putting it on all fours with the motions of the
inanimate elements (12,9-11; cf. 23,26-24,3). He then runs through the gamut of
activities, starting with the vital functions shared by plants (12,12-13); the cogni-
tive, emotional, and appetitive activity of animals, (12,13-19); and finally thought
(12,20-1). Regarding the latter, he shows no signs of hesitation: it falls under the
general thesis equally with the rest. In all of these cases, the activities are
described as occuring either through the body (dia tou smatos, 12,14.17-18.21;
compare Sens. 1, 436b6-7) or when the body undergoes the corresponding change
(12,13.15.16.17.19). It is not clear if the difference between these phrases amounts
to anything more than elegant variation, but it is worth noting that Arist. is willing
to use the former phrase, which if anything is the stronger of the two, with the
cognitive functions of perception, representation, and thought.
Alex. also introduces an important nuance at 12,9-10 (cf. 21-2) when he pairs
92 Notes to page 40
activities (energeiai) of the soul with the changes (kinseis) in the body, instead of
Arist.s vaguer and more neutral talk of modifications (path) common to the soul
and the body. Arist. does draw a distinction between activities and changes in
Metaph. 9.6, 1048b18-35, and it is plausibly behind his treatment of perception in
DA 2.5 as being a kind of completion and so either not an alteration or an alteration
of a different sort (417a21-b16). Although this is compatible with there being a
change in the body corresponding to the activity in the soul, the text does not
explicitly state it. Alex., in contrast, insists on both an activity in the soul and a
corresponding change in the body, and he applies this across the board to all
modifications of the soul, not just those involved in cognition, as an instance of a
general pattern that holds throughout the whole of nature down to the four
elements. This distinction is critical for Alex., who wants to uphold Arist.s view
that the soul does not itself undergo change (21,22-24,17; 78,24-79,21), while still
maintaining, as he does here, that all psychological activity involves change in the
body.
109. [12,10-11] Although Alex. could conceivably have in mind natural objects
here quite generally, since their natures are internal principles of change and rest
(Arist. Phys. 2.1), he is more likely alluding to the tendency of the four elements to
move towards their natural places: Arist. Cael. 4.3-5, passim; Phys. 8.4, 255b13-20.
Alex. identifies the natural tendencies (kata phusin } rhopai) of natural bodies
as lightness and heaviness at in Metaph. 72,34-6; cf. in Sens. 73,26-9. At one point,
he draws a general analogy between the form of living things and of the four
elements at in Metaph. 357,25-9, as the origin and principle of changes they
naturally undergo, and specifically identifies the former as souls and the latter as
tendencies; he explicitly contrasts these two principles of change at Quaest. 2.3,
49,1-4 and in the Arabic translation of a treatise attributed to Alexander, against
Galens theory of motion: see 67a14-22 (Rescher and Marmura 1965, 17).
110. [12,14-15] 76,15-78,2; Arist. DA 3.10, 433b16-21; MA 10, passim. On desire
generally, see Alex.s discussion of striving (horm), DA 73,14-80,15, since he
regards desire as a type of striving (74,1).
111. [12,17] Although commonly translated as imagination, phantasia is
something that has a wider and more central role in psychology, for Alex. as well
as for Arist. It designates a form of representation that is generated initially from
perceptions and possesses content that is perceptual in character, but also serves
as a basis for the content of memories, dreams, desires, passions, and even thought
(although in the latter case, further conceptual operations are required). It thus
plays a key role in Arist.s and Alex.s explanations of most higher mental states.
For more on the term and its use, see Alex.s main discussion of it at 66,9-73,13, as
well as Arist.s in DA 3.3. passim, along with my 2006, 331-5; for more detail, see
my 1996 and 1998.
112. [12,17-18] 68,4-10; 69,14-70,5; in Metaph. 312,2-5; Arist. DA 1.1, 403a8-10;
Mem. 1, 450a27-b11; 2, 453a14-30; cf. DA 3.3, 428b11-14.
113. [12,19] Although Arist. uses the term horm (see Bonitz 525b47-526a10)
Alex.s use is endebted to the Stoics, for whom it is a key technical term in their
moral psychology: horm is an intentional state directed at the pursuit or avoid-
ance of something and initiates the corresponding motion. The common rendering
of the term as impulse wrongly suggests a primitive and instinctual drive or even
a blind force. For more on the term and its use, see Alex.s main discussion of it at
73,14-80,15 below.
114. [12,20-1] The conditional here is not merely hypothetical: on three differ-
ent occasions, Arist. insists that we cannot think without a representation, thus
Notes to page 40 93
affirming the antecedent: DA 3.7, 431a16-17 (cf. b2); 3.8, 432a8-10; Mem. 1,
449b31. Alex. uses phronein here in its broad sense, for thinking in general,
instead of understanding (noein), which Arist. uses in the parallel passage at
403a8-10, in order to make clear that the point about dependence on phantasia and
the body is not limited to some specific form of thinking, but holds for all thinking
quite generally. It would weaken Alex.s argument if he were appealing mainly or
exclusively to practical thinking, as AD suggest (125).
115. [12,21-4] Alex. develops this pivotal claim of Arist.s (DA 1.1, 403a10-12)
in several interesting ways. As before, he frames it in terms of the pairing of
activities in the soul with changes in the body (see n. 108 on 12,9-13,8 above). But
he also states it in terms of conceivability: the soul is inseparable from the body
provided that the activity of the soul cannot even be conceived (labein) apart from
a body. Given how strong the antecedent is, the claim is somewhat weaker than it
might otherwise appear, since it would not address cases where the correlations to
the body are merely nomically necessary, not conceptually. Finally, Alex. offers an
explanation for the claim, not found in Aristotle: without a body, such a power
could not be exercised and so would be otiose or superfluous (matn). But as both
Arist. and Alex. often say, nature does nothing superfluous: see n. 247 on 27,9; also
27,13; 28,4; 98,9 below.
116. [12,24-5] In the first half of this statement, Alex. ties the inseparability of
the soul to its being the form of the body. But the second half is stronger still, in so
far as it emphasises the explanatory role of the bodys capabilities. On the meaning
of epitdeiots, see Todd 1972.
117. [13,1-8] Of the three cases Arist. presents as evidence for his thesis that
the states of the soul are enmattered logoi at DA 1.1, 403a19-24, Alex. only
presents the first two, which involve variations in strength, in a slightly general-
ised form. Note that Alex. takes Arist.s use of pathmata at 403a20, as many
modern commentators do, to be referring to external provocations (13,1; cf. 13,4),
rather than conditions in the body. He also takes the condition of the body in the
second case merely to be one of being primed and predisposed towards anger (13,5),
rather than the same state one is in when one is angry as Arist. says (hotan orgai
to sma kai houts ekhi hsper hotan orgiztai, 403a22). The phrasing of Alex.s
last remark here, about the body being in a certain state (ps ekhein, 13,8), could
be an allusion to Stoic terminology for their third category, but it is hard to see
what point that would serve.
118. [13,10-14,3] Arist. Phys. 4.3, 210a14-24; cf. Metaph. 5.23, 1023a23-5. Since
Alex. is going to use this division of senses as the basis for an extended argument
by elimination (cf. 15,26-8), he has to assume that it exhausts all the relevant
alternatives. So it is worth noting that while his list substantially overlaps Arist.s,
it does not do so entirely: Arist. includes several causal senses (efficient and final)
which Alex. omits, while Alex. includes the presence of ingredients in a mixture.
(The list from Arist.s Phys. does not include the inherence of extrinsic charac-
teristics either, though an example of it is given at Metaph. 5.23 1023a13 and it of
course underlies the central distinction of the Categ. at 2, 1a23-b3.) Given the
range of senses here, Alex. cannot plausibly dismiss these others as being too far
afield, esp. since Arist. argues the soul is also an efficient and final cause (DA 2.3,
415a8-12). Alex. might think that these senses are to be understood in terms of the
sense in which form is in matter, but it would still require argument.
Alex. offers additional arguments against (c), (e), and (g) at Mant. 115,28-116,1,
on the assumption that the body which has a soul is animate throughout. See n.
104 on 12,2-3.
94 Notes to pages 40-42
119. [13,22] 15,7; see n. 101 on 11,17-18 above (on honeyed water).
120. [14,2-3] 9,14-18; 23,29-24,3.
121. [14,4-5] Alex. alludes here to Arist.s cryptic remark at the end of DA 2.1
that it is unclear whether the soul is a completion or entelekheia of the body as a
sailor is of a ship (hsper pltr ploiou, 413a8-10). Alex.s remark suggests that,
far from being unclear, it is straightforwardly covered by the senses of in some-
thing he has already enumerated, although he will still discuss it on its own below
at 15,9-26 for good measure; see also 20,28-21,10; 79,19-20. The analogy itself
suggests a form of substance dualism, in so far as the pilot, who governs and
controls the ship, is a distinct, independent entity that can be separated from it
(although Descartes himself explicitly rejected this analogy in the Sixth Medita-
tion, AT 7.81). On the analogy see Tracy 1982.
122. [14,5-7] 12,3-4. Alex. assumes here that every part of a body is a body. But
he cannot accept this without further qualification, since he rejects the unqualified
version as confused at 18,10-27 (cf. 6,6-12). If the principle holds at all, it holds only
for those parts which contribute to the bodys size and continue to exist after they
are carved (18,15-17).
123. [14,8-9] The phrase body passing through body (sma dia smatos
khrein) refers to the interpenetration of bodies, which Alex. associates above all
(though not exclusively) with the Stoic theory of total blending and consequently
with their view that the soul pervades the entire body (see n. 101 on 11,17-18
above). Alex., like Arist., regards the idea that two bodies can be in the same place
as absurd and argues against it at length: see n. 189 on 20,8-19 below.
124. [14,10-11] Alex. offers no argument for this claim: he may just take it to
be patently absurd that the soul is composed from parts of the body; or, alterna-
tively, he may be assuming that the compositional principle mentioned earlier at
14,6 goes hand in hand with the converse principle, that anything composed of
bodies is itself a body. Neither hypothesis quite explains his silence, though, since
elsewhere he does not hesitate to offer an argument against a position he thinks
absurd or to work over again a point he has recently covered.
125. [14,12-14] The first objection (12-14) is to the point so long as individual
souls and bodies are our sole concern. But it is silent about the possibility that soul
in general is a species of body in general, which would have the consequence that
any given soul is an individual body. The second objection (eti te, 14) moves to close
this gap, by implicitly denying that soul is a body, as it surely is not a body on the
view Alex. has been developing, since forms in general are not bodies (5,22-6,2;
Mant. 113,26-8). But in the present context, Alex. cannot help himself to the view
that the soul is a form without begging the question, since the purpose of the entire
argument is just to establish that the soul is the form of the body (13,9; 15,26-9);
and given that he makes a point of starting from agreed upon common ground
(13,10), he ought not to simply take it for granted that the soul is not a body, since
others (like the Stoics and Epicureans) insist that it is. Nevertheless, Alex. relies
on this premiss repeatedly in the reductio arguments that follow: 14,18; 15,5-9.14-
15. He does, however, offer a battery of arguments for the claim elsewhere, though,
independent of his view that the soul is a form: see Mant. 113,25-118,4.
126. [14,14-16] Presumably Alex. is referring to the claim at 14,12-14 that
species and genera are universals, whereas the soul and the body are particulars.
127. [14,16-17] Whatever is predicated truly of the genus is predicated truly
of the species: Categ. 5, 2b17-22; cf. 3, 1b10-15. But in the present case, the
consequence is manifestly unacceptable, since if no soul is a body on Alex.s view,
then it follows that no body is a soul either, by conversion. Even on his opponents
Notes to page 42 95
views, not all bodies are souls; and that is all that is needed for this objection to
strike home.
128. [14,18] Arist. Phys. 4.5, 212b7, b28-9; 4.1, 209a26-7; cf. 4.4, 211a17-23.
Alex. again needs the premiss that the soul is not a body to be established
independently: see n. 125 on 14,12-14 above. Plotinus 4.3.20 develops and builds
on the arguments here.
129. [14,19-20] Alex.s objection seems to be that everything has absurdly
swapped roles: if (e) were right, the soul would be a body and the body would be a
place, that is, something that contains bodies. The simple disjunction of alterna-
tive views on place he mentions does not appear in Arist. The first alternative,
empty extension (diastma kenon), seems to be an imprecise way of referring to
space (khra), which Arist. describes as the extension between the boundaries of
the containing body, considered as something over and above any bodies that
might occupy it (4.4, Phys. 211b14-29, 212a3-5). Alex. may possibly be conflating
this notion with the common conception of emptiness or void (to kenon) as
extension with nothing in it, according to Arist. (4.6, 213a27-9; 4.7, 213b31,
214a5-6, a20). Alex. might also have Platonic and even Stoic conceptions of space
in mind here: Tim. 52A-B; Sext. Emp. Adv. Math. 10.3-4 (= SVF 2.505 = LS 49B);
cf. Ps.-Galen Qual. incorp. 2, 19-23 Giusta (= SVF 2.502 = LS 49E). The second
alternative, the boundary of what contains the occupant, is of course Arist.s own
favoured definition of place: Phys. 4.4, 212a20; Cael. 4.3, 310b7-12. For recent
treatments of Arist.s concept of place, see Algra 1995, chs 4-5; Morison 2002.
130. [14,20-2] The claim that the animate bodys differences are due to the soul
is just an instance of Alex.s more general position that the differences between
bodies are due to their forms (7,9-11); even the contribution of matter is through
the forms of the underlying bodies which consititute it (7,24-8,13).
131. [14,23-4] Arist. Phys. 4.2, 209b28-9; 4.4, 212a14-18, which both describe a
vessel as a transportable place (topos metaphortos).
132. [14,24-15,1] In a number of passages, Alex. discusses the suggestion that
the soul might be in the body as an extrinsic characteristic or accident (sumbe-
bkos) inheres in a subject and emphatically rejects it, devoting many arguments
against it: Mant. 119,21-122,15 (esp. 120,9-17); Quaest. 1.8 passim, esp. 17,34-18,4
and 19,9-15; 1.17 passim, but esp. 30,14-22; 1.26 passim; also ap. Simpl. in Cael.
279,5-14; in Phys. 270,26-31.
Despite these repeated denials, Moraux nonetheless takes Alexander to be
committed to the view that the soul is just an accident of the body and so posterior
to it, in contradiction with both himself and with Aristotle (1942, xix, 42-4; cf. 9-10,
33, 45, 170, 177, 178). But this is obviously an uncharitable and willful distortion
of what Alexander actually says and should be resisted. Moraux confuses being
supervenient with being causally or metaphysically posterior (see 49 n. 2 and 43-4,
respectively). But neither in fact follows: on the contrary, Alex. himself consis-
tently argues against such inferences, as in the present passage. See also nn. 90
and 92 on 10,17-19 and 10,24-6.
Alex.s first argument here connects Arist.s classification of the soul as a
substance or ousia, in so far as it is a form (DA 2.1, 412a19-20, cf. a6-9), with the
claim in Categ. 5 that individual substances alone can receive contraries (4a10-
b19). By taking this characteristic of concrete substances to apply to substantial
forms, Alex. accepts the consequence that the soul can itself be the subject of
extrinsic characteristics (Mant. 121,25-7). One might worry whether this has the
further consequence that the soul can itself undergo change, and not just extrinsi-
cally, against his express position (21,22-24,17).
96 Notes to page 42
133. [15,2-5] Mant. 119,31-120,17; cf. Quaest. 1.17, 29,31-30,22. For something
to inhere in a subject, the subject must already be some specific thing (tode ti) on
its own. But without its form, the matter of a substance is not the specific thing it
becomes in the substance, except potentially. Hence, substantial forms must be in
matter in some way other than inhering in a subject in the way an extrinsic
characteristic does. For translations of Alex.s other texts and reactions to them by
later commentators, see Sorabji 2004a, 3.122-5; on Alex.s view, see Ellis 1994; Sharples
1999. Alex. may be reacting here to earlier Peripatetics, such as Boethus, who held
that form does not belong in the category of substance, but in non-substantial ones, such
as quality or quantity (ap. Simpl. in Categ. 78,17-20), and so would inhere in matter.
Alex.s denial that the soul inheres in the body in the way an extrinsic charac-
teristic does has a further consequence relevant to current debates regarding
Arist.s psychology. If the soul were an extrinsic characteristic of the body, which
the body could equally have or lack, then the same body could exist as such either
before or after its serving as the matter of the living thing. In denying that the soul
is extrinsic to the body, Alex. insists that the organic body gets its nature as an
organic body from the soul (15,3-5; Mant. 120,13-17; Quaest. 1.17, 30,14-22): it is a
living body essentially and so only comes to be such together with the soul when
the living thing first comes into being. When the body ceases to function and to be
alive, it remains a body only in name or homonymously (Arist. DA 2.1, 412b20-3;
Meteor. 4.12, 389b31-390b2; PA 1.1, 640b34-641a5, a17-21; GA 1.19, 726b22-4; 2.1,
734b23-7, 735a7-8; Pol. 1.2, 1253a20-5; Alex. in Meteor. 223,25-224,11 and 225,3-33).
It is precisely this feature of the body/soul relation on Arist.s theory that J. L. Ackrill
finds problematic in his landmark article (1972/73), since it is in tension with the
conception of matter as something that underlies all change, including substantial
change, and so pre-exists the substances coming to be and survives its passing away.
134. [15,7] 13,22; see n. 101 on 11,17-18 above (on honeyed water).
135. [15,5-9] Alex. is operating from within his own framework here rather
than a rival one, like the Stoics. According to the Stoics, the soul is blended with
the body (see n. 101 on 11,17-18); but on their conception of blending, the ingredi-
ents and their natures are preserved, not destroyed, and they are indeed all bodies.
What Alex. describes here the Stoics would call fusion (sunkhusis; see n. 102 on
11,18-20). So Alex.s first objection is not effective against the Stoic view of
blending when taken on its own terms. The second objection would be effective, on
the other hand, provided Alex. can establish independently that the soul is not a
body (see n. 125 on 14,12-14).
136. [15,9-10] Alex. takes up again the ships pilot, something he mentioned
briefly before at 14,4-5 and will return to below at 20,26-21,21. In his view, this
suggestion is ambiguous between two cases: (1) where the pilot is understood as a
form of the body and so falls under (h); and (2) where the pilot is taken to be in a
place or vessel, and so falls under (e). He can thus divide the question: if the
suggestion is just a variant of (h), the conclusion he is arguing for in the present
section, it does not require rebuttal (though he still rejects the analogy: see
20,26-21,21); but if not, it encounters objections. As AD note (128-9), Alex. puts the
possibility of the first construal in the optative (15,11), whereas he puts the second
in the indicative (15,14), to indicate he regards it as the real target.
In discussing whether the soul is in the body in this way, Plotinus (4.3.21.5-21;
cf. 4.3.17.21-31; also 1.1.3.21-3) seems to be responding directly to details in the
present passage (see n. 128 on 14,18 above). Against Alex., some commentators
even take Arist. to endorse a version of the pilot view: Simplicius in DA 96,3-10
and Philoponus in DA 224,28-37. (Themistius leaves it unresolved: in DA 43,30-5.)
Notes to pages 42-43 97
Porphyry embraces the metaphor and develops it at length, even allowing for a
succession of pilots at ad Gaurum 10.4-6, 47.5-48.8 Kalbfleisch (on which James
Wilberding has valuable notes, with further references, in his translation in this
series, 2011).
137. [15,10-13] On construal (1), what is important about the pilot is simply
that he guides and regulates the behavior of the ship as a ship, and not that there
is some actual homunculus deliberating and operating the ship in effect, con-
strual (2). Although it may sound odd to modern ears, a very Aristotelian way of
framing this option is in terms of art, just as Arist. sometimes speaks of the art
itself as the efficient cause, rather than the artisan (e.g., Phys. 2.3, 195a3-8; GA
1.22, 730b19-23; 2.1, 734b36-735a4). For a full elaboration and defence of the
comparison with art, see Menn 2002, esp. 95-6, 99, 114-17, 136-9.
For the inseparability of dispositions, see Mixt. 9, 222,31-2. For the inseparabil-
ity of forms, see DA 17,9-15; cf. Mixt. 13, 228,17-20. Alex. may use both terms here
for dialectical reasons: the same objection will serve regardless of the specific
version this view assumes, since his point will hold more generally for anything
which is not a substance in its own right, but belongs to a substance: cf. Arist.
Categ. 5, 2b5-6; Phys. 1.4, 188a6, a12-13; GC 1.10, 327b22 (cf. 1.3, 317b10-11,
b31-3; 1.5, 320b24-5).
138. [15,14-15] If the point of the comparison is that the soul is like a person,
then it too will be embodied. Note the stronger use of meta + genitive here: if it
merely meant conjoined with, as it often does (e.g., 4,12-17; 5,20), Alex. could not
validly infer that the soul is a body. If an opponent accepts this consequence, Alex.
again needs independent grounds to argue that the soul is not a body (see n. 125
on 14,12-14). But his purpose here may be more tactical and ad hominem: such a
result would be an embarrassment to substance dualists and shows that they
cannot afford to take the comparison too literally.
139. [15,15-17] 21,10-11; for the general assumption, see n. 104 on 12,2-3. This
second objection is the more important one: even if one does not take the pilot to
be literally embodied, the comparison still implies that the soul has a specific
location in the body, like the pilot at the helm; otherwise, it becomes difficult to
distinguish this option from (1).
On the responsiveness (sumpatheia) of the whole body, see 100,1-8; for the claim
that different parts of the body have an accompanying awareness (sunaisthsis) of
each other, see in Metaph. 363,21-4 (cf. 321,24-30). Contrast this with the absence
of such awareness between different persons or even unconnected senses (in Sens.
163,11-14; cf. 36,12-17). A difficulty for the present argument is that Alex. else-
where accepts Arist.s view that parts of the body like hair, bones, and nails are not
sensitive at all: 74,29-75,2 and 97,4-6.
Both notions have Stoic antecedents, though. According to the Stoics, the
responsiveness of one part of the body is a feature of unified bodies generally, and
so of the universe as a whole, which is itself a unified animate body (Sext. Emp.
Adv. Math. 9.78-80 = SVF 2.1013; cf. Alex. Mixt. 3, 216,14-17 = SVF 2.473; 10,
223,7-9.30-3 = SVF 2.441; 12, 227,8-9 = SVF 2.475; Cleom. Motu circ. 2.164.4
Ziegler (cf. 1.8.16-21) = SVF 2.534). Cleanthes even appeals to the responsiveness
of body and soul to each other as an obvious datum, from which he can argue that
the soul must be corporeal (Nemes. Nat. hom. 2, 21.6-9 Morani = SVF 1.518 = LS
45C). The awareness of the entire body plays an important role in their theory of
oikeisis as well: the Stoic Hierocles (first century CE) explains that the initial
awareness an animal has of its own body is due to the blending of the soul with the
entire body (Hierocles Elem. eth. 4.39-53 Long = LS 53B 5-9). Superficially, this
98 Notes to page 43
might seem to conflict with Chrysippus assertion that although a pain occurs in
the foot, the awareness that accompanies it is in the chest (that is, the heart),
where the governing part is located, and in no other part (Galen PHP 3.7.4,
212.14-18 De Lacy = SVF 2.900; cf. 3.1.25, 172.20-6 De Lacy, = SVF 2.886). But
Hierocles also describes the awareness as reaching the governing part (4.49-53).
These positions can be reconciled by distinguishing between act and object: al-
though the animal has awareness of all the parts of the body, the act of awareness
itself occurs solely in the governing part. Galen also employs the notion of the
responsiveness of different parts of the body extensively. For the general notion,
see Siegel 1968, 360-72; and specifically with regard to vision and referred pain,
Siegel 1970, 121-2 and 187-9, respectively.
140. [15,17-22] 21,11-13; the suggestion that the soul on this view does not
move the animal naturally, but by force, is more fully elaborated at 21,13-16. Arist.
uses a similar reductio at DA 1.3, 406b2-5 (cf. a22-7), regarding the possibility of
the souls displacement, and draws the further conclusion that dead animals might
be reanimated. On such a view, the body would not be essentially animated, but
would be the sort of thing it is independently of the presence of soul.
141. [15,22-6] The precise nature of this objection is unclear, since an infinite
regress hardly seems inevitable. In fact, it seems perfectly coherent to go back just
one step: if, e.g., one took the soul to be like the governing part (hgmonikon) of
the soul in Stoicism, which is a body and could be regarded as like a pilot. Alex.s
objection cannot rule such a case out, but at most urges a polemical correction,
namely, that such an explanation is inadequate unless it appeals to the form of this
body (in which case the theory has changed its stripes in favour of Aristotelianism).
142. [15,28-9] Mant. 116,15-18; cf. 6,21-9; 10,7-10.
143. [16,1-2] 9,11-14; 10,4-14; 10,28-11,7; 12,6-8.
144. [16,2] Quaest. 2.10, 55,9. In this treatise, when Alex. speaks of enmattered
forms, he typically has in mind the forms of natural objects that serve as the object of
thought (83,19; 87,5; 88,4.11; 89,13; 90,2; cf. Mant. 108,3.21; Quaest. 1.25, 39,28). In
the present passage, though, he is likely to be alluding to Arist.s characterisation of
states of the soul as enmattered accounts (enhuloi logoi) in the programmatic
argument in DA 1.1 (403a3-25 at a25), which Alex. has just reprised at 12,7-13,8.
145. [16,2-4] 10,26-11,5.
146. [16,5] 6,29-7,3.
147. [16,5-6] Although Arist. occasionally uses the term culmination (teleiots)
see n. 64 on 6,30 above he does not use it in this context. What Alex. is saying
here is rather that where he himself uses culmination, Arist. typically uses the
term completion or entelekheia. Completion occurs prominently in Arist.s general
definition of the soul in DA 2.1, and again in 2.5, to distinguish two levels of realisation:
the first, where something actually has the power to perform certain activities, and
the second, where it is actually performing them: 2.1, 412a22-7 (cf. 412b27-413a1); 2.5,
417a21-b16. On Alex.s use of these terms, see Todd 1974, 213-14.
148. [16,7] Arist. DA 2.1, 412a27-8.
149. [16,8-10] 9,15-26; Mant. 103,11-20; Arist. DA 2.1, 412a22-7 (cf. 412b27-
413a1). As AD rightly note (130), Arist. does not use either of the terms disposition
(hexis) or power (dunamis) for the first completion in DA 2.1. But he explicitly
uses power in this way at DA 2.5, 417a21-b16 and Phys. 8.4, 255a30-b3. AD also
suggest that disposition can be derived from 417a32, but the parallel is not exact.
For further discussion of the Aristotelian antecedents, see BD 246-7.
150. [16,10] Arist. DA 2.1, 412b5-6.
151. [16,11-12] Quaest. 2.8, 54,9-10; also see n. 93 on 11,1 above.
Notes to pages 43-44 99
152. [16,12-13] The traditional rendering of the dunamei as potentially
wrongly suggests that Arist. is concerned with the form of bodies which are not yet
alive, but could become so. But Arist. is quite clear that he is describing the body
that actually has the soul (412b25-6) and so has life by or in virtue of its power.
153. [16,12-18] Mant. 104,11-17; Quaest. 2.8, passim. Life, according to Arist.,
is said in many ways, and amounts to performing the various activities Alex. lists
here, which distinguish the animate from the inanimate (DA 2.2, 413a21-5). But
something that possesses these abilities is alive even when it is not exercising
them: being inactive or asleep is different from being dead or inanimate, since in
the former case one actually has the power to function in these ways and not in the
latter (cf. 2.1, 412a23-6, 412b28-413a1). Hence, the soul is defined in terms of the
powers to perform these various activities, rather than the activities themselves
(2.2, 413b11-13).
154. [16,18] This answers the first question on the agenda Alex. sets out at the
very beginning of the treatise. See n. 3 on 1, 2-4.
155. [16,18-17,1] For the order of three main types of souls, see Arist. DA 2.3,
esp. 414b20-415a12.
156. [17,1-8] 28,14-29,1; 30,17-20; cf. 40,4-10; Quaest. 1.11a and b, passim. As
Arist. himself emphasises, the definition given in DA 2.1 is only an account of the
most general sort (koinotatos logos, 412a5-6; cf. b4-5) and merely a sketch in
outline (413a9-10). The underlying problem, according to Arist., is that different
types of souls form an ordered sequence (2.3, 414b20-415a1) and such sequences
lack a genus prior to the individual species: Metaph. 3.3, 999a6-11; EN 1.6,
1096a17-23; EE 1.8, 1218a1-15; Polit. 3.1, 1275a35-b2; cf. DA 1.1, 402b5-9. For a
detailed examination of the Aristotelian view, see Lloyd 1962.
Moraux attacks Alex.s interpretation in the present treatise as tendentious and
erroneous (1942, 50-62). He misrepresents Alex., though, as denying the possibility
of a general definition and of an essence common to all souls. But Alex. begins the
present passage by affirming that there is an essence of the soul (16,18) and goes
on to discuss its general characterisation. What he denies is that such a general
definition can be fully perspicuous (saph, 17,1) or reveal the nature of each type
of soul intelligbly (gnrims, 30,18). For a detailed critique of Moraux, see AD
131-2.
157. [17,9] Alex. devotes the arguments of the two previous sections (11,14-13,8
and 13,9-17,8) to demonstrating that the soul is a form. But the view has been
explicitly in play since at least 9,11-12 and in preparation since 2,25. Cf. also
31,20-5.
158. [17,9-10] Showing that each of these three characteristics belongs to form
sets the agenda for what immediately follows: after a brief argument to show that
the soul is inseparable (17,11-15), Alex. argues at greater length that the soul is
non-bodily (17,15-20,26; see also Mant. 113,25-118,4), before turning to the claim
that the soul cannot undergo change on its own (21,22-24,17).
159. [17,11-12] 4,8-9; 5,21; Mant. 117,20-1; Arist. Metaph. 8.1, 1042a29-31. The
key to Alex.s argument here is his general view that bodies are composites and so
already have the matter and form needed to exist on their own (4,22-5; 5,21). Since
the soul is a form, it cannot exist on its own (5,22-6,2), but needs the body of a living
thing as its matter. Alex. passes over in silence here the awkward fact that on his
own view the body of a living thing, in so far as it is organic, also cannot exist on
its own as such apart from the soul or the living thing (15,3-5 and n. 133 on 15,2-5;
Mant. 120,13-17; Quaest. 1.17, 30,14-22), despite the fact that it is a body and so
something that, according to 17,11-12, should be able to exist on its own.
100 Notes to page 44
160. [17,12-15] Although the form in question is a substance, it cannot exist on
its own as the composite body can. Instead it is always the form of something and
so cannot exist apart from the matter of the composite. This requirement, in Alex.s
view, is common to both Arist. and the Stoic doctrine of principles or arkhai, and
is to be contrasted with the Platonic view of form (in Metaph. 178,15-19 = SVF
2.306; I am grateful to Ian Flora for this reference). But it is also to be contrasted
with the Stoics view of the relation between the soul and the body. As BD rightly
point out (249), the soul can be separated from the body on their view, precisely
because it is a body: see esp. Nemes. Nat. Hom. 2, 22.3-6 Morani (= SVF 2.790 =
LS 45D). This is what allows them to endorse some form of post mortem existence:
Arius Didymus fr. 39, Dox. Gr. 471.18-24 = Euseb. PE 15.20.6, 2.385.9-12 = SVF
2.809 = LS 53W; Aetius 4.7.3, Dox. Gr. 393.1-7= SVF 2.810.
Regarding the inseparability of limits, Arist. recounts a Platonic argument in
Metaph. 3.5, which claims that limits are prior to bodies because they can exist
without bodies, and not vice versa (1002a4-8), contrary to what Alex. claims here.
But in commenting on the passage, Alex. reaffirms that limits cannot exist without
bodies in reality, but only in thought (in Metaph. 229,31-230,1).
161. [17,15] If every body is something that can exist on its own (17,10-11), and
the soul cannot (17,14-15), then the soul cannot be a body.
162. [17,15-17] Repunctuating with von Arnim (= SVF 2.394), as AD rightly
urge (133), although I would offer a different justification. With Bruns punctua-
tion, the sentence reads: The form cannot be a body along the lines of those who
claim, like the Stoics, that everything, whether it is matter or from matter, is a
body. This not only misrepresents the Stoic view, which includes non-bodily items,
but also makes no sense of the argument that follows in the text. What Alex. offers
is an ad hominem argument against a Stoicising view: if every body is either
matter or composed from matter (as the Stoics claim), and a form is neither, then
no form is a body. The minor premise of this argument presupposes von Arnims
construal and so confirms his punctuation, against Bruns.
The premise that every body is either matter or composed from matter is very
likely an inference on Alex.s part from the fact that one of the two Stoic principles
from which the entire cosmos is composed is matter, and is itself a body, unlike
Aristotelian prime matter (D.L. 7.134 = SVF 2.299 = LS 44B 3, retaining the MSS
reading see LS 1.274, 2.266; cf. Alex. Mixt. 225,3-5 = SVF 2.310). The Stoics,
moreover, explicitly maintain that qualities (poiottes) are bodies (Plut. Comm.
Not. 50, 1085E-1086A = SVF 2.380; Galen Nat. Fac. 2.4, 168.12-14 Helmreich =
SVF 2.410; Simpl. in Categ. 217,32-3 = SVF 2.389 = LS 28L 1; 271,20-2 = SVF 2.383
= LS 28K; cf. Alex. Mant. 124,9-10 = SVF 2.432). This is surely due to their view
that qualities like prudence, temperance, and even life are causes and only bodies
can be causes: see esp. Stob. 1.138.14-22 Wachsmuth (= SVF 1.89 = LS 55A 1-3).
If Stoic qualities are to be analysed as breath in a certain state (pneuma ps
ekhon), as Alex. takes them to be (in Top. 360,10 = SVF 2.379), then they will in
general be composed of matter.
Alex. offers a battery of arguments against the view that qualities are bodies at
Mant. 122,16-125,4; similar arguments can be found earlier in Plut. Comm. Not.
50, 1085E-1086B; Alcinous Didask. 11, passim; Ps.-Galen Qual. incorp., passim.
On this tradition, see Kupreeva 2003, esp. 304-15.
163. [17,17-18] The dialectic is complex. Alex.s overall aim is to examine
different ways of construing his own theoretical terms. But the option he is
considering and arguing against here is one that incorporates distinctively Stoic
views (n. b. Stoikn paides, Ps.-Galen Qual. incorp. 1, 10-11 Giusta = SVF 2.377).
Notes to pages 44-45 101
Matter, for example, has to be understood in the Stoic sense, since on Alex.s own
view only prime matter is qualityless (on prime matter as qualityless, see refer-
ences in n. 26 on 3,28-4,4). Forms, on the other hand, can be viewed as a kind of
qualification, even on an Aristotelian view, at least in a broad sense (see n. 105 on
12,3-4); in fact, Boethus had claimed that forms belong in non-substantial catego-
ries, such as quality or quantity (ap. Simpl. in Categ. 78,17-20); see n. 133 on 15,2-5
above.
164. [17,18] Notice Alex.s reinsertion of form here. One might worry that this
undercuts the effectiveness of Alex.s argument, since an opponent might well
reject a further analysis into matter and form. But Alex.s objection has a more
general force. His rationale is presumably that if the second half of the original
disjunction, being composed of matter, is to differ from the first, then it cannot
mean being composed solely of matter, but must involve some other element,
whether we regard this as an Aristotelian form or a Stoic quality or something else.
The question is then whether this additional element can also be a body.
165. [17,18-19] It is difficult to make good sense of this objection. It is cavilling
if Alex. is claiming that the second form, as a proper part of a whole, is both the
same and not the same as the whole (as Kupreeva 2003, 317 appears to take it),
since then it would merely equivocate on same. For there is nothing absurd in (a) the
second forms being a proper part of the composite (namely, the first form) and so not
identical with it, while also (b) being the same kind of thing, namely, a form. Alex.
might instead be making a more general complaint about the views overall strategy:
if parts of composites such as forms are themselves composites, then there is no longer
a clear difference in their respective theoretical roles. But so construed, the objection
is obviously not fatal and, worse, it can be raised against Alex.s own conception of
matter: apart from prime matter, all other matter on his view is a body (3,23-5) and so
a compound of form and matter; but he does not think that this prevents such a
compound from serving as matter in some even more complex body.
166. [18,1] The MSS read to eidos here, which would have the form somehow
possessing a form not different from itself. Bruns reasonably emends it to to einai
(as followed here), referring to 88,12-13. But the relevant parallel is surely 18,6.
167. [18,2-7] Alex. has presented his opponents with a series of dilemmas:
either forms are bodies or they are not; if they are bodies, then either they are
matter or they are composed of matter and form; if they are composed of matter
and form, then either the component form is also composed in this way, leading to
an infinite regress, or at some point such composition comes to a stop. But if
composition comes to a stop, the last form will neither be matter nor composed of
matter. Why, then, should it still be considered a body? Alex. now completes his
reductio with a final dilemma. If the last form is a body because it has its being
with matter and so cannot exist apart from it, something analogous would have to
be said of the last matter, namely, that it has its being with form; but this would
conflict with the Stoic definition of matter as qualityless (18,2-3; see the penulti-
mate paragraph of n. 26 on 3,28-4,4). But if, on the other hand, matter is
qualityless because form is not part of its nature, though inseparable from it, then
something analogous will have to be said of the last form too: even though the last
form is inseparable from the corresponding matter, it is definitionally independent
of it and so not a body, against the initial hypothesis. Alex.s opponents have thus
been forced to accept Alex.s own position (5,18-6,2 and 6,6-20; though compare
Mant. 115,12-14). For related criticisms, see Plut. Comm. Not. 50, 1085E-1086B.
For a somewhat different analysis of the argument, see Kupreeva 2003, 317-20.
168. [18,7-10] The objection appears to be that any change in quality, on this
102 Notes to pages 45-46
view, would entail a change in mass or quantity, since if the new form is itself
embodied, it will add some matter of its own to the composite: cf. Mant. 116,1-2.13-
15. But this does not follow, since the former quality or form which is lost in the
change would be embodied too and might well depart with its own matter in tow;
so no general conclusions can be drawn about a net gain or loss.
169. [18,10-14] Von Arnim includes the present passage as a fragment of
Chrysippus (SVF 2.793), but we have no other testimony for the argument as a
whole, much less an explicit attribution to Chrysippus. The mereological principle
offered as the first premise (18,10-13; cf. 14,6) is attributed almost verbatim to the
Stoic Apollodorus (second-century BCE), down to the examples (fr. 7, SVF vol. 3, p.
260.12-14 = Stob. 1.19.5, 167.9-12 Wachsmuth = Arius Didymus fr. 24, Dox. Gr.
460.14-16). But the second premise of Alexanders argument can hardly be attributed
to the Stoics; it again seems to be part of the Stoicising interpretation of form which
Alex. has been critiquing. See nn. 162 and 163 on 17,15-17 and 17,17-18 respectively.
Since the argument is valid if, at any rate, one adds the reasonable assump-
tion that an animal is a body and Alex. accepts both this tacit premise and the
second premise above (cf. 6,2-4), he should reject the first premise, that the parts
of a body are necessarily a body. But while he clearly rejects any unqualified,
general version of this principle as involving a fallacy of division (6,6-12), he does
in fact employ a version of it himself earlier on (see n. 122 on 14,5-7 above). For a
different account of this argument, see Kupreeva 2003, 321-4.
170. [18,17] As AD point out (136), this claim does not hold for organic parts of
living bodies, at least not without further qualification, since a dead or detached
hand is no longer a hand except in name or homonymously (DA 2.1, 412b20-5;
Meteor. 4.12, 389b31-390b2; PA 1.1, 640b34-641a5, a17-21; GA 1.19, 726b22-4; 2.1,
734b23-7, 735a7-8; Pol. 1.2, 1253a20-5). Alex.s main point will still hold, though,
for the compositional matter that makes up the organs and the body. What matters
most to the kind of division here is its being something quantifiable, as Alex. makes
clear in his discussion of the different meanings of part in commenting on Metaph.
5.25: when animal is divided as something with size (hs poson), it is divided into
head, trunk, and so on (in Metaph. 424,17-19).
171. [18,17-21] Arist. Metaph. 5.25, 1023b19-22.
172. [18,21-3] For the contrast, cf. 12,3-4; 14,6-7.
173. [18,24] 18,15-17.
174. [18,25-7] 6,24-7,8, together with 5,19-6,6.
175. [18,25-19,1] Von Arnim also includes this passage as a fragment of
Chrysippus (SVF 2.793), but again we do not have any other testimony or attribu-
tion for the argument as such. The first premise is obviously related to the first premise
of the previous argument above, which is Stoic in origin (n. 169 on 18,10-14). But they
are nonetheless distinct: the earlier principle concerns mereological division, whereas
this one concerns composition, and neither entails the other. It is reasonable to think
that Stoics like Apollodorus would have endorsed both, but we have no independent
evidence. We have no direct evidence that the Stoics endorsed the second premise of
the present argument either, although again it is reasonable, in so far as they speak
of the sense organs (aisthtria) as parts of the soul: Aet. 4.21.2-3, Dox. Gr. 410.30-
411.13 = SVF 2.836 = LS 53H 3; D.L. 7.110 (cf. 157) = SVF 2.828. For discussion of this
argument, see Kupreeva 2003, 324.
176. [19,1-3] Alex. does not challenge the validity of the argument, but its
soundness. In his view, the second premise is never true, no matter how we
understand the word sense (aisthsis): taken in one way, a sense is a part of the
soul, but so understood, it is not a body; taken in the other way, it is a body, but
Notes to page 46 103
not a part of the soul. Alex. relies here on the Aristotelian view that the sense
(aisthsis), taken as a part of the soul, is distinct from the perceptual organ
(aisthtrion), which is a body in fact, a sense is the form of a perceptual organ:
DA 2.12, 424a24-8; cf. 2.1, 412b17-25. If sense is used more loosely to speak of the
organs, it no longer designates a power of the soul.
177. [19,3-5] Alex. seems to be relying on the position he attributes to the Stoics
above (17,15-17), that every body is either matter or composed from matter, with
the added assumption that the second alternative, if it is to go beyond the first,
must include the addition of some quality or form. See n. 164 on 17,18.
178. [19,6-9] See 26,15-17 (= SVF 2.786), with accompanying note; also Mant.
115,6-12 (= SVF 2.785). According to the Stoics, not all breath or pneuma is
animate, since it pervades the entire cosmos and underlies all things (Mixt. 3,
216,14-17). It constitutes soul only with the addition of a certain quality: Galen
Hipp. Epid. 17b.251.1-6 Khn = SVF 2.715; QAM 4, 45.5-8 Mller = SVF 2.787;
D.L. 7.157 = SVF 1.135; Sext. Emp. Pyrr. Hyp. 2.70 = SVF 1.484. More specifically,
the breath turns into soul at parturition, when it is chilled by the outside air: Plut.
Stoic. Rep. 41, 1052F, 1053C-D = SVF 2.806; Hierocles Elem. eth. 1.5-27 Long; Plot.
4.7.83.1-3 = SVF 2.804. For a comparison of Stoic embryology with Alex.s own
views, preserved at Simpl. in Phys. 311.1-19, see Kupreeva 2003, 327-30.
179. [19,9-16] On Alex.s view, the soul is a substance and therefore has a
distinct essence of its own, which breath as such does not share, since on the Stoic
account it is not always alive. It cannot become a soul, therefore, unless there is a
change in its essence, in which case it would no longer be breath, but would have
transformed into soul. Compare this, as Alex. suggests, with other natural bodies:
when air becomes fire, it is no longer air. The alternative is to view soul merely as
an extrinsic characteristic of the underlying breath, which it can equally have or
lack (19,14-16). Alex. is in effect pressing the Stoics on their holism here. To the
extent that there is a single substance, underlying all things, which is merely
differentiated in different places and times by different qualities, the soul and all
the phenomena associated with it would simply be phases or states of breath and
living things simply parts of the whole cosmos. These are consequences the Stoics
might well embrace, but which Alex. takes to be highly counterintuitive and to
undermine the souls status as a substance. On this argument, see also Kupreeva
2003, 325-30.
180. [19,18-20] Another dialectical possibility, which Alex. countenances only
briefly, is that there are two types of breath, one inanimate and the other animate,
so that there is a sense in which soul is breath, namely, it is a kind or species of
breath. But a genus, Alex. points out, cannot exist as such on its own. The Stoics
own corporealism in any case precludes this option: for them breath is not an
abstract kind or type, under which they fall, but a concrete body which pervades
the cosmos and exists on its own. They cannot abandon such a core claim without
completely changing their theory.
181. [19,21] Alex. again seems to be critically examining a hybrid Aristotelian
view, as in the previous section (see nn. 162 and 163 on 17,15-17 and 17,17-18),
which accepts the letter of the doctrine that the soul is the form of the body, but
construes form in a corporealist way. This might seem otiose, in so far as the target
here is only a special case of the one in the previous section: any of the general
objections there against taking form to be a body should apply a fortiori to any
specific kind of body, such as fire or air. But the discussion here is not redundant,
as Alex. limits himself for the most part to objections which are peculiar to this
variant. The different theories may also have different provenances. The view
104 Notes to pages 46-47
discussed in the immediately preceding section concerns Stoic influences, and
while the mention of fire and air here might also suggest the Stoic notion of breath
or pneuma or even Presocratic views, as Ian Flora has suggested to me, such as
Democritus (Arist. DA 1.2, 405a8-13 = DK 68 A 101) or Heraclitus (Arist. DA 1.2,
405a25-6 = DK 22 A 15) or still earlier Anaximenes (Aet. 1.3.4, Dox. Gr. 278.12-15
= DK 13 B 2) the appeal to form specifically suggests a hybrid view which may
be Peripatetic in origin. We do not have evidence of anyone advancing precisely
this claim. But given the way Alex. reports the view (hoi legontes, 19,21), the
commitment to form (19,21; 20,23) and the details Alex. later adds (19,26-8), it
seems unlikely that it is a purely dialectical construction. AD tentatively suggest
(137-8) that there are resemblances to the view reported at Mant. 112,11-16, which
they (following Moraux 1967, on the basis of Mant. 110,4) take to derive from
Aristotle of Mytilene (second century CE), who may have been one of Alex.s own
teachers: according to this view, fire or something fire-like supervenes on the blend
of ingredients in each body and serves as an instrument of the divine under-
standing; but as AD note, this passage does not explicitly say that this fire is the
soul or form of the body. Opsomer and Sharples 2000, moreover, have argued that
the Mant. passage refers not to this later Aristotle from Mytilene, but to the
original one, from Stagira. For a discussion of corporealist views held by later
Peripatetics, see Kupreeva 2003, 331-5.
182. [19,24] 2,25-7,8, esp. 2,25-3,15.
183. [19,24-8] Alex.s inference does not turn on the notion of a culmination by
itself, since he considers the form of even a simple body, such as the heaviness of
earth, to be a culmination (9,17-19), and it does not result from some mixture. The
consequence Alex. draws in the text above follows instead from the fact that the
soul is the form and culmination of a particular kind of complex body. Therefore,
if the soul is fire or air or the like, as claimed, then an apparently elemental body
would instead be the form and culmination of this complex body and would come
into being, as the soul does, when the complex body is created from a mixture of
ingredients.
184. [19,31-6] On Alex.s view, although the ingredients in a blend are not
present in it in actuality, they do not perish entirely, but persist in potentiality
(Mixt. 14, 230,17-19). In particular, their qualities and matter are unified into
those of the blend and are present in actuality: 230,26-34; 231,1-4.10-12. Conse-
quently, fire would have to exhibit the qualities of its ingredients on this view,
which it manifestly does not. For more on Alexs view, esp. as compared with the
Stoic view, see n. 102 on 11,18-20. On Alex.s use of mixis and krasis in the DA, see
n. 76 on 8,16.
185. [19,36-20,1] Alex.s criticism suggests that the proponents of this view did
not offer any explanation for why fire or air held a privileged position. Instead, he
takes the view simply as a claim about one of the elements and in effect sets up an
indifference argument as a challenge: unless there is a relevant difference between
elements, either they each should be forms of the remaining elements or, if they
cannot all be, then none should be a form of the others. Alex. presupposes that they
are all elements or principles and so on a par with one another.
186. [20,4] For this use of mden mallon, where we would use no less, see also
Fat. 175,17 and 176,10 (Sharples 1975, 251 n.8; 1983, 134 and 242 ad loc.).
187. [20,1-6] Alex. continues to examine how, on the present proposal, the body
that is supposed to be the soul is related to the other bodies as their form. Just
above he questioned whether it could be a mixture of them, since it would then
have to exhibit their powers as well, which it does not (19,31-6). Here he considers
Notes to page 47 105
the body in question simply as a body produced out of the others: if a new body is
produced, then only its powers should remain in the animal (20,1-4). But the
powers of the other bodies are in evidence no less than fires, so it cannot be a new
body produced from them (20,4-5). Alex. concludes that it cannot be a form and so
not a soul either (20,6), but it is not clear whether this is ad hominem viz. on this
proposal, it would not be a form or a soul or whether it conflicts with the nature
of forms on anyones view, including Alex.s. The ad hominem reading seems more
plausible, since Alex. does not elsewhere seem committed to the view that the
forms powers are the sole powers in the animal (20,1.5).
188. [20,6-7] Alex. sets up a dilemma here, each of whose horns he regards as
leading to a reductio: either (a) the body which constitutes the soul on this view
does not completely pervade the body of the living thing or (b) it does and so
requires that one body passes through another body. The first option is one he has
already objected to before (see n. 104 on 12,2-3; also Mant. 104,21-2) and he offers
no further argument here. The second is the target of the remainder of this
paragraph (20,7-19).
189. [20,8-19] 14,8-9; Mant. 139,29-141,28; Quaest. 2.12, passim; ap. Simpl. in
Phys. 530,9-30. Alex. regards the view that one body can extend or pass through
another body to be one of the most distinctive and unacceptable consequences of
the Stoic theory of blending, which applies not only to the relation of the soul to
the body, but of divine reason to the matter of the entire cosmos. Although Alex.
often characterises their view simply by reference to this consequence, he recog-
nises elsewhere that they did not themselves fully appreciate the importance of
this commitment to their overall theory (Mixt. 12, 227,1-10). See Todd 1976, 73-88,
who argues even more strongly that this phrase is not used by the Stoics at all (74),
despite von Arnims use of it as a subtitle for a section on their theory of blending
(SVF vol. 2, p. 151, l. 7); in Todds view, it is an independent notion imposed on
them rather than an actual doctrine of theirs (228). But it may not be an unjust
inference for all that: see the apparently direct reports of Stoic views at Mixt. 3,
216,14-17; 4, 217,27-9.33-6; 218,1-2; 6, 219,9-10. For a quick overview of the Stoic
theory of blending, see LS 1.292-4 and M.J. White 2003, 146-9. For more detailed
studies, see Todd 1976, 21-88; Sorabji 1988, ch. 6; E. Lewis 1988.
Alex.s reference at the end of this paragraph (20,18-19) to a more lengthy
treatment is almost certainly to his de Mixtione, which is largely devoted to
opposing the Stoic theory of blending and above all the notion that body passes
through body. Some of the criticisms he makes here can also be found there: it is
counterintutive that what is full can receive something further (5, 218,17-19.21-4);
and if two bodies can pass through each other, nothing prevents indefinitely many
from doing so or even the largest body (6, 220,13-16; cf. Mant. 140,22-5; 141,9-22).
The second criticism appears to be Academic in origin, tracing back perhaps even
to Arcesilaus himself (third century BCE): see Plut. Comm. Not. 37, 1077E-1078E,
with the reference to Arcesilaus at 1078C. Alex. is no doubt also influenced by
Arist.s frequent references to the absurdity of two bodies being in the same place,
which he takes an opposing position he is criticising to be committed to: e.g., DA
1.5, 409b2-4; 2.7, 418b13-18; Phys. 4.1, 209a4-7; (cf. 4.5, 412b25); 4.6, 213b5-12;
4.7, 214b6; Cael. 3.6, 305a19-20; GC 1.5, 321a5-10. Todd even goes so far as to
suggest (1976, 85) that Alex. DA 20,6-19 says little more than what Alex. would
have said in commenting on Arist.s argument at DA 1.5, 409b2-4.
190. [20,19-22] Although Alex. had earlier spoken of an analogia or proportion
between the forms of living things, namely, souls and the forms of the simpler
bodies that make up the bodies of living things (10,15 and 11,12), here he seems to
106 Notes to pages 47-48
have in mind just the analogy or parallel between souls and other forms in general,
including the simplest ones, those of the elements, such as the lightness of air or
the heaviness of earth (5,11-15; 9,14-26; 22,7-12.18-21), and even the forms of
artefacts (3,2-15), against AD 140. Alex.s objection appears to be that in none of
these other cases is the form plausibly a body like fire or air, or itself a mixture of
other bodies. If so, then Alex. is arguing that an advantage of his naturalistic
approach is its uniformity in the treatment of form: souls are just one kind of form
among many, differing only in the specific powers and activities they concern; their
general characteristics, in contrast, are shared by all forms, from those of the
simplest natural bodies to the products of art.
191. [20,22-6] Alex. aims to drive a wedge between his opponents corporealism
and their commitment to the notion of form: it would be more consistent for them
to hold that the soul was the form of fire, rather than the elemental body itself.
Alex. of course would not accept this modification either, because of the immense
complexity which on his view distinguishes the form and matter of living things
from that of simple bodies (8,25-11,13).
AD (140) take Alex.s use of the optative here to indicate that his opponents did
not explicitly identify the soul as form with a simple body. But 19,21-3 leaves little
doubt that they spoke of soul both as a form and as a body like fire or air; so the
only thing left open to question is whether they regarded fire and air as simple
bodies, which they might not have, if they regarded them as mixtures (19,26-9).
192. [20,26-30] 14,4-5; 15,9-26; 79,19-20. Although Alex. has already discussed
and rejected the notion that the soul might be like the pilot of a ship, he mentions
it here as another way of developing the corporealist suggestion that the soul,
though a form of the body, is itself a body; it is possible that this was part of his
opponents actual view. Of the two options he considered earlier, this corresponds
to the second (15,13-26), which regards the pilot as a concrete individual, which
can exist separately from the ship, rather than a completion or entelekheia (15,10-
13). Alex.s first objection is simply that the pilot, so construed, is not necessary for
the ships being a ship and so cannot be its culmination and form.
193. [21,4-5] 16,7; Arist. DA 2.1, 412a27-8.
194. [21,1-10] The second argument considers a more nuanced conception of the
pilot: on this view, he is part of the completion of the ship, in so far as he is
necessary to its activity as a ship. But Alex.s objection turns on the same point as
before: since the pilot is separable, the ship remains a ship without the pilot; hence
the pilot contributes at most to the second completion of the ship, not the first. But
there cannot similarly be an animal without a soul; and it would be an animal only
in name or homonymously: see n. 133 on 15,2-5. The soul is thus a first, and not a
second, completion and so is like the shape of the ship rather than its pilot.
195. [21,10-11] Alex.s line of reasoning here and in what immediately follows
is very similar to his earlier treatment: see 15,15-17. The duplication of arguments
suggests a certain amount of editorial indecision. But it is difficult to decide where
the arguments are better suited. Systematically, they fit quite naturally in his
earlier discussion of how the soul is in a body and whether it constitutes a distinct
sense of being in; but the arguments also look as though they were inserted there
as an afterthought. In the present passage, in contrast, there seems to be little
systematic reason to raise the pilot analogy again. But this suggests that the
reason it was included here is that the actual proponents the theory now under
discussion invoked it.
For the general assumption that the soul pervades the entire body, see n. 104
on 12,2-3; also 20,6-7; Mant. 104,21-2.
Notes to page 48 107
196. [21,11-13] 15,17-21. AD detect in the use of the aorist participle khristhe-
isan here a suggestion of a possible resurrection after death (142), similar to the
parallel passage in Arist. (DA 1.3, 406b4-5). But nothing in the present passage
requires that the separation in question occur at death.
197. [21,13-16] This objection is only hinted at in the earlier version, when
Alex. states that the soul, so construed, would move the body by force (biai, 15,17).
The final remark, that on this conception the soul would be a body (21,16), marks
Alex.s disagreement, but it hardly constitutes an independent objection, since his
opponents here are openly corporealist. AD 143 suggest that it is meant instead to
be an embarrassment to Platonic or, more likely, Platonising opponents, possibly
even the Peripatetic Xenarchus (first century BCE; cf. Aet. 4.3.10, Dox. Gr.
388.16-20), whom they conjecture are the source of the comparison with the pilot
(140-2). But this is just speculation; there is nothing in the text that requires the
hypothesis. Moreover, if Platonists were the target, Alex.s objection at 21,11-13
and 15,17-21 would completely misfire, since from their perspective the possibility
of the souls exiting and entering the body, which Alex. rejects out of hand as
absurd, would be regarded as a positive advantage of the theory, not a flaw.
On forced motion and bodily resistance, cf. 79,10-11 below. BD also cite a highly
relevant fragment of Alex. quoted by Simplicius (in Cael. 379,18-23), which simi-
larly argues, regarding the soul of the heavens, that if a soul is moved by force, it
must be endowed with a body, since it is impossible for a body to be moved by
something else without contact and resistance, which are bodily features (20-1).
For further discussion, see the rest of their comment at 254.
198. [21,16-21] Depending on how we construe the phrase in this way (houts,
21,17) in the first clause, Alex. might be challenging either the view that (a) the
soul is like a pilot, or (b) the soul is one of the simple elements, or more broadly
still the view that (c) the soul is a body. Since this last claim is the one most recently
mentioned in context (21,16), and none of Alex.s criticisms here turn on any
specific form this view might take, (c) is the most reasonable option: Alex. is raising
a general worry about how the soul and the body can constitute a single, unified
living thing on corporealist theories, especially given how unlike each other they
are in nature (tas phuseis diapheronta, 20,19). Although Arist. himself does not
press this issue against his predecessors in this exact form, he does complain that
they give little thought to what enables soul and body to interact together (DA 1.3,
407b13-26), in contrast with his own theory, where there is no more question of
why they are one than there is with wax and its shape (2.1, 412b6-9). This question
about the soul-body relation is quite distinct from his worry about what makes the
soul one or the body one, and whether either gives the other its unity (1.5,
411b6-19). Still less does it make an explicit statement about the mortality of the
soul (pace AD 141, 143).
199. [21,22] Alex. has been arguing at some length that the soul is the form of
the living body (see n. 157 on 17,9 above) and so as a consequence has three
characteristics (17,9-10): it is (i) inseparable from the body (17,11-15), though (ii)
non-bodily (17,15-20,26), and (iii) incapable of undergoing change on its own. In
the present section, he argues for the last of these three characteristics.
AD, we should note, reject the section break in MS V (preserved above and by
Bruns), taking 21,22-4 instead as the end of the previous section, where it is
supposed to make a merely apparent concession, intended ironically, to Platonic
opponents (143). But 21,22-4 can be much more straightforwardly understood as
exploring the consequences of Alex.s own position, since he himself holds that the
soul is inseparable. The sentence that immediately follows at 21,24-5, moreover,
108 Notes to pages 48-49
draws out the consequences of 21,22-4 and links them directly to the subject of the
present section, and hence should not be separated from it.
200. [21,24] Alex. addresses the question of individual survival much more
directly than Arist., who generally speaks only about whether the soul or its parts
are separable (khristos) from the body (DA 2.1, 413a4-7; 2.2, 413b27-8). Alex.
refrains from denying personal immortality here outright, because on his view the
human understanding (nous) lacks a dedicated organ (84,10-24) and so is not a
form of part of the body; whether this enables it to survive death is left unsettled
here. In this respect, Alex. follows Arist., who frequently hedges his claims with
the suggestion that the understanding might prove an exception: DA 1.1, 403a8-12;
1.4, 408b18-25; 2.2, 413b24-7; GA 2.3, 736b21-9; cf. DA 2.3, 414b18-19, 415a11-12.
In the end, Arist. distinguishes two understandings, one of which he declares to be
immortal and eternal, while the other is perishable (DA 3.5, 430a22-5); but he does
not make fully explicit what this implies for individual humans. In contrast, when
Alex. draws a similar distinction at DA 90,11-91,6, he states explicitly that the
immortal intellect only comes to be in us from outside and is not itself a part or
disposition of our soul (90,23-91,4); what he calls the material understanding
perishes along with the rest of the human soul (90,13-16). The question of human
mortality is arguably present in the background throughout Alex.s treatise, hinted
at already at the very beginning: see n. 8 on 1,4-2,4.
201. [21,25-6] cf. 78.24-5. Alex. is now beginning to pick up the sequence of
Arist. DA again with 1.3, 405b31-406a4. (His last extensive comment on Arist.s
DA was at 12,9-13,8, concerning Arist. DA 1.1, 403a3-25.) For the argument that
not everything responsible for a change need undergo change itself, and that some
of them in fact must not be changed, see Arist. Phys. 8.5, 256b13-24, 258b4-9 (cf.
258a1-4); GC 1.6, 323a13-14; 1.7, 324a30-4; cf. Metaph. 12.7, 1072a25.
202. [21,29-22,2] For the claim that what produces change by contact must
itself undergo change, see Arist. Phys. 3.2, 202a5-9; GC 1.6, 322b32-323a12; cf.
Alex. Quaest. 2.23, 74,5-6. The four types of change Alex. mentions correspond
exactly to Arist.s list of the four basic types of motion produced by something else
(hup allou phora), in terms of which all other spatial motions can be accounted for:
Phys. 7.2, 243a16-244a4; for the details of the reduction, see Wardy 1990, 127-9.
Alex. gives a slightly enlarged list of bodily motion through contact, as BD note
(255), at Mant. 122,22-123,3. AD question (144) whether Alex. might be diverging
from Arist., who considers all types of change and not just spatial change at DA
1.3, 406a13-14. But there is little divergence, if any: Arist. himself immediately
goes on to argue that all of these changes occur in space and so therefore would
imply that the soul is in place itself (406a16-22).
203. [22,5-6] That is, they are all inseparable on their own from the things that
have them. The soul was argued to be a form at length above: see n. 157 on 17,9.
204. [22,5-12] Arist. argues prominently and at length that the soul does not
move the body or produce change of any sort by undergoing it itself in DA 1.3 and
portions of 1.4 (408a4-b18, cf. 408b30-409a30) in fact, he regards it as an
impossibility (406a2), a position that sets him apart from materialists like De-
mocritus and dualists like Plato (see Menn 2002). But Arist. nowhere offers the
general reason that Alex. does here, that this is a characteristic of powers and
dispositions generally, and hence of forms like the soul. The closest Arist. comes is
to say that arts or tekhnai produce their effects without being modified themselves
(GC 1.7, 324a34-b6), and they are powers and principles of change (Metaph. 9.2,
1046b2-4) and dispositions (EN 6.4, 1140a6-10, a20-1). Similarly, he regards the
endpoint of every change, whether a form, a modification, or a place, not to be
Notes to page 49 109
subject to change itself, since it is what the change is a change to (Phys. 5.1,
224b11-13, cf. 4-7). Ian Flora has also suggested to me that the point about
separability is implicit in the contrast between natural and mathematical sciences
at Metaph. 6.1, 1026a13-16; 11.7, 1064a31-3. (Arist.s point in DA 2.5, that certain
activities of the soul like perception are not modifications is quite different, since
it does not turn on the proper subject of change, as here, but on the type of change
undergone.)
Alex.s point in the present passage is fundamentally metaphysical. To claim
that the soul undergoes change is in effect to make a category mistake: it is the
wrong sort of thing to undergo change. For something to undergo change, it must
be a subject in its own right and hence separable; anything which is inseparable
cannot undergo change even if it is responsible for that change. Alex.s point is thus
much more general: it applies not just to the soul, but to all forms, powers, and
dispositions, animate or inanimate, all the way down to the elemental level. Here
as elsewhere, Alex. underscores the continuity of psychology with the rest of
natural philosophy by invoking the example of earths heaviness or fires lightness:
see nn. 40 and 190 on 5,5-6 (cf. n. 43 on 5,11-15) and 20,19-22 respectively, as well
as 22,18-20 and 23,29-24,1 below.
Alex. is reported elsewhere (Simpl. in Phys. 964,9-23 and Parisinus suppl. gr.
643, 101r, as reported in Rashed [1997] 2007, 146-7) to have argued that the soul
cannot undergo change because it is without parts and according to Arist., nothing
without parts can undergo change (Phys. 6.4, 234b10-20; 8.6, 258b24-6). Prima
facie, this appears to conflict with the present treatise, which assumes that the
soul has parts: see 18,26-19,3 above and more generally under meros ts psukhs
and morion ts psukhs under psukh in the Greek-English Index. But these are
not parts in the sense at issue in the argument. Arist. himself, it should be noted,
does not apply the argument from Phys. 6.4 to the soul, although he does employ
a version of it against Xenocrates theory of the soul at DA 1.4, 409a1-3.
205. [22,13-15] Arist. distinguishes between somethings undergoing change on
its own or intrinsically (kath hauto) and undergoing it in virtue of something else
(kath heteron) or extrinsically (kata sumbebkos) and argues that the soul is
changed only extrinsically: DA 1.3, 406a4-22; 1.4, 408a30-4; for the general distinc-
tion, cf. Phys. 8.4, 254b7-12. Although he rarely uses the expression, he does speak
of joint change in the relevant sense at Top. 2.7, 113a29-30.
Alex. continues to be more general and metaphysical in his reasoning than
Arist. (see n. 204 on 22,5-12 above): not just the soul, but every form undergoes
change jointly (sunkineisthai) with its body, precisely because the form is insepa-
rable from it; it is necessarily a part of the changing thing. But talk of joint change
amounts to little more than saying that the composite of matter and form under-
goes change, since Alex. clearly does not take it to imply that the form itself
undergoes change intrinsically.
206. [22,15-23] Alex. again stresses that what holds for the soul holds equally
for inanimate things too, right down to the elemental level (22,18-20; see n. 204 on
22,5-12 above). In fact, he is careful to point out that this is so not merely with
regard to changes that animate and inanimate objects share, such as forced motion
due to external causes, but also with respect to changes that belong to living things
precisely in virtue of being alive. The distinction between different types of motion
unsurprisingly has Aristotelian antecedents: see Phys. 8.4, 254b15-20 (although
note the still finer distinctions at 8.6, 259b1-14).
207. [22,24-23,6] Alex. need not have a specific opponent in mind here, since
Arist. represents his predecessors as generally committed to this view, from
110 Notes to pages 49-50
Democritus to Plato, though in so far as it specifically targets the claim that change
is part of the essence of the soul (22,27), he may have Plato especially in mind, who
uses this to argue for the souls immortality (Phdr. 245C-E). AD 144-5 rightly note
that Alex. uses the first Stoic inference schema (modus ponens) as part of a reductio
ad absurdum, since he will reject both alternatives at 23,6ff.
208. [22,23-7] Arist. takes the natures of things to be principles that govern
both the initiation of a type of change and its cessation (Phys. 2.1, 192b8-23), and
so generally takes change and rest to be correlative and on a par with one another
(5.6, 229b23-8). But if the soul can equally be at rest or changing, then change does
not belong to the essence of the soul: cf. Alex. in Top. 321,12-16. On the meaning
of apoklrsis, see Sharples 2004, 75 n. 236.
It is worth noting how much simpler and more direct Alex.s criticism of Platos
position is, using familiar Aristotelian doctrines, than Arist.s own fairly baroque
argument about change and the essence of the soul at DA 1.3, 406b11-15 (discussed
by Alex. in Quaest. 2.2, 46,22-47,27 and most recently in Shields 2007, esp. 152-4).
209. [23,2-3] On several occasions, Arist. characterises coming to know and
understand as coming to rest, in so far as arriving at a conclusion involves the
dissolution of uncertainty or disturbance: DA 1.3, 407a32-4, but also esp. Phys. 7.3,
247b10-248a6 (cf. 5.1, 224b13).
210. [23,4-5] All of the conditions Alex. describes here are bodily ones (see also
n. 114 on 12,20-1 above), even though acc. to Arist. understanding is the activity
that seems to belong most properly to the soul (DA 1.1, 403a8). But Alex.s
description here of bodily disturbance in the young and old, the drunk and the
sober, as an impediment to knowledge itself draws on Arist. Phys. 7.3, 247b13-
248b6; cf. DA 1.4, 408b22-5.
211. [23,6-24] This passage is an exposition and development of Arist. DA 1.4,
408a34-b18, which J. Barnes memorably described as the celebrated Rylean
passage ([1971/72] 1979, 33-4) because it seems to share Ryles basic insight (1949,
16-24) that to treat the mind as a subject of mental states in its own right, rather
than the person, is a category mistake. Even if Shields reservations (2007, 156-60)
about attributing Rylean concerns to Arist. were justified, it is plain that Alex.
understands Arist. in this way: the soul cannot be the subject of such activities
precisely because it is a power of the body and inseparable from it (22,2-6; cf.
23,24-8); to think it can be a subject, rather than the composite to which it belongs,
is just a category mistake. Thus the Rylean reading of Arist. goes back at least to
Alex.
212. [23,9-11] Notice that Alex.s claim here is completely general and un-
hedged: every activity that is characteristic of living things as such belongs to the
living thing and not the soul. The soul is a fortiori not the subject of cognitive
activities either, such as perception or even understanding, which on Alex.s view
lacks a dedicated bodily organ. The soul, as a form, is the power in virtue of which
living things engage in those activities, not what engages in them itself. Cf. 23,26-8
below.
213. [23,11-13] Because of the absence of manifest, external changes in the
body, we are misled in certain cases into thinking that a psychological activity
could only belong to the soul cases, presumably, like those listed at 23,15-17.
Alex. seems to suggest that if we had better knowledge of the inner workings of
living bodies, we would be less tempted to make this mistake. Arist. is actually
more explicit here, saying that the heart undergoes a specific change in anger and
fear and that thinking (dianoeisthai) either involves something of this sort or
perhaps something else, whether it involves movement or alteration or some other
Notes to page 50 111
kind of change (408b8-12). But Arist. likewise complains, in a quite different
context, of the faulty conclusions others draw from ignorance of the internal
workings of the body: Respir. 3, 471b23-9.
214. [23,15-21] Alex. expands Arist.s list of activities, by bringing together
other examples mentioned in the larger context of Arist.s remarks (408b5-30).
215. [23,20] The aulos is a double reed instrument much like a shawm or oboe
(West 1992, 81-5), especially classical and baroque oboes, which lack the elaborate
mechanical system of keys we are familiar with today.
216. [23,18-22] Alex. modifies Arist.s point slightly here. Arist. appeals to
artistic activities simply as clear, parallel cases where we take the human being to
be the agent, and so concludes we should treat psychological activities similarly.
Alex., in contrast, draws an analogy directly between art and the soul: both are
powers and dispositions, but neither is itself the agent. (Arist. regards art as
responsible for change, but not subject to change itself: for references, see n. 204
above on 22,5-12.)
Alex. also modifies Arist.s examples: he keeps weaving, but replaces house-
building with oboe-playing and wrestling. It may be to avoid making the point turn
on incidental features of these arts, by offering a broader range of examples: unlike
weaving, neither oboe-playing nor wrestling results in a concrete product, and
wrestling further does not use any other instrument than a persons own body.
217. [23,26-8] This sort of instrumentalism is emblematic of Platonic views
(Alc. I 129C-130A), and can be found in first- and second-century CE authors such
as Plutarch (de Pyth. Or. 21, 404B; Sept. Sap. 21, 163D) or Galen (UP 1.2, 1.13-2.2
Helmreich; see Moraux 1984, 778). But even Arist.s own characterisation of the soul
as the form of a body that is organikon (DA 2.1, 412a28-b6) could be taken as implying
that the living body is itself a tool or organ which the soul uses, and in fact Arist. does
occasionally say that the soul uses the body as an art or an artisan uses tools: DA 1.3,
407b25-7; EE 7.9, 1241b17-19, b22; Protr. B59 Dring = Iambl. Protr. 41.15-20 Pistelli.
For a reading of Arist. along these lines, see Menn 2002; also Plut. Quaest. Plat. 8.2,
1006D. (Thanks to Jan Opsomer for the Plutarch references.)
Alex. rejects this understanding of organikon, however (see n. 93 on 11,1 above)
and here he explicitly argues against an instrumentalist characterisation of the
souls relation to the body. The soul is comparable to an art on his view, but neither
art nor the soul can be said to use any instrument or organ. To think that they can
is a category mistake: only a separable substance, like an artisan, can use tools,
and the soul is not like an artisan in this regard (against EE 7.9, 1241b17-19, b22).
218. [23,29-24,4] Alex.s characterisation of the soul as a power, form, and
completion, which comes from a mixture and blend has not been stated this
succinctly before, but it was developed at length at 7,21-11,13; see esp. 8,10-13.15-
17; 9,11-12; 10,17-19.24-6; and 11,11-13, with their corresponding notes. The
example here of an elemental form, heaviness, again underscores the quite general
nature of this claim: see n. 204 on 22,5-12 above.
219. [24,5] Alex. maintains the Aristotelian view that the governing part is
located in the heart and its immediately surrounding region (39,21-2), a thesis he
defends at length in the final section of this treatise (94,7-100,17), as a kind of
appendix (see n. 3 on 1,2-4). In speaking of a leading or governing part (to
hgmonikon) of the soul Alex. is adopting the popular terminology of the Stoics,
who prominently use it for the central faculty of the soul, which they also locate in
the heart: see 98,24-5 below (= SVF 2.839); also Aet. 4.21, Dox. Gr. 410.25-9 = SVF
2.836; D.L. 7.159 = SVF 2.837. Mansfeld (1990, 3109 n. 222) takes the adoption of
Stoic terminology to be significant and an indication that the question has already
112 Notes to page 50
become a scholastic topos. On the Stoic conception generally, see Annas 1992,
61-70. This consensus appears to have held until Diogenes of Babylon; on later,
heterodox Stoics, see Mansfeld 1990, 3095 n. 145.
The Peripatetic Strato (d. 269 BCE), who succeeded Theophrastus as third head
of the Lyceum, is reported to have had views about the hgemonikon: fr. 111 Wehrli
= Plut. de Lib. et Aegr. 4, 53.13-54.9 Ziegler-Pohlenz at 53.22-3. But the terminol-
ogy in our sources is more likely due to the influence of Stoic terminology, than
evidence of independent Peripatetic terminology.
The core notion of a central, governing part of the soul, though, is undeniably
in Arist. himself. He says that in deliberation, we reason until we can trace back
the beginning of action to the governing part of oneself (hautou eis to
hgoumenon), and it is this that makes decisions (prohairoumenon, EN 3.3,
1113a5-7). He locates this principle, moreover, in the heart (or in bloodless animals
in what is analogous to the heart), which he speaks of as the chief or controlling
part of the animal (kurion, kuritat, Iuv. 3, 469a4-10 (cf. 4, 469b11-13); cf. Metaph.
7.10, 1035b25-7) which operates from a governing location (arkhontos khra, Iuv.
4, 469a33-4). He identifies it more specifically as the source of each of the animals
principal activities. It is, as we have seen, the source of the animals movements:
PA 2.1, 647a24-6; 3.3, 665a10-15; 3.4, 666b14-15; MA 7, 701b24-32; 9, 702b12-28.
But it also serves as the source of perception and life more broadly: Somn. 2,
455b34-456a6; Iuv. 1, 467b18-27; 2, 468a20-3; 3, 469a5-20; 4, 469a23-34, b1-6; PA
2.1, 647a24-31; 2.10, 656a27-9; 3.3, 665a10-15; 3.4, 666a11-18, a34-5; 4.5, 678b1-4;
MA 11, 703b23-4; cf. Metaph. 5.1, 1013a4-6; 7.10, 1035b25-7. Even thinking
(dianoeisthai) depends crucially on internal parts such as the heart: cf. DA 1.4,
408b7-12, cf. b25-7. For a discussion of the hearts central role and its compatibility
with Arist.s mature hylomorphism, see Tracy 1983; in fact, BD 257 explicitly
compare Tracys compatibilist position with Alexanders.
220. [24,5-6] In describing the part of the body where the governing part is
located as animate prts and in its own right, Alex. may be claiming (a) that it is
the first part of the animals embryo to be formed and so to be animate on its own,
in contrast with a parents seed, which is animate only in so far as it is part of
another living thing; or (b) that the source and principle of the activities of animate
things as such is itself pre-eminently animate; or (c) both of these to some extent;
and depending on which is in play, Alex.s explanation (gar) of where the governing
part is located will differ accordingly. In favor of (a) is Arist.s frequent claim that
the heart is the first part of the animal to be formed and so neither prior nor
posterior to the new living being: Iuv. 3, 468b28-31; PA 3.4, 666a10-11, 20-2; GA
2.5, 741b15-16; 2.6, 742b1-2; Metaph. 7.10, 1035b25-7. Alex. appeals to this tem-
poral priority below as evidence that the source and principle of nutritive activities
is located in the heart (95,6-8). But (b) cannot be ruled out either or hence (c). Alex.
uses prts in the sense of pre-eminently at 89,2 in explicating one of his central
tenets about explanation and the sorts of things that can serve as grounds or
principles (88,26-89,8). Arist. himself remarks that the essence and defining
character of a living being in some sense resides in the first, governing [part],
such as the heart or brain (Metaph. 7.10, 1035b25-7).
221. [24,6-7] It is not the governing part of the soul, but a part of the living body,
the heart, in which the governing part resides, that uses the organic or instrumen-
tal parts (tois organikois meresin). As Alex. goes on to make clear, this includes
sense organs as well (24,9-11). For Arist., this phrase primarily designates the
limbs, which are constituted out of bones, muscle, and flesh (PA 2.2, 647b22-5) and
are contrasted with the senses as heterogeneous structures that subserve motion
Notes to pages 50-51 113
and other activities (PA 2.1, 647a2-5, a24-30; cf. HA 1.6, 491a26). See also DA 3.9,
432b25-6; IA, 3, 705a20; 4, 705b22; EN 3.1, 1110a15-18; cf. GA 2.6, 742a28-b6.
Alex.s use of this phrase elsewhere (Eth. Prob. 9, 129,22; 12, 133,12) is in keeping
with Arist.s usage.
222. [24,7-9] Alexanders comparison with art relies on the original sense of
organon as a tool. Like art, nature uses tools as well, except that in this case the
tools or organs are an intrinsic and congenital part of the natural substance: see
Arist. GA 2.1, 735a2-4; cf. 1.22, 730b11-24.
223. [24,13] Alex. has argued at length in this treatise that the soul is a form
(see n. 157 on 17,9). But he may have in mind the specific point here that forms
are responsible for the differences between things with regard to causal powers
(7,9-11.13).
224. [24,11-15] The passage closely follows Arist. DA 2.4, 415b8-28, which
likewise characterises the soul as what is responsible for the living body as its
origin and prinicple (b8) in three of the ways that something can be responsible
(b8-12): as the essence of the living thing (b12-15); as that for the sake of which the
body functions (b15-21); and as the source from which change originates (b21-8).
225. [24,16-17] 28,8, with corresponding note; see also in Metaph. 121,16-18;
160,17-18; 349,14-16; Quaest. 3.12, 102,31-103,1; cf. in Top. 171,21-172,3. Al-
though this statement might sound like an affirmation of universal teleology, it
contains an important restriction. For an Aristotelian, not everything in the
natural world is produced by nature: this subset contrasts not only with things
produced by art (and so by reason), but also with those which arise by chance or
spontaneously (Arist. Phys. 2.5, 196b17-33). Arist. even denies that eclipses have
a final cause (Metaph. 8.4, 1044b8-12). For Alex.s discussion of these distinctions,
see esp. Fat. 4, 167,16-168,24.
226. [24,18-20] Alex., of course, is one of the people who subscribe to (1): on
supervenience, see n. 229 on 25,2.
The claim that the soul is a harmonia, (2), first appears in Platos Phaedo, when
the Pythagorean Simmias suggests that the soul is a blending or tempering of
elemental forces in the body, similar to the harmonia or tuning of a lyre, in
opposition to one of Socrates arguments for immortality (85E-86C). The English
cognate harmony is to a certain extent a false friend, since a harmonia is not in
the first instance something heard in fact, Alex. explicitly denies that it is below
at 25,24-26,2 but rather the harmonious fitting together of the parts of the
instrument that can produce such sounds, its being tuned in a certain way. The
Greek term originally applied even more broadly still, to any artefact that has been
well joined or fitted together, as when Alex. speaks of the harmonia of a bench or
a temple below at 25,16-19; and it was common to think of harmonia in natural
cases too, such as health of the body (see 25,4-7 below). Given the range of cases,
it would be awkward to use the English tuning, even though more correct,
especially since we speak of non-musical items as being in harmony or harmoni-
ously constructed. For more on the meaning and etymology of the term, see my
1997, 219-22, esp. nn. 21, 23, and 27 for references.
Both Plato and Arist. reject the view that the soul is a harmonia, although not
for entirely the same reasons (Plato Phaedo 91E-95A; Arist. frr. 37-47 Rose3; DA
1.4, 407b27-408a28). One of their shared concerns is mental causation. In Arist.s
view, a harmonia cannot be responsible for change, whereas he regards this as one
of the most characteristic features of the soul (DA 1.4, 407b34-408a5). The concern
with causation is obviously important for Alex. too, who identifies the soul as a
power and disposition (see, e.g., 22,5-6 above). The blend of the underlying body
114 Notes to page 51
determines which powers supervene on it (see 10,17-19 and 24-6, with correspond-
ing notes), but these powers are not thereby identical with or reducible to the
blend. They are genuinely new causal powers that emerge (see n. 40 on 5,5-6). On
the harmonia theory and the subsequent debate over it, including Alex. and Galen,
see my 1997.
227. [24,22-3] On emergence, see n. 40 on 5,5-6 above; also 25,7-9; 26,27-9; cf.
24,28.
228. [24,23-25,2] Alex.s observation about drugs is meant to be simple and
direct: in their case, he thinks, it is just evident that the power of the mixture is
one thing and the proportion of ingredients quite another; if the soul is a power,
then it too will differ from any underlying proportion or harmony. BD offer an
alternative reading (258), based on differences in the power of different dosages,
which nonetheless maintain the same proportion of ingredients, but this is unnec-
essary and much more complex than what the text actually says.
Medical drugs are used as a star example by the Stoics and above all by Galen,
who often speaks of the powers of the four-fold drug (tetrapharmakos), which is
not the same as any of its ingredients. For the Stoics: Alex. Mixt. 3, 216,4; 4, 218,6
(= SVF 2.473); Philo Confus. Ling. 187, 2.265.11 Cohn-Wendland (= SVF 2.472).
For Galen, see Hipp. Elem. 1.455 K; MM 10.281, 352, 353, 882, 883 K; MMG 11.84,
138 K; SMT 12.328 K; Comp. Med. Loc. 12.602, 604, 610 K; Comp. Med. Gen. 13,
593 K; HNH 15.18 K; adv. Lyc. 18a.240, 241 K; Hipp. Fract. 18b.598 K, in addition
to those cited by Moraux (1984, 740, n. 214), namely, Hipp. Elem. 1.452 K; HNH
15.32 K; CAM 1.242 K; CC, CMG Suppl., 2.3.2. For discussion, see my 1997, 350.
229. [25,2] The verb rendered here as supervenes, epigignesthai, is not an
innovation on Alex.s part: Arist. himself speaks of form supervening on matter at
Metaph. 7.10, 1035a5, a12; 7.11, 1036a31-2, b6. For Alex.s conception of superven-
ience, see n. 40 on 5,5-6, n. 92 on 10,24-6, and n. 90 on 10,17-19; cf. 24,18-20 above
and 25,7-9 and 26,22 below.
230. [25,2-4] In articulating his position, Alex. may be distinguishing himself
carefully from his predecessors. Galen reports (QAM 4, 44.12-20 Mller, transla-
tion available in Sorabji 2004a, 1.187, 6a, text 13) that Andronicus of Rhodes, the
first century BCE Aristotelian and editor of Aristotles works, says that the soul
is either a blend or a power that follows on the blend (toi krasin einai phasin
dunamin hepomenn ti krasei, 18-19) and criticises Andronicus for adding power
rather than just identifying the soul with the blend, as Galen does himself
(44.20-45.3; cf. 37.5-23). Alex.s position can be seen as directly opposing Galens
criticism, by defending the second of Andronicus two alternatives (against the
harmonising interpretation of Michael of Ephesus, in Parva nat. 135,22-30). For
discussion on these differences, with further references, see my 1997, 351-2; on
Andronicus views about the soul, see Moraux 1973, 132-4.
231. [25,4-7] Arist. DA 1.4, 408a1-5 (though, as Ian Flora points out, Alex.
thankfully sidesteps Arist.s pun here). Arist. takes health to be constituted by a
blend and balance (en krasei kai summetriai) of hot and cold: see esp. Phys. 7.3,
246b3-10; cf. also Top. 6.2, 139b21; 6.6, 145b7-11; PA 3.12, 673b26. At Metaph. 7.7,
1032b7 and 18-20 he suggests that it consists in an equable state (homalots). On
Arist.s view within the context of the Greek medical and philosophical tradition,
see Tracy 1969, esp. 157-63.
232. [25,2-9] Alex.s strategy here is that of a non-reductive materialist: al-
though the soul supervenes on materials composed in a certain way, it is not
identical with these materials, their composition, or the materials composed in this
way (in the way that health might be); rather, the soul is a distinctive new power
Notes to pages 51-52 115
that emerges from this combination. The ontological distinction between what
supervenes on certain base conditions and the base conditions themselves is what
allows him to block the inference at 24,18-20: the soul depends on matter, but it is
not identical with matter. The firmness of this distinction allows Alex. a more
eirenic and nuanced view. In his overall account, the harmonia of the body its
being composed, mixed, and blended in appropriate ways has a legitimate and
important role, as part of the base conditions on which the soul supervenes. It is
just not what the soul itself is.
Arist.s response to the harmonia theory, in contrast, is more aggressive and
less forthright. He issues a broadside attack on the theory, disambiguating harmo-
nia into two senses and then categorically rejecting each possibility with a barrage
of rhetorical questions, suggesting the hopelessness of any such identification (DA
1.4, 408a10-13 and a13-26, respectively). But he is silent on whether there is
anything like a harmonia in his theory, leaving him open to the suspicion that he
protests too much.
In what follows, Alex. accepts Arist.s disambiguation, but as part of a positive
exploration of what a harmonia is (25,9-23), not as a criticism. He reserves his
objections solely for the view that the soul is identical with a harmonia (26,3-13),
something he thinks other theorists come closer to than Arist. does himself (26,13-30).
233. [25,9-15] Arist. DA 1.4, 408a5-9. Alex. follows Arist.s characterisation of
the two senses of harmonia almost verbatim, adding only (i) that the elements of
a harmony in the first sense cannot have anything of the same sort in between
them (which, as BD 259 note, is how Arist. characterises things in succession at
Phys. 5.3, 226b34-227a6 and 6.1, 231a23); and (ii) the examples of specific ratios.
But unlike Arist., who proceeds directly to criticisms of the harmonia theory (see
n. 232 on 25,2-9 above), Alex. just incorporates these two senses of the term into
his explanation of what harmonia might mean, without using them as a key part
of his objection.
234. [25,15-23] Regardless of whether we take a harmonia to be the concrete
compounding or composition (sunthesis) of elements or instead the abstract pro-
portion (logos) by which they are composed, the harmonia will be distinct from the
compound that arises from (to ginomenon ek) this mixture. This ontological distinc-
tion does not directly challenge the harmonia theory, which never claimed that the
harmonia is the living being as a whole. It is simply part of Alex.s exploration of
what precisely a harmonia is.
235. [25,23-26,3] The harmonia of a musical instrument, like a cithera or lyre,
is not any melody or sound it produces, but the underlying tuning which is
responsible for those sounds. In this respect, the English harmony is a false
friend: see n. 226 above on 24,18-20.
236. [26,3-7] The initial objection is that having a certain proportion is not
sufficient for being in harmony, since a given proportion is a harmony in some
cases and not others. Musical cases of harmony are clearly regarded as paradig-
matic here. What is unclear is whether the objection contests the earlier,
non-musical cases of harmonia given at 25,4-7 and 16-19, or only certain cases like
honeyed wine (on which, cf. 13,22 and 15,7). Rejecting all non-musical cases would
involve a narrowing of the original semantic range of harmonia: see n. 226 on
24,18-20 above. But the objection does not require such a narrowing: the argument
would work just as well so long as there are certain cases, like honeyed wine, where
there is a proportion, but a native Greek speaker would not use the word harmo-
nia. Alex. does not seem to share this linguistic intuition in any case: in what
preceded, he does not mark off wider uses of harmonia as infelicitous.
116 Notes to page 52
237. [26,7-13] Despite Alex.s use here of the explanatory particle for (gar), this
objection is logically independent of the preceding one. For even if a given
proportion always constituted a harmony, wherever it occurred, it would not
follow that all proportions are harmonies, the point contested here. The sug-
gestion that only determinate proportions are harmonies would be familiar
from Greek music theory, where all the tones of musical scales are taken to be
ratios of whole numbers, and so determinate and commensurable: Arist.
extends this requirement to colours at Sens. 3, 439b25-440a3, suggesting that
the generally recognised ones might all be due to determinate ratios of white
and black, analogous to musical tones, while indeterminate ones are infinite in
number; see also Alex. in Sens. 54,4-55,8. If this is right, then the soul could
not be identified with a proportion tout court, but only with certain commensu-
rable proportions.
Alex. immediately goes on to develop two further objections, which effectively
constitute a dilemma, depending on whether or not one accepts the claim that
every harmony is a determinate proportion:
If (1) harmonies are restricted to determinate proportions, there is a further
disanalogy with musical harmonies (26,9-11): the mixture of elemental powers in
living things is not limited to a single, fully determinate proportion, but can vary
within a certain range; hence the soul cannot be identified with any one of the
proportions in this range rather than another.
This variation has greater significance than Alex. acknowledges here, since it
should pose an analogous difficulty for his own theory: though not identical with
the harmony and proportion of elemental powers mixed in the body, the soul
nevertheless supervenes on them (see n. 229 on 25,2 above); therefore, if living
things can tolerate a certain amount of variation in these proportions, the soul
must supervene on a range of proportions, rather than just one. This is not so much
an inconsistency for Alex. as a complication, since supervenience allows for the
possibility of multiple realisation: a higher-order power can supervene on a num-
ber of different underlying bases, each of which is sufficient individually for the
presence of the higher-order power. By acknowledging that the underlying powers
in the living beings can vary in strength as any Aristotelian surely must, given
the ordinary changes in heat and cold that occur in digestion and the movement of
limbs (a point well made by AD, 149) Alex. is committed to multiple realisation:
if the soul is to supervene on the underlying mixture, it must be the case that any
proportion within certain limits (hes tinos, 26,10) will suffice for the soul and its
powers. But only within these limits: if the underlying conditions range beyond
them, they will no longer support the existence of a soul.
If, on the other hand, (2) harmonies are not restricted to determinate propor-
tions, the harmonia theory will face a different problem (26,11-13): there will be
too many harmonies, since every composite will be a mixture of elements or powers
in some proportion, whether commensurable or not. But then, the worry seems to
be, there would be nothing distinctive about the souls being a proportion, beyond
perhaps being some specific one. On Alex.s view, in contrast, there is an important
difference: even if all composite things exhibit proportions, different powers will
supervene on different mixtures; and the soul will be identified by specific powers
and not the proportions as such on which it supervenes.
238. [26,13-15] If the opposing view is to be distinct from Alex.s own, as AD
rightly point out (150), it must be understood in such a way that the soul is
constituted from the mixture and so in some sense the same as it, rather than (as
on Alex.s view), a distinct power which emerges from it.
Notes to pages 52-53 117
239. [26,15-17] On the place of 26,13-20 generally in the doxographic tradition,
see Mansfeld 1990, 3108-9.
On the Stoic view that the soul is a type of breath or pneuma, already attributed
to Zeno (D.L. 7.157 = SVF 1.135) and Cleanthes (Sext. Emp. Pyrr. Hyp. 2.70 = SVF
1.484), see n. 178 on 19,6-9 above. The claim that this pneuma is specifically a
blend of fire and air, however, appears to trace back to Chrysippus: see Galen QAM
4, 45.5-25 Mller (= SVF 2.787), which, as Ian Flora points out to me, is an
especially close parallel to Alex.s report, and immediately follows the passage
about Andronicus which Alex. may have been responding to, discussed above in n.
230 on 25,2-4. For the more general view that pneuma is a mixture of air and fire,
see Alex. Mixt. 10, 224,14-15 (= SVF 2.442); 11, 225,5-8 (= SVF 2.310); Galen PHP
5.3.8, 306,23-7 De Lacy (= SVF 2.841); D.L. 7.156 (= SVF 2.774); cf. Simpl. in Categ.
218,1 (= SVF 2.389). For a discussion of the evidence, and in particular the value
of Alex.s reports, see Sorabji 1988, 85-9.
240. [26,17-18] Epicurus is reported to have said that the soul was a blend
(krama) of four things, one of which is like fire, another like air, another like breath
or wind and a fourth without any name, which is responsible for awareness: Aet.
4.3.11, 388.21-389.8 = Usener 315 = Arrighetti2 159 = LS 14C; cf. Plut. Adv. Colot.
20, 1118D-E = Usener 314= Arrighetti2 158. See Kerferd 1971, 90. For Epicurus
conception of blending, see Alex. Mixt. 2, 214,28-215,8 = Usener 290.
241. [26,18-20] Alex. may have in mind here Platos description of the creation
of the world soul at Tim. 34C-35B, which is constituted from a complex mixture of
being, the same, and the different. But more likely he is simply relying on Arist.s
report that in the Timaeus Plato made the soul from the elements (ek tn
stoikhein, DA 1.2, 404b16-17). As Mansfeld points out (1990, 3109), Calcidius
reports that anti-Platonists had claimed that the Timaeus and the Phaedrus were
in conflict on just this point, of whether the soul was composite (in Tim. 228,
243.13-18 Waszink).
242. [26,21-2] See n. 230 on 25,2-4; also 24,19-23; 25,7-9.
243. [26,26-30] See n. 40 on 5,5-6 above; see also 24,22-3 (cf. 24,28); 25,7-9. Here
Alex. makes an even finer distinction than he had before, regarding the soul and
the bodies that underlie it. The underlying mixture of bodies on his view is in some
sense responsible for the new power that emerges from it and supervenes on it,
which he identifies with the soul: see 10,17-19 and 24-6. But neither the underly-
ing mixture nor its proportion constitutes the essence or being of the soul, that is,
what it is to be a soul. The soul gets its essence solely from the power itself, not the
mixture from which the power emerges.
244. [27,1-4] This sentence signals the end of the first part of the programme
laid out at the beginning of the treatise (1,1-3): Alex. indicates at 16,18 that he has
stated the essence of the soul, and then discusses various consequences of this
characterisation, including the fact that it is (i) inseparable from the body (17,11-
15), while nevertheless being (ii) non-bodily (17,15-20,26), (iii) incapable of
undergoing change in itself (21,22-24,17), and (iv) not a harmony (24,18-26,30).
Alex. now proceeds to the second part of his programme, concerning the parts or
powers of the soul, starting with the claim that there is more than one part
(27,4-28,3), though not infinitely many (28,4-13), and then characterising the
differences between them (28,13-30,17).
Alex. takes it for granted that the parts of the soul (mern auts, 27,3) are
powers (dunameis, 27,5). Nevertheless, more is at issue than simply differentiat-
ing powers: he is concerned more with how powers are grouped together as parts
within the whole soul and how they are distributed across different kinds of living
118 Notes to page 53
thing: see nn. 246 and 249 on 27,4-8 and 27,8-16 respectively. This distribution
shows that there are many parts or powers of the soul, even though there is only
one soul in each living thing: 30,2-3; cf. 28,3; Mant. 118,6-7. An independent set of
arguments for the plurality of powers is offered at Mant. 118,6-119,20.
245. [27,6-7] The reference to Democritus (= Luria fr. 459) may be little more
than an inference, as AD suggest (152), from Arist.s assertion that Democritus
says that the soul (psukh) and the understanding (nous) are the same (DA 1.2,
404a27-8, 405a8-9 = DK 68 A 101) and that he makes all the objects of senses into
objects of touch (Sens. 4, 442a29-b1 = DK 68 A 119); the evidence from Philoponus,
as AD point out, is itself based on Alex. and so not independent (in DA 35,10-16 =
DK 68 A 105; in DA 71,19-34 = DK 68 A 113). But we should note that in context
Arist.s claims are dubious conjectures themselves, made for polemical purposes.
Democritus in fact speaks in a number of fragments about psychological conflict
and even akrasia (DK 68 B 214, 236, 234). By Alex.s standards (27,7-8), this should
commit Democritus to distinct parts of the soul in conflict with each other,
although his words only speak of a conflict between the individual and a part of
himself. On this point, see Kahn 1985, 16-17.
Democritus, however, is probably just a beard for the Stoics, who are more likely
Alex.s main target here. The Stoics attribute a strong form of unity to the soul.
Thus, while they recognise many parts of the soul, these are physical extensions
of the breath in the governing part to the various bodily organs (Aet. 4.21.1-4, Dox.
Gr. 410.25-411.24 = SVF 2.836 = LS 53H; Iambl. ap. Stob. 1.368.12-20 = SVF 2.826
= LS 53K); and for Chrysippus, at least, the characteristic activities of living things
are in an important sense activities of the governing part itself (Sen. Ep. 113.23; cf.
Aet. 4.21.1, Dox. Gr. 410.25-9) and so of a single thing being exercised in different
ways (Sext. Emp. Adv. Math. 7.307, 359 = SVF 2.849; Plut. de Virt. Mor. 3, 441C =
SVF 3.459). This is true, significantly, even in cases of psychological conflict:
struggles between reason and the passions are to be understood as the governing
part vacillating between different objects, being drawn one way and then the other
sequentially over time: Plut. de Virt. Mor. 7, 446F-447A = SVF 3.459 = LS 65G.
Alex. explicitly attributes this sort of vacillation to the Stoics, interestingly, with
regard to perceiving objects in different sense modalities at in Sens. 167,6-8.
246. [27,4-8] Mant. 118,6-9 (= SVF 2.823). Alex. here refers to the experience
of psychological conflict, of being torn in different directions, something which
occurs in the enkrats, someone who overcomes this conflict and maintains control,
as much as in the akrats, the person who loses control and does something he
believes he should not. The latter term is often translated as having a weak will,
but the Greek does not mention the will and is not limited to cases where a
specifically moral weakness is involved: any failure to do what one thinks should
be done, for whatever reason, is relevant.
Arist. does not make this first argument about parts. In general, his approach
is taxonomical: he emphasises that different types of living things along the scala
naturae are capable of different activities; and he takes this to imply that there are
distinct powers of the soul, some of which can exist apart from the rest (DA 2.2,
413a20-414a3; 2.3, 414a29-b1, 414b32-415a12). But such arguments would be
inadequate against the Stoics, who also differentiate living things along the scala
naturae, but regard each level as having a distinct and unified form of pneuma of
its own, which includes the powers of those beneath it (see Inwood 1985, 25-6): thus
plants have a nature, while animals have soul, and humans and gods possess
reason (Philo Quod Deus sit Immut. 35-46, 64.1-66.15 Cohn-Wendland = SVF
2.458 = LS 47Q; Leg. Alleg. 2.22-3, 1.95.7-20 Cohn-Wendland = SVF 2.458 = LS
Notes to page 53 119
47P). Arist.s arguments would likewise be ineffective against the sort of opponents
Alex. targets below (27,12-28,2), who take all souls to be endowed with the same
powers, even if some are unable to exercise them on account of the body they
currently inhabit. The fact that living things exhibit different activities is thus not
decisive for either group of philosophers.
Perhaps with this shortcoming in mind, Alex. chooses a quicker and more
decisive argument, drawn from Plato. In Republic IV, Socrates argues that the soul
has three parts: the appetitive, the competitive, and the rational (434D-441C). He
makes the crucial point that such conflict is not merely diachronic, but synchronic:
it is possible to want contrary things at a single time, something that would be
impossible, he claims, if the soul were undivided (436B-437A; see also Alcinous
Didask. 24, 176.41-177.3 Whittaker). Alex. plainly does not accept Platos triparti-
tion of the soul, but he does little more than allude to Platos argument here. He
does not explain which parts of the soul on his own account would be in conflict or
make further use of psychological conflict in the present treatise. It is striking that
he does not even name Plato: perhaps it is to distance himself from the specifics of
Platos account or otherwise diminish it. Alex. does, however, endorse the critical
point elsewhere that a single power cannot act in different ways at a single time
(Mant. 118,31-5). This claim is in fact stronger than Platos, which only denies that
a power can act in contrary ways at once, and as Richard Sorabji rightly points out,
even more controversial.
247. [27,9] Alex., like Arist., uses this principle about the natural order to argue
primarily for the absence of certain features or powers, which would be to no
purpose or superfluous (matn) in a particular species, were they present. For
Alex., see DA 28,4-6; 98,8-9; Mant. 157,15-18; 164,25-7; 183,25-6; Fat. 11, 178,11-
12; 179,25-6, but esp. Mant. 163,24-30, where Alex. bases this principle on an
analogy with art, characterising nature as a kind of divine art (tekhn tis theia).
The sage and virtue are also said to choose nothing superfluous: Mant. 165,29-30;
166,11. For Arist.s use of this principle, see Cael. 1.4, 271a33; 2.11, 291b13-14; DA
3.9, 432b21-6; 3.12, 434a31-2; Resp. 10, 476a11-15; PA 2.13, 658a8-9; 3.1, 661b23-
4; 4.11, 691b4; 4.13, 695b19; IA 2, 704b12-18; 8, 708a9-12; GA 2.5, 741b4-5; 2.6,
744a36-7; 5.8, 788b20-4; Polit. 1.8, 1256b20-2. For discussion of Arist.s use of this
principle and its variants, see Lennox [1997] 2001.
In a somewhat different usage, a process can be said to be to no purpose or in
vain when in a particular case it does not issue in its natural outcome. Arist.
appeals to this usage to provide an etymology for a fortuitous outcome (to
automaton) at Phys. 2.6, 197b22-30, which Alex. adopts at Mant. 178,26-9. But
that clearly is not the sense at issue here.
248. [27,9-11] Bruns inserts zia after tina at 27,9, following the Hebrew
translation, which restricts the point to variations between different kinds of
animals, presumably with regard to their different abilities to perceive objects in
different sense modalities. But Alex.s point is intelligible as it stands, and even
more poignant, as it would apply to plants as well as animals: the idea that plants,
as well as worms and anemones, might have the power of sight and hearing, but
just lack the organs, is quite obviously absurd. AD 153 add extra arguments for
this reading based on the exact form of the Greek expression.
249. [27,8-15] This second argument is also not present in Arist., but the target
here seems somewhat different than in the first. The issue is no longer simply
whether a theory distinguishes between different powers, but whether they are
distributed differently across living things, as Alex. and Arist. and even the Stoics
think (see n. 246 on 27,4-8). The opponent Alex. has in mind here thinks that all
120 Notes to pages 53-54
living things have the same powers, in virtue of having a soul, and differ only as
to which they can exercise, a difference which is to be explained by the body,
whereas Alex. and Arist. think that there are different types of soul, which only
overlap partially in their parts or powers. There is no reason to think that the
Stoics are still in view here, much less the primary target.
Alex.s objection to this view is not simply that we would be ascribing powers to
living things that could not exercise them: Arist. himself allows that a living thing
could continue to have a power, even though the organ had failed (DA 1.4,
408b19-25). The problem rather is that on this theory certain things could not
exercise these powers by nature, because of their proper bodily constitution, and
not by becoming deformed, injured, or decrepit. To have a power by nature that
cannot be exercised by nature is superfluous and violates a teleological conception
of the soul and its powers. On Arist.s view, a difference in powers implies a
difference in organs (GA 4.1, 766a22-3) and which organs something has is to be
explained by reference to its powers and activities, not vice versa: see esp. PA 4.10,
687a10-14; GA 1.2, 716b23-5; cf. also PA 1.5, 645b14-20; 4.12, 694b13-14.
250. [27,15-28,2] This argument is an extension of the preceding one, height-
ened by the comparison with metempsychosis or the transmigration of souls, which
Arist. also rejects out of hand: DA 1.3, 407b13-27; cf. 2.2, 414a22-5. The objection
does not turn so much on the religious elements, or its commitment to the
disembodiment of the soul and its successive reincarnations, but rather on the fact
that it severs any correlation between what powers a soul has and its bodily
organs.
Arist. associates this view with the Pythagoreans in the first passage cited
above, as it commonly is: Xenophanes quip about beating a puppy presupposes
that it is a view for which Pythagoras was already known at the time (D.L. 8.36 =
DK 21 B 7); the attribution is explicitly confirmed later by Aristotles student,
Dicaearchus (Porphyry, Vita Pyth. 19 des Places = fr. 40 Mirhady): for a review of
the evidence, see Huffman [2005] 2009, sec. 4.1, and more generally Huffman 2009.
Herodotus (2.123) suggests that the doctrine is Egyptian in origin. But the view is
certainly found later in Empedocles and Plato. For Empedocles: DK 31 B 115, 117,
126, 137, B 147, along with testimony from Diogenes of Oenoanda (fr. 42 Smith);
for a general discussion of Empedocles view, see Inwood [1992] 2001, 55-68; for
discussion of the Diogenes of Oenoanda fragment, see Inwood 2009. For Plato:
Phaedo 81B-82C, 107D-E, 113A; Phaedrus 248C-249C; Rep. 10, 614B-621B, esp.
617E-618A and 620A-C; Tim. 41D-42D, 76D-E, 90E-91A and 91D-92C; Laws 10,
903D; cf. Meno 81A-B; Phaedo 70C-D, 72A-D; Laws 9, 870D-E; 10, 904A-C; for a
general discussion, see Ehnmark 1957. The view can also be found in later
Platonists like Alcinous (Didask. 25, 178.33-9 Whittaker).
251. [28,3-13] Alex. takes up the second preliminary question mentioned at
27,3-4, before a discussion of the specific parts of the soul. The trichotomy of one
power vs finitely many powers vs infinitely many powers is purely systematic in
motivation: no known philosopher maintained that there were infinitely many.
The latter position might be suggested, as AD (154) and BD (261) note, by Arist.
DA 3.9, 432a23-b7. But Arist. just raises it there as an obvious absurdity, solely for
purposes of a reductio. By offering arguments against it, Alex. takes it more
seriously as a theoretical possibility.
252. [28,4-6] If a living thing cannot exercise an infinite number of powers, then
the powers it does not and cannot exercise will be superfluous, which violates
Arist.s teleological principle that nature does nothing superfluous (on which, see
n. 247 on 27,9). More interesting, perhaps, is the supporting argument Alex. offers
Notes to page 54 121
for premiss (2), namely, that an animal cannot exercise an infinite number of
powers. The exercise of different powers does not itself seem to present a logical
difficulty, as the exercise of a single power in contrary ways might (see 27,4-8
above, together with Mant. 118,31-5). So the problem is presumably that there
might be an infinite number of them: perhaps no finite creature can engage in an
infinite number of activities, either simultaneously or over time. Alex. does not
elaborate further.
253. [28,6-8] This argument relies on a general correspondence between the
functional parts of the body and the parts or powers of the soul. But it need not be
one-to-one. For while Arist. thinks that nature in general follows the better course
of assigning one organ for one function, this is not always possible, in which case
a single organ can subserve two or more functions and powers: see esp. PA 4.6,
683a20-6 and Pol. 1.2, 1252b1-5; also PA 2.16, 659b34-660a2; 3.1, 662a16-24; DA
2.8, 420b16-22; Sens. 5, 444a25-7; GA 5.8, 789b9-12. But so long as no part of the
body can serve an infinite number of functions, and there are only a finite number
of such parts, as Alex. claims, there will not be an infinite number of functions and
hence powers corresponding to them.
A more pertinent objection would be that the correspondence does not hold in
the opposite direction, from functions to parts: Alex. allows that in some cases there is
no bodily part corresponding to a given function, since on his view, like Arist.s, the
understanding lacks a dedicated organ (84,10-24; for further references, see n. 200
on 21,24). But this need not open the floodgates: so long as the understanding is
unique in this regard, or even one of a limited number of cases, Alex. can uphold
his conclusion that there will not be an infinite number of powers of the soul.
254. [28,8] See 24,16-17 above, with corresponding note. Arist. DA 2.4, 415b16-
17; Somn. 2, 455b17-18; cf. Metaph. 1.9, 992a29-31; 11.8, 1065a26-7.
255. [28,8-13] The central notion on which this argument rests namely, that
the parts of the soul constitute an order or sequence derives from Arist.s
extended comparison of the different types of soul with the series of geometrical
figures in DA 2.3, 414b20-31, an argument which Alex. goes on to develop imme-
diately below at 28,14-29,1. But his treatment of it here is very different in a
number of ways. First, Alex. claims quite generally that the earlier members in
series of this sort bear a reference to subsequent members, and not (as one might
expect) vice versa. In this respect, such series differ from cases of focal meaning,
where many things are characterised by reference back to a single, primary
member (pros hen), as Alex. elaborates at in Metaph. 263,25-33: in series of the sort
at issue here, earlier members contribute to later ones (suntelein, 263,30.32),
which are higher or more advanced (teleiotera, 263,33). Second, the order Alex.
appeals to here is specifically a teleological one, where earlier parts of the soul are
for the sake of the subsequent ones: cf. 75,31-76,6 below. Arist. never makes such
claims. Third, Alex. claims that this specific type of series cannot go on infinitely,
in order to show that the parts of the soul must be finite in number. But this is
certainly not true of this type of series in general: Arist.s examples, of polygons
and counting numbers, both continue indefinitely. But one thing cannot be for the
sake of another, and that thing for another, and so on indefinitely: a series of
teleological explanations must ultimately reach a final good, in terms of which all
the earlier ones must be understood; otherwise, it is unintelligible what they are
all for. This assumption, which underlies Arist.s reasoning about our practical
aims and aspirations and the highest good in EN 1, is explicit at EN 1.2, 1094a18-
22 and defended at Metaph. 2.2, 994a8-10 and esp. b9-16, and Alex.s discussion of
the latter at in Metaph. 159,27-160,27.
122 Notes to pages 54-55
256. [28,15] 16,18-17,1. See further n. 155 on 16,18-17,1; also 30,17-20; 40,4-10;
Quaest. 1.11a and b, passim.
257. [28,17-20] 17,2-5.
258. [28,23-5] 92,15-94,6; cf. 40,10-15. The kind of separation (khrizesthai) at
issue is a special one Arist. elaborates in this context: see esp. DA 2.2, 413a31-b7;
cf. 1.5, 411b27-30. It concerns the separation of parts or powers of the soul, not
within an individual these cannot be separated from one another (1.5, 411b19-27)
but across different kinds of living thing: some powers occur in certain species
without others. Plants, for example, have nutritive powers, but no perceptual or
rational ones. This sort of taxonomic separability is non-symmetrical, moreover:
even though nutritive powers can be found without perceptual powers, the latter
never occur without the former. Alex.s point in the passage above is that this sort
of separability runs in one direction only, from lowest to highest: the nutritive
powers are separable from the perceptual, and the perceptual from the rational,
but not vice versa in either case: see 30,3-4. For more on this notion of separability,
see my 1999, 208-10.
259. [28,25-9] Alex. refers to the opening line of the treatise, where he restricts
the scope of his inquiry to the souls of things that come to be and perish: see 1,2
with accompanying note. The reason he insists on the qualification here is that if
the gods have souls in the same sense as other living things, they would pose a
counterexample to the claim he has just made at 28,23-5 about the separability of
powers, since in their case, a higher power does occur without a lower one. The
clearest case is the Prime Mover or First Cause, which for Alex., like Arist., is a
non-bodily understanding (89,9-15; Quaest. 1.25, 40,8-9; cf. 2.19, 63,20-1; Arist.
Metaph. 12.7, 1073a3-7) and so has a rational power, but cannot have the nutritive or
perceptual powers, which require organs. For the possibility that Arist. himself might
have been concerned about this case, see my 1999, 210 (for reservations, see Accattino
1992, 42). Arist. certainly draws the relevant contrasts at DA 2.2, 413a31-2 and 2.3,
415a8-12, which AD 155 offer as a possible source of Alex.s remarks here.
But Alex.s use here of the plural gods may indicate, as AD 155 suggest, that
he is thinking of the celestial spheres as well, which he, like Arist., regards as
animate: Arist. Cael. 2.2, 285a29-30; 2.12, 292a18-21; Alex. Quaest. 1.1, 3,10-14;
1.10, 20,30; 1.25, 40,9-11; 2.19, 63,18-20; see Accattino 1992, 40-51. Unlike the
Prime Mover, the celestial spheres are embodied. But being eternal, they will not
have any of the nutritive powers or, according to Alex., perceptual powers either:
this, at any rate, is how he interpreted Arist. DA 3.12, 434b4-7 in his lost
commentary on the DA, according to Ps.-Philop., in DA 595,37-596,14, although
Alex.s original wording was no doubt closer to Them. in DA, 123,24-9, which is
clearly drawn from the same passage (cf. also 44,22-4; 125,21-30). Alex. does,
however, attribute desire and striving to the celestial spheres: Quaest. 1.1, 3,15-19;
1.25, 40,10-23.)
To exclude both kinds of counterexamples, then, it is necessary to restrict the
characterisation of the soul here to things which are not eternal, but subject to
coming-to-be and perishing and thereby exclude all divine beings, whether embod-
ied or not. In support of this restriction, Alex. suggests that when one speaks of the
gods as having a soul, it will be a soul in a different sense than the general
characterisation of its essence given above (16,10.18-20). It will thus be a soul only
in name or homonymously, since although it will share the same name, it will not
share the same definition (Arist. Categ. 1, 1a1-6). The general characterisation
above claims that the soul is the first completion of a natural organic body and so
straightforwardly fails to apply to the Prime Mover, who has no body. The case is
Notes to page 55 123
less obvious for the celestial spheres, since they have a body which moves and is
animate, and so ought to be a natural body (Quaest. 1.10, 20,23-30). But presum-
ably Alex. would deny that their body is organic. According to Simplicius, Alex.
claimed that while divine beings were animate, they could not be said to be animals
(zia) except homonymously, since they lack the power to perceive and to nourish
themselves (in Cael. 463,3-6).
Arist. is less forgiving. He does not speak of the general characterisation given
in DA 2.1 as a definition (horismos), as Alex. does (28,26), but maintains instead
that life is said in many ways (DA 2.2, 413a22; cf. 1.1, 402b5-9) and so is
homonymous even when applied to plants and animals (Top. 6.10, 148a27-31, a
passage on which Alex. fails to say anything in his commentary). Homonymy is
much less pervasive for Alex., who assumes that his general conclusions apply
univocally to all souls in the sublunary realm. The crucial divide for him is between
the mortal and the eternal. (BD 263 explain the homonymy in a way that, if correct,
would make the argument at 28,26-9 circular.)
260. [28,28-29,1] Arist. DA 2.3, 415a12-13.
261. [29,1-3] This power of the soul is first because it can exist without the
others, while the others cannot exist without it, at least not in the sublunary realm,
where things are born, grow, and decay note Alex.s repetition of this crucial
restriction (see n. 259 on 28,25-9). Thanks to Ian Flora for the point about first.
Alex. speaks of the various abilities to be nourished, grow, and reproduce as
powers that are linked (sunezeuktai). This appears to have an almost technical
sense here (which it lacks in Arist.), such that one thing cannot be separated
(khrizesthai) from another if they are linked. If so, it is a non-symmetrical
relation: although all three powers named here are mutually inseparable from
each other, the inferences at 29,16-22 suggest that some can be linked to another,
without the other being linked to it.
Arist. likewise claims that the power for growing and decaying is inseparable
from the power for nourishing oneself: DA 3.12, 434a22-6. But he occasionally
makes an even stronger claim: he says that the power for nourishing oneself and
the power for reproducing are the same power at DA 2.4, 416a19; cf. GA 2.5,
740b36-7. It is possible to read Alex. this way as well: AD seem to think that there
are not three really distinct powers here, but only three expressions for the same
power. Much depends, though, on what is required for being really distinct. Alex.
does claim these powers are inseparably linked and speaks of them conjointly as
the first power. But he nowhere asserts that they are same and he could not in
any case claim that they are same in being; at most they would be the same in
number. Arist. himself, it is worth noting, distinguishes the being of these three
powers, at DA 2.4, 416b11-17.
It is also worth noting the terminology Alex. begins to use here. He constructs
adjectives for different powers and souls from the verbs for particular activities
together with the suffix -ikos, denoting ability. I have chosen to emphasise the
verbal notion in my translations, rather than use the corresponding abstract
construction in English: thus, I have used the power for growing, for example,
rather than the augmentative power (auxtik dunamis) and the soul for desiring
rather than the desiderative soul (orektik psukh). Both forms of expression are
equally awkward in English. But emphasising the verbal notion makes the
straightforward nature of many of Alex.s inferences more evident.
262. [29,3-10] Arist. DA 2.1, 412a13-15; cf. 2.2, 413a20-b4. It is not clear that
Arist. would define life as Alex. does at 29,10, though, since Arist. holds that being
alive is said in many ways (pleonakhs, 413a22), which suggests he thinks it is
124 Notes to page 55
homonymous between the different forms of life and so does not permit the sort of
univocal account Alex. gives here (see the end of n. 259 on 28,25-9). In fact,
elsewhere Arist. seems comfortable accepting his predecessors view that the
animate is distinguished from the inanimate by (self-) movement and perception:
DA 1.2, 403b25-7; cf. 3.3, 427a17-19; 3.9, 432a15-16.
263. [29,12-14] Arist. DA 2.2, 413a31-b4. On the notion of separation, see n. 258
on 28,23-5 above. For Alex.s use of representation (phantasia), see n. 111 on 12,17
above; for his use of striving (horm), n. 113 on 12,19.
264. [29,14-15] Alex. clearly is developing Arist.s claim at DA 413b2, but the
stronger form that Alex. gives it here, involving definition, is frequently made in
Arist.s corpus: for references, see n. 87 on 10,8-10 above.
265. [29,16-18] Arist. DA 2.2, 413b22-4; 2.3, 414b1-15; 3.7, 431a13-14; 3.11,
433b31-434a5; Insomn. 1, 459a15-17; MA 8, 702a17-19; cf. DA 3.10, 433b28-9,
though compare DA 3.3, 428a9-11. Alex. explicitly makes the point here that these
powers mutually entail each other, which would be otiose if linking (sunezeuktai,
29,17) were symmetrical (see n. 261 on 29,1-3 above). The powers named here,
then, cannot occur without the others, but come as a package for anything capable
of perception and so are shared by all animals.
For Alex.s discussion of the power for representing, see 66,9-73,13 below.
266. [29,18-22] 30,13-15; Arist. DA 2.2, 413b2-4. In contrast with the preceding
lines, the powers named here do not mutually imply each other. Anything capable
of (self-) motion must have the power for perceiving and representing, but not vice
versa. Some animals are stationary, even though they are capable of perception
and some crude form of representation (on this last point, see n. 265 on 29,16-18).
The basic powers of perception are thus separable from the power of locomotion,
but not vice versa. For the claim that some animals have perception, but are
stationary: Arist. DA 1.5, 410b18-20; 2.2, 413b2-4; 2.3, 415a6-7; 3.9, 432b19-21 (cf.
b13-14); HA 1.1, 487b6-11; PA 4.7, 683b4-11; cf. DA 2.3, 414b17; 3.12, 434b7-8; HA
8.1, 588b12-17.
If the power for striving (hormetik dunamis) is always connected with the
power of locomotion, as 29,18 suggests, then there will be a similar asymmetry
between the power for striving and the power for desiring: the power for striving
will be inseparably linked to the power for desiring, but not vice versa. The fact
that Alex. treats them very much in tandem in his extended discussion of the
power for striving below (73,14-80,15) only shows that he thinks the power for
striving is linked to the power for desiring (73,26-8) and that striving is even a kind
of desire (74,1); it does not show that anything capable of desire is capable of
striving, as the converse would require. On the term striving (or impulse, as
horm is often misleadingly translated), AD 156 rightly note a possible Stoic
influence, but Arist. himself also uses the term (see Bonitz 525b47-526a10). On the
origin of the expression, see also BD 264. Alex.s notion of striving is in any case
different from the Stoic one. See n. 113 on 12,19 above and more generally
73,14-80,15 below.
267. [29,22-4] Arist. EN 6.1, 1139a6-17; cf. DA 3.3, 427b24-6 and 428a3-5.
268. [29,22-30,2] The rational powers, all of which involve the use of concepts
in some way, are inseparable from the lower powers, at least in mortal beings, even
though the lower powers are separable from them: for the non-symmetric notion
of separation, see n. 258 on 28,23-5; and for the restriction of the claim to mortal
beings, see n. 259 on 28,25-9. For the use of teleioteros to mean more advanced in
cross-species comparisons, see n. 71 on 7,23.
It is noteworthy that Alex. includes the powers of practical and theoretical
Notes to page 56 125
reason together under one head, as AD 156-7 point out. For antecedents they cite
EN 6.1, 1139a6-17 and 6.5, 1140b25-30, even though Arist. speaks of these two
kinds of reason as distinct parts of the soul (1140b25; cf. 1139a3-6).
269. [30,2-6] Alex. emphasises the unity of the rational soul here, as it com-
prises all of the psychological powers and not just rational ones. Because the
rational powers are inseparable from the others, they cannot be found apart from
them in sublunary beings and so must in some way together constitute a whole, of
which they are parts (30,5). AD 157 worry unnecessarily whether Alex. intends
his remarks about unity to be restricted to rational souls. But nothing in the text
precludes the perceptual soul, for example, from exhibiting a similar sort of unity.
If anything, Alex.s remark that the lower powers are expanded and developed
(auxsin te kai epidosin } lambanousn, 30,6) by the addition of higher powers
suggests that organic unity is to be found in more complex souls quite generally,
where the higher powers are not merely added on as distinct modules or compo-
nents, but instead enrich and augment the lower powers through more complex
interactions. Feeding and reproduction in animals, unlike plants, can manifest
itself in quite varied and complex hunting and mating behavior, for example,
precisely because perception is involved.
BD 264-5 take Alex. to be making a further point here, namely, that in each
living thing there is a single soul, composed of multiple powers, rather than a
plurality of souls. Although this is no doubt Alex.s view, he does not explicitly
reject a plurality of souls here.
270. [30,10-12] 36,13-15.
271. [30,6-17] Alex. is not simply making the claim that there is a scala naturae
among living things in the sublunary realm, from plants to animals to humans, but
that this is grounded in the corresponding hierarchy of souls and powers. In this
case, it is clear that the order of explanation is the reverse of the order of discovery:
we posit various powers of the soul in response to the range of activities we observe
in living beings and then establish the hierarchy of powers based on how they are
distributed conjointly or independently among plants, animals, and humans. But
we make these theoretical posits precisely in order to account for the patterns we
observe among living things, and so accord the souls and soul powers explanatory
priority.
272. [30,17-20] 17,1-8; 28,14-29,1.
273. [30,20-6] Alex. makes clear that the present treatise is intended only as a
systematic overview of the subject, unlike (presumably) his lost commentary on
Arist.s DA or the more detailed treatments of particular topics or puzzles that we
find in the Quaest. and Mant.
274. [30,26-31,2] The two senses of part Alex. rules out here are both quanti-
tative, in line with the first sense of part given in Arist. Metaph. 5.25 at
1023b12-15; see Alex. in Metaph. 423,33-9. But Alex. is more specific. By ruling out
(i) continuous extensions, Alex. effectively rules out the souls being a body, as on
the Stoic view, where the parts of the soul are parts of the breath or pneuma
extending from the governing part of the soul (Aet. 4.21.1-4, Dox. Gr. 410.25-411.24
= SVF 2.836 = LS 53H; Iambl. ap. Stob. 1.368.12-20 = SVF 2.826 = LS 53K). For
Alex. the parts of the soul are not arrayed spatially themselves, except extrinsi-
cally in virtue of the organs (in Metaph. 417,14-19). And by ruling out (ii) numbers,
he also rules out any conception of the soul as an aggegrate of discrete parts
(composed from parts into which it is divided). For Alex., the powers of the soul
must exhibit a stronger unity or integrity.
Arist. discusses critically the view that the soul is a magnitude and a number
126 Notes to pages 56-57
at, respectively DA 1.3, 407a2-19 and 1.2, 404b27-405a7, but neither passage
provides a close parallel.
275. [31,2-6] A very similar argument, at least in outline, is attributed to the
first century BCE Aristotelian, Nicolaus Damascenus (Porph. ap. Stob. 1.353.12-
354.6 Wachsmuth): he argues that the parts of the soul are not quantitative, but
qualitative parts, like the parts of art or of philosophy (see Moraux 1973, 481-5).
Alex.s use of a body as an example instead may have a polemical purpose, since
the apple example also appears in a report of the Stoic view, which similarly treats
the parts of the soul as qualities (Iambl. ap. Stob. 1.368.12-20 Wachsmuth = SVF
2.826 = LS 53K). Stoic qualities are bodies, however (see n. 162 on 17,15-17), and
Alex.s point here is precisely that the qualities of the apple, though parts of a body,
are not bodies themselves. (In point of fact, the parts of the soul are not even
qualities for Alex., stricitly speaking, but parts of the form: in Metaph. 424,29-31.)
It is impossible to determine whether the apple example, which also occurs at in
Sens. 165,26-166,2, comes from the Stoics originally or from an earlier Peripatetic,
or whether Iamblichus is actually borrowing it from Alex. in reporting on the
Stoics. For discussion, see Inwood 1985, 30-2, esp. n. 55.
276. [31,7] 29,1-10, esp. 4-5; 31,28; 92,17; Mant. 105,12; 118,19; cf. 103,9-10.
277. [31,8-9] 30,15-17.
278. [31,7-10] 9,11-14.23-6; Mant. 118,19-21. BD 266 suggest that this argu-
ment derives from the more general argument Arist. offers at DA 2.2, 414a4-14.
Ian Flora has suggested to me that 413a25-b2 provides an even closer parallel.
279. [31,10-14] This statement is striking for the way in which it acknowledges
the role of matter, even if secondary to form. Generally, Alex. simply says that it
is in virtue of the form that bodies are the kinds of things they are and do the kinds
of things they do (6,21-3; 7,9-14). But the difference may just be one of emphasis,
since it is also Alex.s view that matter contributes to what a body is and can do
indirectly, in so far as differences in the underlying matter are responsible for
which forms supervene (10,14-26).
280. [31,16-20] N.B. Bruns edition misnumbers the lines here, with only three
lines between those marked 31,15 and 31,20. I will treat 31,19 as missing and
otherwise respect the numbering printed. Bruns suggests a lengthy insertion at
31,18 to bring the use of primary and secondary in line with those above. But this
is unnecessary, since deuters here is not a precise parallel to deuteron at 31,13
above: it marks the derivative sense in which the body can itself be said to be
healthy, in contrast with the animal, which is healthy in a primary or fundamental
sense (kuris). For the use of deuters to mark a derivative application of a
predicate, see above 8,22.
At 25,4-7, as AD 160 point out, Alex. contrasts the soul and health, suggesting
that the latter is more like a harmony and a balance, which the soul is not, much
as Arist. does, we might add: see n. 231 on 25,4-7. But Arist. also treats health as
an example of a form received by a body, as AD rightly observe: see next note.
281. [31,20-5] The claim that the soul is the form of the body has been one of
Alex.s central points in the treatise so far: see n. 157 on 17,9. For the comparison
between health and the soul, as forms of the body, see Arist. DA 2.2, 414a4-14:
though one is an accidental form and the other a substantial form, each is that by
which (hi) something lives and is healthy, respectively, in contrast with the
underlying subject and matter.
282. [31,25-32,1] Mant. 105,4-12; 118,12-16.19-25. It is sometimes thought that
Alex. is objecting here to a Stoic view, since they think that the breath or pneuma
in plants constitutes a nature (phusis) but not a soul (psukh): see esp. Philo Leg.
Notes to pages 57-58 127
Alleg. 2.22-3, 1.95.7-20 Cohn-Wendland = SVF 2.458 = LS 47P; Quod Deus sit
Immut. 35-46, 2.64.1-66.15 Cohn-Wendland = SVF 2.458 = LS 47Q; Galen PHP
6.3.7, 374.18-19 De Lacey = SVF 2.710; Foet. Form. 3, 68.9-17 Nickel = SVF 2.712;
6, 104.16-19 Nickel; Hipp. Epid. 17b.251.1-6 Khn = SVF 2.715; cf. Origen Orat.
6.1, 311.16-312.10 Koetschau = SVF 2.989 = LS 53A. For discussion of the Stoic
scala naturae, see Inwood 1985, 21-7. But if the Stoics were Alex.s target, his
objection would actually misfire: on the Stoic view, simple bodies and stones belong
to a lower echelon: they do not have a nature, as plants do, but only cohesion
(hexis). So the Stoics would be indifferent to Alex.s objection that simple bodies are
not alive at 31,27-8, as an ignoratio elenchi. Galen is more likely to be Alex.s target
here. Galen not only begins his Nat. fac. by defending the distinction between
nature and soul (without mentioning cohesion), but claims that the Aristotelian
view amounts to the same thing in different and unusual language, while his
distinction conforms more closely to ordinary usage: 1,1, 101.1-15 Helmreich; cf.
adv. Iul. 5, 18a.266.9-13 Khn; Caus. Sympt. 7.129.9-10 Khn; Subst. Nat. Fac.
4.759.9-15 Khn; HNH 15.123.14-124.7 Khn. Alex. is concerned to show that the
difference is not merely verbal and to give these terms their proper extensions:
things which are not alive still have a nature, but nothing without a soul is alive.
283. [32,1-4] The powers by which seeds operate will be powers of the progeni-
tor that produced them, presumably because seeds are originally proper parts of
the progenitor. (2) is then offered as an instance of (1), with seed now used in its
proper, biological sense. The connection between animate and soul power in (2)
is more plausible in the Greek, since both key terms are based on the word soul
or psukh: the powers of animate things (tn empsukhn) are themselves soul
powers (psukhikai).
AD 160-1 give Mant. 118,25-6 as a parallel and argue that this argument is
directed at the Stoics, accusing them effectively of inconsistency, since they hold
that the power for reproducing is a part of the soul in contrast with the power for
nourishing oneself and growing, which merely belong to nature (see n. 282 on
31,25-32,1 above): Aet. 4.21.2-3, Dox. Gr. 410.30-411.13 = SVF 2.836 = LS 53H 3;
Aet. 4.4.4, Dox. Gr. 390.5-13 = SVF 2.827; D.L. 7.110 (cf. 157) = SVF 2.828,
although Panaetius appears to have assigned all of these powers to nature (fr. 86
van Straaten; for discussion, see also Sharples 2004, 58 n. 168). ADs remarks
apply straightforwardly to the argument in the Mant., which explicitly concerns
the part for reproducing (to genntikon morion) and reports the view of unnamed
philosophers who might be the Stoics (they say). But the argument here does not
mention the power for reproducing at all, only the power active in seeds (32,2) and
the power for nourishing oneself (32,4-5), and so does not object to inconsistent
treatment of different powers. Seeds figure in the argument here primarily to
establish certain powers of the progenitor, by (1); and then Alex. crucially claims
in (2) that the seeds of living things are active in virtue of soul powers. It is not
clear that the Stoics would have agreed with (2): they can easily argue that the first
activities of the developing embryo are expressions of nature, not soul.
284. [32,4-6] Arist. GA 2.1, 735a15-21 (cf. a2-4).
285. [32,6-7] Although Alex. is usually scrupulous about distinguishing powers
from the souls that are constituted from them, here he explicitly states that the
power for nourishing oneself is a soul. But the soul for nourishing oneself is a
special case, since it is constituted by just one kind of power, the power for
nourishing oneself, and so the distinction between the power and the soul is a very
fine one (much like the difference between a singleton set {a} and its sole member,
128 Notes to pages 58-59
a). Whether the power for nourishing oneself can itself be subdivided is a further
question: see nn. 261 and 288 on 29,1-3 and 32,15-22, respectively.
286. [32,8] For Alex.s use of linked (sunezeuktai), see n. 261 on 29,1-3.
287. [32,11-15] cf. 36,6-9. The lines above (apart from the example of the
eunuchs) are a very close verbal echo of Arist. DA 2.4, 415a26-b2; see also GA 2.1,
731b31-732a1, 735a17-19; cf. Oecon. 1.3, 1343b23-5. Arist. makes a similar claim
about the endless cycle of elemental transformation at GC 2.10, 336b26-33 (cf 2.11,
338b6-19); Metaph. 9.8, 1050b28-9. But the point is made most generally by Plato,
beginning with the reproduction of living things at Symp. 206E-207A, 207D-208B;
cf. also Laws 4, 721B-C. Alex. may also have in mind the middle Platonist theme
of likening oneself to God (homoisis thei), as AD 161 suggest: cf. Quaest. 1.25,
40,17-19. But there are also parallels in Arist. himself: EN 10.7, 1177b33-1178a3.
The example of eunuchs may derive from Arist. Metaph. 5.12, 1019b15-19, on
which Alex. comments at in Metaph. 393,2-17.
AD 161 point out that the Greek in 32,15 is ambiguous, perhaps intentionally:
panta here can either be the subject (as they take it in their translation) or the
object of prattei (as translated above). The first reading is weaker by making its
meaning nearly the same as the previous clause: } and everything naturally acts
for its sake.
But however we construe it, the statement at 32,14-15 goes well beyond Alex.s
earlier claims about natural teleology at 24,16-17 and 28,8. The natural activity of
natural objects is not merely directed at an end intrinsic to itself namely, its
complete realisation as a member of its kind but at participation, in some limited
degree, in Gods perfection and eternity.
288. [32,15-22] Arist. GA 2.1, 735a16-19; 2.4, 740b29-37; DA 2.4, 416a19.
Alex.s argument here leans heavily on (3), the claim that the soul which plants
have is only a soul for nourishing oneself. But he is entitled to this premiss, given
the notion of separability on which his taxonomy of souls rests (see n. 258 on
28,23-5): unless there are living things that have one of these powers, but not the
other, they belong to the same type of soul. Alex. here speaks of the soul for
nourishing oneself as a power again (see n. 285 on 32,6-7), but it is even more
noteworthy that he speaks of the power for reproducing as a part (morion, 32,22)
of the power for nourishing oneself: compare 29,1-3, with corresponding note on
the relation between these powers.
289. [32,23-4] Alex.s methodological remark here seems to be a simplification
of the familiar Aristotelian principle that we proceed from things which are more
evident and intelligible to us (gnrimteron hmin) to things which are less so, but
more intelligible in themselves and without qualification (gnrimteron hapls),
and so can be used to illuminate the obvious facts from which we started: this is
the principle briefly alluded to at DA 2.2, 413a11-12, but more fully developed at
Phys. 1.1, 184a16-b14; An. Post. 1.2, 71b33-72a5; EN 1.4, 1095b2-4; Metaph. 7.3,
1029b3-12; and as applied to definition, Top. 6.4, 141a26-142a16. But Alex. does
not draw the distinction between a relative and an absolute use here, and more
importantly he speaks about the starting points simply as clearer (dia tn
saphestern), in keeping with non-Aristotelian usage rather than Aristotles more
nuanced phrasing, that we start from what is less clear, but more obvious (ek tn
asaphn men phanertern de, DA 2.2, 413a11-12).
290. [32,25-7] Arist. DA 2.4, 415a16-20; cf. 1.1, 402b10-14. In claiming that
activities are clearer (saphesterai) to observation, presumably than the powers
from which they arise, Alex. goes beyond Arist., who only claims that we should
study them first because they are prior in account (kata ton logon, 415a19-20).
Notes to page 59 129
Arist. does say elsewhere that in cases where one thing is prior to another in this
way, knowledge (gnsis) of the first precedes knowledge of the second (Metaph. 9.8,
1049b16-17), but he has definitional knowledge in mind there, which is unlikely to
be at issue in these lines; in fact, Arist. claims that what is prior in knowledge (to
ti gnsei proteron) can come apart from what is prior in account at Metaph. 5.11,
1018b29-34.
291. [32,29-30] This sentence is difficult. On the most natural way of construing
the Greek pronouns in 32,29-30, the sentence would read those things are first in
definition which include in their own definition [other] things in whose definition
they do not enter. That is, it would claim that A is prior in definition to B just in
case B is part of the definition of A, but A is not part of the definition of B. (In a
note, Bruns makes clear this is how he takes it.) But this gets things the wrong
way around, since it makes the definition of A dependent on B. As priority in
definition is standardly formulated, A is prior because A is not dependent on B, but
B is dependent on A: A is part of the definition of B, but B is not part of the
definition of A. See Arist. Metaph. 7.10, 1035b4-6; 13.2, 1077b3-4; also 10.8,
1049b12-17; cf. Alex. in Metaph. 386,33-5. To get around this difficulty, AD propose
translating kata ton logon as in essence, while construing the pronouns in the
same way (162-6). But this is no improvement, even if we grant the rendering of
logos, since it still has the dependence relations running in the wrong direction: A
cannot be prior in essence to B if it includes B in its essence, while B does not
include A. They are forced to look for a non-standard reading of priority in essence,
which is essentially teleological in nature, based on Philoponus report of Alex.s
comment on DA 2.4, 415a20 at in DA 264,15-22. (For another possible reading of
Alex.s comment, see n. 292 on 32,30-33,5.)
The sentence at 32,29-30 is in line with the standard doctrine, however, if we
construe the pronouns as BD do (267-8), taking tina to be the subject of periekhei
and the genitive hn to depend on prta, as rendered in my translation above.
There is a very similar difficulty in the construal of the same point at Arist.
Metaph. 13.2, 1077b3-4, where Ross in his notes similarly takes the genitive hosn
at b3 as dependent on protera. This is not the most obvious reading of the Greek,
but it is a possible construal and the best way to make sense of the text as it stands.
292. [32,30-33,5] At first glance, Alex.s explanation (gar, 33,1) seems quite odd.
In order to support his claim that being nourished is prior in definition to the
nutritive power (32,27-9), he seems to change the subject twice, by arguing that
powers have a kind of existential priority to their corresponding activities. But
another way to take Alex.s line of thought would be as follows: the way in which
activities are prior has to be in definition, since they are not prior in being, but
rather powers are.
293. [33,7-10] Arist. DA 2.4, 415a20-2; cf. 1.1, 402b14-16. Like Arist., Alex.
holds that the activity of each of the souls powers involves an object, and therefore
cannot be understood or defined without reference to that object and without some
understanding of what makes it a suitable object for that activity.
294. [33,13-15] in Meteor. 186,21-6; cf. in Sens. 108,10-14. Arist. draws the
same distinction at DA 2.4, 416b3-7, in connection with the issue of whether like
is nourished by like or contraries by contraries (see n. 295 on 33,15-16 just below);
GC 1.5, 321b35-322a3. (The distinction between two types of nourishment at GA
2.6, 744b32-6 is quite different.)
The notion of cooking (pepsis), sometimes rendered concoction, is central in
Arist.s natural works (see Lloyd 1996) and applies to many processes we would
not ordinarily describe as cooking: in fact, Arist. explicitly states that it is an
130 Notes to pages 59-60
extended use of the term, for lack of any better (Meteor. 4.2, 379b14-17). As he
defines it, cooking is the process of completing (teleisis) a thing by heats acting
on its matter (379b18-20). In many cases, this results from the objects own natural
heat, as when fruit ripens, or food is broken down and digested (PA 3.14, 674b9-13,
cf. b26-34), or in the formation of the homogenous materials that make up the body,
such as bone, hair, or sperm (PA 2.3, 650a2-32). For the formal account of cooking,
see Arist. Meteor. 4.2-3; Alex. in Meteor. 185,24-198,3; on the different uses of the
term, see esp. the references in Lloyd 1987, 205 nn. 119-24.
I have accepted Bruns deletion of troph legetai, instead of retaining it with AD
and inserting a te. It does not, in any case, affect the overall meaning.
295. [33,15-16] Arist. DA 2.4, 416a29-34 (cf. a21-2); Phys. 8.7, 260a30-2; Alex.
also mentions the point briefly at in Meteor. 186,24-6.
Both Anaxagoras and Empedocles seem to be committed to the view that like is
nourished by like. For Anaxagoras, see Simpl. in Phys. 460,15-20 = DK 59 A 45; cf.
Aetius 1.3.5, Dox. Gr. 279.12-17 = DK 59 A 46. In fact, nourishment appears to
have been one of the most striking cases for his general thesis that there is a
portion of everything in everything (DK 59 B 6 & 11): the accretion of like by like
in growth is possible because the similar homogeneous substances are already
present in food. The evidence for Empedocles is somewhat more tenuous, but see
esp. Plut. Quaest. Conviv. 4.1, 663A = DK 31 B 90; Aet. 5.27.1, Dox. Gr. 440.4-10 =
DK 31 A 77; also cf. Simpl. in Phys. 381,31-382,3 = DK 31 B 62.
It is not clear who held the view that contraries are nourished by contraries.
This may just be an application of Arist.s claim regarding agent-patient interac-
tions in general that most of his predecessors agreed (hoi pleistoi } homonotiks
legousin) that only dissimilar things could have an effect on one another (GC 1.7,
323b3-7).
296. [33,17-18] Repunctuating the text in line with the Hebrew translation
(followed also by AD and BD).
297. [33,18-24] Arist. DA 2.4, 416a33-4. See also GC 2.5, 332a7-8; Phys. 8.7,
261a32-b1; Metaph. 4.7, 1011b34-5; 10.7, 1057a31; 11.11, 1068a1-4; 12.1, 1069b3;
Alex. DA 59,11-12; Quaest. 1.7, 15,26, 16,6; Eth. Probl. 30, 162,7-9; in Metaph.
329,18-21; 423,28.
298. [33,25-9] Alex. here applies the general argument Arist. makes in GC 1.7,
with regard to all agent-patient interaction: in order to interact, they must be
initially contrary to each other, but potentially alike, since they will come to be
actually like each other as a result of the interaction. The case is particularly
strong with nutrition, since nourishment produces growth in what is nourished by
adding to its mass; and it can do this only if it comes to be like the body that absorbs
it as a result of digestion.
299. [33,28-34,8] Arist. DA 2.4, 416a21-5. Alex. elaborates Arist.s argument
further, though not fully. A distinction is needed between growth and alteration,
which also involves a transformation from one contrary to another. In alteration,
however, there is a single underlying subject, which changes in quality; in growth,
a separate body is transformed and added to another, which does not change its
nature or quality, but only increases in size as a result.
300. [34,8-14] Alex. is reconciling two Aristotelian passages here: DA 2.4,
416a25-9, which allows that water is nourishment for fire, and GC 1.5, 321a9-29,
which seems to argue that the transformation of the elements is not growth, but
generation and destruction. Alex. reads the first passage as being true only in a
broader sense of growth, while the second is true in the fundamental sense. This
is reasonable, since 416a25-9 only makes claims hedged by it seems and it
Notes to pages 60-61 131
appears, and Arist. corrects this half a column later at 416b9-11, by denying that
anything which is not alive can nourish itself, as Alex. emphasises at 34,13-14.
Arist. elsewhere denies that fire can genuinely be nourished by moisture because
it does not retain its identity from one moment to the next, but is in constant flux:
Meteor. 2.2, 355a3-11; cf. Alex. Mixt. 16, 233,30-234,11. On water as nourishment
for fire, see Mant. 149,11-13, where Alex. notes that it is the moisture in logs that
burns; see also in Sens. 46,9-12.
301. [34,16-17] The aphorism points to a duality similar to the one Alex. has
been emphasising: in a broad sense, food is nourishment, but more strictly it is only
the part of it that nourishes and is incorporated into the body which is nourish-
ment, the rest being waste. Accattino 1993 points out that Philoponus also quotes
this phrase in this context (in DA 284,11-12) and explicitly attributes it to medical
writers (hoi iatroi); and that it can be found in MS M (Marcianus gr. 269) of
pseudo-Hippocrates de Alim. 8. The closest parallels from Galen, in contrast, all
reflect the quite different reading in MS A of de Alim. (troph de to trephon, troph
de to hoion). We can add that the phrasing Alex. quotes is also attributed to
Hippocrates by Meletius de Nat. Hom. 101.13 Cramer.
302. [34,14-18] These remarks are Alex.s gloss on Arist.s solution at DA 2.4,
416b4-9. For a similar solution applied to perception, see 39,12-18 below (parallel
to DA 2.5, 417a18-20).
303. [34,18-19] 34,14. Mixt. 16, 233,26-7.
304. [34,18-26] Arist. GC 1.5, 321a29-b10. Cf. Alex. DA 36,9-13; Mixt. 16,
234,11-15; 236,19-30; cf. 238,10-20. Alex. argues that there is a crucial asymmetry
here between these two like things, in spite of their symmetry in other respects.
With respect to growth, what nourishes (in the second sense) and what is nour-
ished are on a par, since the two are similar in kind, and so both are equally
increased through the accretion of like by like. But with regard to being nourished,
they are not symmetrical, since one is transformed into the other and not vice
versa, and so only one is nourished by the other. The ground of this asymmetry is
the presence of a soul (34,18-19), which Alex. had argued just above distinguishes
nourishment specifically from the more general category of growth (34,8-14).
305. [34,27-35,2] Alex. is following the order of explanation set out at 32,23-
33,12 (based on Arist. DA 2.4, 415a16-22): having finished his discussion of
nourishment (33,13-34,26), he turns briefly to the activity of being nourished
(34,27-35,8), along with the activities of growing (35,8-17) and reproducing (35,17-
25), before finally discussing the powers underlying these (35,25-36,9).
On the need to transform the external nourishment before it can be assimilated
and incorporated, see 33,18-19 (cf. 25-6) above.
306. [35,2-3] 96,5-8; in Top. 376,27-30; in Sens. 78,6-12; in Meteor. 186,35-
187,3. In this passage, Alex. is surely continuing his elaboration of Arist. DA 2.4,
by developing the claim at 416b28-9 that being nourished requires heat, which
must therefore belong to all living beings (cf. also Iuv. 4, 469b11-12; Resp. 8,
474a25-8). For the specific claim that heat is a tool or instrument (organon) of
the soul, see GA 2.4, 740b29-36, where Arist. says that the soul uses heat and cold
as its tools in nourishment, growth, and reproduction; cf. also PA 2.7, 652b9-15.
According to Philoponus (in DA 288,5-19), Alex. discussed an interpretation of
DA 416b25-9 presumably in his lost commentary on the DA whereby both heat
and the soul for nourishing oneself are that by which (to hi trephetai) an animate
thing is nourished, the soul as an unchanged changer and heat as a changed
changer. But it is offered as an interpretation of an alternative manuscript
reading, which Alex. may not himself endorse. See n. 323 on 36,9-13 below.
132 Notes to pages 61-62
307. [35,3-5] Arist. DA 2.4, 416a9-15, which also cites certain unnamed philoso-
phers as holding that the nature of fire (h tou puros phusis) is what is productive
(to ergazomenon) in animals and plants and responsible for nourishing oneself and
growing. Simplicius ascribes this view to Hippasus of Metapontum and Heraclitus
at in Phys. 23,33-24,12, esp. 24,6-8. Arist. himself holds that heat simply as such
is not responsible for these processes, but is only a kind of contributing factor
(sunaition ps, 416a14).
308. [35,5-8] These lines are a close reformulation of Arist. DA 2.4, 416a13-18,
with some differences in emphasis. Alex. claims that all natural compounds (in
contrast with the elements) observe a definite order in their activities and natural
limits, which he attributes to their form: see esp. in Metaph. 103,31-104,18; also
Fat. 4, 168,3-7.14-15; 6, 170,9-11; in Metaph. 32,1-5. Cf. Mant. 185,14-15; in
Metaph. 101,28-30.
309. [35,8-9] 34,5-9.
310. [35,10-11] 74,23-5; cf. in Top. 167,8-11; Arist. GC 322a24-5; Top. 2.4,
111b24-6; cf. Arist. Somn. 1, 454b32-455a3; EN 1.13, 1102b4-5. AD 168 note that
Alex. has animal (zion) at 35,11, where Arist. has the broader animate being
(empsukhon) at 416b13, and conjecture that Alex. might have written living thing
(zn). Though possible, this is unnecessary: Alex. can use animals as an example
without thinking that his claim applies only to them; or he may be using zion as
he sometimes does for living things in general.
311. [35,9-17] Arist. DA 2.4, 416b11-15; GC 1.5, 322a20-8; cf. Alex. Mixt. 16,
236,24-6. Since the power for nourishing oneself and growing are always coinstan-
tiated (see n. 261 on 29,1-3) and both utilise nourishment (as he states here), Alex.
needs to show that they are different in other ways: they can operate inde-
pendently at different times (35,10-13), have different functions (35,13-15), and
consequently involve different powers (35,15-17). For a more extended discussion
of Alex.s views on growth, see Mixt. 16, 233,23-238,23.
312. [35,19-20] 32,11-13.18-19; cf. Arist. DA 2.4, 415a26-8; 3.9, 432b23-5. On
the face of it, this claim might seem to conflict with Alex.s earlier assertions that
the power for reproducing is inseparable from the other powers of the soul for
nourishing oneself (29,1-3, with corresponding note; cf. 32,15-22). But they are
compatible if we distinguish between types and tokens. The present claim concerns
individuals within a species: even if the species as a whole has all three powers of
the soul for nourishing oneself, a given individual may not, if it is immature or
deformed. Claims about the separability of powers, in contrast, are based on
whether there are any species where one power is found without the others. If there
is no species that has the powers for nourishing oneself and growing, but lacks the
power for reproducing, the power for reproducing will be inseparable. In the DA,
Alex. simply treats them as inseparable, though it would be interesting to know
whether he regards mules, which are infertile, as a species. Alex. does not appear
to discuss the problem mules pose in his extant works, unlike Arist. (GA 2.8).
313. [35,20-2] On the reproductive changes during puberty (in humans), see HA
7.1 passim, which puts the first appearance of sperm at the age of 14 (581a12-14)
though it is infertile until the age of 21 (582a16-17); cf. also HA 7.5, 585a34-b2; GA
1.19, 727a5-8; 1.20, 728b21-7. The appearance of the sexual organs is earlier, of
course. He observes that a penis in a male embryo can already be detected by the
fortieth day of gestation: HA 7.3, 583b14-20; cf. GA 4.1, 766b4-7 (though compare
2.4, 737b8-12).
314. [35,22-3] 32,11-16. Note that Alex. claim here, as in the earlier passage,
Notes to page 62 133
is not limited to living things (as BD seem to assume), but in some sense is meant
to apply to all natural things.
315. [35,24] For the distinction between first and subsequent souls, see 16,18-
17,1; 29,1-6; and esp. 38,12-13, where the first soul is also described as for
vegetation (phutik), confirming Victorius emendation of the MSS here, which
read natural (phusiks), to phutiks. AD 169 rightly cite Arist.s pertinent use of
the term for one side of the non-rational part of the soul at EN 1.13, 1102a32 and
b29; BD adds EE 2.1, 1219b37.
316. [35,24-5] In line with the methodology given above (32,23-33,12), Alex.
now turns from the activities of this type of soul to discuss the powers underlying
it briefly (35,24-36,9): see n. 305 on 34,27-35,2. Since powers are defined by
reference to the activities they give rise to, there is little Alex. has to add here
beyond what he has already said.
317. [35,26] 29,2.
318. [35,26-7] Arist. DA 2.4, 416b14-19.
319. [36,2-3] Notice the omission of the phrase through its own activity (dia
ts oikeias energeias) which Alex. has used in the previous two cases (35,27; 36,2),
as sexual reproduction requires the activity of another progenitor as well.
320. [36,3-4] Arist. Somn. 3, 456a34-b6; PA 2.3, 650b2-13; 2.4, 651a14-15; GA
1.18, 725a11-13, 726a26-7; 4.1, 766b7-9 (also b12-14); cf. 1.19, 726b1-5, b9-11; 2.3,
737a18-20. Alex.s view, following Arist., is that the useful, nourishing part of food
is extracted and refined through a process of cooking (pettesthai, see nn. 301 on
34,16-17 and 294 on 33,13-15 respectively), and then transformed into blood and
distributed to the different parts of the body, which it nourishes and causes to
grow. Through further cooking, the blood the last or final nourishment is
turned into seed for producing another animal of the same sort: see 40,2-3 and
94,26-95,8 on blood (which is described as the last stage of nourishment at 94,28)
and 92,19-21 on seed.
Although Alex. states that this view has been shown elsewhere, there appears
to be nothing in the extant works that corresponds to it. AD 169 reasonably suggest
he is not referring to his own writings, but to Arist.s GA (much as 97,26-7 refers
to the MA).
321. [36,7] cf. 32,17; Arist. GA 2.1, 735a17-19.
322. [36,6-9] 32,11-15, together with accompanying note for references to Plato
and Aristotle. The phrase reinvigorated immortality (athanasias episkeuasts) at
36,8 comes from Plato Polit. 270A, which states that the heavens receive a
reinvigorated immortality when in the eternal cycle the craftsman god causes the
heavens to move forward again, after having moved backwards under their own
power, thus restoring forward moving life-processes to them. The details of this
great myth seem unimportant here, apart from the idea that something which is
mortal in its own right might nonetheless adventitiously receive an extension of
its life through something else, as part of a cycle. In particular, there is no
suggestion in Alex. of divine intervention.
323. [36,9-13] Alex. here virtually reproduces Arist. DA 2.4, 416b20-3, with
only minor changes. BD 271 suggest that Alex. interpreted this passage differently
in his lost commentary on the DA, based on Philoponus in DA 288,5-19 (discussed
above in n. 306 on 35,2-3). But Philoponus is concerned with a passage several lines
later at 416b25-9, where there is a textual crux. Alex. may only be reporting there,
as he often does, alternative readings and the consequent differences in interpre-
tation, rather than endorsing it outright, and Philoponus wording does not impute
any stronger commitment on Alex.s part than that. Themistius, in contrast,
134 Notes to pages 62-63
clearly does endorse the alternative reading: see his in DA 53,26-33. On the
commentators differing readings of this passage, see Moraux 2001, 333-4.
324. [36,13] Alexander is oversimplifying here: the threefold schema does
not extend straightforwardly in the case of reproduction, since the soul that
makes reproduction possible is only the same in species as the form of the body
that is generated; they are not the same in number, as two distinct individuals
are involved. Alexander does regard the seed, though, as that by which (di
hou) something is reproduced and as a refinement of nourishment: see 36,3-4
above.
325. [36,14-18] Alex.s argument is an expansion of Arist. DA 2.4, 416b23-5,
with very close verbal echoes in (1), (2), and (3). A principle similar to (1) is
attributed by Cicero (in a different context, concerning excellence and fulfillment)
to Antiochus of Ascalon at Tusc. Disp. 5.22; Fin. 5.92. But it concerns the largest
or greatest part, rather than the most advanced or complete one.
326. [36,16] 32,11-16.
327. [36,19-21] Arist. GA 2.1, 735a12-14; cf. 2.3, 736a35-b1. It is important to
emphasise, here and in the lines that follow, both that (1) Alex.s view is entirely
faithful to Arist.s account of reproduction in GA and (2) it is not in any way
incompatible with his own account of the emergence of the soul from the body, as
Moraux 1942 alleges (31-4, 38-43; cf. also Donini 1971, 81). On (1), see esp.
Accattino 1988; on (2), see Sharples 1994b, Accattino 1995 and BD, 26-34, esp.
32-4.
The emergence of the soul and other complex forms from the various forms in
the matter underlying them need not be a diachronic process, as Moraux seems to
assume, despite the language of emerging or begetting. It may instead refer to
the supervenience that holds synchronically at any given time: in order for there to
be a soul or other higher-level form at a given time, there must also be at the same
time certain kinds of underlying material conditions that are sufficient nomologi-
cally for the higher-level forms to exist. The language of emergence and generation
just underscores the novelty and irreducibility of the higher-level forms, which
cannot be found apart from conditions like these. See, along with their correspond-
ing notes, 5,5-6; 9,4-7; 10,17-19; 10,24-6; 23,29-24,4; 24,18-23; and 26,26-30.
What is crucial to observe is that none of these passages says anything about
how the underlying material conditions are brought about: a supervenience thesis
of this sort is, in the abstract, compatible with any number of causal stories.
Nothing in Alex.s emergentism, therefore, need contradict Alex.s or Arist.s view
that man begets man, that is, that each living thing is always produced by another
living thing of the same type, by means of its seeds. Reproduction is not the
transmission of a soul that pre-exists in actuality, much less the magical genera-
tion of a new form by purely formal means. On the contrary, the seed of a living
thing has the power to produce another living thing with a soul of its own by
effecting changes in matter, which eventually lead to the appropriate combination
of homogeneous materials and the constitution of organs from them. Clearly none
of this precludes there being underlying material conditions which are a sufficient
condition for the type of soul involved. On the contrary, one might even argue that
such conditions are required if there is to be an explanation of why these material
changes result in a new substance with a form of its own.
Finally, one should note that the agent which produces this underlying mixture
is not the soul for nourishing oneself in the seed, as BD 33 suggests, since initially
this soul is present only as a power: it does not attain the level of first completion
until the heart is formed in the embryo and begins to draw nourishment. The agent
Notes to page 63 135
is instead the seed itself, which produces the changes in the material furnished by
the mother, which in turn leads to the production of the relevant homoeomerous
materials and eventually the anhomoeomerous organs.
328. [36,22-3] Arist. GA 2.3, 736a31-6. Arist. makes clear that neither the seed
nor the embryo has this soul in actuality before it has the organs necessary to
perform the corresponding functions it has soul only potentially, that is, through
its power (dunamei): GA 2.3, 736b8-15, 737a16-18; cf. 2.1, 735a9-11; Metaph. 7.9,
1034a33-b1. Alex. also hedges on the status of an animal embryo: see 38,7-8 below.
329. [36,23] Repunctuating the text epitdeiou labtai pros touto, sunhistsin
with AD; see Accattino 1988, 81 n. 10 on Alex.s use of epitdeios with pros + acc.
330. [36,23-6] Arist. GA 2.3, 737a18-22; cf. 2.1, 735a12-14.
331. [37,1-3] A living thing cannot possess a power in actuality that is, at the
level of a first actuality or completion (prt entelekheia) unless it is in a position
to exercise that power (absent something extrinsic hindering or preventing it). But
the powers of the soul require bodily organs for their exercise and so cannot be
exercised before the relevant organs have developed. Prior to that time, then, a
living thing can have such powers only potentially: Arist. GA 2.3, 736b22-4. The
powers for nourishing oneself are present in actuality as soon as the heart or its
analogue is formed, the first of the bodily organs: GA 2.1, 735a14-26; 2.5, 741b15-
17. Other powers appear later, as the formation of the corresponding organs is
staggered over the embryos development: no power for perceiving can attain first
completion before the relevant perceptual organs have been formed (736b1-5). For
Alex. on the gradual emergence of powers, see below, 74,17-23; for the requirement
that the organs must be present in order to possess the corresponding power, see
27,8-15 with note.
What is noteworthy here is Alex.s vague reference to the other powers (37,1).
If this includes the rational powers, as it seems to suggest, then Alex. holds that
the rational powers do not differ from others powers in this regard: they have not
yet appeared because there are parts of the body on which their functioning
depends. Contrast this with the central Aristotelian text on which Alex. depends
here, GA 2.3, which takes the rational powers to pose a puzzle that needs to be
solved (736b5-8). Unlike other parts of the soul, the understanding lacks a dedi-
cated organ and so is not the form of any part of the body (DA 3.4, 429a18-27; cf.
2.1, 413a4-7): Arist. even claims that its activity is not a bodily one (GA 2.3,
736b29). Consequently, he regards it alone, of all the powers of the soul, as not
involving a material contribution from the mother and so alone comes from
without (monon thurathen epeisienai, 736b27-8), that is, from the father. (For a
more extended defense of this reading of Arist., see my 1999, 215-16.) Alex. agrees
that the understanding is not the form of any part of the body; but he insists that
it cannot function without bodily activity either (see n. 44 on 12,20-1 and n. 200 on
21,24 above). This might explain why the rational powers are the last to develop: they
cannot be exercised until other powers, especially perception and representation, have
themselves reached first completion; but these in turn depend on the development of
organs that only emerge over time. (Alex. develops an entirely different reading of the
understanding from without below, however: see 90,19-21 and 91,2.)
332. [37,4-5] Arist. appeals to this fact repeatedly, but draws a more nuanced
distinction than Alex. states here. In Arist.s view, not just plants, but also certain
insects like the centipede can survive dismemberment, at least for a little while;
some animals like the tortoise can even survive the loss of an essential organ, like
the heart: DA 1.4, 409a9-10; 1.5, 411b19-27; 2.2, 413b16-21; Long. 6, 467a18-23;
Iuv. 2, 468a23-b15; Resp. 3, 471b203; de Vita et Morte 1 (= Resp. 18), 479b1-7; IA
136 Notes to page 63
7, 707a27-b5; PA 4.6, 682b30-2. On Aristotles discussions of dismembered insects,
see Sprague 1989.
Alex. oversimplifies Arist.s view throughout most of this section, maintaining
a strict dichotomy between plants and animals. At the end (38,8-11), he does
mention that animals often survive dismemberment briefly, but it seems little
more than an afterthought to bring himself in line with Arist.s stated views. Alex.
does not specify here whether all animals are capable of this or only some; nor does
he explore how or in what circumstances animals manage to do this. At the end of
the present treatise, he mentions that the tortoise and also the chameleon can
survive the loss of their heart and adds, for dialectical purposes, that other animals
can survive the loss of their head (100,9-13). But again he does not integrate this
observation with his views here.
333. [37,5-11] Alex.s argument here is elaborated more fully below at 37,18-
38,4 (see accompanying note). But the key point is that nourishment is
accomplished in plants without highly differentiated organs in disparate locations;
and such differentiation as there is can be regenerated from most of the plants
parts, as they were originally generated from the seed.
334. [37,11-15] The Greek in this passage has been found difficult to construe.
Bruns takes there to be a lacuna, inserts two clauses, and repunctuates the
subsequent lines, altering the argument structure. But the text is grammatical as
it stands and only requires a minor emendation of oute to oude at 37,14 (following
Bruns suggestion). The real difficulty is to identify the unexpressed antecedents
for various terms. (i) It is not clear what paresparmenai at 37,12 modifies. AD take
it to refer to souls and accordingly insert <hai psukhai eisin> before oukh homois
at 37,12; but this would make the singular h tnde psukh in the next line
awkward. Like BD, I prefer to take powers (dunameis) as understood, since it is
prominent in the immediately preceding lines and so would not need to be
repeated. (ii) The antecedent for ekeinn at 37,14 and 15 must be the same (against
BD), but then it cannot refer to animals (against Bruns and BD), since the
antecedent is said to be what is received at 37,15 (dektika). Like AD, I take these
pronouns instead to refer to souls, explicitly mentioned at 37,13.
335. [37,11-18] Alex. is making two distinct points in these lines, which he
interweaves throughout: (1) that in animals the soul is not homogeneous (ho-
moiomers) in each part of the body in the way that it is in plants, and (2) that a
separated part of an animals body cannot regenerate the missing parts or acquire
functions that were not originally developed in it from the seed.
(1) The lack of homogeneity is due to the close relationship between the actual
possession of a power and the organs that subserve it: see esp. n. 331 on 37,1-3
above; cf. also n. 249 on 27,8-15. This is true not only of different types of powers,
but even just the power for nourishing oneself by itself. Although this power is
common to both plants and animals, it is realised in different ways in each: as one
ascends the scala naturae, there is an increasing complexity in the way living
things carry out even simpler functions and correspondingly a greater articulation
in their bodily organs, which reflects a more highly specified division of labor:
Mant. 103,38-104,5. As these organs are inevitably located in different parts of the
body, physically separating them from each other destroys their ability to function
together as a system. An animal can lose its eyes or limbs and still live (cf. PA 3.4,
665b24-6). But the permanent loss of parts of vital organs, as are required for
digestion, eventually spells death: Mant. 104,5-8.
Alex. departs somewhat from Arist. here, who insists that the soul present in
the divided parts is one in form (homoeideis) both with each other and the soul of
Notes to pages 63-64 137
the original animal, despite the loss of organs, which can even remain sensitive for
a time (DA 1.5, 411b19-27; Iuv. 2, 468b1-9, b12-14). In his view, the loss of organs
explains why the divided parts fail to survive, but does not tell against the type of
soul present in them.
(2) The dismemberment of animals generally implies a permanent loss: unlike
plants, they cannot regenerate the missing organs. Arist. holds that this is the case
with most animals and even insects (Long. 6, 467a20-2; cf. PA 4.6, 682b30-2),
although he sometimes allows that simple insects like centipedes can survive
because, like plants, the principle (arkh) of their vital activities is present in each
part (PA 4.5, 682a2-9).
336. [37,18-38,4] Plants can regenerate roots and stalk from any portion, since
the principle (arkh) is present in all parts potentially: Long. 6, 467a18-30; Iuv. 2,
468b5-9; PA 4.6, 682b30-2.
337. [38,4-8] 74,17-23. The question of whether the embryo is an animal is a
set topic in the doxographical tradition already by the first century CE (see Aet.
5.15, Dox. Gr. 425.14-426.14; cf. also Clement Strom. 8.10.3-4; Ps.-Galen Def. Med.
445, 19.451-2 Khn), which typically represents Aristotelians as midway between
Platonists and Stoics. The Platonists think the embryo is an animal, whereas the
Stoics think that it is not, but like a plant (Plut. Stoic. Rep. 41, 1052F = SVF
2.806.1; Aet. 5.15.2, Dox. Gr. 425.18-22 = SVF 2.756; Ps.-Galen Def. Med. 445,
19.452.5-10 Khn = SVF 2.757) in fact, the Stoics regard nourishment and
growth as due to nature, not soul (see n. 282 on 31,25-32,1 above). The Aris-
totelians, in constrast, think the embryo has a soul, but is only an animal in
potentiality. See Moraux 2001, 363-3 n. 201, as well as Mansfelds invaluable and
meticulous discussion in 1990, 3186-90.
According to Arist., embryos actually possess the power of nourishment once
they begin to draw nourishment (helkei tn trophn, GA 2.3, 736b8-13). But prior
to the development of perceptual organs, it only lives the life of a plant (phutou
bion, GA 2.3, 736b12-13; 3.2, 753b27-9; 5.1, 778b32-779a2; cf. EE 1.5, 1216a3-9); it
is an animal only in its power and is still incomplete (GA 2.4, 740a24-7). Alex.s
view is somewhat stronger. It is not simply that the embryo at this stage is an
animal only in its power; it is also not even receiving nourishment as an inde-
pendent being in its own right, but is just a part of another being, its mother. Arist.
himself only says more vaguely that it receives its nourishment from something
else because it is an outgrowth (ti pephukenai, GA 3.2, 753b27-9). It is the Stoics,
in fact, who seem to speak of the embryo as a part of the womb (Aet. 5.15.2, Dox.
Gr. 425.18-22 = SVF 2.756). It is perhaps worth noting in this context that Arist.
thinks abortion is permissible before perception and life (aisthsin } kai zn)
life, presumably, as an animal or a human have been realised (Pol. 7.16,
1335b19-26). Arist. believes this is 40 days for a male child and 90 for a female one
(HA 7.3, 583b9-25).
Galens position is broadly similar. He also compares the embryo to plants, like
Arist. and the Stoics: see esp. Foet. Form. 3, 68.10-70.21 Nickel. But Galen is closer
to Arist., since he does not hesitate to speak of the embryo as having a soul and
regards it as becoming animal-like prior to birth, as soon as a pulse is present: 3,
74.11-16 Nickel.
338. [38,8-11] Arist. DA 1.5, 411b19-27; Iuv. 2, 468b1-9, b12-14. Alex. here
qualifies the oversimplified position he takes throughout most of this section,
though without further exploring its theoretical ramifications (see n. 332 on 37,4-5
above). On traces (enkataleimmata), see the note on. 62,22-63,4 in the second
volume of this translation.
138 Notes to page 64
339. [38,12-13] For the first soul as the soul for life, see 29,1-10; as the soul for
vegetation, see n. 315 on 35,24, where Alex. identifies its activities as nourish-
ment, growth, and reproduction.
340. [38,13-15] The argument for the first soul as the form of the body occurs
at 31,7-25. For the three types of powers involved, see 35,26-36,9 (cf. 32,6-23); the
activities themselves are specified at 34,27-35,25.
341. [38,16-18] 29,12-22; 29,26-30,6. For Alex., this claim is true only if re-
stricted to mortal beings: see the discussion of Alex.s reading of Arist. DA 3.12,
434b3-8 in his lost commentary, ap. Ps.-Philop. in DA 595,37-596,14 (cf. n. 259 on
28,25-9 above).
342. [38,18-19] See 10,8-10 with corresponding note.
343. [38,19-20] In addition to DA 2.5-3.2, Arist. devotes an entire essay to
perception (Sens.), on which Alex. has an extant commentary (in Sens.) and to
which he may be referring here.
344. [38,21-39,2] cf. Mich. in EN 556,37-557,8, which contains a doublet of the
entire passage (38,21-39,8). Alex. synthesises elements from several different
discussions of perception in Arist. in this compact opening formulation:
(1) Perception is a form of cognition (krinein), in which one discerns or discrimi-
nates objects. For the meaning of this term, see n. 346 below on 39,4-5.
(2) Perception is understood as something relative to (pros) its objects and vice
versa. This is explicit in Cat. 7 (6b2-8, b33-6, 7b33-8a12) and clearly implicit in DA
3.2, 425b25-426a27. Recognising the relational nature of perception does not
imply, however, that the change involved is a merely relational one, a mere
Cambridge change, where an object is said to undergo a change solely in virtue of
changes in its relation to other objects rather than due to any modification intrinsic
to itself (Geach 1969, 71-2): see Alex. in Sens. 127,5-128,6, esp. 127,5-12 and
128,2-6; more generally, in Sens. 16,15-16; 17,13; in Metaph. 324,33-325,1; 402,8-
13; 406,25-407,15; 409,26-36; cf. in Metaph. 323,26; in Top. 71,4-7; 104,2-6;
218,21-2; 407,4-5.
(3) Perception comes about through a kind of alteration (dia tinos alloises),
as a result of which the perceptual organ that has the power is assimilated or
likened (homoioumenon) to the object; Alex. expands on this below at 39,10-18.
Arist. classifies perception as this kind of modification or alteration in his first
extended discussion of perception in general in DA 2.5. Like other agent-patient
interactions, he regards perception as a case where the agent (the perceptible
object) modifies or alters the patient (the perceiving subject), making it like itself,
having been unlike before (417a14-21, 418a3-6). The kind of modification it under-
goes does not destroy its nature, though, as other alterations might, but instead
preserves the power to perceive by exercising it. Arist. concludes that perceiving is
either not altering or an alteration of a different type (heteron genos, 417b2-16;
see also 3.7, 431a4-6). This explains Alex.s qualification above, that perception is a
kind of alteration (tinos alloises, 39,2) or a kind of modification (paskhein ti), a
qualification Arist. consistently makes himself: DA 2.4, 415b24; 2.5, 416b34; 2.11,
424a1; Insomn. 2, 459b4-5; MA 7, 701b16-18; Phys. 7.2, 244b10-12 (though n. b. 7.3,
248a6-9); it is also said to be a kind of change (kinsis tis) at Somn. 1, 454a9-11.
Alex. discusses whether perception is an alteration in more depth in Quaest. 3.2
& 3 (the former also refers explicitly at 81,17-18 to a discussion in his commentary,
most likely his lost commentary on the DA, though see Sharples 1994a, 128 n. 197).
Alex.s considered position seems to be that (a) perception is not itself an alteration
in the primary sense, but (b) it nevertheless occurs through (dia) an alteration in
the perceiver. For the contrast between (a) and (b), see Quaest. 3.3, 84,4-6.12-13;
Notes to page 64 139
3.9, 98,2-8; cf. also the analogy to the exercise of practical know-how (tekhn) at
Quaest. 3.2, 81,27-9. At in Sens. 9,5-6 Alex. suggests that Arist. says that percep-
tion is a modification because it comes about through a modification (dia pathous).
He frequently makes the second point, (b), by itself in the DA: see 39,12; 50,11;
53,28; 54,21; 55,14; cf. 44,12; 59,8; also Quaest. 3.3, 83,3; 86,29-31; 3.9, 96,35. For
(a) specifically, see Quaest. 3.3, 83,18; 84,18; 85,34-86,2; 86,23-5; cf. 84,8-12.
(4) It is interesting that Alex. does not mention the famous doctrine of receiving
form without the matter in his general characterisation here, but only a little later
at 39,13 below. If the formulation above is a kind of formal definition, as AD
suggest (175), its omission could be significant. But there is nothing that requires
us to treat this formulation as anything more than a general summary, which can
be filled in more as we go on.
345. [39,2-4] 33,7-10, with corresponding note; also Arist. DA 2.4, 415a20-2; cf.
1.1, 402b14-16.
346. [39,4-5] Alex. uses this pair of descriptions in defining most of the senses:
each is a power for being aware of its objects and having cognition of them
(antilptik kai kritik). See 46,20-1; 50,9-11; 53,26-9; 53,30-54,2; 55,12-14; cf.
61,24-7; also Mant. 107,6-7; in Sens. 166,3. Both terms are central to Alex.s
treatment.
The verb for the first activity, antilambanesthai, and its corresponding noun,
antilpsis, are often rendered with cognates of apprehend. But this is too abstract:
in the DA Alex. never uses the term for understanding or thought, but only for
perception. (Part A of the de Intellectu uses both terms for understanding at Mant.
107,7 but as part of a strong parallel with perception.) Alex.s use of antilpsis thus
suggests the kind of direct, perceptual awareness we have of an object facing us. It
is not used in these contexts to emphasise a connection with knowledge specifi-
cally, as the more abstract apprehension might suggest, but something
phenomenal and qualitative.
The verb for the second activity, krinein, and its corresponding noun, krisis,
have been rendered by other scholars in various ways. Often cognates of judge
have been used, which would imply that it signifies a propositional attitude with
a conceptual content as well as the subjects endorsement or commitment, much
like belief. But it has been persuasively argued that, at least in Arist., this family
of terms is used for a much more basic activity, of discriminating or discerning
between different qualities, especially those on the spectrum of qualities that a
given sense is sensitive to: see esp. Ebert 1983. Most of Alex.s uses of this term are
not decisive: he often speaks of engaging or being able to engage in krinein without
further specification, beyond the mention of an object. Of the remaining cases,
many can be understood in terms of perceptual discrimination, even when they
involve content that seems more judgment-like, as when we perceive that contrar-
ies are contrary to one another (64,12-20). But not all of them can be understood
in terms of discrimination and discerning. At 78,13-21 Alex. notes that there are
many kinds of krisis other than perception: representing, endorsing, being aware,
assuming, believing, knowing, reasoning, thinking, and understanding; cf. also
66,9-19; 72,3-5; 76,8-12; cf. Arist. DA 3.3, 427a20-1, 428a3-5. Discrimination may
be what perceptual cognition consists in; but there is no reason to think all of these
other sorts of cognition consist in discrimination for Alex. Because of this broad
semantic field, I have rendered these terms with cognitive and its cognates to
express the general notion of a mental activity that has content and is involved in
some way with knowing, in contrast with desiderative or practical attitudes. Alex.
explicitly draws this contrast himself, splitting the capacities of the soul into two
140 Notes to page 64
broad types, the cognitive (to kritikon) and the practical (to praktikon), at
73,20-6; 75,13-15; 76,1-3; 99,17-24. Emilsson also notes this contrast (1988, 122),
but claims that the traditional rendering judgement will still serve as well as
cognition for Alex.s use of krinein. But this is not so. Judgement presupposes the
subjects endorsement or rejection (cf. Salles 2001, 6 on the Stoic notion), whereas
it is clear that some of the attitudes Alex. mentions, such as representation, do not.
The content of judgements, moreover, is both propositional and conceptual, and so
presupposes logos, whereas Alex. clearly believes that non-rational animals, who
lack logos, nevertheless have kritikai powers too. What is essential is that all these
activities are involved in the acquisition of knowledge, broadly understood, in
contrast with desires or inclinations. If this is right, then Alex.s identification of
perception and cognition at Quaest. 3.9, 94,32-95,1 sounds less odd: for there is no
independent reason to think that Alex. regards perception in general as a kind of
judgement.
The term gnstikos, which some translators render as cognitive, has a more
limited semantic field, in contrast: gnstikos and its cognates are used for higher
epistemic achievements and in connection with knowledge in a narrower sense (see
esp. 82,18 and its cognates at 1,6-9; 2,1; 66,17; 80,8; 82,13; 84,12). I have therefore
used the more epistemic word apprehend to render gnstikos and its cognates.
347. [39,5-6] The term perception (aisthsis) is ambiguous: it can designate
either the activity of perceiving or the power for it (cf. Arist. DA 2.5, 417a9-14; 3.2,
426a23-4). Before a perceptible object acts on it, it is perception only in power
(dunamei), at the level of first completion or actuality (cf. 425b29-31). Perception
is always of objects which are present (parontn) to the perceiver, unlike anticipa-
tion and memory, which are of future and past objects respectively: Arist. Mem. 1,
449b13-15, b27; cf. Rhet. 1.11, 1370a32-5; Alex. DA 69,11-14; Quaest. 3.3, 86,8-10;
in Top. 116,28-9; cf. DA 66,24-8; in Metaph. 312,1-2.8-9; Quaest. 3.3, 85,15.
348. [39,6-8] Arist. uses the example of someone who possesses knowledge, but
is not actively using it at the moment, to explain the difference between having the
power to perceive and actually exercising it (DA 2.5, 417a21-b19), much as he had
earlier to explain the idea that the soul is a first completion or actuality (entelekheia
prt, Arist. DA 2.1, 412a22-8; see above, Alex. DA 16,7-18). But if (a) actually
exercising a power stands above (b) merely possessing it, actually possessing a power
still stands above (c) merely having the power to acquire the power (on the latter, Arist.
DA 2.5, 417a22-8, b29-32). (b) is thus Janus-faced: in comparison with (a), a perceptual
power such as sight is a first completion, like the soul (412b27-413a3); but in
comparison with (c), it can also be viewed as something higher than something else, a
second power (deutera dunamis), as Alex. calls it at Quaest. 3.2, 81,9; 3.3, 84,33-6.
On the neuter pronoun ekeina at 39,8, see Bruns note in the critical apparatus
and his entry in the index for Alexanders occasional practice of using neuter
pronouns for nearby masculine and feminine antecedents.
349. [39,9-10] 71,26-72,5; 92,23-93,24; in Sens. 10,15-26; cf. in Meteor. 148,9-11.
Alex.s explanation for why perception is directed at external objects has no precise
parallel in Arist. The chapter Alex. is summarising, DA 2.5, does mention that
perceptibles are external and uses this to explain why perception is not up to us
(417b19-25). But it does not refer to the final cause of perception or its role in action
at all. Arist. does address the issue elsewhere, though, in Sens. 1. He. argues that
the distance senses are for the sake of preservation (strias heneken) because
they can detect beneficial and harmful situations in advance, and so allow an
animal to pursue or avoid what it ought (436b18-437a3; cf. DA 3.12, 434b24-7,
though compare 3.13, 435b19-21). Pursuit and avoidance rely on our ability to
Notes to pages 64-65 141
discriminate between what is pleasant and painful, esp. in food, a sensitivity
already present in the contact senses (Sens. 1, 436b12-18). Without the distance
senses, mobile animals would not be able to find their food from a distance or evade
predators and other dangers (DA 3.12, 434a32-b1); without the contact senses,
which perceive food (2.3, 414b6-9; cf. 3.12, 434b18-22; 3.13, 435b22-3), the animal
would not be able to seize or avoid what it ought (3.12, 434b13-17). Alex. adds to
this observations on how the different distance senses contribute to preservation:
smell allows animals to detect food, sight to detect predators, and sound both (in
Sens. 10,16-25).
Arist. also argues that the distance senses are for the sake of the animals
well-being (tou eu einai heneka, 3.12, 434b24-7; 3.13, 435b19-21), esp. sight and
hearing in animals with intelligence (Sens. 1, 437a1-17). Alex. greatly expands on
Arist.s brief remarks about the different contributions of sight and hearing to
thought and differences in what they can discriminate: see in Sens. 10,26-14,6.
350. [39,13-14] Alex. briefly alludes here to a doctrine from a later chapter, DA
2.12, in which Arist. compares the senses to sealing wax: both receive the form of
the object impinging on them without the matter (aneu ts huls, 424a17-24 at
a18-19). Exactly what this phrase means and whether (and how) it qualifies the
kind of likening involved is a matter of considerable dispute. Alex. will take this
up later in the treatise, esp. in his comparisons of perception and understanding:
see esp. 83,13-84,10; also 60,3-6; 66,13-14; 78,7-8; 92,21-3; Quaest. 3.7, 92,27-31 (cf.
92,35 and 93,9-11); 3.9, 95,20-1. In the present passage, Alex. does not seem to
think that such qualifications affect his point about likening.
351. [39,10-18] Alex. elaborates the claim above (39,1-2) further. Even if
perception should turn out not to be an alteration in the primary sense itself see
(3) of n. 344 on 38,21-39,2 above it nevertheless occurs in virtue of a kind of
assimilation or likening (kat homoisin) to the perceptible object, which comes
about through an alteration (di alloises, 39,12). Alex. is clear that what under-
goes alteration and is likened is what possesses the soul (to ekhon autn, 39,1), that
is, the body that primarily possesses the perceptual soul (ginetai to pathos touto
en ti prti tn aisthtikn psukhn ekhonti smati, 39, 18-19).
In the passage above, Alex. develops one of the main conclusions of the argu-
ment in DA 2.5, namely, that this change is understood along the same lines as
alteration in general (cf. 417a1-2). The agent, in acting on the patient, makes the
patient like itself, with respect to the active quality, so that in one way they must
be unlike (namely, before the change), and in another alike (after the change): see
DA 2.5, 417a18-20 and, for the general principle, GC 1.7, passim. In this respect,
perception does not differ from nourishing oneself, as Alex. notes (39,16), glossing
Arist.s own back-reference at 417a19-20: at any rate, Arist.s discussion of nour-
ishment in DA 2.4 employs the same causal framework to reach an analogous
result (416a29-b9). See Alex.s remarks above, 33,15-34,18.
352. [39,18-21] Alex. begins to develop here the notion of a perceptual system,
with the contrast between a central or primary perceptual organ, where percep-
tion actually occurs, and the various peripheral organs which relay the
modifications they undergo from perceptible objects to it. Alex. later famously
compares the perceptual system to a circle, with the five peripheral sense organs
at points on the circumference, relaying their modifications along radii that
terminate in the common sense at the circles centre: 63,8-64,11; cf. 64,17-65,2;
78,10-13; see Henry 1960. Alex. mentions the internal relaying (diadosis, diadi-
donai) of perceptual stimulations from the peripheral organs to the central organ
on a number of occasions: 39,21; 41,5; 63,16; 64,8; Quaest. 3.9, 97,5.12; in Sens.
142 Notes to page 65
59,12-15; cf. also parapempn at DA 50,15-18 below and diapempomena at Quaest.
3.9, 96,33.36. In light of this, AD are surely right to correct the MSS eit at 39,21
to ep, a simple majuscule error. They point out that the construction diadidonai ti
epi ti occurs several times in the treatise: 41,4; 42,19; 63,16; 64,8. Ironically, Alex.
uses the verb diakoneisthai here (39,20) in its more common sense of serve,
although elsewhere in this treatise he uses it most frequently to designate the
transmission of a message from perceptible objects to the peripheral organs: see n.
362 on 41,1-3.
Alex. is systematising Arist.s views here, based on scattered remarks across the
corpus, but especially from the Parva naturalia. Several passages confirm that
there is a single, common perceptual organ (koinon aisthtrion), to which all the
other perceptual organs lead (pros ho sunteinei talla) and which is in charge of
them (tou kuriou tn alln pantn aisthtriou), receiving all perceptibles (dektikon
pantn tn aisthtn): Somn. 2, 455a12-b2; Iuv. 3, 469a10-16; PA 2.1, 647a24-30;
cf. 2.10, 656a26-33; GA 5.2, 781a20-3. Arist. describes at some length the transmis-
sion of changes due to perceptual stimulations (hai kinseis hai apo tn
aisthmatn ginomenai), which travel to the principle and origin of perception (epi
tn arkhn ts aisthses katapherontai) at Insomn. 3, 460b28-461a8 (cf. a8-b20).
The Stoics also claim that perception arises when an external impulse has been
relayed to the governing faculty (Galen PHP 2.5.35, 134.23-6 De Lacy = SVF
1.151, 2.882, cf. also 7.7.17, 472.29-33; Plot. 4.7.7.7-11 = SVF 2.858), as does
Ptolemy (de Judic. 8, 13.6 Lammert). But we can find this usage in Plato already
as well (toutn tas kinseis diadidon eis hapan to sma mekhri ts psukhs, Tim.
45D; cf. Tht. 184D).
Alex.s initial reference at 39,18 to a primary body may derive from Arist.s
definition of a perceptual organ (aisthtrion) as the primary thing in which this
sort of power resides (prton en hi h toiaut dunamis, DA 2.12, 424a24-5). But
if so, then Alex. does not take this to refer to the peripheral organ that external
stimuli first impinge upon, but rather the ultimate (eskhaton) perceptual organ,
inside (entos) the animal, much as the organ of touch is described in Arist. as
located not on the bodys surface, but in something primary beneath the flesh: DA
2.11, 423b29-31 (cf. b22-3); also 422b21-3; 3.2, 426b15-17; PA 2.10, 656b34-6 (2.8,
653b23-30, though compare 2.1, 647a19-21); and in the Parva naturalia Arist.
describes the common sense as the primary thing that can perceive (prton
aisthtikon): 450a12, 451a17, 458a28. In any case, Alex. also uses both adjectives
for the central organ (or sense), sometimes in close proximity: it is first at 60,6;
65,8; 68,6; 69,22; 97,6; cf. Mant. 142,32; and ultimate at 63,15.18; 64,8.18; Quaest.
3.9, 96,31; 97,5.10; in Sens. 168,3. For discussion, see Sharples 2005b, 349-50 nn.
17-18.
353. [39,21-40,3] On the soul for perceiving being situated in the area around
the heart (hidrumenon peri tn kardian, 39,21-2), see esp. 96,25-97,14; 98,1-7;
99,19-30). The location of the governing part and the souls other powers is the
subject of a lengthy appendix at the end of the present treatise (94,7-100,17). On
the heat and moisture of the heart, see 94,17-20; in Sens. 40,22-3.
These details about the heart are entirely orthodox. Arist. locates the principle
and origin of perception (arkh ts aisthses) in the heart: see esp. Iuv. 3,
469a5-20; 4, 469a23-7, b3-6; Somn. 2, 455b34-456a6; also PA 2.1, 647a24-33; 2.7,
653b5-8; 2.10, 656a27-31 (cf. b24); 3.4, 665a10-13; 3.4, 666a11-18, a34-b1; 3.5,
667b21-31; 3.10, 672b14-19; 4.5, 678b1-4; GA 2.6, 743b25-7; MA 11, 703b23-4 (cf.
9, 702b20-1). The heart is described as a well-spring (pg) at PA 3.4, 666a7-8 (cf.
666a11-18, a34-b1). On the heart as responsible for both the moisture and heat of
Notes to page 65 143
blood, in connection with perception, see PA 3.5, 667b19-34; cf. 2.4, 650b22-3;
Somn. 2, 456a6-10; heat and moisture are said to belong to animals naturally at
Long. 5, 466a18-19. On heat and the heart more generally, Iuv. 4, 469b6-20; Sens.
2, 439a1-4; cf. Resp. 15, 478a24; PA 3.7, 670a23-6. On the heart as the origin of
breath or pneuma, Somn. 2, 456a3-10; MA 10, 703a14-16.
The commitment to a central organ need not imply that perception only occurs
there, however; still less does it reduce the peripheral organs to a purely interme-
diary role like the external medium. But there are passages where Alex. seems to
take extra steps in this direction. At in Sens. 34,3, for example, he says that neither
the eye nor the vitreous humour sees and at 36,9 that the power of sight is not in
the eye, but rather in the central organ; in fact, none of the senses can be in the
peripheral organs, he argues, or we would not be able to compare their reports (in
Sens. 36,11-19). But he does not make these denials expressly in the DA, which
may therefore allow a more accommodating position, where, for example, the
peripheral organs are first affected differently than the medium, but then transmit
or even report (64,2.20) their modification to the heart: see DA 60,14-17; 62,16-20;
63,28-65,2; also 44,6 where the eyes are said to be that through which (di hou)
the awareness of colour arises. For a sensitive and thorough discussion of the
evidence for both Arist.s and Alex.s position on this question, see Sharples 2005b,
esp. 352-9.
354. [40,4-5] Alex. evidently has in mind the five senses here: sight, hearing,
smell, taste, and touch. But in speaking of them as parts (mer) of perception, he
is thinking of them as different powers, much as perception is part of the soul: see
n. 244 on 27,1-4 above; cf. 19,2-3. The Stoics also speak of the five senses as parts,
though as parts of the soul: Aet. 4.21.2-3, Dox. Gr. 410.30-411.13 = SVF 2.836 = LS
53H 3; D.L. 7.110 (cf. 157) = SVF 2.828.
355. [40,5-6] On the serial order of souls, see 16,18-17,1; 28,20-5; on the serial
order of senses, see Alex.s remarks just below at 40,11-15. Bruns inserts phutiks
before ts psukhs at 40,5, in line with the Hebrew translation, thus making the
comparison to a serial order specifically in the soul for vegetation, rather than the
soul in general (as I have taken it). But the difficulties Alex. immediately goes on
to discuss at 40,6-9 (see n. 357 below) are peculiar to the soul in general: since only
the most rudimentary powers are possessed by living things universally, while the
subsequent powers are not, a general characterisation could only mention the
more rudimentary powers. The different senses present the same difficulty be-
cause touch is the only sense possessed by animals universally. The three powers
of the soul for vegetation (35,23-36,9), in contrast, are shared universally by all
living things (or at any rate, those which have not been maimed) and so in fact
permits a common account: 32,9-23.
356. [40,6-9] See esp. 17,1-8, with corresponding note; also 28,14-20; 30,17-20;
Quaest. 1.11a and b, passim. I take the imperfect n here to be a philosophical
imperfect expressing a general truth, rather than a back-reference (as AD 177 do).
357. [40,9-10] Just as Alex.s discussion in this section is patterned after Arist.s
in DA 2.5, his procedure for the next 26 pages largely follows Arist.s sequence in
DA 2.6-12, including a preliminary discussion of the different kinds of perceptible,
digressions on light and echoes, and a final discussion of perception in general:
kinds of perceptible (40,20-42,3); sight (42,4-43,8) and light (43,8-46,19); hearing
(46,20-47,24) and echoes (47,25-51,4); smell (51,5-53,29); taste (53,30-55,14); touch
(55,15-59,24); and perception in general (60,1-66,8). Arist. offers no reason for
proceeding in this way, though: the justification at 40,4-10 is wholly Alex.s own.
144 Notes to pages 65-66
358. [40,11-15] Alex. again brings together several different, but related doc-
trines here:
(1) The notion of separation involved is a taxonomical one: for details, see n. 258
on 28,23-5 above; 92,15-94,6; cf. in Sens. 10,12-15. Arist. draws the exact same
parallel between the separability of the different senses and of different kinds of
soul himself at DA 2.2, 413b4-7. He frequently states that touch can be found apart
from the other senses, but not vice versa: DA 2.2, 413b4-7; 2.3, 415a3-5; 3.13,
435a12-13, b2; cf. Sens. 1, 436b13-14. Alex. frequently makes the following, some-
what weaker claim instead.
(2) Touch is necessary for there to be an animal at all and so belongs to all
animals: Alex. DA 59,20; 92,23-93,19; 96,22-3; in Sens. 9,7-11; in Top. 116,25;
455,17; Arist. DA 2.2, 413b4-5, b8-9, 414a3-4; 2.3, 414b3; 3.12, 434a28-9, b9-14,
b22-4; 3.13, 435b4-7, b16-19; Sens. 1, 436b13-14; Somn. 2, 455a27; HA 1.3, 489a17-
19; 4.8, 535a4-5; PA 2.8, 653b22-4; cf. EN 3.10, 1118b1.
(3) For the claim that taste is a kind of touch (haph tis), see 51,26 (cf. 54,1);
93,4; in Sens. 39,23; 78,2; 105,4-5; also Arist. DA 2.9, 421a19; 3.12, 434b18, 21;
Sens. 2, 438b30-439a1; 4, 441a3; PA 2.17, 660a21. But it is important to note that
Arist. also denies that taste just is touch (DA 2.10, 422a10; 2.11, 423a20-1). His
view seems to be that taste is of nourishment and so must be of a kind of tangible
quality (DA 2.3, 414b6-9; 3.12, 434b18-19; cf. Sens. 4, 441b24-442a12). But if so,
then we can only perceive it by being in contact with it literally, by touching
and not through the medium of a foreign body (DA 2.10, 422a8-9; cf. Alex. DA
54,1-5; in Sens. 168,20-2).
On a philological note, AD 177 persuasively argue that opsis at 40,11 should
be taken as corrective (or rather sight), since the frequent occurrence of horasis
in the doxographical tradition is out of line with the strict distinction Alex. draws
between the activity of seeing (horasis) and a capacity like sight (opsis), which is
more appropriate here.
359. [40,15-19] Alex. relies here on Arist.s methodological principle (DA 2.4,
415a14-23) that in order to understand a given capacity, we must look first to the
activity or exercise of that capacity, and to understand that activity, we must look
still before that at the kinds of objects on which it is exercised: see 32,25-7 and
33,7-10 above, with corresponding notes. But this order makes even more sense in
the case of perception, where the animals senses are in some way likened to the
objects of perception, unlike nourishment, where the object is assimilated to the
animals body. By learning about the nature of perceptible objects in themselves,
we therefore learn something about the nature of the effect they have on us.
360. [40,20-1] Alex adopts Arist.s two-stage division of perceptibles at DA 2.6,
418a8-11, first broadly into those which are intrinsically perceptible (40,21-41,5)
and those which are extrinsically perceptible (41,5-10), but then subdividing the
first into those which are intrinsically perceptible to one sense exclusively (40,21-
41,1) and those which are intrinsically perceptible to several senses in common
(41,1-5). Arist. also discusses different kinds of perceptible at DA 3.1, 425a14-b10;
3.3, 428b18-30; Sens. 1, 437a3-17; cf. Insomn. 1, 458b4-6.
Arist. offers no independent characterisation of things which are perceptible
intrinsically (kath hauto), literally, in virtue of themselves or on their own. But
presumably any kind of thing will be intrinsically perceptible just in case it is
perceptible as such, that is, it is perceptible in so far as it is an F or in virtue of
being F (for some value of F), something that is not true of things which are
perceptible only extrinsically (DA 3.1, 425a25-7). It is tempting, furthermore, to
construe this distinction causally: Fs are intrinsically perceptible because they can
Notes to page 66 145
bring about perception (poitikon ts aisthses) in virtue of being F (Sens. 6,
445b4-8; cf. 3, 439a17). The senses are not affected, in contrast, by extrinsic
perceptibles in so far as they are such (DA 2.6, 418a23-4). But a causal reading is
not unproblematic. While it applies straightforwardly to things which are intrin-
sically perceptible to one sense exclusively (cf. DA 2.6, 418a24-5), it is not clear
whether or how this would apply to common perceptibles. Arist. never explains the
causal mechanism by which we perceive common perceptibles and there are points
where he speaks of them as if they might even be extrinsic to perceptibles exclusive
to one sense: DA 3.1, 425a13-20, b4-10; 3.3, 428b22-4 (though contrast 3.1, 425a21-
30). There is little discussion, or even awareness, in the literature on Arist. of this
apparent conflict. But the expression in virtue of can be understood in many ways,
including definitionally (see, e.g., Sorabji [1971] 1979), and so its precise meaning
in this context, for either Arist. or Alex., is very much open to debate.
The most Alex. says about what is intrinsically perceptible is negative. Time is
not intrinsically perceptible because it is not some underlying nature which we
perceive (hupokeimen tis phusis hs aisthanometha), but rather that in which the
things we perceive are and occur (in Sens. 147,5-8). Similarly, small undetached
parts of bodies, like the ten thousandth part of a millet seed, are not intrinsically
perceptible, except potentially, because they do not as parts intrinsically exist, that
is, on their own as such (and once they do, they are too small to be seen by
themselves): see esp. in Sens. 116,21-3 and 116,7-122,23 passim. But these uses of
intrinsically perceptible seem to be different from Arist.s sense or in the passage
above, since both time and the colour of a ten thousandth part of a millet seed are
both intrinsically perceptible in the sense given here (colour being perceptible on
its own exclusively to sight and time common to several senses).
361. [40,21-41,5] Although usually translated as special, peculiar, or even
proper perceptibles, the meaning of the phrase idia aisthta is best understood
through its contrast with koina aisthta (see n. 362 on 41,1-3), that is, as simply a
question of whether something can be intrinsically perceived by just one sense
exclusively or is shared by several; expressing the contrast in this ways avoids some
of the epistemological connotations of the more traditional renderings. When Arist.
first introduces exclusive perceptibles, he characterises them in two ways: as
things (a) that cannot be perceived by another sense and (b) about which we cannot
be mistaken (418a11-12). Alex. mentions only (a) in the present passage, deferring
discussion of (b) until 41,13-15, which suggests that (a) is what defines exclusive
perceptibles, while (b) may simply be a distinctive property they all possess. Alex.s
wording at 40,26 includes an important qualfication, which is absent from Arist.s
formulation, but is nonetheless required. Arist.s considered position is not that
exclusive perceptibles can be perceived by only one sense tout court, as his words
at 418a11-12 might suggest, since each of these perceptibles can also be perceived
by the other senses extrinsically (kata sumbebkos, DA 3.1, 425a30-b3; see Alex.
Quaest. 3.8 passim). What distinguishes them is that they are intrinsically percep-
tible (kath hauta) to only one sense.
The examples Alex. gives here are the customary ones, with the exception of
touch. Arist. holds that touch is sensitive to a number of tangible contrarieties, and
the list differs on different occasions: see Arist. DA 2.11, 422b25-7; GC 2.2,
329b18-20. One might wonder whether the list of exclusive perceptibles is exhaus-
tive for the other senses as well: Alex., at any rate, is willing to consider the
possibility at 55,26-56,3 (commenting on Arist. DA 2.11, 422b27-33).
Arist. says that exclusive perceptibles are perceptible in the primary or funda-
mental sense (kuris) and that the essence of each sense is determined by its
146 Notes to page 66
relation to them: DA 2.6, 418a24-5. This explains why the discussion of the
different senses in DA 2.7-11 and Sens. 2-5 is so focused on these qualities, since
they serve to differentiate the activities of the various senses and so help to define
them, in line with his general methodology (DA 2.4, 415a16-23): see Alex. in Sens.
41,15-21. The tight connection between exclusive perceptibles and the individual
senses is also meant to explain the authority that each sense has regarding the
presence of such qualities in the perceivers environment: Metaph. 4.5, 1010b14-
17.
362. [41,1-3] Alex.s division of things which are intrinsically perceptible is
dichotomous: they are either intrinsically perceptible to exactly one sense (idia) or
to more than one sense (koina). This cleans up Arist.s presentation, which usually
characterises the common perceptibles more narrowly, as things which are intrin-
sically perceptible to all the senses (DA 2.6, 418a10-11, a18-19). This leaves open
a gap: anything that is intrinsically perceptible to more than one sense, but not to
all, would be neither an exclusive nor a common perceptible by this charac-
terisation. And it seems likely that several of the perceptibles Arist. calls common
would in fact belong to this middle group: it is hard to see how size or shape can be
perceived by hearing, smell, or taste, as several commentators have pointed out
(Simpl. in DA 126,27-37; Hicks 1907, 362, ad 418a19; cf. Alex. in Sens. 85,25-86,1;
Them. in DA 57,36-58,5); and Alex. even denies that touch can be aware of number
or distance (in Sens. 85,16, although this is somewhat surprising, since Arist.
speaks about the relative authority of sight and touch regarding the number of
objects perceived at Insomn. 2, 460b20-3). Arist. himself seems to recognise the
issue elsewhere, though: at Sens. 4, 442b4-7, he says that perceptibles such as size,
shape, roughness, and sharpness are common to all the senses (b4) and then goes
on to correct this, saying common to the senses, if not to all of them, then at least
to sight and touch (b6-7). Alex. adopts this more permissive construal as Arist.s
considered view. Such a distinction is already implicit in Aet. 4.8.6, Dox. Gr.
395.12-15. Galen goes even further, denying that size and shape can be perceived
by any sense other than sight, except sometimes touch extrinsically, and that
none of the others can perceive position and distance (PHP 7.5.35-8, 460.11-16; cf.
7.6.24, 466.26-7 De Lacy).
Later Alex. will deny that it is the five senses which actually have cognition
(krinei, krisis) of common perceptibles this belongs instead to the common sense
alone (65,10-21). It is significant, then, that he speaks of the five senses here as
transmitters (diakonoi, 41,3), which is how they are also described in the pre-
Galenic Def. Med. 116, atttributed to Galen (19.379.7 Khn). As Alex. immediately
goes on to say, the perceptual organs relay (poietai diadosin) the common percep-
tibles to the sense for them (epi tn aisthsin autn, 41,4-5), that is, to the common
sense. The Greek diakonos and its cognates are often used elsewhere simply to
indicate a servant or the functions a servant performs taken quite broadly, and this
can be the case even when Alex. is talking about the senses (see esp. 39,20 above,
also Mant. 104,8). But diakonos can also refer to a specific kind of servant, a
messenger (LSJ cite Aesch. PV 942 and Soph. Phil. 497). This sense predominates
in Alex.s works, esp. where the transmission of perceptible forms is concerned,
often through an external medium: 41,3.20; 42,9; 52,20; 59,14; 62,5.12; 94,28; 95,1;
100,16; Mant. 145,27; in Sens. 25,21; Quaest. 1.2, 5,14-16; 6,9.13.20; 7,12; also cf.
a report from his lost commentary ap. Philop. in DA 519,29 (= fr. IIIa Moraux). This
sense is confirmed by Alex.s repeated description of the medium as something that
can receive and transmit (dektikon te kai diakonon) perceptible forms: Mant.
145,26-7; Quaest. 1.2, 5,14-15; 6,9; 7,12. The last citation is the most telling. Alex.
Notes to page 66 147
uses elegant variation in a parallel between sight and smell: just as an odourless
medium can receive and can communicate (mnutikon, 7,10) odours, so the
transparent medium can receive and transmit (diakonon, 7,12; cf. 5,18-19) col-
ours; we find the two terms used together again at DA 59,14 below. Thus, by
undergoing a kind of alteration due to the object, the transparent medium reports
the form from the object that can be seen to sight (to eidos to apo tou horatou ti
opsei diangellontos, Mant. 141,35-6; cf. 142,31-143,1).
The Stoics refer to the senses as messengers that report back to the governing
faculty, which acts like a monarch: Calc. 220, 233.5-6 Waszink = SVF 2.879 = LS
53G 7. But other philosophers make use of the comparison as well: Epicurus
regards the senses as messengers of the truth (veri nuntios, Cic. de Nat. Deor.
1.70); Ptolemy has the senses relaying their reports to the understanding, but like
a messenger not interfering further (outhen epipolupragmonei kathaper angelos
tis, de Judic. 8, 13.5-8 Lammert); and Galen compares breath or pneuma to a
messenger (hsper angelou tinos), although it goes in the reverse direction, from
the central organ to the peripheral senses (PHP 7.4.1, 448.7-9 De Lacy).
363. [41,3-5] As mentioned in n. 362 on 41,1-3 above, Alex. later identifies the
sense to which the common perceptibles are relayed as the common sense, which
alone has cognition of them (65,10-21). What is striking here is Alex.s insistence
that common perceptibles accompany exclusive perceptibles at each stage of the
process, not only in the distal object, where for example colour and shape always
occur with each other (Arist. DA 3.1, 425b8-9), but also in the transmission from
the peripheral organs to the central sense. Our awareness of common perceptibles
is not the result of reorganising perceptual data at the final stage (as seems to be
the case in Aet. 4.8.6, Dox. Gr. 395.1-12). Rather, common perceptibles are inter-
woven (sumpeplegmena) from the start (DA 83,17-22), affecting the peripheral
sense along with exclusive perceptibles like colour (65,11-16; cf. in Sens. 11,14-19).
Alex. offers detailed explanations of how we discern shape, size, distance,
number, movement, rest, and number in vision, by means of the angle of the visual
cone extending to the object, variations in colour and other visible qualities, at
Mant. 146,1-147,16 (quite possibly in response to Galens complaints, PHP 7.7.4-8,
470.17-472.2 De Lacy).
364. [41,5] in Sens. 11,12-17; 84,11-16. Alex.s list of common perceptibles here
is exactly the same as Arist.s at DA 2.6, 418a17-18; Sens. 1, 437a8-9; cf. Insomn.
1, 458b4-6. Arist. adds one (hen) to the list at DA 3.1, 425a14-16 and roughness,
smoothness, sharpness, and bluntness at Sens. 4, 442b4-7. Alex. argues that these
last four qualities cannot be subsumed under shape, because they do not fit its
definition (in Sens. 84,13-16), but are common to very different perceptibles (in
Sens. 84,16-18; DA 56,1-3). Alex. himself adds distance to the list at DA 65,13-14;
in Sens. 84,11-13. One might have thought that distance could have been sub-
sumed under extension (megethos). But Alex.s explanation below of how we are
aware of the distance of an object through hearing, sight, and smell suggests
otherwise: DA 50,18-51,6 (with accompanying notes).
These lists are often compared to Plato Tht. 185A-D, where Socrates focuses on
our awareness of certain common features of perceptible objects, such as their
being and not-being, or their being like and unlike, same and different, one and
many, and somewhat further on admirable and shameful, and good and bad
(186A). But Socrates point in context is precisely that these features are not
perceived by the senses, but rather by the soul, which alone compares and calcu-
lates (185D-186C; cf. 185B-C); Galen similarly claims that size, shape, motion, and
number cannot be perceived by perception alone, but by reasoning and memory
148 Notes to page 66
(PHP 7.6.24, 466.25-8 De Lacy). Arist. and Alex.s point is thus quite different.
While they agree that there must be some single thing which has access to the
reports of the various senses, they insist that we are nonetheless aware of these
features by perception (hence the importance of a central or common sense).
Moreover, they include quite mundane physical characteristics like size, shape,
and change, while omitting the more transcendental characteristics that concern
Plato, like being, sameness, and goodness. In fact, the only intersection between
their respective lists concerns oneness and number.
365. [41,6-8] Alex. returns to the primary division here, between those things
which are intrinsically perceptible (kath hauta) and those which are extrinsically
(kata sumbebkos) perceptible (or, as commonly rendered, accidentally, concomi-
tantly, or incidentally perceptible). These are characteristics which are extrinsic
to something perceptible more literally, they happen to belong to it (sumbe-
bkenai), perhaps even necessarily, as opposed to being part of its essence and
just because of this can be counted as perceptible themselves. So A is extrinsically
perceptible just in case (i) A is an extrinsic characteristic of B and (ii) B is
perceptible (intrinsically, one assumes). Arist. only offers examples of extrinsic
perceptibles, rather than a fully general characterisation. But each time he does,
he offers an explanation along the lines just sketched: with seeing Diares son (DA
2.6, 418a20-3); seeing Cleons son (3.1, 425a24-7); and even seeing a universal,
colour (Metaph. 13.10, 1087a19-20 cf. An. Post. 2.19, 100a17). Alex.s explanation
conforms closely to Arist.s first example, even down to the phrasing, especially if
Bruns clever suggestion is right: that the Greek for foam (aphron) is a corruption
of the name Euphronius (Euphron), which the Hebrew translation appears to have
taken to be a form of the word thoughtful (euphrn); Euphronius is perceptible
because he happens to be white or pale (similar to Diares and Cleons sons), and this
colour is perceptible. The suggestion that a substance like Euphronius can be an
extrinsic characteristic of a perceptible like white might be surprising, as it seems like
a reversal of the ontological priority developed in the Categories. But Arist. allows such
locutions, so long as we do not take them to imply that the perceptible in these cases
is the underlying subject: see An. Pr. 1.27, 43a33-6; An. Post. 1.22, 83a1-14.
Nothing prevents something from being intrinsically perceptible and extrinsi-
cally perceptible, so long as it is so to different senses: sweetness is intrinsically
perceptible to taste, but extrinsically perceptible to sight. In fact, every exclusive
perceptible will be extrinsically perceptible to the other senses: DA 3.1, 425a30-1.
But according to Alex. it need not be the case that every extrinsic perceptible be
perceptible to each sense. In Quaest. 3.8 (passim), he argues that an extrinsic
perceptible can be perceived by every sense only if the object has some qualities
that are perceptible exclusively to each sense. He thus assumes we cannot perceive
something extrinsically with a particular sense unless it has some perceptible
quality exclusive to that same sense. I may be able to perceive a toxic gas
extrinsically by smell and, if it has a colour, by sight too; but if it lacks a colour, I
cannot perceive it by sight, either intrinsically or extrinsically according to Alex.
366. [41,8-10] Taken at face value, Alex.s denial that extrinsic perceptibles are
perceptibles au fond or at bottom (tn arkhn) is quite strong. It suggests a view
we find expressed in other authors from time to time, that extrinsic perceptibles
are only indirectly perceived (e.g. Hicks 1907, 360-1, cf. 363) or even that they are
not really or strictly perceived at all (Ross 1961, 271). This reading is supported by
Alex.s hedged expressions in the previous line: things called perceptible as well
and if someone were to say that foam is perceptible (41,7).
Arist. never quite says such things himself, though. He does say at DA 2.6,
Notes to page 66 149
418a23-4 that no sense is affected by this sort of perceptible in so far as it is such.
This implies that extrinsic perceptibles are both ontologically extrinsic and caus-
ally extrinsic: they are both extrinsic characteristics of some perceptible quality
and extrinsic to the quality that is causally efficacious. But it does not follow that
we do not perceive things as having these characteristics, or still less that they are
not genuinely perceived at all (although that seems to be Alex.s inference: hoti,
41,9). When Arist. says at 418a24-5 that exclusive perceptibles (idia) are kuris
perceptible, he is not claiming that the other types are not strictly perceptible (as
e.g. Everson 1997, 18 renders it), but only that they are not what is fundamentally
perceptible, as the exclusives are. In general, to say that something is extrinsically
F is not for Arist. a way of denying that it is really F or is F only homonymously.
On the contrary, it is a way of affirming that something is genuinely F, while
noting that it is not so in virtue of itself (kath hauto) it is F in virtue of some
extrinsic characteristic (kata sumbebkos) and so not due to its own nature. It is
thus open to Arist. to say not only that I genuinely perceive a human being, Callias,
but that my perception is of a human being (An. Post. 2.19, 100a16-b1), that I see
him as a human being. This is a genuine kind of perception, as much as seeing red
or feeling somethings roundness is.
Alex. may take a more moderate position elsewhere. In Quaest. 3.8 he insists
that it is not the case that something which is only perceptible extrinsically by all
of the senses is not perceptible at all (oud hols aisthton, 94,4-5); if it were not
perceptible by any sense, he continues, it would be imperceptible (94,5-6), against
our hypothesis. In this text, then, something that is merely perceptible extrinsi-
cally is nonetheless perceptible. It is unclear, though, whether this represents a
change of mind or only a change of emphasis. Moraux, for example, thinks the
positions do not deviate much from each other (2001, 340).
367. [41,10-13] 70,10-12 (cf. 20-3). Alex. is reordering and reworking Arist.s
material here into a more systematic presentation. He has deferred speaking
about error until this point, even though it is something Arist. raises towards the
beginning of DA 2.6, when he speaks about the infallibility of perceiving exclusive
perceptibles (see n. 361 on 40,21-41,5 above). But Alex. also emphasises the
fallibility of perception here, unlike Arist., with both common and exclusive
perceptibles, and thus things that can be perceived intrinsically. Arist. does not
even raise the issue of error with common perceptibles or place a limit on the
infallibility of exclusive perceptibles in DA 2.6, but only later in DA 3.3 (cf. also
Sens. 4, 442b8-10; Alex. in Sens. 84,9-10.22-3). Alex. is thus consolidating Arist.s
various claims and synthesising them to offer what he regards as the best Aristo-
telian position, or perhaps even Arist.s considered position on error in perception.
Most of DA 3.3, 428b17-25 is concerned with how truth and falsehood in
representations corresponds with different types of perception (which Alex. dis-
cusses below at 70,5-71,5). But at 428b22-5 Arist. claims that the greatest errors
occur over common perceptibles (and therefore more, he seems to be claiming, than
extrinsic perceptibles, which one might have expected were the most subject to
error). But Arist. does not explain why. Alex. offers an explanation at in Metaph.
313,20-32: each sense only has authority over its exclusive objects, about which it
reports truly, but not over extraneous objects (allotria); and common perceptibles
are in a way extraneous to each sense (313,30). But the latter claim surely runs
counter to Arist.s division in DA 2.6, which makes common perceptibles intrinsi-
cally perceptible to all of the senses.
In the present passage, Alex. does not attempt to explain the difference in error,
but only offers a remark about perceptual error in general: it occurs when the
150 Notes to pages 66-67
modification of the sense is not like (m } hs ekhei, m homoion) the object
perceived. It is unclear, though, what the phrase modification (pathos) of the
sense refers to. (1) If Alex. has in mind the modification of the perceptual organ
that underlies perception, then he takes veridical perception to require a signifi-
cant degree of likeness between the qualities of the organ and the qualities of the
object (though he need not assume exact similarity, as so-called literalist interpre-
tations demand; similarity in certain respects, such as the underlying proportions
of the qualities perceived, might conceivably suffice as well). Alternatively, (2)
Alex. might not have in mind a modification of the organ, but rather a state of the
sense itself, that is, the power of perception. But this state could be like the object
only in so far as its content corresponds to it: the only way for a power to be red is
for it to be in an intentional state about something red. If so, then Alex.s explana-
tion above expresses a rudimentary correspondence principle: a perception is in
error just in case it takes objects in the world to be a certain way when in fact they
are not. (Compare Alex.s characterisation of a false statement at in Metaph.
433,10-11.14-15.)
Arist. himself gives a closer characterisation of how this correspondence goes
awry in Insomn. 1: when we mis-see or mis-hear, we see or hear something real,
but not what we take it to be (ou mentoi touto ho oietai, 458b31-3). This comes
about, he notes in the next chapter (while discussing a mistake about the number
of objects perceived and hence a common perceptible), whenever a sense is affected
in the same way as it would be due to perceptible objects in such cases, what
appears to us is the same as when we are perceiving veridically (Insomn. 2,
460b23-5). For more detailed discussion of the second passage, see my 1998, 272-9,
esp. 276-7. For Arist.s characterisations of truth as correspondence, see Metaph.
5.29, 1024b21-4; 9.10, 1051b4-5, b33-5; cf. Int. 9, 19a33. A very fine-grained
discussion of Arist.s commitment to a correspondence theory of truth can be found
in Crivelli 2004, 129-38.
For the use of hupokeimenon to indicate the object of perception, as underlying
or corresponding to it, see 56,5-9; 60,14-16; 61,26; 69,6; 81,5; and in Arist., see esp.
DA 3.2, 426b8-11; also 2.11, 422b32-4; 3.2, 425b14; GC 2.2, 329b13; GA 5.1,
781a15-17; Metaph. 10.1, 1053b1-3. This use is just a special case of the terms
more general use for the object of a cognitive attitude or subject-matter of a field
of study: see Bonitz, Ind. Arist. 798b60-799a27.
368. [41,13-42,3] Alex.s qualifications here are significant. In general, Arist.s
position is that a sense is never wrong about those perceptibles exclusive to it, and
in fact when he first introduces exclusive perceptibles in DA 2.6, he says it is
impossible for a sense to be mistaken about them (peri ho m endekhetai
apatthnai, 418a12), but only about what has the quality in question or where it
is found: 2.6, 418a14-16; 3.3, 428b21-2; 3.6, 430b29-30; Metaph. 4.5, 1010b19-21.
On the truth of such perception more generally, see 3.3, 427b12 (cf. 428a11); Sens.
4, 442b8-10; Metaph. 4.5, 1010b2-3. In one passage, however, Arist. qualifies this
claim: the perception of exclusive perceptibles is true or has the least possible
falsehood (alths estin hoti oligiston ekhousa to pseudos, DA 3.3, 428b18-19). It
is not clear whether Arist. only has in mind the related errors he mentions earlier,
about the bearer of the quality or its location; or whether he is instead allowing
that there can be errors about the qualities themselves, and if so, in which
circumstances. He might have after-images in mind (Insomn. 2, 459b7-13); but it
is not clear if these errors occur without representations (phantasiai) being in-
volved. For discussion, see my 1998, 272 esp. n. 56.
Alex. sometimes repeats Arist.s unqualified claim of infallibility about exclu-
Notes to page 67 151
sive perceptibles: in Sens. 84,9-10.22-3; in Metaph. 311,31-2 (cf. 28-9); 313,20-1.
But in the present passage, Alex. instead makes a comparative claim that percep-
tion of these perceptibles is the most true (altheuousi malista, 41,14) and later on
that they are true only for the most part (hs epi to pleiston, 66,15). He argues at
in Metaph. 313,32-314,1 that differing conditions matter to the accuracy of percep-
tion: sight is correct about colour only when it is located at a commensurate
distance and not far away (apo summetrou diastmatos kai m porr, 313,34-5; cf.
in Sens. 50,22-4), and similarly for hearing and smell. Although Alex. mentions the
iridescence of the pigeons neck in different lighting and angles, he does not
mention error in this context (in Sens. 51,1-3; cf. Sext. Emp. Pyrr. Hyp. 1.120). But
he does discuss the mistakes weavers make in the choice of threads due to
illumination and surrounding colours (in Meteor. 158,27-159,3). In general, Alex.
concludes, the senses are not capable of cognising exclusive perceptibles accurately
in just any condition or situated in any location whatsoever (hopsoun ekhontn
hopoudpote keimenn, in Metaph. 313,36; cf. in Sens. 127,20-1). Even Arist.
acknowledges some of these facts himself at various points: that colours appear
differently at different distances (Metaph. 4.5, 1010b5-6; cf. Sens. 3, 439b5-6); that
dyed textiles appear to be of different colours when juxtaposed differently and in
different illumination (Meteor. 3.4, 375a22-8); that wine tastes different depending
on ones own condition (1010b21-3; cf. DA 2.10, 422b7-10); that the same things do
not seem sweet to a healthy person and a sick one, or to a strong person and a weak
one (EN 10.5, 1176a13-16); and in one passage, he even argues that there is not
any variation in how things taste, unless a persons organs are damaged or
maimed (diephtharmenn kai lelbmenn, Metaph. 11.6, 1062b36-1063a3). But
nowhere does he try to account for these facts in a systematic way.
This is precisely what Alex. sets out to do. He states the general position at in
Metaph. 312, 24-31: things appear as they actually are most of all to the person
who is in a natural state (ti kata phusin ekhonti, 312,29): to the perceiver who is
at a commensurate distance, in good health and vigour, and awake. Natural
conditions are not so much those which are normal or standard statistically, but
those which are suited to the optimal, natural functioning of the perceptual
apparatus. In the present passage, Alex. identifies five such conditions: (i) the
health and condition of the organ, (ii) the direction in which the object lies, (iii) a
suitably commensurate distance, (iv) the condition of the medium, and (v) the
absence of interference. It is not clear where these details first originated. An
almost identical list, though adumbrated with only five keywords, is sometimes
attributed to the Stoics based on attributions in neighbouring contexts (Sext. Emp.
Adv. Math. 7.424 = LS 40L). But it involves a modification of the old Stoas criterion
of truth: in addition to having a secure representation (katalptik phantasia), it
must also not be undermined by any obstacle (enstma), such as the failure of any
of these conditions would present. The later Stoics credited with this strategy
(Sext. Emp. Adv. Math. 7.253-8) are thus likely to be making a concession to
objections raised by Academic sceptics: at any rate, we find the point elaborated
very fully by Carneades, with a list of nine conditions, at Sext. Emp. Adv. Math.
7.183 (= LS 69E 4 in part) and more briefly in a presentation of Antiochus of
Ascalons position at Cic. Luc. 19. We have no way of knowing whether Carneades
is the first to explore these ideas or whether he is adapting and developing the
views of others.
369. [42,4] Although there is no subtitle in our MSS On Sight, the discussion
of sight clearly starts here, beginning with the objects of sight (42,4-43,4). Alex.
then focuses primarily on the medium of sight (43,4-45,5), with brief digressions
152 Notes to page 67
along the way on how sight comes about (43,18-44,1), the eye (44,1-9), and the
power of sight (44,9-13). At 45,5 he returns to the objects of sight and the nature
of colour in particular (44,5-46,19).
370. [42,4-5] Arist. DA 2.7, 418a26-8. In the passage above, Alex. is paraphras-
ing the opening of DA 2.7, which identifies both colour and something which can
be described, but lacks a name what we now call phosphorescent objects as
visible. Arist. discusses them further at 419a1-7 as fiery and shining objects that
can bring about perception in the dark (en ti skoti poiei aisthsin), though what
they exhibit is not their own colour (to oikeion khrma, a2), that is, the colour they
would show in the light. He identifies fungi, horn (Ross reads meat instead,
following Themistius), and the heads, scales, and eyes of certain fishes as having
this ability (419a5), as well as the ink of the cuttle-fish (Sens. 2, 437b4-7; cf. a31-2).
Alex. adds the eyes of lions and rabbits and the glow-worm (pugolampis) at in Sens.
18,12-13, fish bones and glow-worms at Mant. 128,5-7, stones and glow-worms
(lampuris) at 138,15-16.22-3. On glow-worms and fireflies, see Arist. HA 5.19,
551b23-6 (cf. 4.1, 523b21); PA 1.3, 642b33-4; for the Greek terms and further
ancient references, see Davies and Kathirithamby 1986, 158-9.
Alex. points out at in Sens. 17,5-7 that these objects do not in fact illuminate the
medium and so do not make other objects visible, but merely shine themselves
(though contrast Mant. 144,19-23 which says certain animals and stones illumi-
nate the transparent). We know that an earlier 2nd c. CE Peripatetic, Sosigenes,
whom Alex. refers to as our teacher (ho didaskalos hmn, in Meteor. 143,13-14;
cf. Them. in DA 61,23; Ammon. in An. Pr. 39,24-5), discussed just these points in
the third book of his On Sight (Them. in DA 61,22-32).
371. [42,5-7] 43,16-17. What Alex. refers to as the definition of colour (horis-
mos, 42,6) is not the definition of colour found in Arist. Sens. 3 (439b11-12, cf.
a28-9), which Alex. discusses more fully below at 45,5-46,19, but rather a verbatim
quotation of the beginning of DA 2.7, which states the very nature of colour (tout
estin autou h phusis, 418a31-b2) and is repeated later in the chapter as the
essence of colour, what it is to be a colour (to khrmati einai, 419a9-10). Since
phosphorescent objects are luminescent in darkness, when the medium is not
illuminated and actively transparent, their apparent hue will not strictly be
colours by this definition, or at any rate the objects own colour (oikeion khrma,
DA 2.7, 419a1-6).
Although this may appear to be a merely verbal difference, the claim that colour
can only be seen in the light has an important consequence. For what can be seen
(to horaton) cannot then be defined as colour, since other things can be seen as well,
but only as the correlative of what can see (to horatikon). So when Arist. says that
colour is an exclusive perceptible of sight, he should not be understood to be
making a definitional claim, but a causal one: colour is what is responsible for
certain objects being visible (to aition tou einai horaton, DA 2.7, 418a31)
That which is actively (kat energeian) transparent is that which has fully
realised its nature as something transparent and so allows colours to be seen
through it. Transparent materials achieve this state when they are illuminated
and have light: n. 373 on 42,8-11 below. On the meaning and etymology of
transparent itself, see n. 378 on 43,5-8 below.
372. [42,7-8] Alex. continues with his summary of the opening of DA 2.7. At
418b4-9 (cf. Sens. 2, 438a12-15), Arist. mentions air, water, and many solids as
examples of what is transparent, which he characterises generally as that which
cannot be seen intrinsically or on its own (ou kath hauto horaton), but only through
an extraneous colour (allotrion khrma). Alex. takes this to show that it lacks a
Notes to page 67 153
colour of its own (oikeion khrma): see also DA 44,3-4; 46,7-9; in Sens. 43,20-2;
45,17-18; 50,4-7; cf. DA 59,14-15; Mant. 106,30-107,1; in Meteor. 200,25-6.
Alex. offers somewhat different explanations for why transparent bodies lack a
proper colour, depending on whether they are (in effect) liquids and gases with only
indefinite boundaries or solids with definite boundaries. Air and water, which
serve as the medium of sight, have only indefinite boundaries that are determined
by whichever bodies happen to be in contact with it. Alex. argues that it is precisely
because air and water lack a definite boundary that they lack a colour of their own,
since colour is the boundary of a definite body (in Sens. 48,22-49,4; Quaest. 1.2,
5,11-15; cf. DA 46,7-9); because of their looseness (dia manotta), they are unable
to possess or retain any colour (Quaest. 1.2, 6,7-8.24-5). Such colour as they have
comes, just like their boundaries, from other bodies (in Sens. 45,18-20; cf. 50,4-5).
But this account obviously will not apply to the solids mentioned in the passage
above (tn steren, 42,8), which do have definite boundaries: elsewhere Alex.
mentions glass, horn, certain transparent stones (mica, stelenite), and tortoise
shell as examples (Mant. 133,18-19; 149,26-9; in Sens. 26,27-27,1; 29,5-7; 44,1;
46,19; several of these are already mentioned in Arius Didymus fr. 17, Dox. Gr.
456,15). What explains transparency in their case, as well as for bodies with
indeterminate boundaries, is the absence of earth (in Sens. 45,23-6), which ob-
structs and prevents bodies from admitting light (46,12-16; cf. DA 45,9-10).
Transparent solids are solid because they have more earth in them than bodies
with indeterminate boundaries (in Sens. 45,25); but they are still transparent
because they nevertheless have more water than earth (46,6-9.19-21). Solids that
contain larger amounts of earth have the ability to retain colours previously
acquired, and this accounts for their having some colour of their own (47,8-15;
contrast 46,21-47,8). So Alex. should not endorse the claim, without qualification,
that colour is the surface of a determinate body: by his own admission, some of
these bodies are transparent.
Later, Alex. will state that all bodies have transparent material in them on
Arist.s theory, some more, some less (see n. 399 on 44,20-1). But those with
sufficient amounts of earth mixed in will be opaque (see n. 407 on 45,7-10). The
only ones we can see through (diopta), as the above references make clear, are
those which have much more water than earth in them.
373. [42,8-11] Colours affect the eye by affecting the transparent medium in
between them, in such a way as to cause us to become aware of them: Arist. DA
2.7, 419a13-15; cf. Sens. 2, 438b2-5; 3, 440a18-19. On this transmission (diak-
oneisthai), see n. 362 on 41,1-3 above. Colours can affect the medium, though, only
when it is actively (kat energeian) transparent, which occurs only when it is
illuminated through the presence of light: see esp. Arist. DA 2.7, 419a11; Alex. DA
43,11-16 (cf. 4-5); Mant. 141,36-142,16; 143,35-144,6; 145,2-7; in Sens. 35,6-9;
59,4-12; 135,7-8; cf. also Mant. 146,21-5; in Sens. 105,9.
374. [42,11-19] Mant. 145,25-30; in Sens. 50,9-11; ap. Averroem Comm. Magn.
in DA, 2.67.39-45, 231-2 Crawford. Alex. takes the phenomenon of tingeing to show
that the transparent medium is changed in a way (kineitai ps, 42,12) by colour
in general, even though in most cases the effect on the medium will not itself be
visible to us. But it is precisely because the medium is affected (ti paskhein) that
it can relay colours to other things (diadidontos, 42,18-19; cf. sunanapheron,
42,14). On this qualification, see further n. 375 on 42,19-22. The example is
potentially significant, because tingeing might seem to involve the mediums
literally taking on a colour, at least in the sense that it visibly comes to be of that
colour, even if this is not a change in its own proper (oikeion) colour. But Alex. will
154 Notes to page 67
later deny that in ordinary cases the medium undergoes literal alteration when it
relays colour or that the eye does either (62,3-7).
Both the phenomenon of tingeing and the specific examples Alex. uses to
illustrate it can be found in other authors, most notably in Galen, who mentions
the effect of a trees foliage colouring the air for those beneath it and of a coloured
wall on the air and other bodies, especially if it is blue or yellow or any other
brilliant colour (PHP 7.7.1-2, 470.5-11 De Lacy). The similarity naturally suggests
that Alex. is depending in some way on Galen or possibly a common source (as
Moraux claims, though without argument or positive evidence, at 1984, 758).There
are some earlier parallels extant, although none as close. Lucretius, for example,
describes the effect of awnings in early Roman theatre tents in DRN 4.72-89 as
evidence of effluences, and the Peripatetic Strato is reported to have discussed the
phenomenon before that, although we do not have further details (Aet. 4.13.7, Dox.
Gr. 403.26-8). In Arist., the closest antecedent is Sens. 3, 439b1-2, which suggests
that the occurrence of glare (aug) in air and water shows that they become
temporarily coloured (khromatizomena). Differences in illumination (en augi
toiaidi toiaidi) further have an effect on how the colour of weavers thread
appears (Meteor. 3.4, 375a22-8).
375. [42,19-22] The comparison is, in the first instance, between the production
of light in the medium and the mediums being tinged by the colour of surrounding
objects: see Mant. 142,10-16, where the actively transparent medium is said to
take on colour as a kind of second light } and a kind of second activity (hs
deuteron ti phs } lambanei deuteran tina tautn energeian dekhomenon). But in
both passages the mediums becoming tinged seems to be taken as evidence (gar,
42,11) for how colour affects the medium in ordinary cases, even when the medium
does not itself become tinged, in bringing about sight of coloured objects. If so, then
the reception of colour by light or the illuminated medium is an essential part of
the transmission of colour through the medium for seeing in general. Alex. seems
to have the transmission of colours specifically in mind when he draws the parallel
again at in Sens. 59,7-15; 135,17-20 (expanding on Arist. Sens. 6, 447a10-11);
Mant. 142,2-4; 144,31-145,7. His most explicit statement comes below at 43,18-
44,1.
In passing, Alex. rejects two alternative accounts of these phenomena. (1) The
first holds that light and sight are brought about by effluences (aporrhoai), that
is, by streams of matter flowing from objects. This basic idea is found in various
authors, either alone or in combination with other mechanisms: it is attributed, for
example, to Democritus (Theophrastus Sens. 50, Dox. Gr. 513.20 = DK 68 A 135;
cf. Arist. Div. Somn. 2, 464a5-6, a9-11 and Alex. in Sens. 124,6-7) and invoked by
Plato to account for colour in the Timaeus (67C; cf. 58C). It is standardly associ-
ated, though, with Empedocles, who might have been the first to coin the term: it
occurs, at any rate, in his claim that effluences are emitted from all things (Plut.
Quaest. Nat. 19, 916D = DK 31 B 89). For Empedocles claim that light specifically
is an effluence, see Philop. in DA 344,34-345,2 = DK 31 A 57; cf. Arist. DA 2.7,
418b13-16, b20-3; Sens. 6, 446a25-8. For his claim that colour vision is due at least
in part to effluences, see Plato Meno 76C-D = DK 31 A 92; cf. Arist. Sens. 2,
438a4-5; 3, 440a15-16; GC 1.8, 324b26-35; Theophr. Sens. 7, Dox. Gr. 500.28-9; Aet.
1.15.3, Dox. Gr. 313.8-20 = DK 31 A 92); Alex. in Sens. 24,7-9.
Arist. rejects any such account. He denies that light is an effluence and travels
through the medium at DA 2.7, 418b14-15 and b20-6, arguing against the latter
claim at length in Sens. 6 (446a20-b2, b9-13, 446b27-447a3, a8-10). He likewise
denies that sight comes about by effluences at DA 2.10, 422a14-15; Sens. 3,
Notes to page 67 155
440a15-20. It is not surprising, then, to find Alex. defending a similar position: on
light, see Mant. 128,14-15; 138,14-23; on sight, see esp. in Sens. 31,20-9; 56,10-
58,22; 61,17-21; Mant. 130,10-12; 136,29-138,2; 141,33; cf. in Sens. 83,3-7; Mant.
134,28-136,28 (esp. 134,27). In these passages, Alex. is sometimes attacking
specific, hybrid versions of an effluence theory, for example, one that involves
simulacra or eidla (in Sens. 56,10-58,22) or one which involves some extromission
from the eye in addition to the effluence from the object (Mant. 136,29-138,2).
(2) More interesting is Alex.s dismissal of the second alternative, that the
transparent takes on light or light takes on colour in the way that matter takes on
a form. One possibility, suggested by Accattino (1992, 52 n. 45) is that (a) Alex. is
rejecting a rival view according to which light or colour, respectively, is a body, a
view he explicitly rejects in the case of light: see 43,11 just below and the
corresponding note for further references. These views are generally associated
with the Stoics, but the only testimonia for this precise claim in von Arnim (SVF
2.432) are among the passages just cited and none identifies the Stoics by name.
According to other sources, the Stoics maintain that light is not an effluence, but
a change of the entire medium at once due to contact with and stimulation by the
sun: Plut. Fac. in Orb. Lun. 18, 930F (= SVF 2.433b); cf. 5, 922E (= SVF 2.570);
Galen Hipp. Epidem. 4.11, 214.24-215.3 Wenkebach-Pfaff (= SVF 2.433a). These
claims could be reconciled if the Stoics think of light as a quality that pervades the
illuminated medium, and so, given their materialist view of qualities, is also a body
(Mant. 124,9-10). The qualified object that results, day, would also be a body, as
they notoriously claim: see Plutarch Comm. Not. 45, 1084CD (= SVF 2.665).
Another possibility is (b) that Alex. is concerned here not with what kind of
object light or colour is, but with the type of change involved. By rejecting efflu-
ences in the previous clause, he denies that light and colour are due to something
travelling through space. To add that the medium does not receive them in the
way matter does (oute hs huls, 42,21) is to reject a different type of change,
namely, alteration (alloisis). In this respect, these changes would be like perceiv-
ing, which on Alex.s view is not itself an alteration, even though it involves
alterations (see n. 344 on 38,21-39,2 above). This seems to be the clear sense of the
phrase at 62,3-7, where Alex. denies that the eye receives colours as matter and
draws an explicit parallel with the illumination of the transparent medium; see
also his explanation of why perception does not involve receiving a form as matter
at 83,23-84,6. Alex. explicitly denies that illumination is an alteration elsewhere
(in Sens. 133,13-14.19; 134,11) and insists that the medium does not undergo or
receive a modification (ou paskhei, ou pathtiks kineisthai/andekhetai/ekhei)
when it receives light or colour (DA 62,11-13; Mant. 144,34-145,2 (cf. 144,31-2);
145,6-7; in Sens. 42,26-7; 47,3-7; 50,16-18 (cf. 20-2); 52,2; Quaest. 1.2, 6,8-10.24-5;
cf. Mant. 147,16-17) this is one respect in which the transmission of colours
differs from the transmission of odours or flavours (in Sens. 135,9-13). Alex. may
have this point in mind earlier in the present passage when he twice hedges the
claim that the medium undergoes change with the qualification in a way (ki-
noumena ps, DA 42,8; kineitai ps, 42,12; cf. tis kinsis below at 43,1-2). Similar
qualifications can be found elsewhere: when the transparent medium undergoes
such changes, it is said to alter in a way (alloioumenou ps, in Sens. 141,34-5) or
to vary and be modified in a way (trepetai ps kai paskhei, Mant. 142,2), a point
Alex. immediately explains by distinguishing this variation (trop) from what
happens to things that are genuinely altered (Mant. 143,4-10). What the transpar-
ent medium undergoes is only a seeming alteration (ts dokouss alloises,
156 Notes to page 67
143,17-18). (He does not always add such qualifications, though: see n. 383 on
43,11-16 below.)
Alex. characterises this change more positively in two ways. First of all, the
illumination of the medium should be understood as a completion of its nature, the
activity of a transparent material capable of being illuminated (see below, 43,7-8),
and so not an alteration in which its nature and capacities are destroyed. The same
could be said of the mediums becoming tinged by surrounding colours and even of
the effect colours normally have on the medium when transmitted to the eye, if this
change is to be understood on the same model (see the beginning of current note).
But he also understands it as a kind of relational change, which he sharply
contrasts with alteration: see n. 376 on 42,22-43,4 below.
376. [42,22-43,4] 43,8-9; 45,1-4. Arist. maintains that light is not brought about
through the local motion of a body, but rather by the mere presence (parousia) in
the transparent medium of fire or something like it (DA 2.7, 418b13-20; Sens. 3,
439a19-21; cf. 6, 446a20-447a11). Alex. develops this view further in a number of
passages throughout his corpus and extends it here by applying the same claims
to the tingeing of the medium by colour (and so presumably to the action of colour
on the medium in general; see the beginning of n. 375 on 42,19-22 and also n. 385
on 43,18-44,1 below). In saying that these phenomena arise from both sources (ap
amphotern, 43,2), Alex. has in mind the illuminant and the colour respectively,
rather than the illuminant together with the medium and the colour together with
the light, as Schroeder claims (1981, 217-18). Alex. has been treating the two cases
in tandem from 42,19 on and is using both here to refer back directly to the things
that produce these effects (tn tauta empoiountn, 42,22) a few lines before, which
are explicitly identified as the colour (42,23) and the light (43,1). For a full defense
of this reading, see Accattino 1992, 51-9, esp. 57.
Of Alex.s claims about light (and on occasion colour), the following are espe-
cially relevant to the present passage:
(1) The presence of an illuminant in a transparent medium is logically speaking
a sufficient condition for the illuminatation of the medium and is responsible for
it (h aitia autou, Mant. 144,19). But according to Alex. the medium is not altered
or changed in the primary sense of the word when it is illuminated or tinged. Both
phenomena come about solely in virtue of the relation (kata skhesin) of the medium
to the illuminant or colour and so are at best mere Cambridge changes, much like
somethings coming to be to the right of something as a result of the other things
being moved (see n. 344 on 38,21-39,2 above): see esp. Mant. 143,4-21; but also DA
43,9-10; 45,1-3; Mant. 142,3-8; 144,31-145,7; in Sens. 31,11-18; 42,25-43,1; 43,5-7;
52,8-12; 132,2-16; 133,23-6; 134,11-19; 135,12-13; Quaest. 1.2, 6,10-11.25-6; 1.21,
35,12-14. Alex. does speak in one passage of the medium being changed by the
presence of an illuminant (kinoumenon hupo ts tou phtizein pephukotos
parousias, in Sens. 59,8-10). But he immediately goes on to say it is only changed
and disposed in a way (ps, 59,10-12), a qualification he uses elsewhere to indicate
that it is not an alteration or change in the primary sense (see n. 375 on 42,19-22),
much as he speaks here of the sort of change involved in both phenomena (tis h
} kinsis, 43,1-2).
These claims go beyond what Arist. explicitly says at Sens. 6, 446b27-447a11.
In that passage, he only denies that illumination is the motion of a body passing
through space (here kinsis, 446b28, but cf. 446a25-b2 and the contrast between
alloisis and phora at 446b28 ff.). In fact, Arist. appeals to the example of
alterations like freezing as a comparison to show how illumination occurs and even
speaks there of the mediums being modified by light (paskhei, 447a9-10), without
Notes to page 67 157
making the crucial qualifications which Alex. in contrast insists on at in Sens.
133,12-134,19. But Arist.s position elsewhere is that something does not undergo
change at all, except extrinsically (kata sumbebkos), merely because its relations
to other things shift: Phys. 5.2, 225b11-13 = Metaph. 11.12, 1068a11-13; Phys. 7.3,
246b11-12; Metaph. 14.1, 1088a29-35; cf. Phys. 3.1, 200b33-201a9. In his lost
commentary on the Physics, Alex. appears to have developed this view further, by
allowing that in these circumstances something could be said to change in a
broader sense of the word, in so far as there is a realisation of one of its powers, in
line with Arist.s initial definition of change (Phys. 3.1, 201b4-5, cf. a10-11). But he
then praises Arist.s subsequent refinement and narrowing of the definition at 3.2,
202a7-9 (cf. 3.1, 201a27-9), so that not every realisation is a change in the primary
sense of the word, but only changes in quality, quantity, place, and being. See esp.
Alex. apud Simpl. in Phys. 437,18-28, together with 436,24-35 and 437,1-6 (=
409,27-32). Mere Cambridge changes, then, are not changes in the primary sense
of the word, and Alex. elaborates Arist.s views on light and colour with this
systematic consideration in mind. For more on mere Cambridge changes in Aris-
totle and post-Aristotelian philosophy, see Sorabji 2004a, 2.64-8; 3.79-88.
(2) Alex. takes the twin facts that light and colour are present immediately
(euthus) whenever their sources come to be present and disappear immediately
together with them to be evidence (smeion) that the presence of light is due to
relational facts: see esp. Mant. 143,11-29 and in Sens. 134,11-19; 135,9-17; also in
Sens. 31,15-18; 47,4-6; cf. Mant. 145,2-5; Quaest. 1.2, 6,17-19; in Sens. 52,10-12.
Such transitions are instantaneous (akhrons), occuring simultaneously (hama)
throughout the medium as a whole (athroon), both far away from the source and
nearby: in Sens. 31,13-15; 132,11-16; 133,18-27; 135,21-2. This is not true of spatial
motion or even alteration, at least not when the latter takes place over a suffi-
ciently large area: Mant. 138,16-18 (cf. 139,17-19); in Sens. 123,6-11; 132,17-133,6.
For more on instantaneous transitions, see n. 382 on 43,11 below.
This is a point, interestingly, on which Galen would have agreed. He contrasts
the effect of heat in the air with light: the former lingers after the source has been
removed, while the latter does not, but disappears along with it (PHP 7.5.11,
454.27-9 De Lacy).
(3) The presence of an illuminant or colour is a specific kind of relation (poian
skhsin, 43,3), of one things being spatially located in another. Alex. standardly
uses the example of somethings being to the left or to the right of something to
illustrate the notion of a relation. But not all relations are spatial: at one point,
Alex. uses the relation between a father and a son alongside what is to the left or
right as examples of a skhsis (in Sens. 134,17-19; see Accattino 1992, 56 n. 50). At
56,8-10 below, Alex. characterises the terms for perceptible qualities what can
be touched (hapton), seen (horaton), or heard (akouston) as the names of a
relation, rather than the underlying object.
377. [43, 4-5] 42,9-10.
378. [43,5-8] Alex. offers a similar etymology at Mant. 148,7-9 and in Sens.
45,20-1: although all bodies have a share in transparency (diaphaneia) on Arist.s
view, Alex. notes that the only ones we call transparent are those which are
see-through (diopta, Mant. 148,7); we specifically (idis) call transparent those
through which (di hou) colours appear (phainetai); cf. DA 44,23-5 below. See also
the further etymological elaboration of appearing (phainomenon) in terms of
light (phaos) at in Sens. 45,11-14 (accepting Diels emendation of the lacuna).
Alex. describes light as the form (eidos) of the transparent at in Sens. 35,13-15.
But he also questions this at Quaest. 1.21, 35,9-16, on the grounds that the
158 Notes to pages 67-68
transparent can exist without it, and suggests that it is instead a kind of modifi-
cation (pathos ti), which occurs in virtue of relational facts, involving the presence
of something like fire. On light as the culmination (teleiots) of the transparent, see
n. 379 on 43,7-8.
379. [43,7-8] DA 2.7, 418b9-10; 419a11. On light as the activity (energeia) of the
transparent, DA 44,13-15; Mant. 142,1-2 (cf. 148,14-16); in Sens. 35,8-9; 39,20-1;
see also n. 373 on 42,8-11 above; as the culmination (teleiots) of the transparent,
see Mant. 142,8-10; cf. Quaest. 1.2, 7,1-4; and as the completion (entelekheia) of the
transparent, in Sens. 35,13-15; 42,25; Quaest. 1.2, 7,3-4; 1.21, 35,9-10; cf. Mant.
144,29.
380. [43,8 title] The section title in manuscript V seems out of place, coming
almost at the end of the discussion of light (42,19-43,11). Alex. does return below
to the nature of light briefly at 44,13-15 and 44,23-45,5. But the principal topic in
this section is the nature of colour in bodies (44,16-46,19), after a brief recapitula-
tion of how colour affects the eye via the medium (43,11-44,1), the nature of the eye
itself (44,1-9), and the power of sight (44,9-13). Light is discussed only in so far as
its treatment has points of contact with that of colour. For an overview of the entire
section on sight, see n. 369 on 42,4.
381. [43,8-11] 43,2; 45,1-3. On the relational nature of light, see n. 376 on
42,22-43,4 above. On light as the presence of fire or something like it, see 45,1-3;
Mant. 144,18-19.23-4; 145,2-3; in Sens. 43,5-7; 132,7-8 (quoting DA 2.7, 418b16-
17). Fire is not the only kind of thing naturally suited to illuminate the medium.
Alex. also takes the divine body (theion sma) that is, the celestial sphere which
carries the fixed stars to be a source of illumination: DA 45,3; Mant. 142,5. He
refers to it at Mant. 144,19-22 as the aetherial body (to aitherion sma), along with
various phosphorescent bodies, as a source of illumination.
382. [43,11] On light not being a body, Arist. DA 2.7, 418b13-15; Alex. Mant.
124,9-10; 132,37;138,3-139,28; in Sens. 31,13-17; 33,13. See also n. 375 on 42,19-22
above.
On illumination as instantaneous (akhrons), see Alex. in Sens. 135,13-14.
Although it is often translated as atemporally, the Greek akhrons does not imply
that something occurs timelessly or outside of time, since the events in question
can be dated. The word indicates rather that, unlike events which require an
extended duration (deisthai khronikou diastmatos, Mant. 143,31-4), these events
are without duration and so instantaneous: in fr. 16 Giannakis, Alex. explicitly
compares them to the now and to points, which begin or cease to exist without any
process of generation or destruction (1995/96, 169 and 177). Illumination occurs
throughout the medium all at once (athros, Mant. 143,19-23.26-8; 145,2-5; in
Sens. 132,8-16; 135,16-17; cf. Mant. 139,17-19; 142,13-16): as soon as a source of
illumination is present, so is light, and as soon as it disappears so does the light
(see n. 376 on 42,22-43,4 above). Alex. further argues that seeing itself and indeed
perceiving in general occurs instantaneously and so apart from time (khris
khronou) in the sense of duration. Nonetheless, colours are the only perceptibles
which are transmitted through a medium instantaneously: in Sens. 135,13-17; in
Meteor. 129,3-14; fr. 16 Giannakis (1995/96, 169 and 177); cf. Mant. 143,29-33. For
Alex.s view on these issues, see de Groot 1983, 180-2 and 1991, 37-8, cf. 52-3. For
Arist.s views on the temporal character of illumination, see DA 2.7, 418b20-6;
Sens. 6, 446a20-447a11 (446b9-13, 447a9-10).
According to Alex. the fact that these events are instantaneous shows that they
are not changes (kinseis) in the primary sense, since all such change is in time in
the sense that it has duration: Mant. 143,23-8; Quaest. 1.22, 35,19-21; in Sens.
Notes to page 68 159
59,25; 124,9-10; in Meteor. 129,3-9; cf. Mant. 139,22; 143,6-7; in Sens. 30,1-2;
135,9-11. For discussion, see Sharples 1998, 78-9.
383. [43,11-16] Alex. recapitulates here the causal story of how vision is
produced by visible objects through a medium, first outlined at 42,7-12, but
without adding any of the qualifications he often adds (see n. 375 on 42,19-22). He
speaks here simply of colours being able to affect the medium (kintikon, 43,12.14;
cf. 42,6-7) and of the mediums being modified (paskhein) by the colour and the eye
by the medium (43,13-14). He likewise fails to mention the stimulation in the eyes
being relayed or transmitted to a primary or central sense, as he had before
(39,18-21), although nothing he says here precludes such a process either. More
interesting, though, is the opening inference, which makes fully explicit the causal
assumptions underlying Alex.s general approach: a colour can be seen in the light
only if it can affect the light in some way, since it is by doing this that seeing comes
about: compare Arist. Sens. 2, 438b2-5. In speaking of colour affecting light, as he
does earlier at 42,11-12, Alex. is thinking of light in effect as the actively transpar-
ent medium (43,7-8), which colour is equally said to affect (42,6-7.9-10): cf. in Sens.
35,6-12.
Although Alex. standardly uses the Greek opsis for the power of sight, as
opposed to the body which has this power, the opthalmos or omma for the
contrast, see e.g. 44,9-11 below and 57,13-14 (or indeed Arist. DA 2.1, 412b18-20)
it is clear that opsis here must signify the eye, as it can in Greek, since otherwise
the reference to its transparency at 43,15 and the plural opsesin at 43,20 would not
make sense (see also 62,4). It is relevant, as AD rightly point out (184), that Arist.
himself uses opsis in this way at DA 2.7, 419a13, in departure from his more
common usage of the term for the power; see also Sens. 2, 438a13 and Alex. in Sens.
26,10.
For Alex.s reference to inquiries into how we see, see n. 384 on 43,16-18 below.
384. [43,16-18] On phosphorescent objects, see 42,4-7 above, with correspond-
ing note. In the present passage, Alex. refers twice (43,16.18) to writings on sight,
which examine How We See, and he refers to them again at in Meteor. 141,11-12
together with On the Soul. Both Rose (1863, 374-5) and Moraux (2001, 365 n. 216)
construe these as references to a work of Alex.s and so are forced to take a phrase
in the in Meteor. passage, which he also recalls in On the Soul (hn kai autos
emnmoneuse, 141,12), to be a gloss inserted by a scholiast. But Moraux also
argues against identifying it with Mant. 141,29-147,25, since this passage does not
discuss phosphorescent objects at any length (although it does mention them at
144,21-3; see Sharples 2004, 88-9 esp. n. 284). So we would then have to posit a
lost work of Alex.s.
There is an alternative, though, if (as BD have argued, 282) it is not a scholiast
who says which he also recalls in On the Soul at in Meteor. 141,12, referring to
Alex., but rather Alex. himself referring to Arist. This reading is far more natural
and economical. And we need not take it to refer to another possibly lost work of
Arist.s either, as BD seem to, on the basis of Arist.s remark at DA 2.7, 419a6-7
that phosophoresent objects are the subject of another discussion (allos logos), a
phrase which may just indicate that it is a topic for another day. Consider a
passage from Alex.s in Sens., where he describes how visible objects situated along
the line of sight affect the medium and thereby the eyes, in terms much like those
above at 43,18-44,2, something he claims that Arist. said when he spoke about how
we see in On the Soul (eirken hote elegen en tois Peri psukhs peri tou ps hormen,
in Sens. 25,21-6). All of these references, that is, may be to nothing more than
Arist. DA 2.7. It is striking that Themistius (in DA 61,21-4), who takes Arist.s
160 Notes to page 68
remark that phosphorescent objects are the topic of another discussion to refer to
another treatise, can only cite Arist.s Sens. (which speaks about these objects only
in passing at 2, 437b5) and the third book of On Sight by Alex.s teacher, Sosigenes
(see n. 370 on 42,4-5 above). He is not aware of a separate treatise How We See by
either Arist. or Alex. On Sosigenes book, see Moraux 1984,358-60. Moraux conjec-
tures that Themistius is entirely dependent here on Alex.s lost commentary on the
DA for his references. But this is just a guess on Morauxs part, offered without
argument or evidence.
385. [43,18-44,1] Mant. 146,20-8; in Sens. 25,21-6; cf. in Meteor. 141,12-16.23-9.
Alex.s language here suggests that in some sense tingeing (anakekhrsmenon,
43,19; cf. hsper khrses, Mant. 146,17-18) is indicative of a general phenomenon
and that a colour affects the transparent medium in this way not only when the
effect on the medium is itself visible, as in the cases described at 42,12-19, but also
in quite ordinary cases when it is not visible, unlike the distal object: see discussion
in n. 375 on 42,19-22. Tingeing does not in any event involve the mediums being
modified or altered except in a qualified sense: see again n. 375 on 42,19-22. Later
Alex. will explicitly deny that the illuminated medium receives colours as matter
does when they are relayed to sight (62,3-7).
For the line of sight, see Arist. PA 2.10, 656b29-31, which he also uses to explain
the location of afterimages at Insomn. 2, 459b13-18; Alex. points out that the line
of sight is not deflected even by strong winds: Mant. 129,21-4; in Sens. 28,27-8.
The exclusive modification (idion pathos) Alex. refers to here is presumably a
transferred epithet: it is the modification due to a perceptible exclusive to sight,
namely, colour: see 40,21-41,5.
386. [44,1-2] The reflection one sees in anothers eyes might seem irrelevant
here and likewise for the parallel passage at Mant. 146,20-8. But it is explicitly
part of the dialectical context for the parallel passage at in Sens. 25,21-6, which
criticises Democritus for in some way identifying sight with the reflection of an
object in the eye (Arist. Sens. 2, 438a5-6). In the course of elaborating Arist.s
rejection of this theory (438a7-12), Alex. states what he thinks is the correct
explanation of vision and of the reflection seen in anothers eyes (in Sens. 24,14-
26,7), as he also does in the passage above and at Mant. 146,20-8. Both take the
reflection to be due to the fact that the eyes contents, which they call the kor (see
n. 392 on 44,7 below), are not only transparent, but smooth and bright, especially
at the aperture we refer to as the pupil, in which the reflection appears. On the
difference between transparent materials generally and those which are reflective,
see Mant. 142,21-31; cf. in Sens. 26,21-5; 36,2; also Sorabji 1991, 230 n. 14.
387. [44,2-3] cf. 44,10 below. Alex.s claim about sight being due to the reception
of colour and a kind of likening is just a particular instance of the general
characterisation he offered at 38,21-39,2 above. He again makes it clear that it is
the organ rather than the power which receives a perceptible quality like colour,
and that there is some sense in which the organ takes on the colour (whether this
is literal colouration or not). On the notion of likening, see 39,10-18 above with
corresponding note.
388. [44,3-4] 59,12-20; Mant. 106,30-107,1; Quaest. 1.2, 7,12-13; Arist. DA 2.7,
418b26-9; cf. 2.11, 424a7-9; Sens. 3, 439b6-10. This claim appears to be an instance
of a more general principle, namely, that something cannot become F unless it is
not F to begin with. But if (as Arist. assumes) we are able to see every colour, then
the organ must be able to receive every colour; therefore, it cannot have any colour
of its own. An analogous claim can be made for every sense except touch, which
must have tangible qualities of its own, as no organ can be intangible (2.11,
Notes to page 68 161
423b31-424a10). An even stronger claim will be made for the understanding,
which Arist. thinks is capable of understanding everything whatsoever and so
cannot be anything in actuality at all (beyond the ability to understand), and so a
fortiori cannot have any organ (3.4, 429a18-27). Alex. discusses the contrast
between touch and the other senses at 59,1-20 below and Mant. 106,30-107,7 and
the case of understanding at 84,10-85,5; Mant. 106,24-9 and 107,7-20.
389. [44,4-5] See esp. 46,6-12 below; also 44,23-45,1; 45,7; in Sens. 47,3-8; 48,17;
Quaest. 1.2, 5,10-19; 7,12-13; Arist. Sens. 3, 439a26-7 (cf. b1-5). As we shall see
below, coloured bodies also contain transparent material, even though they have
a definite shape of their own (hrismena, 44,18-23; 45,5-46,6; 46,12-19). Those that
lack a definite shape (ahorista) not only get their boundaries from surrounding
objects, but get their colour from them as well (46,6-12). Alex. explains that these
sorts of bodies receive, but do not retain, colours because of their moisture and lack
of definite shape: in Sens. 47,3-8; Quaest. 1.2, 6,7-8. He adds at Quaest. 1.2, 5,10-19
that it is precisely because such materials are capable of receiving colour, while
lacking one of their own, that they are apt for relaying the colours of things truly,
precisely because there is no admixture with a colour of their own.
390. [44,5] 42,7; 52,18-19; Mant. 147,27-9; in Sens. 26,13.22; 44,4; 50,8-9;
52,19-20; 89,2-3.9. Arist. Sens. 2, 438a14-15.
391. [44,6] Arist. Sens. 2, 438a13-15, b5-8. The perceptual organ Alex. has in
mind here is clearly the eye. But notice that he only says that the eyes are that
through which (di hou) the awareness of colours occurs, rather than that in which
it occurs. Elsewhere he is clear that this awareness occurs in the heart, which is
the primary perceptual organ, and not in the eyes. But he does not make this denial
explicitly in the DA: see n. 353 on 39,21-40,3.
Alex. discusses in great detail at in Sens. 26,10-27,6 why the eyes consist of
water rather than air, as he goes on to claim just below at 44,7.
392. [44,7] 59,14; Quaest. 3.6, 91,13-14; in Sens. 26,10-11; 27,6.7; 31,4; 35,17-
36,4. Arist. DA 3.1, 425a4; Sens. 2, 438a16-22, b5-11, b19; HA 1.9, 491b20-1; GA
5.1, 780b23.
The Greek kor is often translated as pupil, due no doubt to the derivation of
the English pupil from the Latin pupilla. But these are false friends. The Greek
kor, like the Latin, can be used for dolls or figurines (lit. a little girl) and so by
extension for the part of the eye where one sees a tiny reflection of oneself, as Plato
explains at Alc. I 133A. But where precisely does one sees the image? It is not
unnatural to think of the doll-like figure as inside the eye, rather than at the
aperture, much as we instinctively think of a mirror image as being at some
distance behind the surface of the mirror; and it is surely the geometry of appear-
ances that is relevant here, not whether there is an actual figurine at either of
these locations (since there isnt any).
In fact, neither Arist. nor Alex. uses kor for the aperture in the iris, as we do,
but for the interior contents of the eye that are visible through this aperture. This
is clearly how Alex. understands Empedocles famous lantern fragment (DK 31
B84.10-11): see in Sens. 34,13-15. Arist. explicitly identifies the kor as the moist
[material] inside of the eye, by which we see (to d entos tou ophthalmou, to men
hugron, hi blepei) at HA 1.9, 491b20-2, in contrast with the dark part of the eye
(to melan, the iris and pupil, properly speaking) and the white of the eye (to leukon,
the sclera). See also HA 3.18, 520b3-5; 4.8, 533a8-9; DA 2.8, 420a14; Sens. 2,
438b5-16; GA 5.1, 780b22-9 (cf. a25-36). On this point, see Sorabji [1974] 1979, 49
n. 22; Johansen 1998, 56. For references to uses of kor in other ancient authors,
see Mugler 1964, 226-8. Although Mugler regards the aperture as the primary
162 Notes to page 68
meaning, which is extended pars pro toto to the whole eye, none of his texts
decisively support this, while many explicitly support the meaning I have argued
for here.
For Alex., Sharples appeals to Arist.s usage (2005b, 348 n. 12) in generally
translating kor as eye jelly. But he thinks that sometimes it must be rendered
pupil (2004, 130 n. 441), even alternately within the same passage: see his
translations of Mant. 128,25 and 129,2.5.10.21; 142,17.21.31; for pupil, cf. also
136,14.16.18; 146,19.25. But it is possible to take virtually all of the terms
occurrences, both in these passages and in the rest of Alex.s corpus, as consistently
referring to the interior contents of the eye. This includes references to the vertex
of a visual cone being at the centre of the kor, which should be construed as the
centre of the eyeball rather than the centre of the aperture at or near the surface
(Mant. 130,15; 146,19.31-3; in Meteor. 149,28) this understanding of the geome-
try is made explicit in Damian Opt. 2,19-21 and 12,12-16 Schne (even though he
himself uses kor for what we call the pupil). It might be thought that Alex. Mant.
128,25 is an exception and must refer to the pupil. But it can refer here to the
interior of the eye as well: on the theory under consideration, the interior of the eye
as a whole would constitute a passageway from the optic nerve to the pupil for the
body that sees. A more plausible exception is Mant. 135,6-10, which concerns the
relative sizes of the kor and the incoming Epicurean simulacra of the object, since
presumably the simulacra will have to be small enough to fit through the aperture
of the pupil. But the wording of Alex.s objection looks garbled, and if what he
means is that the kor can be no smaller than the simulacra, then it will be true a
fortiori of the interior of the eye no less than the pupil, and so does not constitute
a clear exception to the rule.
393. [44,6-9] Mant. 142,26-32; in Sens. 26,15-27,6; 35,25-36,4. BD claim (283)
that Alex.s appeal to the density and consistency (pakhutta te kai sustasin,
44,8-9) of the vitreous jelly has no parallel in Arist.s work. But in fact Arist. claims
at Sens. 2, 438a15-16 that water is more easily retained and more condensed than
air (euphulaktoteron kai eupiltoteron), if at any rate we follow the reading in MSS
EMY, with most modern editions. Ironically, the last term does not occur in the
text Alex. used for his commentary, which has more easily contained (euhupolp-
toteron) instead. But Alex.s develops his explanation along much the same lines:
he refers to waters superior consistency (mallon sunestanai, in Sens. 26,19;
sustasin mallon ekhei, 26,21) and goes on to explain that water is a mean between
the sparseness of air and the solidity of glass and other transparent bodies, making
it ideal for perception (26,26-27,6). The idea that the density of the vitreous jelly
keeps it inside the eyeball is not peculiar to Arist. either. Plato suggests at Tim.
45BC, using similar terms, that the denser material in the eye is kept in (to men
allo hoson pakhuteron stegein pan) because of the way the eyes are compacted
(sumpilsantes).
More generally, Arist. explains different animals different sensitivity to light
and colours in terms of the amount of moisture in their eyes: GA 5.1, 780a1-5.
394. [44,9-13] Alex. recapitulates here (44,9-11) what he said earlier about the
activity of seeing (44,2-3), which provided the basis for his conclusions about the
eye, the organ of sight, in the preceding argument (44,3-9; cf. 43,14-15). He then
extends those comments by elaborating briefly on the power of sight, which
underlies this activity (44,11-13), before returning again to questions about light
and colour. As AD 185 point out, there is little that can be said about the power
itself beyond defining it in terms of its activity. Alex. preserves here the distinction
between the alteration (alloisis) produced by visible things and the activity
Notes to page 68 163
(energeia) by which we have awareness and cognition of them, even though he does
not emphasise it: for more on this distinction, see n. 344 on 38,21-39,2.
395. [44,13-14] 44,23-5; 89,1-2; Mant. 148,23; 149,32-5; in Sens. 43,13-15;
46,21-47,3; 47,12-13; in Metaph. 142,13-16. Alex. also says such things about
sources of illumination such as fire, at DA 46,1-4; in Sens. 45,26-46,3; cf. Quaest.
1.2, 6,33-7,1. But there is no genuine inconsistency here, as AD 185-6 seem to
allege: both light and the sources of illumination can be visible in the highest
degree without contradiction. Superlatives do not in general preclude ties.
Alex.s statement here seems to go beyond Arist. in several ways, for which he
was vigorously criticised by Moraux (1942, 89-92), first for characterising light as
visible in the highest degree (malista horaton) and second for the underlying
causal principle that seems to be presupposed here. Both will play a critical role
later in the treatise in Alex.s discussion of the productive understanding (nous
poitikos), where he enunciates a fully general version of the principle: for any
attribute F, whatever is F in the highest degree is responsible (aition) for anything
elses being F (88,23-89,11). This principle has seemed to many to be fundamen-
tally Platonic and non-Aristotelian Merlan declares, in fact, that it is as close to
what causality is in Neoplatonism as possible. (1969, 14-15, 38-40 at 39; see also
Donini 1974, 42-8.) For detailed discussion of the principle, including its relation
to Platonism, see the notes on 88,23-89,11 below. The present note is concerned
with Alex.s claims only in so far as they concern light.
Morauxs objection that light is not visible for Arist., because the transparent is
colourless, is simply mistaken. As AD rightly point out (185-6), colours are not the
only things which are visible on Arist.s view: phosphorescent objects also are,
despite not having colour of their own (DA 2.7, 419a1-6; cf. 418a26-8), a point Alex.
also makes above (42,4-7; cf. 43,16-18). AD suggest (186) that Alex. is here
assimilating light to its source, fire, which Arist. already recognises as visible (DA
2.7, 419a23); and given that fire and other illuminants are responsible for the
visibility of colours, they might themselves be regarded as having this quality
themselves in the highest degree, based on the general causal principle Arist.
enunciates at Metaph. 2.1, 993b24-6 (see also An. Post. 1.2, 72a29-32).
This response does not go to the heart of the matter, though, since an objection
similar to the original one can easily be constructed. Light, as Alex. immediately
goes on to say (44,15; cf. 43,7-8), is the activity of the transparent medium, in so
far as it is transparent, that is, of the transparent medium when it is illuminated.
The medium in this condition, however, is something that one can see through
(diopton) in Sens. 45,11-21 (cf. 43,20-2) and also Mant. 148,7-10 (though the text
here requires supplementation, either with Bruns or Bessarion, see Sharples 2004,
144 n. 485; 2008, 204 ad loc.); cf. Quaest. 1.2, 5,15-19. in which case light might
not seem to be the sort of thing that can be seen itself. As Arist. says himself, the
transparent material is not visible on its own, to put it simply (ou kath hauto de
horaton hs hapls eipein, DA 2.7, 418b5). So we still need an explanation of why
light, as the activity of the transparent medium, is visible and indeed to the highest
degree. ADs interpretation, moreover, poses an additional difficulty. If correct, it
would invert Alex.s explanatory strategy: his conclusions about light would
merely be an application of the general causal principle stated later, rather than
a case he regards as intuitive on its own, which can then be offered as evidence for
the general principle (as seems to be his strategy at the crucial juncture, 88,26-
89,2).
In fact, Alex. has entirely orthodox grounds within Arist.s philosophy for
thinking that light is visible in the highest degree. Later in the present passage
164 Notes to pages 68-69
and elsewhere, he systematically elaborates Arist.s claim at Sens. 3, 439b11-12
that colour is the limit of transparent material in a body with a definite shape (see
44,18-19 below with corresponding note). On this view, all bodies are transparent
to a greater or lesser extent, not simply those we can see through (see 44,20-1 with
corresponding note); and their colour is due precisely to the amount of transpar-
ency present in each, in so far as it is transparent. So a body which has
transparency in the highest degree, such as air or water, thereby has the basis of
visibility within itself to the highest degree (Mant. 149,34-5; in Sens. 45,16; 46,1-2).
It is rather coloured bodies which are imperfect in this regard: they are not fully
transparent, due to the presence of earth in their composition, and so opaque (see
45,7-11 below with corresponding note) even flame is not fully transparent due
to the admixture of earth in smoke (Mant. 149,24-6; in Sens. 46,17-18). Colour thus
ranges between white or, better, bright (leukon), which is the most luminous
(phtoeides) colour, and black or dark (melan), which is the privation of light (see
45,10-19 with corresponding note). At Quaest. 1.2, 6,26-30 Alex. explains that
those solid bodies are bright which have the most transparent body present in
them, which he later identifies with fire (7,1), and he characterises white or bright
in terms much like light in the present passage: it is colour in the highest degree
and visible in the highest degree (malista gar touto khrma kai malista horaton)
and responsible for bodies being more coloured and being more visible (tou te
mallon kekhrsthai tois smasin aitia kai tou mallon horatois einai). The excep-
tional visibility of light and bright colour is thus due to the same cause, namely,
the high degree of transparency of the body which exhibits them.
Arist. himself also seems to regard light as not only visible, but visible in the
highest degree. The transparent medium may not be visible in itself, but it is
nonetheless visible it is just visible through an extraneous colour (ho esti men
horaton } di allotrion khrma, DA 2.7, 418b4-6), that is, its colour is due to
something else, such as the presence of an illuminant. It is thus no contradiction
to claim that its activity, light, is itself highly visible: the glare (aug) in air and
sheen on water are offered as clear evidence that both mediums are coloured in
some way (Sens. 3, 439b1-2) and light is even spoken of as like a colour (hoion
khrma) of the transparent medium (DA 2.7, 418b11), that is, as the kind of thing
that we ordinarly see. However we interpret these passages (for further discussion,
see n. 397 on 44,16 below), they suggest that light is visible in some way. More
strongly still, light seems to be the kind of extreme or intense perceptible (tn
aisthtn hai huperbolai, sphodra aisthta) which Arist. often speaks of as damag-
ing to our perceptual organs precisely because of the superlative degree of its
perceptible quality. For the general point, see DA 2.12, 424a28-32; 3.2, 426a29-b8;
3.4, 429a31-b3; 3.13, 435b7-9. For what is excessively bright (lian lamprou) in
particular, see 2.10, 422a21-8; GA 5.1, 780a7-14. On other extreme perceptibles,
cf. also 2.9, 421b23-5; 2.10, 422a29-32; 2.11, 424a10-15.
396. [44,15] 43,7-8 with corresponding note.
397. [44,16] Arist. himself nowhere says that light is a colour without qualifi-
cation. The closest he comes is to say that it is like a colour of the transparent
(hoion khroma tou diaphanous) at DA 2.7, 418b11 and that the glare in air and
sheen on water show that these mediums are coloured or tinged (khrmati-
zomena, Sens. 3, 439b1-2). But when he speaks more precisely, he says that light
is only an extrinsic colour of the transparent medium (khrma tou diaphanous kata
sumbebkos, Sens. 3. 439a18-19) the transparent is not visible on its own (ou
kath hauto horaton), but only through an extraneous colour (di allotrion khrma),
that is, it is visible due to something elses colour (DA 2.7, 418b4-6). Richard Sorabji
Notes to page 69 165
adds (in his editorial comments) that there are good reasons to distinguish light
from colour. Colour, glare, and sheen are all surface phenomena; illumination is
not, but suffuses an entire volume. Unlike colour, furthermore, light is not seen,
but rather seen through (Sorabji 2004b, 134).
Alex. vacillates on whether light is a colour. Immediately below at 45,1 and
elsewhere he qualifies the claim. Light is like a colour, not simply as such, but
extrinsically, because the transparent does not receive light in a way that modifies
it (hsper khrma tou diaphanous, oukh hapls, alla kata sumbebkos, hoti m
pathtiks anadekhetai to diaphanes to phs, in Sens. 42,26). The transparent
receives light as if it were its own colour (oikeion khrma), although in fact it is not
(Quaest. 1.2, 6,16-17; cf. 5,12-19; Mant. 144,16; 150,2; DA 42,7-8). It is colour only
in a way (khrma ps on, in Sens. 52,9) or analogous (analogon) to it (Quaest.
1.21, 35,14-15). Illuminants colour it in a way (khrnnusin tropon tina) and light
is like a colour of the transparent medium (Mant. 144,5-6).
In other passages, though, he makes the stronger, unqualified claim he makes
here. At in Sens. 43,12-17 he insists that light, because it is a colour (khrma on),
is visible in the highest degree and pre-eminently, and that the other colours (tn
alln khrmatn) are perceived because of it, a phrase which recurs again at 47,2-3
and below at DA 46,3. Sharples believes (2004, 134 n. 453; 2008, 200 ad 144,6) that
Alex. goes further and claims that light is the transparent mediums own colour
at Mant. 144,16 and 150,2. But these last remarks are both qualified with hsper:
each should be rendered as though it were the transparent mediums own colour,
as in his fuller statement at Quaest. 1.2, 6,16-17 (with Sharples note).
398. [44,16-19] Arist. Sens. 3, 439a26-30, b10-18. Alex. DA 45,5-46,6; 46,12-19;
Mant. 148,2-6 (cf. 150,13-18); Quaest. 1.2, 5,2-3.21-2; in Sens. 49,4-8; 51,21-4; cf.
44,20-1; 47,24-5. Alex. is aware that this is a counterintuitive view and that we
typically only speak of substances we can see through as transparent (in Sens.
45,11-21)
Bruns suggested emendation at 44,17-18 is extravagant and unnecessary. AD
(187) suggest that Alex.s back reference (in the way just mentioned) is to lights
being the activity of the transparent (44,15). But in any case colour is nowhere
described as an activity of the body to which it belongs. Alex. might be taken more
plausibly to be referring to the fact that colour, like light, is in transparent
material and, more importantly, that it is responsible (aition) for things in general
being seen (44,14).
399. [44,20-1] Arist. describes transparency as a kind of common nature and
power (tis koin phusis kai dunamis) at Sens. 3, 439a23; more generally, see
439a21-5; b8-10. Alex. follows suit: Mant. 147,27-30; 148,1-2.9-12; 149,1-3; in Sens.
44,4-9; 45,9-11; 45,21-47,20; Quaest 1.2, 6,1-3 (though compare 5,22-3). By taking
both light and colour to be due ultimately to the same underlying cause, the
Aristotelian reconceives the notion of what is transparent along the lines of its
etymology, in Greek as in English it is that through which things appear (see DA
43,5-8 above; cf. 44,23-5 below) rather than that which is seen through, for which
he uses the term diopton. See also n. 400 on 44,21-3 below.
400. [44,21-3] Arist. Sens. 3, 439a26-8, b7. Alex. Mant. 147,30-148,2; 148,29-
149,3; Quaest. 1.2, 5,30-6,1; in Sens. 44,22-45,3. If the claim here about matter is
applied to the transparent medium, then it seems to contradict 42,21-2, as BD
point out (285), and so needs to be understood in some broader sense of reception
that does not imply literal alteration; cf. also AD 187. Both sets of commentators
appeal to the notion of the material intellect at 81,24-5 as an example of such a
sense.
166 Notes to page 69
The lines above (44,20-3) give us the underlying rationale for the Aristotelian
conception of transparency (see n. 399 on 44,20-1): every body is transparent in so
far as it can in some sense receive colour, that is, in so far as it is susceptible in
some way to colour as such (cf. Arist. Sens. 3, 439b7). The transparent medium may
not be genuinely modified or changed by colour and so does not acquire a colour of
its own (see n. 375 on 42,19-22 above; also 62,5-7 below). But it is such that it can
relay whatever effect colour has on it to the visual organs of animals, which are
sensitive to it (42,8-11; 43,11-16; 43,18-44,1). The transparent nature mixed with
earth in solid bodies, in contrast, is able to possess and manifest colour and in turn
affect the medium surrounding it. See 42,5-7 and 43,12 above and 45,5-46,6 below.
401. [44,23-5] in Sens. 45,11-16. On the etymology of transparent, with further
references, see n. 378 on 43,5-8 above. Notice the implicit qualification in Alex.s
remark: the transparent medium, which we can see through, is just something that
is transparent in the highest degree. As we have just seen (44,20-1), all bodies are
transparent to some degree, including those we cannot see through. Cf. Mant.
148,7-9.
For the claim that the transparent medium is coloured or tinged by something
outside it, see esp. in Sens. 49,8-50,5 (cf. 50,7-13); Mant. 144,5-6; as well as
42,11-19 and 43,18-20 above. The contrast between extraneous colours and a
colour of ones own is implicit at Arist. Sens. 3, 439a31-b6.
402. [45,1] 44,16, with accompanying note.
403. [45,2-3] 43,8-11, with accompanying note; cf. 43,2.
404. [45,3-5] 42,22-43,4, with accompanying note.
405. [45,5-6] cf. Arist. Sens. 3, 439a33-b1. Alex. DA 45,18-19; in Sens. 50,5-7;
51,25-7. The contrast here is with the transparent medium, which has colour only
extrinsically and from an extraneous source: see n. 397 on 44,16. BD (286) seem to
think mistakenly that this passage concerns translucent solids like gems and so
contradicts 42,7-8. But those are solids that lack a colour of their own, whereas the
present passage is more plausibly about solids that have a colour of their own (see
45,17-19 below and n. 372 on 42,7-8 above).
If we take Alex.s words here in their most straightforward sense, as claiming
that the interior of bodies is coloured, he may be departing from Arist., who says
only that colour is on the surface of things which are intrinsically visible (to epi tn
kath hauto horatn, DA 2.7, 418a29-30). In Sens. 3, Arist. does insist that one
must think that the same [nature] that is coloured outside is also within (alla tn
autn phusin dei nomizein, hper kai ex, tautn kai entos, 439a33-b1); and Alex.
might conceivably have taken this to mean that this nature is coloured within: in
his commentary on the passage, Alex. in fact claims that Arist. says that these
objects are coloured inside (entos kekhrsthai legei, in Sens. 50,6). But strictly
Arist.s actual words need only mean that this nature, transparency, is present
within. And this strict reading may well be all that is justified: as Richard Sorabji
points out (in his editorial comments), the interior of a solid object might be the
same colour as its surface, a different colour, or even colourless.
In saying that the colour is inside, however, Alex. may have meant only that
colour is intrinsic to the solid: it is due to the material composition of the body itself
rather than borrowed from surrounding bodies, as the tingeing of a transparent
medium is. This, in fact, is how Alex. goes on to explain the claim from his
commentary on Sens.: Arist. says things that have a colour of their own from
themselves are coloured inside because (dihoti) they have a colour of their own and
what is responsible for colour (to aition tou khrmatos) within themselves (in Sens.
50,5-7). If so, then Alex.s position might be compatible with a surface only reading
Notes to pages 69-70 167
of Arist. after all. The colour of a solid will, as Arist. says, be present in the
boundary (Sens. 3, 439a31-2). But the interior depths of bodies (ta en bathei tn
smatn) will have colour at most potentially, through their power, instead (du-
namei khrma ekhein, 51,25-6).
406. [45,6-7] cf. Arist. Sens. 3, 439b1-6. On the impermanence of light in the
medium, see n. 376 on 42,22-43,4 above. For a similar contrast between how
transparent materials, with and without a definite shape, look from different
distances and positions, see in Sens. 50,16-26; compare Arist. Sens. 3, 439b3-5 with
440b16-18.
407. [45,7-10] Mant. 148,11-14; 149,15-32; 149,36-150,6; in Sens. 45,23-6;
46,6.12-19; 47,8-15; Quaest. 1.2, 7,4-7. Cf. also DA 46,4-6; Quaest. 1.2, 6,26-33.
408. [45,10-19] cf. Arist. Sens. 3, 440b14-23; 442a12-16; Alex. Quaest. 1.2,
6,26-7,7; in Sens. 53,5-8. Alex. draws an even more fine-grained distinction at both
Mant. 150,7-9 and in Sens. 47,15-20, where he sets apart those bodies which are
more transparent and truly luminous, from those which are just close (engus) to
these and so merely bright. As BD 287 rightly point out, the objects in the passage
above are not in any case phosphorescent: they still need light in order to be seen
(46,19).
AD argue (188) that the earlier explanation above at 45,5-10 is more limited
than the present one, applying only to solid, sublunary bodies, on the grounds that
earth is not part of transparent mediums like air and water, or the heavenly body
at all, whereas the explanation here (45,10-11) applies to all bodies. But there is
no fundamental conflict here. The earlier explanation does in fact apply to sublu-
nary bodies of air and water, since Arist. holds that every sublunary body (or those
near the middle anyway) contains each of the four elements, including earth he
is explicit on this last point, GC 2.8, 334b31-3 (see also 335a22-3 and 2.3, 330b21-5)
ordinary bodies of air and water are not pure samples of the corresponding
elements (GC 2.8 passim; for brief comments, see Sorabji 1988, 70 nn. 38 and 40).
The heavenly body is still an exception, but it is a marginal one, since anywhere
fire can be invoked, earth can too and vice versa. The two explanations are not in
competition with each other, moreover, since the underlying principle turns on the
proportion of several ingredients to each other in a mixture: Arist. Sens. 3,
440b14-23. If it is the proportion of earth to fire, as Alex. explicitly states at Quaest.
1.2, 6,26-7,7 (and as AD later argue themselves, 189), or even of earth possibly to
all three remaining elements, which are all transparent in varying degrees (as
Sorabji 2004b, 130 argues; cf. 1972, 293), then both of the explanations offered
above could be in play.
AD also worry that fire will not be involved in all cases, especially the heavens.
But Alex. offers fire as just an example of something bright (as BD 287 rightly
note). This broader characterisation allows him to extend the second explanation
to the case of the heavenly body, which may be a reason why he does not mention
either sublunary element at 46,4-6.
409. [45,20] 45,25-46,1; Arist. Sens. 3, 439b11-18 (cf. a27-32). For further
references to Alex., see n. 398 on 44,16-19 above.
410. [45,21-2] 46,12-17; Quaest. 1.2, 5,3-8; 7,13-17; in Sens. 44,18-20. A given
body and the transparent material that makes it up are, in Arist.s terminology,
one and the same in number. But they differ in being or in account what it is
to be a body is not the same as what it is to be transparent. Consequently, what it
is to be the boundary of one is not the same as what it is to be the boundary of the
other either. See n. 416 on 46,12-17 below.
168 Notes to page 70
411. [45,22-4] Mant. 148,4-5; Quaest. 1.2, 6,3-7. For appearance (phantasia)
used in this way, see Sens. 3, 439b5-6.
412. [45,24-46,1] 45,20; Mant. 150,13-19; in Sens. 48,10-13; 51, 21-4. This
sentence amplifies the reason underlying the previous one: things are visible to us
in virtue of their surfaces facing us, and these are visible because they have the
power to affect the intervening medium so as to be seen. This power, which in the
central cases just is the colour, must therefore reside in the surface. The sub-
sequent move to identify this with the surface of the body in so far as it is
transparent still requires further justification.
413. [46,1-6] AD 191 not unreasonably describe these lines (46,1-6) as the
summary and foundation of Alex.s entire theory of light and colour. On fire as
something visible in the highest degree, see 44,13-14 above (with accompanying
note for further references). On fire as a source of light, 45,2-5. On light as
contrasted with the other colours, see n. 397 on 44,16. On colour as belonging to
transparent material with a definite shape, see 44,17-19; 45,20; cf. 46,6-7. On
colour as due to the proportion of mixture (whether of earth, fire, or what is
bright), see 45,7-19.
The most striking feature of this passage, though, is how full and explicit its
causal claims are. In saying that fire and other illuminants are such as to produce
(poitikois) light and thereby make (poiei) other things visible and so become
responsible (aitia) for their being seen (cf. 44,13-14), Alex. does more than just
elaborate Arist.s summary remark at Sens. 6, 447a11, that light produces seeing
(to phs poiei to horan; though compare DA 3.5,430a16-17). It makes clear that in
Alex.s mind these are not merely sufficient conditions, but explanatory ones as
well.
414. [46,6-11] For the argument here, see also Quaest. 1.2, 5,11-15; in Sens.
48,22-49,4. On the transparent medium as only being coloured in a way, by
something extraneous to it, see 44,23-45,1 with accompanying notes; on its not
having a colour of its own, see 44,3-4 and n. 397 on 44,16.
415. [46,11-13] Quaest. 1.2, 5,19-23; in Sens. 49,4-8. AD 192 rightly reject
Bruns emendation here (following the Hebrew translation) as making nonsense of
the passage: the successive sentences beginning hn gar and hn de are links in a
continuous argument concerning coloured solids, as the parallel passage from the
Quaestiones makes clear, not a contrast between transparent mediums and solids.
416. [46,12-17] 45,21-2, with accompanying note. At in Sens. 47,25-48,10, Alex.
argues that bodies do not differ in degree as to whether they are bodies, but they
do differ in how transparent they are; therefore, even if the boundary of the body
and the transparent are one and the same in number, they will still differ in
account (48,5). At in Sens. 49,15-21 he adds a further battery of arguments to show
that colour is not the boundary of the body as a body: colour is a quality, surface a
quantity (15-16); every body has a surface, but not every body is coloured (17); a
single surface may have many colours (17-19) and a single colour might be on many
surfaces (20-1).
417. [46,17-19] The second half of this claim is surprising: for while it is true
that the colour of a solid could not be seen unless it were illuminated, there is no
reason to think that it would not still have its colour even when it was not
illuminated, esp. given Arist.s own realism about colours: objects still have their
colours even when they are not seen in DA 3.2, 426a20-7.
Bibliography

Accattino, Paolo. 1988. Alessandro di Afrodisia e la trasmissione della forma nella


riproduzione animale. Atti della Accademia delle Scienze di Torino, Classe di
scienze morali, storiche e filologiche, 122: 79-94.
Accattino, Paolo. 1992. Alessandro di Afrodisia e gli astri: lanima e la luce. Atti
della Accademia delle Scienze di Torino, Classe di scienze morali, storiche e
filologiche, 126: 39-62.
Accattino, Paolo. 1993. Ps. Hippocr. in Alex. Aphr. De anima, in F. Berger, C.
Brockmann, G. de Gregorio et al. (eds), Symbolae Berolinenses fr Dieter
Harlfinger, 169-73. Amsterdam: Verlag Adolf M. Hakkert.
Accattino, Paolo. 1995. Generazione dellanima in Alessandro di Afrodisia, De
anima 2.10-11.13? Phronesis 40: 182-201.
Accattino, Paolo and Pier Luigi Donini. 1996. Alessandro di Afrodisia: Lanima.
Traduzione, introduzione e commento. (= Biblioteca Universale Laterza, v.
447.) Rome: Laterza.
Ackrill, J.L. 1972/73. Aristotles Definitions of psuch. Proceedings of the Aristote-
lian Society 73: 119-33.
Algra, Keimpe. 1995. Concepts of Space in Greek Thought. Leiden: E.J. Brill.
Annas, Julia. 1992. Hellenistic Philosophy of Mind. Berkeley: University of Cali-
fornia Press.
Asmis, Elizabeth. 1984. Epicurus Scientific Method. Ithaca: Cornell University
Press.
Barnes, Jonathan. [1971/72] 1979. Aristotles Concept of Mind. Proceedings of the
Aristotelian Society 72 (1971/72): 101-14. Reprinted in Barnes, Schofield, and
Sorabji 1979, 32-41.
Barnes, Jonathan, Susanne Bobzien, Kevin Flannery, and Katerina Ierodiakonou.
1991. Alexander of Aphrodisias: On Aristotles Prior Analytics 1.1-7. Ancient
Commentators on Aristotle. London: Duckworth.
Barnes, J., M. Schofield, and R. Sorabji (eds). 1979. Articles on Aristotle, 4 vols, vol.
4: Psychology and Aesthetics. London: Duckworth.
Bergeron, Martin and Richard Dufour. 2008. Alexandre dAphrodise: De lme.
Texte grec introduit, traduit et annot. Paris: Librairie Philosophique J. Vrin.
Berryman, Sylvia. 2002. The Sweetness of Honey: Philoponus against the Doctors
on Supervening Qualities, in C. Leijenhorst, C. Lthy, and J.M.M.H. Thijssen
(eds), The Dynamics of Aristotelian Natural Philosophy from Antiquity to the
Seventeenth Century, 65-79. (= Medieval and Early Modern Science, vol. 5.)
Leiden: Brill.
Bonitz, Hermann. [1870] 1955. Index Aristotelicus. Graz: Akademische Druck- u.
Verlagsanstalt. (Originally published as vol. 5 of I. Bekker (ed.), Aristotelis
Opera.)
Caston, Victor. 1996. Why Aristotle Needs Imagination. Phronesis 41: 20-55.
Caston, Victor. 1997. Epiphenomenalisms, Ancient and Modern. Philosophical
Review 106: 309-63.
170 Bibliography
Caston, Victor. 1998. Aristotle and the Problem of Intentionality. Philosophy and
Phenomenological Research 58: 249-98.
Caston, Victor. 1999. Aristotles Two Intellects: A Modest Proposal. Phronesis
44:199-227.
Caston, Victor. 2005. The Spirit and the Letter: Aristotle on Perception, in R.
Salles (ed.), Metaphysics, Soul, and Ethics: Themes from the work of Richard
Sorabji, 245-320. Oxford: Oxford University Press.
Caston, Victor. 2006. Aristotles Psychology, in M.L. Gill and P. Pellegrin (eds),
The Blackwell Companion to Ancient Philosophy, 316-46. Oxford: Blackwell
Publishing.
Chaniotis, Angelos. 2004a. New Inscriptions from Aphrodisias (1995-2001).
American Journal of Archaeology 108: 377-416.
Chaniotis, Angelos. 2004b. Epigraphic Evidence for the Philosopher Alexander of
Aphrodisias, Bulletin of the Institute of Classical Studies 47: 79-81.
Charlton, William. [1970] 1983. Aristotles Physics, Books I and II. Translated with
Introduction and Notes. Oxford: Clarendon Press, 1970. Reprinted, with correc-
tions, 1983.
Close, A.J. 1969. Commonplace Theories of Art and Nature in Classical Antiquity
and in the Renaissance. Journal of the History of Ideas 30: 467-86.
Courcelle, Pierre. 1974-75. Connais-toi toi-mme: De Socrate Saint Bernard. 3
vols. Paris: tudes Augustiniennes.
Crivelli, Paolo. 2004. Aristotle on Truth. Cambridge: Cambridge University Press.
Davies, Malcolm and Jeyaraney Kathirithamby. 1986. Greek Insects. London:
Duckworth.
de Groot, Jean Christensen. 1983. Philoponus on De anima II.5, Physics III.3, and
the Propagation of Light. Phronesis 28:177-96.
de Groot, Jean Christensen. 1991. Aristotle and Philoponus on Light. New York:
Garland Publishing.
Dillon, John. [1977] 1996. The Middle Platonists. Revised edition with a new
afterword. Ithaca: Cornell University Press, 1996. (1st edn, 1977.)
Donini, Pier Luigi. 1971. Lanima e gli elementi nel de anima di Alessandro di
Afrodisia. Atti della Accademia delle Scienze di Torino, Classe di scienze
morali, storiche e filologiche, 105: 61-107.
Donini, Pier Luigi. 1974. Tre studi sull Aristotelismo nel II secolo d. C. (= Historica,
politica, philosophica: Il pensiero antico, studi e testi, vol. 7.) Turin: G.B.
Paravia & Co.
Ebert, Theodor. 1983. Aristotle on What is Done in Perceiving. Zeitschrift fr
philosophische Forschung 37: 181-98.
Ehnmark, Erland. 1957. Transmigration in Plato. Harvard Theological Review
50: 1-20.
Eichholz, D.E. 1949. Aristotles Theory of the Formation of Metals and Minerals.
Classical Quarterly 43: 141-6.
Ellis, John. 1994. Alexanders Defense of Aristotles Categories. Phronesis 39:
69-89.
Emilsson, Eyjlfur Kjalar. 1988. Plotinus on Sense-Perception: A Philosophical
Study. Cambridge: Cambridge University Press.
Everson, Stephen. 1997. Aristotle on Perception. Oxford: Clarendon Press.
Fontenrose, Joseph Eddy. 1959. Python: A Study of Delphic Myth and its Origins.
Berkeley: University of California Press.
Frede, Dorothea and Burkhard Reis (eds). 2009. Body and Soul in Ancient Philo-
sophy. Berlin: Walter de Gruyter.
Bibliography 171
Frede, Michael and Gnther Patzig. 1988. Aristoteles Metaphysik Z. Text,
bersetzung und Kommentar. 2 Bnde. Munich: Verlag C.H. Beck.
Geach, Peter. 1969. God and the Soul. London: Routledge & Kegan Paul.
Giannakis, Elias. 1995/96. Fragments from Alexanders Lost Commentary on
Aristotles Physics. Zeitschrift fr Geschichte der arabisch-islamischen Wissen-
schaften 10: 157-87.
Gill, Mary Louise. 1989. Aristotle on Substance: The Paradox of Unity. Princeton:
Princeton University Press.
Glucker, John. 1978. Antiochus and the Late Academy. (= Hypomnemata, Heft 56.)
Gttingen: Vandenhoeck & Ruprecht.
Gottschalk, H.B. 1986. Boethus Psychology and the Neoplatonists. Phronesis 31:
243-57.
Gottschalk, H.B. 1987. Aristotelian Philosophy in the Roman World from the Time
of Cicero to the End of the Second Century AD, in Aufstieg und Niedergang der
rmischen Welt, ed. Wolfgang Haase, 2.36.2: 1079-1174. Berlin: Walter de
Gruyter.
Graf, Fritz. 2009. Apollo. London: Routledge.
Hahm, David E. 1977. The Origins of Stoic Cosmology. Columbus: Ohio State
University Press.
Hahn, Johannes. 1989. Der Philosoph und die Gesellschaft: Selbstverstndnis,
ffentliches Auftreten und populre Erwartungen in der hohen Kaiserzeit. (=
Heidelberger Althistorische Beitrge und Epigraphische Studien, Band 7.)
Stuttgart: Franz Steiner Verlag.
Henry, P. 1960. Une comparaison chez Aristote, Alexandre et Plotin, in Les
Sources de Plotin, 429-49. (= Entretiens sur lAntiquit Classique, vol. 5.)
Vandoeuvres-Genve: Fondation Hardt.
Hicks, R.D. 1907. Aristotle, De anima. With translation, introduction and notes.
Cambridge: Cambridge University Press.
Huffman, Carl. [2005] 2009. Pythagoras. Revised version. Stanford Encyclopedia
of Philosophy. http://plato.stanford.edu/entries/pythagoras/ (First version,
2005.)
Huffman, Carl. 2009. The Pythagorean Conception of the Soul from Pythagoras to
Philolaus, in Frede and Reis 2009, 21-43.
Inwood, Brad. 1985. Ethics and Human Action in Early Stoicism. Oxford: Claren-
don Press.
Inwood, Brad. [1992] 2001. The Poem of Empedocles. A Text and Translation with
an Introduction. Revised edition. (= Phoenix, suppl. vol. 39) Toronto: University
of Toronto Press, 2001. (1st edn, 1992.)
Inwood, Brad. 2009. Empedocles and metempsychsis, in Frede and Reis 2009,
71-86.
Joachim, Harold H. 1903. Aristotles Conception of Chemical Combination. Jour-
nal of Philology 29: 72-86.
Johansen, T.K. 1998. Aristotle on the Sense-Organs. Cambridge: Cambridge Uni-
versity Press.
Johnston, Sarah Iles. 2008. Ancient Greek Divination. Malden, MA: Wiley-Black-
well.
Kahn, Charles H. 1985. Democritus and the Origins of Moral Psychology. Ameri-
can Journal of Philology 106: 1-31.
Katayama, Errol G. 1999. Aristotle on Artifacts: A Metaphysical Puzzle. Albany,
NY: State University of New York Press.
Kerferd, G.B. 1971. Epicurus Doctrine of the Soul. Phronesis 16: 80-96.
172 Bibliography
Kim, Jaegwon. 1993. The Nonreductivists Troubles with Mental Causation, in J.
Heil and A. Mele (eds), Mental Causation, 189-210. Oxford: Oxford University
Press, 1993. Reprinted in Jaegwon Kim, Supervenience and Mind: Selected
Philosophical Essays, 336-57. Cambridge: Cambridge University Press, 1993.
Kupreeva, Inna Veniaminovna. 2003. Qualities and Bodies: Alexander against the
Stoics. Oxford Studies in Ancient Philosophy 25: 297-44.
Lennox, James G. [1997] 2001. Nature does Nothing in Vain }, in H.-C. Guenther
and A. Rengakos (eds), Beitrge zur antiken Philosophie: Festschrift fr
Wolfgang Kullmann, 199-214. Stuttgart: Franz Steiner Verlag, 1997. Reprinted
in Lennox 2001, 205-23.
Lennox, James G. 2001. Aristotles Philosophy of Biology: Studies in the Origins of
Life Science. Cambridge: Cambridge University Press.
Lewis, Eric. 1988. Diogenes Laertius and the Stoic Theory of Mixture. Bulletin of
the Institute of Classical Studies 35: 84-90.
Lewis, Frank A. 2008. Whats the Matter with Prime Matter . Oxford Studies
in Ancient Philosophy 34: 123-46.
Lloyd, A.C. 1962. Genus, Species, and Ordered Series in Aristotle. Phronesis 7:
67-90.
Lloyd, G.E.R. 1966. Polarity and Analogy: Two Types of Argumentation in Early
Greek Thought. Cambridge: Cambridge University Press.
Lloyd, G.E.R. 1987. The Revolutions of Wisdom: Studies in the Claims and Practice
of Ancient Greek Science. (= Sather Classical Lectures, vol. 52.) Berkeley:
University of California Press.
Lloyd, G.E.R. 2006. The Master Cook, in his Aristotelian Explorations, 83-103.
Cambridge: Cambridge University Press.
Lloyd, G.E.R. and G.E.L. Owen (eds). 1978. Aristotle on Mind and the Senses.
Proceedings of the Seventh Symposium Aristotelicum, Cambridge 1975. Cam-
bridge: Cambridge University Press.
Long, A.A. and D.N. Sedley. 1987. The Hellenistic Philosophers. 2 vols. Cambridge:
Cambridge University Press.
Lynch, John Patrick. 1972. Aristotles School: A Study of a Greek Educational
Institution. Berkeley: University of California Press.
Mansfeld, Jaap. 1990. Doxography and Dialectic: The Sitz im Leben of the Pla-
cita. Aufstieg und Niedergang der rmischen Welt 2.36.4, 3056-3229. Berlin:
Walter de Gruyter.
McLaughlin, Brian. 1992. The Rise and Fall of British Emergentism, in Ansgar
Beckermann, Hans Flohr, and Jaegwon Kim (eds), Emergence or Reduction?
Essays on the Prospects of Nonreductive Physicalism, 49-93. Berlin: Walter de
Gruyter.
Menn, Stephen. 2002. Aristotles Definition of the Soul and the Programme of the
De anima. Oxford Studies in Ancient Philosophy 22: 83-139.
Merlan, Philip. 1969. Monopsychism, Mysticism, Metaconsciousness: Problems of
the Soul in the Neoaristotelian and Neoplatonic Tradition. (= Archives interna-
tionales dhistoire des ides, vol. 2.) The Hague: M. Nijhoff.
Moraux, Paul. 1942. Alexandre dAphrodise: Exgte de la notique dAristote. (=
Bibliothque de la Facult de Philosophie et Lettres de lUniversit de Lige,
fasc. 99.) Lige: Facult de Philosophie et Lettres.
Moraux, Paul. 1967. Aristoteles, der Lehrer Alexanders von Aphrodisias. Archiv
fr Geschichte der Philosophie 49: 169-82.
Moraux, Paul. 1973. Der Aristotelismus bei den Griechen: Von Andronikos bis
Bibliography 173
Alexander von Aphrodisias. Band I: Die Renaissance des Aristotelismus im 1.
Jh. v. Chr. Berlin: Walter de Gruyter.
Moraux, Paul. 1984. Der Aristotelismus bei den Griechen: Von Andronikos bis
Alexander von Aphrodisias. Band II: Der Aristotelismus im I. und II. Jh. n. Chr.
Berlin: Walter de Gruyter.
Moraux, Paul. 2001. Der Aristotelismus bei den Griechen: Von Andronikos bis
Alexander von Aphrodisias. Band III: Alexander von Aphrodisias. Ed. von
Jrgen Wiesner. Berlin: Walter de Gruyter.
Morison, Benjamin. 2002. On Location: Aristotles Concept of Place. Oxford: Oxford
University Press.
Mugler, Charles. 1964. Dictionnaire historique de la terminologie optique des
Grecs: Douze sicles de dialogues avec la lumire. (= tudes et Commentaires,
vol. 53.) Paris: Librairie C. Klincksieck.
Nussbaum, Martha C. 1978. Aristotles De motu animalium. Text with translation,
commentary and interpretive essays. Princeton: Princeton University Press.
Oliver, James H. 1970. Marcus Aurelius: Aspects of Civic and Cultural Policy in
the East. (= Hesperia, Suppl. 13.) Princeton: American School of Classical
Studies at Athens.
Oliver, James H. 1977. The diadoch at Athens under the Humanistic Emperors.
American Journal of Philology 98: 160-78.
Oliver, James H. 1981. Marcus Aurelius and the Philosophical Schools at Athens.
American Journal of Philology 102: 213-25.
Opsomer, Jan and R.W. Sharples. 2000. Alexander of Aphrodisias, De intellectu
110.4: I heard this from Aristotle. A Modest Proposal. Classical Quarterly 50:
252-6.
Rashed, Marwan. [1997] 2007. A New Text of Alexander on the Souls Motion,
in R. Sorabji (ed.), Aristotle and After, 181-95. (= Bulletin of the Institute of
Classical Studies 68 (1997), supplement.) Reprinted in Rashed 2007b, 143-57.
Rashed, Marwan. 2005. Aristote, De la gnration et la corruption. Texte tabli et
traduit. Nouvelle dition. Paris: Les Belles Lettres.
Rashed, Marwan. 2007a. Essentialisme: Alexandre dAphrodise entre logique, phy-
sique et cosmologie. (= Commentaria in Aristotelem Graeca et Byzantina:
Quellen und Studien, Band 2.) Berlin: Walter de Gruyter.
Rashed, Marwan. 2007b. Lhritage aristotlicien: Textes indits de lAntiquit.
Paris: Les Belles Lettres.
Rescher, Nicholas and Michael E. Marmura. 1965. The Refutation by Alexander of
Aphrodisias of Galens Treatise on the Theory of Motion. Translated from the
medieval Arabic version, with an introduction, notes, and an edition of the
Arabic text. Islamabad: Islamic Research Institute.
Rose, Valentin. 1863. Aristoteles Pseudepigraphus. Leipzig: B.G. Teubner.
Ross, W.D. 1961. Aristotle De anima. Edited, with Introduction and Commentary.
Oxford: Clarendon Press.
Ryle, Gilbert. 1949. The Concept of Mind. New York: Barnes & Noble, Inc.
Salles, Ricardo. 2001. Compatibilism, Ancient and Modern. Archiv fr Geschichte
der Philosophie 83: 1-23.
Sandbach, F.H. 1985. Aristotle and the Stoics. (= Proceedings of the Cambridge
Philological Society, suppl. vol. 10.) Cambridge: Cambridge Philological Soci-
ety.
Schroeder, Frederic M. 1981. The Analogy of the Active Intellect to Light in the
De Anima of Alexander of Aphrodisias. Hermes 59: 215-25.
174 Bibliography
Sharples, R.W. 1975. Aristotelian and Stoic Conceptions of Necessity in the De fato
of Alexander of Aphrodisias. Phronesis 20: 247-74.
Sharples, R.W. 1983. Alexander of Aphrodisias on Fate. Text, Translation and
Commentary. London: Duckworth.
Sharples, R.W. 1987. Alexander of Aphrodisias: Scholasticism and Innovation.
Aufstieg und Niedergang der rmischen Welt, 2.36.2, 1176-1243. Berlin: Walter
de Gruyter.
Sharples, R.W. 1990. The School of Alexander? in Richard Sorabji (ed.), Aristotle
Transformed: The Ancient Commentators and their Influence, 83-111. London:
Duckworth.
Sharples, R.W. 1994a. Alexander of Aphrodisias, Quaestiones 2.16-3.15. Ancient
Commentators on Aristotle. London: Duckworth.
Sharples, R.W. 1994b. On Body, Soul, and Generation in Alexander of Aph-
rodisias. Apeiron 27: 163-70.
Sharples, R.W. 1998. Theophrastus of Eresus: Sources for his Life, Writings,
Thought, and Influence. Commentary vol. 3.1: Sources on Physics (Texts 137-
223). With contributions on the Arabic material by Dimitri Gutas. (=
Philosophia Antiqua, vol. 79.) Leiden: Brill.
Sharples, R.W. 1999. On being a tode ti in Aristotle and Alexander. Mthexis 12:
77-87.
Sharples, R.W. 2004. Alexander of Aphrodisias, Supplement to On the Soul.
Ancient Commentators on Aristotle. London: Duckworth.
Sharples, R.W. 2005a. Implications of the New Alexander of Aphrodisias Inscrip-
tion. Bulletin of the Institute of Classical Studies 48: 47-56.
Sharples, R.W. 2005b. Alexander of Aphrodisias on the Nature and Location of
Vision, in Ricardo Salles (ed.), Metaphysics, Soul, and Ethics in Ancient
Thought: Themes from the Work of Richard Sorabji, 345-62.
Sharples, R.W. 2008. Alexander Aphrodisiensis, De anima libri mantissa. A New
Edition of the Greek Text with Introduction and Commentary. (= Peripatoi:
Philologisch-historische Studien zum Aristotelismus, Band 21.) Berlin: Walter
de Gruyter.
Shields, Christopher. 2007. The Peculiar Motion of Aristotelian Souls. Proceed-
ings of the Aristotelian Society, suppl. vol. 81: 139-61.
Siegel, Rudolph E. 1968. Galens System of Physiology and Medicine: An Analysis
of his Doctrines and Observations on Bloodflow, Respiration, Humors and
Internal Diseases. Basel: S. Karger.
Siegel, Rudolph E. 1970. Galen on Sense Perception: His Doctrines, Observations
and Experiments on Vision, Hearing, Smell, Taste, Touch and Pain, and their
Historical Sources. Basel: S. Karger.
Sorabji, Richard. [1971] 1979. Aristotle on Demarcating the Five Senses. Philo-
sophical Review 80 (1971): 55-79. Reprinted in Barnes, Schofield and Sorabji
1979, 76-92.
Sorabji, Richard. 1972. Aristotle, Mathematics, and Colour. Classical Quarterly
22: 293-308.
Sorabji, Richard. [1974] 1979. Body and Soul in Aristotle. Philosophy 49 (1974):
63-89. Reprinted in Barnes, Schofield, and Sorabji 1979, 42-64.
Sorabji, Richard. [1987] 2010. Philoponus and the Rejection of Aristotelian Science.
2nd edn. (= Bulletin of the Institute of Classical Studies, Suppl. 103.) London:
Institute of Classical Studies, 2010. (1st edn, 1987.)
Sorabji, Richard. 1988. Matter, Space, and Motion: Theories in Antiquity and their
Sequel. London: Duckworth.
Bibliography 175
Sorabji, Richard. 1991. From Aristotle to Brentano: The Development of the
Concept of Intentionality, in Henry Blumenthal and Howard Robinson (eds),
Aristotle and the Later Tradition, 227-59. (= Oxford Studies in Ancient Philoso-
phy, suppl. vol.)
Sorabji, Richard. 2003. The Mind-Body Relation in the Wake of Platos Timaeus,
in Gretchen J. Reydams-Schils (ed.), Platos Timaeus as Cultural Icon, 152-62.
Notre Dame: University of Notre Dame Press.
Sorabji, Richard. 2004a. The Philosophy of the Commentators, 200-600 AD: A
Sourcebook. 3 vols. London: Duckworth.
Sorabji, Richard. 2004b. Aristotle on Colour, Light, and Imperceptibles. Bulletin
of the Institute of Classical Studies 47: 129-40.
Sprague, Rosamond Kent. 1989. Aristotle and Divided Insects. Mthexis 2: 29-40.
Striker, Gisela. [1983] 1996. The Role of oikeisis in Stoic Ethics. Oxford Studies
in Ancient Philosophy 1 (1983): 145-67. Reprinted in her 1996, 281-97.
Striker, Gisela. [1991] 1996. Following Nature: A Study in Stoic Ethics. Oxford
Studies in Ancient Philosophy 9 (1991): 1-73. Reprinted in her 1996, 221-80.
Striker, Gisela. 1996. Essays on Hellenistic Epistemology and Ethics. Cambridge:
Cambridge University Press.
Thillet, Pierre. 1984. Alexandre dAphrodise: Trait du destin. Paris: Les Belles
Lettres.
Todd, R.B. 1972. Epitdeiots in Philosophical Literature: Towards an Analysis.
Acta Classica 15: 25-36.
Todd, R.B. 1974. Lexicographical Notes on Alexander of Aphrodisias Philosophi-
cal Terminology. Glotta 52: 207-15.
Todd, R.B. 1976. Alexander of Aphrodisias on Stoic Physics: A Study of the De
mixtione with Preliminary Essays, Text, Translation and Commentary. Leiden:
Brill.
Tracy, Theodore. 1969. Physiological Theory and the Doctrine of the Mean in Plato
and Aristotle. Chicago: Loyola University Press.
Tracy, Theodore. 1982. The Soul/Boatman Analogy in Aristotles De anima.
Classical Philology 77: 98-112.
Tracy, Theodore. 1983. Heart and Soul in Aristotle, in John P. Anton and Anthony
Preus (eds), Essays in Ancient Greek Philosophy, vol. II, 321-39. Albany, NY:
State University of New York Press.
Wardy, Robert. 1990. The Chain of Change: A Study of Aristotles Physics VII.
Cambridge: Cambridge University Press.
Watts, Edward J. 2006. City and School in Late Antique Athens and Alexandria.
Berkeley: University of California Press.
West, M.L. 1992. Ancient Greek Music. Oxford: Clarendon Press.
Westerink, L.G. 1959. Damascius, Lectures on the Philebus, wrongly attributed to
Olympiodorus. Text, translation, notes and indices. Amsterdam: North-Holland
Publishing Company.
Whittaker, John. 1990. Alcinoos: Enseignement des doctrines de Platon. Introduc-
tion, texte tabli et comment par John Whittaker et traduit par Pierre Louis.
Paris: Les Belles Lettres.
White, Michael J. 2003. Stoic Natural Philosophy (Physics and Cosmology), in B.
Inwood (ed.), The Cambridge Companion to the Stoics, 124-52. Cambridge:
Cambridge University Press.
White, Nicholas. 1979. The Basis of Stoic Ethics. Harvard Studies in Classical
Philology 83: 143-78.
White, Nicholas. 1985. The Role of Physics in Stoic Ethics, in Ronald H. Epp (ed.),
176 Bibliography
Recovering the Stoics, 57-74. (= The Southern Journal of Philosophy 23, Supple-
ment.)
Wilberding, J. 2011. Porphyry: To Gaurus On How Embryos are Ensouled and On
What is in Our Power. Ancient Commentators on Aristotle. London: Bristol
Classical Press.
Wilkins, Eliza Gregory. 1917. Know Thyself in Greek and Latin Literature. Ph.D.
diss., University of Chicago.
Williams, C.J.F. 1982. Aristotles De Generatione et Corruptione. Translated with
notes. Oxford: Clarendon Press.
Zeller, Eduard. 1923. Die Philosophie der Griechen in ihrer geschichtlichen
Entwicklung. Teil 3.1: Die nacharistotelische Philosophie. 5. Auflage. Leipzig:
O.R. Reisland.
English-Greek Glossary

This Glossary lists key terms in the English translation for both Parts I and II and the
Greek words which they standardly represent. It can thus also serve as an aid to
locating terms in the Greek-English Index. In some cases a family of cognate terms is
represented by one or two leading members.

absence: apousia, stersis agree: homologein, sumphnein


abstractions: ta ex aphaireses agreement: sumphnia
absurd: atopos aim: prothesis
absurdity: atopia aim at: stokhazesthai
abundance: euporia, plthos air: ar
accept: suntithesthai aligned: kat euthurian
acidic: oxus alive, be: zn
acidity: oxuts alone: monos, exhairetos
active: drastikos alter: alloiousthai
activity: energeia, ergon alter, able to: alloitikos
accompany: sunhodeuein alteration: alloisis
accomplished: spoudaios amorphous: amorphos
accrue: prosginesthai amputation: aphairesis
accumulated: athroos analogous: analogon
accurate: akribs analogously: analogs
accurately: akribs analogy: analogia
achieve: epitelein analyse: dialambanein
acquire: lambanein, proslambanein ancient: arkhaios
acquired: epikttos anger: org
acquiring: lpsis angry, become: orgizesthai
acquisition: lpsis animal: zion
acrid: drimus animate: empsukhos
act: poiein, prattein apparatus: hupresia
action: praxis apparent: phaneros
action, capable of: praktikos appear: phainesthai
activate: energein appear through: diaphainesthai
active: drastikos appearance: phantasia
actively: kat energeian appetite: epithumia
activity: energeia apple: mlos
actuality, in: energeiai apply: prosginesthai, epharmozein
add: proslambanein appreciate: manthanein
added, be: prosginesthai, prostithesthai apprehending, for: gnstikos
adding: sunteleia apprehension: gnsis
admit: paradekhesthai approach: prosienai, proshiesthai
advanced: teleios appropriate: oikeion
advanced, less: atelesteros appropriately: oikeis, proskonts
affect (n.): pathos apt: epitdeios
affected, be: paskhein aptitude: epitdeiots
affected first, be: propaskhein aquatic: enhudros
affected, by being: pathtiks arbitrariness: apoklrsis
affirmation, make a: kataphanai argument: logos, sullogismos
affirmative: kataphatikos army: stratopedon
age: graskein arrange: paraskeuazein
178 English-Greek Glossary
arrangement: thesis blood, containing: enhaimos
art: tekhn blood vessel: phleps
artefact: to kata technn ginomenon bloodless: anhaimos
artificial: tekhnikos blushing: eruthma
artisan: tekhnits body: sma
ascent: anhodos body of: athroos
ascertaining: heursis bodily: smatikos
asleep, be: koimasthai bone: osteon
assemble: athroizein born, be: apotiktesthai
assimilated to, become: exhomoiousthai bound (v.): peratoun, horizein
assistant: epikouros boundary: peras
assistance: botheia bow: prra
assume: hupolambanein brain: enkephalos
assumption: lambanomenon breath: pneuma
attached to, naturally: prosphuesthai brick: plinthos
attribute (v.): katgorein, anatithenai bright: leukos
audible: akoustos bring together: sullegein
auditory: akoustikos broader sense, in a: koinoteron
augmentative: auxtikos bronze: khalkos
authoritative: kurios build in: enkatoikodomein
avoid: phugein, ekklinein, apostrephein
avoided, to be: pheuktos calm: hsukhia
avoiding: phug cannot be heard: ankoustos
awake, be: egrgorthai cannot be seen: ahoratos
aware, being: antilambanesthai cannot be smelled: anosphrtos
aware, capable of being: antilptikos carry: okhein, pherein
aware conjointly: sunaisthanesthai carry along with: summetapherein,
awareness: antilpsis sunanapherein
awareness, accompanying: sunaisthsis cart: hamaxa
carve: temnein
backwards: opis cease: sumpauesthai, pauein,
bad: kakos methistasthai
balance: summetria, mesots centre: kentron
balance, appropriate: to summetron chameleon: khamailen
ball: sphaira change: kinsis, metabol
basic: prtos change, able to produce: kintikos
basis: hupobathra change, produce: kinein
be together: sunhuparkhein change first: prokinein
becoming: genesis change jointly: sunkinein
being: to einai, on change place: methistasthai
belief: doxa changing place: metastasis
belief, have a: doxazein character: poion
believe: doxazein characteristic: oikeion
believed, can be: doxastos characterisation: logos
believing, for or capable of: doxastikos characterise: katgoreisthai
bench: bathron choice: prohairesis
between: metaxu choice, worthy of: hairetos
bile: khol choosing: hairesis
bitter: austros chosen, be: krinein
black: melan circle: kuklos
blade: sidros circumference: periphereia
blanching: khrots circumscribe: apokrinein
blank: agraphos circumstance: peristasis
blend (n.): krasis, krama cithara: kithara
blend (v.): kerannunai class (v.): perilambanein
blend, good: eukrasia cleanse, able to: rhuptikos
blood: haima clear: saphs, phaneros
English-Greek Glossary 179
clear, able to make: dltikos consider: lambanein
clear, make: dloun consistency: sustasis
clear, making: dlsis constitution: sustasis
clearly: saphs construct: kataskeuazein
cloak: himation construction: kataskeu
close: engus contain: periekhein
close shut: muein contact: haph
clothing: skepasma continuity: sunekheia
cognition: krisis continuous: sunekhs
coincide: sunhaptein, epharmozein contour: morph
cold: psukhros contract: sustellein
cold, grow: apopsukhesthai contraction: sustol
coldness: psukhrots contradistinguish: antidihairein
collide with: prospiptein contrariety: enantisis
colour (n.): khrma contrary: enantios
colour (v.): khrmatizein, khrizein contrast: antitithenai
colour, of the same: homokhroos contribute, make a contribution:
combination: parathesis suntelein, sunergein
combined: koinos contribution: sunteleia
come together: sunhodos control: one who is in control of himself,
coming into being: genesis enkrats; one who is not in control of
coming-to-be: genesis himself, akrats
co-mingle: mignunai converse: antistroph
command: prostassein convert: antistrephein, methistanai
commensurateness: summetria conviction: pistis
common: koinos convinced, be: pisteuein
common sense: koin aisthsis cook: pessein
communicate: mnuein cooking: pepsis
communicate, able to: mnutikos cool down: katapsukhein
communication: mnusis cooling: empsuxis, psuxis
community: koinnia cooling down: katapsuxis
completion: entelekheia correspond: hupokeisthai
complex: poikilos corresponding: antikeimenos, kata
compose: sunkeisthai cough: bttein
composite: sunamphoteros, to cover up: epikaluptein
sunkeimenon culmination: teleiots
composition: sunthesis cultivation: exergasia
compound (adj.): sunthetos cure: hugiazein
compound (n.): sunthesis cut: temnein
compound (v.): sunkeisthai cut across: enallattein
comprehend: perilambanein
comprehension: perilpsis damage: apleia
compresent with: suneinai dancer: orkhsts
conceive: lambanein, ennoein dancing, for: orkhstikos
concept: noma dark: skotos
conception: ennoia, epinoia darkness: skotos
conclusion: sumperasma death: thanatos
concur: suntrekhein decay: anesis
confine: eneilein declare: proagoreuein
conflate: sunhaptein decree: prokruttein
conflict: makh deficient, be: apoleipesthai
conformity: oikeiots define: horizein
confused, be: planan definition: horismos, logos
congenital: sumphutos, sumphus definition to, give: horizein
connect by passages: suntetrainein deliberate: bouleuesthai
consequence: akolouthos, hepomenon deliberation, for or capable of:
consequence, be a: hepesthai, akolouthein bouleutikos
180 English-Greek Glossary
delirium: paraphrosun distinctive: diaphoros
deliver: parapempein distinctly: enargesteron, trans
delivery: parapomp distinctively: diapheronts
demonstrative: apodeiktikos distinguish: khrizein
density: pakhuts distribute: paraspeirein
denial, make a: apophanai distrust: apistein
deny: anteipein disturb: enokhlein
depend: rtsthai disturbed, not easily: duspaths
depth: eisokh diversified: polutropos
derivatively: deuters diversity: parallag
desirable: orektos divide: dihairein, skhizein
desire (n.): orexis divine: theios
desire (v.): oregesthai divisible into parts: meristos
desiring, for: orektikos division: dihairesis
destroy: phtheirein do: prattein
destruction: phthora done, ought to be: prakteos
destructive: phthartikos done, can be: praktos
detach: apoluein, aphairein downwards: kat
detail, in: prosekhs drawing: graph
detail, lack: epipolaios drug: pharmakon
detect: phran drunk, to be: methuein
determinate: aphrismenos, hrismenos dry: xros
develop: teleioun dryness: xrots
developed, highly or fully: teleios duality: to ditton
development: epidosis, teleisis dye, murex: halourgs
deviate: parallattein
diagonal: diametros ear: ous
die: apothnskein earth: g
differ: diapherein earthy: geds
difference: diaphora, heterots echo (n.): kh
differentiate: diarthroun echo (v.): khein
digest: kateirgnunai effect: pathos
digestion: katergasia effluence: aporrhoia
dilate: aneurunein element: stoikheion
directed towards: anaphora eliminate: aposkeuazesthai
direction: diastasis emerge above: gennasthai epi
directly: ep eutheias, ex eutheias emission: anathumiasis
discern: krinein emit: prohiesthai
discernible: gnrimos empty: kenos
discriminate: diakrinein enable: paraskeuazein
discriminating: diakrisis, epikrisis enclose: periekhein, enapolambanein
disparity: parallag encounter: peripiptein
disperse: thruptein end: telos
dispersion: thrupsis end product: perittma
disposed in a certain way: ps ekhein endorse: sunkatatithenai
disposition: hexis endorsement: sunkatathesis
dispositional: kath hexin, en hexei endorsement, for: sunkatathetikos
dispute (n.): amphisbtsis endpoint: peras
dispute (v.): amphisbtein engages in activity: energein
dissection: anatom enlarge: sunauxein
dissimilar in genus: anhomogens enmattered: enhulos
dissimilar types, belonging to: ennumeration: katarithmsis
anhomoeids enquire further: epiztein
distance: diastma, apostma entailed by: hepesthai
distance, be at a: aphistasthai enter: eiseinai, epeisienai
distinct: kekhrismenos entrance: eishodos
distinct, be: kekhristhai error: dihamartia, plan
English-Greek Glossary 181
error, be in: diapseudesthai fire: pur
essence: ousia, logos, to ti n einai fire, set on: proskaiein
esteem, hold in the highest: presbeuein first: prtos
eternal: aidios fit: epharmozein
eunuch: eunouchos fitness: epitdeiots
evidence: smeion, marturion flavour: khumos
evident: enargs, dlos flavoured: enkhumos
evident, be: dloun flavourless: akhumos
examine in detail: parakolouthein flee: phugein
exceed: huperballein flesh: sarx
excess: huperbol, pleonasmos flow: rhein
excess, be in: pleonazein foam: aphros
excited, be: organ foggy, be: epikaluptein
exclusively: idis foliage: khlra
exhale: ekpnein follow: akolouthein, akolouthos,
exhibit: emphainesthai hepesthai, epakolouthein
exist: huphistasthai foot: pous
exist in conjunction with: parhuphistanai foot-long: podaios
exist prior: prohuphistasthai footed: pezos
existence: hupostasis force: bia
expansion: auxsis forced: biaios
experience (n.): empeiria forceful: biaios
experience (v.): paskhein forcibly: biais
explanation: paramuthia foreknow: progignskein
expound: ektithenai fork: ektrop
express: diasmainein, smainein form (n.): eidos
expressing oneself: hermneia form (v.): sunhistanai
extend: diatenein, metapherein formation: sustasis
extend through: diienai, dierkhomai, formless: aneideos
diaphuein forwards: pros
extended sense, in an: kata metaphoran foundation: arkh
extension: diastma fragrance: eudia
extension or something extended: frog: batrakhos
megethos fruit: karpos
exterior: peras fulfil: plroun
external: ektos fulfilment: eudaimonia
extraneous: allotrios full: plrs
extraordinary: amkhanos fully: holoklrs
extreme: akros function: ergon, skopos
extrinsic characteristic: sumbebkos fundamental sense, in the: kuris
extrinsically: kata sumbebkos
eye: ophthalmos, omma, opsis general (adj.): koinos
eyes interior: kor general (n.): stratgos
eyelid: blepharon generally: katholou
exiting: khusis generate: gennan
generation: genesis
facilitate: paraskeuazein gentle: remaios
fall: piptein genus: genos
fall short: elleipein go against: antiprassein
fall under: hupopiptein go away: aperkhesthai
false: pseuds god: theos
falsehood: pseudos gold: khrusos
falsehood, possess: pseudesthai goldish: krusoeids
far away: porr good: agathos
fear: phobos govern: dioikein
fill: plroun governing part: hgemonikon
filled: mestos grace: euskhmosun
182 English-Greek Glossary
grafted onto, be: prosgignesthai idea: noma
grant: keisthai ill, be: nosein
grasp (n.): gnsis illness: nosos
grasp (v.): lambanein, gnrizein illuminate: phtizein
grasping: lpsis imagination: anazgraphsis
greenish: pods imitate: mimeisthai
ground: edaphos immediately: euthus
grow, able to: auxtikos immortality: athanasia
grow, cause to: auxein impact: plg
growing: auxsis impact, make a: plssein
growing, capable of: auxtikos impede: empodizein
growth: auxsis imperceptible: anaisthtos
guide: hodgein imperishable: aphthartos
guided by smell: huposmos impinge: prospiptein
impression: tupsis
habit: ethos impression, make: tupoun
hair: thrix imprint: tupos
hand: kheir in many ways: pleonakhs
hard-eyed: sklrophthalmos inactive: anenergtos
hardness: sklrots inactivity: anenergsia
harm: blab inanimate: apsukhos
harmonise: epharmozein incapable of coming-to-be: agentos
harmony: harmonia incapable of perishing: aphthartos
harmony, be in: harmozein include: periekhein
hate: misein incommensurable: asummetros
head: kephal incorporate: proskrinein
health: hugieia increase (n.): epidosis
healthy: hugiainein increase (v.): epiteinein
hear: akouein incredible: paradoxos
heard, cannot be: ankoustos incredibly: paradoxs
hearing: ako indemonstrable: anapodeiktos
heart: kardia indication: smeion
heat (n.): thermots, to thermon indistinctly: amauroteron, amaurs
heat (v.): thermainein individually: idiai
heaviness: baruts inferior to, be: apoleipesthai
heavy: barus infinitely many: apeiros
help (n.): botheia infinity: apeiria
help (v.): bothein inform: eidopoiein
high-pitched: oxus ingest: prospheresthai
highest: teleutaios, kuritatos inhalation: anapno
hit: plssein inhale: anapnein
hold: pisteuein inherence: huparxis
hold in: katekhein initially: prts
hold together: sunekhein inner organs: splankhnon
hollow (adj.): koilos inquiry: ztsis
hollow (n.): koilots insane, be: parakoptein
homogeneous: homoiomers inscribe: engraphein
homonymously: homnums insecure: akataleptos
honey: meli insensitive: anaisthtos
honey water: melikraton inseparable: akhristos
honeyed wine: oinomeli inside: endon
hot: thermos instantaneously: akhrons
house: oikia instruction: didaskalia
human (being): anthrpos instrumental: organikos
humour: khumos intelligible: gnrimos
intend: boulesthai
I: eg intense: sphodros
English-Greek Glossary 183
intention: boulsis loss: apobol
intention, for: boultikos love (v.): philein, eran
intermediate: metaxu love, object of: ermenos
internal: entos lover: erasts
intervening: metaxu low-pitched: barus
interweave: sumplekein lower: kat
intrinsically: kath hauto luminous: phtoeids
investigate: historein, ztein lungs: pleumn
involve: khrsthai lustre: eukhroia
irrationally: alogs lyre: lura
itself just on its own: auto kath hauto
magnificent: perittos
join: sunagein, epharmozein, magnificently: peritts
epizeugnunai maim: proun
judgement: hupolpsis maimed (n.): prma
juxtaposition: parathesis main: kephalaios
maintain: phulassein
keep: phulassein make: kataskeuazein
know: ginskein, epistasthai, eidenai make up: anaplroun
know, not: agnoein management: oikonomia
knower: epistmn manifest in addition: paremphainein
knowing: gnsis marketplace: agora
knowledge: gnsis, epistm marvel: thauma
known, what is: episttos material: hulikos
mathematical: mathmatikos
lack (n.): stersis, aporia matter: hul
lack (v.): sterein mature (n.): teleios
language: logos mature (v.): teleioun
largeness: megethos measure: metrein
last: eskhatos medical: iatrikos
lead: prohgeisthai medium: to metaxu
learn: manthanein melody: melos
leave along with: sunaperkhesthai melted, can be: tktos
leave behind: kataleipein membrane: humn, khitn
left, on the: aristeros memory: mnm
leg: skelos message: prosangellia
less clear: asaphesteros metal: metalleutos
life: bios, z, to zn metempsychosis: metempsukhsis
light: phs middle, in the: mesos
lightness: kouphots mirror: katoptron
like: homoios missing: leipein
like, become: homoiousthai mistake: apat
likeness: homoiots mistakes, make: apatasthai
likening: homoisis mix: mignunai
limit (n.): peras, horos mix in: enkatamignunai
limit (v.): perainein mixed, can be: miktos
line: gramm mixture: mixis
lingering: paramonos modified, be: paskhein
link: suzeugnumi modification: diathesis
live: zn modify: diatithenai
liver: par moist: hugros
located in: hidrusthai moisten: hugrainein
long for: ephiesthai moisture: to hugron, hugrots
longed for: ephetos moment, for just a: akariaios
longing: ephesis mortal: thntos
look: blepein motion: phora
lose: sterein mould further: prosanatupoun
184 English-Greek Glossary
move: kinein, pherein outline: tupos
multiply: pleonazein outside: ex
murex dye: halourgs outside, from: exthen
muscle: mus overflow: huperballein
mutual: koinos own, its: oikeios
oyster: ostreon
nail: onux oysters, similar to: ostreds
name (n.): onoma ox: bous
name (v.): onomazein
name, call by: onomazein pain, feel: lupeisthai
narrowing: stenots painful: lupros
natural: phusikos painter: grapheus
natural endowment, good: euphuia pairing: sunduo
naturally: kata phusin palpitation: pdsis
naturally less gifted: aphuesteros part: meros, morion
naturally more gifted: euphuesteros particular: kathekaston
nature: phusis pass: khrein
nature, be by: phuein passing: parhodos
nature, in accordance with: kata phusin passion: thumos
necessary: anankaion patent: phaneros
neck: trakhlos path: hodos
needle: belon perceive: aisthanesthai
negative: apophatikos perceive, what can: to aisthtikon
non-bodily: asmatos perceived, able to be: aisthtos
non-bodily entities: ta asmata perceiver: ho aisthanomenos
non-evident: adlos perceiving, capable of: aisthtikos
non-homogeneous: anhomoiomers perceptible: aisthtos
non-rational: alogos perception: aisthsis
nose: rhis perceptive: aisthtikos
nourish: trephein perceptual organ: aisthtrion
nourished, capable of being: threptikos percolation: dithsis
nourishing: trophimos pericardium: perikardios
nourishment: troph perish: phtheiresthai, diaphtheiresthai
now: nun perish along with: sumphtheiresthai
number: arithmos, plthos perishable: phthartos
number, in: kat arithmon perishing: phthora
numerically: kat arithmon persist: menein, summenein,
hupomenein, szesthai
obey: peithesthai perspicuous: saphs
object: pragma persuade: peithein
oboe: aulos phrase: logos
oboe, for playing: aultikos picture: anazgraphma
oboe, play: aulein pile: sros
occur first: prohgeisthai pilot: kubernts
odorous: osmds piloting, of or for: kuberntik
odour: osm piper: aults
odourless: aosmos place: topos
oesophagus: stomakhos place directly on: epitithesthai
ointment: emplastos placing: thesis
old age: gras plainly: phaners
on its own: kath hauto planning: boul
opposed to, be: diaballesthai plant: phuton
opposition: antithesis plausibility: pithanots
order (n.): taxis pleasant: hdus
order (v.): tassein pleasure, feel: hdesthai
organ: organon porous: araios
origin: arkh portrait: eikn
English-Greek Glossary 185
position: thesis Pythian god: Puthios
position, change in: metabasis
power: dunamis, exousia qualify: proskeisthai
power, in its: dunamei quality: poiots
practical: praktikos qualityless: apoios
practise: askein query (v.): ztein
precede: progignesthai quick: takhus
predispose against: prodiaballein
pre-eminently: prts rage: org
prefer: prokrinai rational: logikos
premiss: protasis ready: epidektikos
prepared: hetoimos reason (n.): logos, aition, aitia
preponderant, become: pleonazein reason (v.): logizesthai
presence: parousia reasoning, capable of: logistikos
present (v.): phantazesthai rebound (v.): anaklan
present, be: pareinai rebound (n.): anaklasis
preservation: steria, trsis receive: dekhesthai, lambanein
preserve: szein, trein received, to be: dektos
preserve, able to: sstikos, trtikos receptacle: hupodokh
prevent: kluein receptive: epidektikos, dektikos
prime: paraskeuazein recollect: anamimnskein
primarily: prohgoumenos recollection: anamnsis
primary: prtos recovered: episkeuastos
primary perceptive part: to aisthtikon recovery: apokatastasis
principle: arkh reference, can make: endeiktikos
principle of change or movement: arkh reflect: ephistanai
ts kinses reflect: emphainesthai
probative: deiktikos reflected image: emphasis
probe: pslaphoun refute: elenkhein
proclaim: prolegein regard: phronein
produce: poiein, empoiein regulation: oikonomia
produced, be: gignesthai reinvigorated: episkeuastos
production: genesis reject: apostrephein
productive: poitikos relation: skhesis
progress (v.): proienai relative: pros ti
progression: prohodos relax: anhienai
proof: deixis relay (n.): diadosis
proper: oikeion relay (v.): diadidonai
proportion: analogia, logos release: analuein
proportional: analogon remain: menein, summenein
protect against: phulassesthai remember: mnmoneuein
protrusion: exokh remove: exhairein
prove: deiknunai repel: apthein
provide: khorgein repelled, be: apostrephesthai
provider: khorgos replacement: antiperistasis
proximate: prosekhs report: diangellein, eisangellein
proximately: prosekhs represent: phantasiousthai
pull: helkein represented, what is: to phainomenon
pure: katharos representation: phantasia, phantasma
purplish: porphuroeids representing, for or capable of:
pursue: metadikein phantastikos
pursued, to be: diktos reproduce: gennan homoion hauti
push: thein reproducing, for or capable of: genntikos
put together: parakeisthai reproduction: gennsis
puzzle: aporia resist: antitupein
puzzling: aporos resistance: antereisis, antitupia
puzzled, be: aporein resolution, come to a: suntithesthai
186 English-Greek Glossary
resources: khorgia shape (v.): skhmatizein
responsible: aitios shape, with a definite: hrismenos,
responsible, what is: aitia horistos
responsible factor: to aition shape, without a definite: ahoristos
responsive: sumpaths shapeless: askhmatistos
responsiveness: sumpatheia share in, have a: koinnein, metekhein,
rest: remia, argia metalambanein
rest, be at: remein shift (n.): metastasis
rest, bring to: remizein shift (v.): methistasthai
retain: szein, stegein shine: stilbein
return: anastrephein ship: naus
return back again: palindromein shortage: endeia
reveal: dloun shortfall: endeia
revenge, take: timreisthai show: deiknunai
reverence, have great: presbeuein shrink: meiousthai
rhythm: rhuthmos shudder (n.): phrixis
right, in its own: kath hauto shudder (v.): phrittein
right, on the: dexios side: pleura
root: rhiza sight: opsis
rot: spesthai signet ring: sphragis
roughness: trakhuts similar: homoios
rower: erets similar kind, of a: sungens
rudder: oiax simple: haplous
simply as such: hapls
safe: asphals simultaneously: hama
salty: halmuros sinew: neuron
same in form: homoeids size: poson
sand: ammos slate: pinakis
savoury: liparos slight, make a: oligrein
say: legein smallness: mikrots
sea: thalatta smell: osphrsis
sea creature: thalassios smell, perceive: osmasthai
seat, have ones: kathidrusthai smellable: osphrantos, osphrtos
second: deuteros smelling, capable of: osphrantikos
secure: katalptikos smelling, for: osphrtikos
secure grasp, attain a: katalambanesthai smith: khalkeus
securing: katalpsis smithing, for: khalkeutikos
see: horan smooth: leios
seed (n.): sperma smoothness: leiots
seed (adj.): spermatikos soak: dihugrainein
seeing: horasis sober, to be: nphein
seen, cannot be: ahoratos softness: malakots
sense: aisthsis solid: stereos
sensible: eulogos solidity: pxis, stereots
sensibly: eulogs song: melos
sensitivity, greater: euaisthsia soul (n.): psukh
separable: khristos soul (adj.): psukhikos
separate (adj.): khristos sound (n.): psophos
separate (v.): khrizein sound (v.): khein
separated: kekhrismenos sound, emit a: psophein
separation: khrismos sound, capable of emitting: psophtikos
sequence: hodos, taxis sour: struphnos
sequester: diaspan source: pg
serially: kata meros sow: kataballein
set out: ektithenai sow seed: speirein
shakes: tromos spatially: metabatiks
shape (n.): skhma speak: legein
English-Greek Glossary 187
specific: aphrismenos surround: perilambanein, perikeisthai
specific thing: tode ti survive: szesthai
speech: dialektos sustaining: sstikos
speed: takhos, takhuts sweet: glukos
spicy: pikros synopsis, give a: sullabein
spin: dinein
spontaneous: automatos tactile: haptikos
spot: sunhoran take on in addition: proslambanein
sprouting: blastsis take up: analambanein
star: astr take to be the case: hupolambanein
start: arkhein tangible: haptos
state: legein task: ergon, hupresia
statue: andrias tastable: geustos
stay: menein, diamenein taste (n.): geusis
steep: enapoplunein taste (v.): geuesthai
stem (n.): kaulos taste, what can: geustikos
stem (v.): ekphuein temperate: eukratos
stew: pessein temple: nes
stick: baktria tend: apoklinein
stone: lithos tendency: rhop
stored away: apokeisthai term: onoma
strength: sphodrots terminate: teleutan
stretch over: periteinein testify against: antimarturein
stretching: ektasis theory: dogma
strike: empiptein, tuptein things there are: ta onta
string: khord think: dianoeisthai, phronein
striving: horm thinking, for or capable of: dianotikos
striving, for or capable of: hormtikos this, a: tode
structure: kataskeu thought: epinoia, nosis, phronsis
studied, can be: thertos tinge: khrnnunai
study, engage in: theorein tinge anew: anakhrnnunai
study: theria time: khronos
study, for engaging in: theretikos time, at the same: hama
subject: hupokeimenos, pragma total: holoklros
submit: hupotassesthai tongue: gltta
subsequent: deuteros tool: organon
subsequent, be: hepesthai tortoise: khelon
subserve: hupretein touch: haph
subservient: hupretikos touch, able to: haptikos
substance: ousia trace: enkataleimma
such, a: toionde trace back: anagein
suffice: arkein train: progumnazein
sufficient: hikanos transfer: metapherein
suitable: epitdeios transform: metaballein
suitable condition: epitdeiots transformation: metabol
summary, in: suntoms transition: metabasis
sun: hlios translate: metapherein
superficial: epipolaios transmit: diakoneisthai
supervene: epigignesthai transmit, able to: diakontikos
surface: epiphaneia, epipedon transmitter: diakonos
surpass: pleonektein transmitting, for: diakonikos
superfluously: matn transodorant: diosmos
superior: timiteros transonant: dikhs
supplement: prostithenai transparency: diaphaneia
supplementation: prosthk transparent: diaphans
suppose: hupolambanein trap: apolambanein
surprised, be: thaumazein treat: therapeuein
188 English-Greek Glossary
treatise: logos, pragmateia view: doxa
tree: dendron view, come into: hupopiptein
trigger, able to: metabltikos virtue of, in: kata
true: alths visible: horatos
trumpet: salpinx visual: horatikos
trunk: thrax vital: ztikos
truth: to alethes vividly: sphodros
truth, possess: altheuein vocal: phntikos
turn back: anakamptein voice: phn
twofold: dittos voice, have a: phnein
type: eidos
walk: badizein
unclear: asaphs walk, take a: peripatein
uncooked: apeptos walking, capable of: poreutikos
uncover: apokaluptein wall: toikhos
uncovered: askeps wash, able to: pluntikos
undergo a modification: paskhein water: hudr
underlying (subject): (to) hupokeimon wax: kros
undermine: anhairein weak: amudros
understand: noein, manthanein, akouein weave: huphainein
understand, able to: notikos weaver: huphants
understandable: notos weaving, for: huphantikos
understanding: nous, to noein well-being: to eu einai, to eu
understanding, act of: nosis well-blended: eukratos
undeveloped: atels wet dream: oneirgmos
undigested: akatergastos what is: to on
undispersed: athruptos what is not: to m on
undivided: adihairetos, askhistos whip: mastix
unenmattered: ahulos whole: holos
unify: henoun wide: eurus
unity: hensis wind-pipe: artria
universal: katholou wine: oinos
universe: kosmos wing: pterux
unlike: anhomoios winged: ptnos
unmixed: amigs wish: boulesthai
unmodifiable: apaths withdraw: hupokhrein
unnamed: annumos womb: gastr
unpleasantness: duskhereia wood: xulos
untransformed: ametabltos wool: erion
up to us: eph hmin work: erga
upwards: an workings: erga
use (v.): proskhrsthai, katakhrsthai, worm: sklx
khrsthai wrestle: palaiein
used to, become: ethizesthai wrestler: palaists
useful: khrsimos wrestling, for: palastikos
write: graphein
vain, in: matn write down: antigraphein
vaporous: atmidds writing tablet: grammateion
variation: trop
vegetable: lakhanon yielding: eiktikos
vegetative: phutikos you: su
vessel: angeion youth: neots
vestige: ikhnos
Greek-English Index

The present index lists occurrences of Greek terms of interest, together with their
English translations, for both Parts I and II. Although it is meant to be fairly complete,
it should not be assumed to be exhaustive: for a full listing and accurate count of the
occurrences of a term, the reader should use a computer search on the TLG. Note also
that the general listing at the beginning of an entry in particular is not complete: if a
citation is given under a more specific sub-entry within an entry, or is cross-listed as
being under another entry, the citations are given only under that specific sub-entry or
cross-listed entry. The entries and sub-entries are alphabetised by their English trans-
literation, not the Greek alphabet. If an occurrence crests over a line, only the first line
number is given. A range of lines indicates that there are at least two occurrences of the
term within two lines of each other throughout the range.

aer, air, 6,24; 8,21; 19,22.27.35; 26,17; is perceived, 84,7; to kat energeian
42,7; 44,5.9; 47,1-2.7.11.15.18.22.25; aisthesthai, to perceive actively, 69,4
48,1.4.8-12; 49,10-12.16; aisthsis, perception, perceiving, 30,13;
50,12-14.20.24; 52,18; 53,5.8.14; 57,3; 32,26; 38,20; 39,6-9; 40,4.9; 41,12;
58,21-2; 59,15; 62,5 43,15; 55,2; 56,18-20; 57,7-9.13-16.19;
adlos, non-evident, 72,20 58,2.19; 61,6.22; 62,17; 64,3;
adihairetos, undivided, 48,13; 61,20; 66,11-13.23-7; 67,2-5.8-9.23-6;
63,19.27 68,18.21-3.27-9; 69,12-14; 70,1.7.10;
agathos, good, 89,3-4; to malista kai 71,6; 72,3.16-17; 73,3.9; 74,3-5.22;
prts agathon, what is good 75,17.20-3; 76,10; 77,3.11;
pre-eminently and to the highest 78,10.13-14.18; 83,9.16.19-21;
degree, 89,2 84,4.12; 86,16; 87,4-5.12.19;
agentos, incapable of coming-to-be, 91,12.22-4; 92,23; 93,2; 97,3.11;
6,14 sense, 18,28; 19,1; 39,11.17;
agnoein, not know, 33,7 40,13-14.21.26; 41,2-4.11; 51,9.25;
agora, marketplace, 13,19 52,16; 54,5; 55,3.15.18-19.26-7; 56,5;
agraphos, blank, 84,25-8 57,6.12.18; 58,7-8; 59,8.20.23;
ahoratos, cannot be seen, 52,14 60,1.3.8-10.14-15.19-20.23; 61,8.25-6;
ahoristos, lacking a definite shape, 62,1; 65,4-6.12.20-1; 66,3-5; 68,20;
44,15; 46,8. For ahoriston diaphanes, 73,4-6; 74,26-7; 81,4; 83,2-6; 86,7.10;
see under diaphans 93,18.22-4; kat energeian aisthsis,
ahulos, immaterial.For ahulon eidos, perception in activity, 12,18; 39,14;
see under eidos 68,27.31; 69,2.16; 70,2-3.19; 83,13;
aidios, eternal, 6,14; 81,7; 82,18 87,1; 91,8; koin aisthsis, common
aisthanesthai, perceive 12,13; 16,15; sense, 62,20; 63,6.13; 65,3.9.17;
32,26; 39,13; 40,26; 51,10; 52,21; 78,10.12; 97,13; kuria aisthsis,
54,24-55,1; 56,21; 57,1; 58,14.18; principal perception, 65,8; prt
59,6; 60,20.26; 61,1-2.11-12.18-19; aisthsis, primary perception, 65,8.
62,5; 63,3.21-2; 64,10; 65,3-6.9-10; For geustik aisthsis, see under
66,28; 67,6-7; 78,6.10; 79,9.16; geustikos; for haptik aisthsis, see
83,4.20; 84,6-7; 87,2-3; 91,21; 92,11; under haptikos
93,11-15.23; 96,19-20; ho aisthtrion, perceptual organ, 12,14;
aisthanomenos, the perceiver, 29,6; 19,1; 24,10; 41,4.16-20; 44,2-6; 55,3;
58,6; 65,9; cf. 57,1; 60,26; 65,4; to 57,4.7; 58,3-4.9-12; 59,1.13; 60,4.15;
aisthanesthai, perceiving, 29,7; 40,19; 62,2.14.17; 63,14-18.22-4;
65,10; 84,5; to aisthanomenon, what 64,2-4.10.21-2; 65,1; 66,2-3; 69,12;
190 Greek-English Index
78,8; 84,16; 98,3; 99,29; kathekaston 36,7.19.26; 37,4; 46,6; 66,17; 70,21;
aisthtrion, particular perceptual 73,18.28; 77,9; 80,2.5; 88,22; 89,9;
organ, 78,11; to eskhaton aisthtrion, 98,22; reason, 17,5; 22,22; 45,10
the ultimate perceptual organ, aitios, responsible, 13,1; 15,2; 21,20.25;
63,15.18; 64,8.18; to prton 22,1-2.16; 44,14.25; 46,3; 48,22;
aisthtrion, the primary perceptual 50,17; 57,9; 58,7; 68,10; 73,6.11;
organ, 60,6; 68,6; 69,22 75,15; 84,10; 88,24-6; 89,2-5.10;
aisthtikos, perceptive, able to 94,18; reason, 2,16; 51,9; 58,7; to
perceive, 29,21.22; 36,16; 52,1; 60,8; prton aition, the first cause, 89,9.18
85,24; 92,22; 97,7-8; to aisthtikon, akariaios, for just a moment, 77,9
the part for perceiving, what is akatalptos, insecure. For akatalptos
perceptive, 39,14; 57,8.16.20; 63,7; phantasia, see under phantasia
68,28; 69,13; 73,25-6; 74,15.28; akatergastos, undigested, 33,17; 95,1;
75,1.5-15; 76,1.4.12; 96,25; 97,7
99,3.7.20-2; 100,15; aisthtik akhristos, inseparable, 6,19; 12,22;
energeia, perceptual activity, 69,3; 15,13; 17,9; 21,22; 22,15. For
75,19; aisthtikon organon, akhristos ts hul, inseparable from
perceptual organ, 57,18; prton matter, see under hul
aisthtikon, the primary perceptive akhrons, instantaneously, 43,11
[part], 50,16.19. For aisthtik akhumos, flavourless, 53,9-10.18;
psukh, see under psukh; for 54,7.12
aisthtik dunamis, see under akintos, incapable of undergoing
dunamis; for aisthtik antilpsis, see change, 17,10; 21,24; 78,25
under antilpsis ako, (the sense of) hearing, 40,11.23;
aisthtos, perceptible, able to be 46,20; 50,23-4; 52,11-13.18; 55,20;
perceived, 24,11; 39,1.4.9.11.13.20; 56,6; 59,15; 60,18; 78,9
40,17-18; 40,20-42,3; 52,16; 55,4; akolouthein, follow, be a consequence,
56,5; 57,10-11.15.19; 58,8.11-12; 59,9; be next, 19,26; 77,17
60,2.9.15.20-1.24; 61,1.4-6.9.23-5.29; akolouthos, consequence, what follows
62,1.15-16.19; 63,1-2.5.14-16.28; or comes next, 27,3; 73,17; 96,10;
64,2.20-1; 66,26-8; 68,5.8.22-3.27; 98,28
69,7-11.13.15-16.23-5; 71,1; 72,8-9.12; akouein, hear, 42,2; 48,22-49,3; 50,8-18;
76,13; 78,11; 83,6.10; 84,19-20; 85,21; 51,11-12; 65,3-8; 78,9; 83,3;
86,7.11; 87,2-4.11-13.22; 91,11; 93,12; understand, 72,6; to akouein, (the
97,13; aisthton eidos, perceptible activity of) hearing, 65,8
form, 60,5; 66,14; 78,8; 83,13.16; akoustikos, auditory. For akoustik
87,2.5; 91,9; entos aisthtos, internal psukh, see under psukh
perceptible, 69,1; ektos aisthtos, akoustos, audible, 50,8.11; 51,7; 52,12;
external perceptible, 69,2.5; idion 55,27; 56,10; 65,8
aisthton, perceptible exclusively by akrats, one who is not in control of
one sense, 40,21-41,1; 41,3.13-15; himself, 27,8
42,4; 56,2; 65,11.21; 66,15; 67,4; akribs, accurate, 51,9.25; 52,4; 67,8-9;
83,18; cf. idion pathos, 43,20; akribs, accurately, 28,17; 50,15; 67,5
kathekaston aisthton, particular akros, extreme, 59,10
kind of perceptible, 78,11; kath hauto alths, true, 2,4.25; 7,6; 23,6.24; 57,2;
aisthton, intrinsically perceptible, 66,10; 67,4.10.13-14; 70,8.23;
40,20-41,5; 41,10-42,3; 70,6.14; kata 71,4.9-10.13-14.20; to alethes, truth,
sumbebkos aisthton, extrinsically 67,22; 70,7
perceptible, 40,21; 41,6-10; 70,9; altheuein, possess truth, 41,14;
koinon aisthton, common 66,10-12.15.18
perceptible, 41,1.10; 56,3; 65,10; 70,9; alogos, non-rational, 67,12; alogs,
83,17; sphodra aisthton, vividly irrationally, 85,5. For alogon zoion,
perceptible, 86,10. For aistht ousia, see under zoion; for alogos psukh,
see under ousia see under psukh
aitia, something responsible, 5,16; alloisis, alteration, 12,17; 39,2.12-15;
8,15.17.20; 9,10; 10,17.21.23; 15,20; 50,11; 53,28; 54,21; 55,14; 77,11; 84,13
16,7; 22,7.11.20; 24,12.13; 32,16; alloitikos, able to alter, 54,18
Greek-English Index 191
alloiousthai, alter, 44,12; 59,8 anhairein, undermine, 27,13; 28,12
allotrios, extraneous, 54,24; 84,16 anhinai, weaken, 26,10; relax, 77,1.4
amauros, indistinct; amauroteron, more anhodos, ascent, 97,1
indistinctly, 50,26; amaurs, anhomoeids, belonging to dissimilar
indistinctly, 67,6 types, 61,9
amkhanos, extraordinary, 2,12 anhomogens, dissimilar in genus,
ametabltos, untransformed, 34,24 63,28
amigs, unmixed, 89,12 anhomoios, unlike, 39,11.15.17; 54,22;
ammos, sand, 47,23 55,4; 59,7
amorphos, amorphous, 4,1 anhomoiomers, non-homogeneous,
amphisbtein, dispute (v.), 10,1 98,4-5
amphisbtsis, dispute (n.), 55,17 an, upwards, 5,11; 7,20; upper, 97,27
amudros, obscure, 71,10. For amudra annumos, unnamed, 56,13
phantasia, see under phantasia anosphrtos, cannot be smelled, 52,15
anagein, trace back, 58,28; 99,18 anteipein, deny, 12,19
anaisthtos, insensitive, 75,1; 93,13; antereisis, resistance, 21,15
imperceptible, 77,12 anthrpos, human being, 2,1; 13,13.17;
anakamptein, turn back, 48,9 23,14.15.17; 27,16; 30,12; 36,15;
anakhrnnunai, tinge anew, 43,19 49,25; 51,10; 52,1.6; 71,24; 73,16;
anaklan, rebound (v.), 47,13 74,7; 80,12.22; 81,13.27-8; 83,2;
anaklasis, rebound (n.), 47,14.25 85,16-18; 98,27; 99,11; 100,5; koinos
analambanein, take up, 61,31 anthrpos, human in common, 85,16
analogia, proportion, 10,15; 11,12; antidihairein, contradistinguish,
analogy, 20,20; 21,9; 24,26 13,14; 65,21
analogon, proportional, 10,17; antigraphein, write down, 85,2
analogous, 24,23; 39,6; 51,14.21; 57,4; antikeimenos, corresponding, 52,13.16;
58,3.24; 68,27-9; 75,26; 93,17; 94,26; 86,24
97,10.29; analogs, analogously, 89,23 antilambanesthai, be aware of, 44,12;
analuein, release, 95,10 50,25; 56,15; 59,18-19; 62,8.19;
anamimnskein, recollect, 23,16 78,15.19; 83,9.19; 86,8.11
anamnsis, recollection, 69,19 antilpsis, awareness, 24,11; 41,18-19;
anaphora, in the phrase tn anaphoran 42,9; 44,6; 50,18.23; 51,6.12; 53,2;
ekhei, is directed towards, 28,9; 39,9; 54,2; 55,1.16; 58,5.16.20; 59,3; 60,2;
73,23; 76,12 61,28-9; 62,6; 65,11; 72,9; 83,9.17-18;
anaplroun, fill or make up, 13,17 87,13; aisthtik antilpsis,
anapnein, inhale, 49,9.16.20.23; perceptual awareness, 54,21; 83,15;
49,25-50,3; 52,22-53,1; 98,16 98,14
anapno, inhalation, 52,23 antilptikos, for awareness/being
anapodeiktos, indemonstrable, 66,19 aware, capable of awareness/being
anastrephein, return, 48,11 aware, 39,5; 41,15-17; 46,20; 50,10;
anathumiasis, emission, 3,18 53,28; 54,19; 56,16; 58,15; 61,13.26;
anatithenai, attribute, 79,24 69,8; 93,12. For antilptik dunamis,
anatom, dissection, 96,26 see under dunamis
anazgraphma, picture, 68,6 antimarturein, testify against, 71,3
anazgraphsis, imagination, 69,25; antiperistasis, replacement, 48,10
70,18 antiprassein, go against, 2,19
andrias, statue, 3,7-9; 11,20; 14,2; 16,3; antistrephein, convert, 7,6
18,18-23 antistroph, converse, 51,3
aneideos, formless, 4,1.2 antitithenai, contrast, 71,15
ankoustos, cannot be heard, 52,13 antithesis, opposition, 86,25
anenergsia, inactivity, 74,27 antitupein, resist, 47,19
anenergtos, inactive, 39,8; 75,8 antitupia, resistance, 48,17
anesis, decay, 50,20 aosmos, odourless, 53,10
aneurunein, dilate, 52,23 apatasthai, make mistakes, 41,11; 51,4
angeion, vessel, 13,19; 14,23-4; apat, mistake, 51,1
48,4-5.15-17 apaths, unmodifiable, cannot be
anhaimos, bloodless, 97,4 modified, 59,4; 86,9; 89,11.15-16
192 Greek-English Index
apeiria, infinity, 28,5 Aristotels, Aristotle, 16,5; 50,12; 51,16;
apeiros, infinitely many, 17,19; 54,17; 56,13; 58,4; 59,1; 66,4; 89,18;
28,4.6.12 91,2; ta Aristotelous, Aristotles
apeptos, uncooked, 33,14.17.25; 95,1 writings, 2,4; to peri psukhs, On the
aperkhesthai, go away, 42,22-3; 45,4; Soul, 2,6; 66,4; to peri aisthsen [or
63,1 aisthses] te kai aisthtn, On the
aphairein, detach, 37,4.8.12; 100,11 Senses [or Perception] and
aphairesis, abstraction; amputation, Perceptibles, 51,16; 53,26; 58,21; ta
100,8; ta ex aphaireses, abstractions, peri geneses kai phthoras, On
90,9 Coming-to-be and Perishing, 58,28; ta
aphistasthai, be at a distance, 58,15-16 peri zin kinses, On the Motion of
aphrismenos, determinate, 4,14; Animals, 97,26
specific, 37,17; 38,3 aristeros, on the left, 97,25
aphros, foam, 41,7 arithmos, number, 30,28; 31,1.6; 41,5;
aphthartos, imperishable, incapable of 65,14; kat arithmon, in number,
perishing, 6,15; 89,16; 90,12-14.20.23; numerically, 14,13; 46,14-16; 61,20;
91,1.4 94,8; 99,20
aphuesteros, naturally less gifted, 81,28 arkein, suffice, 84,12
apistein, distrust, 71,20-1 arkhaios, ancient, 91,24
apobol, loss, 4,20 arkh, origin and principle, 9,3;
apodeiktikos, demonstrative, 66,16 10,20.24; 24,12; 35,6; 37,1; 66,19;
apoios, qualityless, 17,17; 18,2 71,27; 72,5; 77,22-3; 84,10; 91,25;
apokaluptein, uncover, 52,23 92,1-3.6-7; 94,30; 95,24; 96,27;
apokatastasis, recovery, 100,6 99,21-3; beginning, origin, 10,6.19;
apokeisthai, be stored away, 86,5 21,21; 36,20; 40,2; 70,19; 95,11;
apoklrsis, arbitrariness, 22,25 97,1-2.19.22-5.28; (acc.) initially,
apoklinein, tend, 45,17 first, in the first place, at the outset,
apokrinein, to separate off, 10,27; 37,15; 40,14; 41,9; 47,1; 48,21;
circumscribe, 15,16 72,9; 85,21-3; foundation, 66,19;
apolambanein, trap, 47,7.10; 48,9; arkh ts auxses, source of growth,
50,24 94,21.24; cf. 9,3; arkh praxes,
apleia, damage, 98,22 principle of action, 73,18; 81,10;
apoleipesthai, be deficient, 52,7; be arkh ts kinses, origin and
inferior, 85,3 principle of change or movement,
apoluein, detach, 21,26 2,23; 5,10; 7,18.19; 9,2; 73,18.25.28;
apophanai, make a denial, 80,1 77,7.15; 97,16.25-6; 98,15; cf. 97,22;
apophatikos, negative, 67,21 arkh tou einai, source of being,
apopsukhesthai, grow cold, 95,10 89,9-10; arkh tou trephesthai, origin
aporia, puzzle, 2,15; lack, 27,12.15; and principle of being nourished, 9,2;
72,11 arkh tou zn, principle of life, 95,13;
aporein, be puzzled, 6,6.7; 30,22.26; 96,2; arkh tou zoiou, principle of the
56,4.14 animal, 77,13; kuria arkh,
aporos, puzzling, 61,24; 62,3 controlling principle, 98,21
aporrhoia, effluence, 42,21 arkhein, start, 48,20
aposkeuazesthai, eliminate, 95,18 artria, wind-pipe, 49,11.16
apostma, distance, 65,14; 93,24 asaphs, unclear; asaphesteros, less
apostrephein, reject, 72,3; (pass.), be clear, 32,23
repelled, 73,29; avoid, 93,3.7 askein, practise, 82,2
apthein, repel, 48,3.6.18 askeps, uncovered, 53,1
apothnskein, die, 29,7 askhmatistos, shapeless, 4,1.12; 34,3
apotiktesthai, be born, 74,21 askhistos, undivided, 95,4
apousia, absence, 30,15 asmatos, non-bodily, 4,18; 6,7; 7,14;
apsukhos, inanimate, 19,9; 29,4; 34,13; 15,13; 17,10; 63,18; ta asmata,
49,5; 79,8.10. For apsukhon sma, see non-bodily entities, 7,14
under sma asphals, safe, 95,13-15; 98,9
araios, porous, 98,19 astr, star, 71,18
argia, rest, 74,26 asummetros, incommensurable, 72,23
Greek-English Index 193
atels, undeveloped, 35,20; 92,14; boulsis, intention, 73,2; 74,2.7-8.11-12;
94,12-15; atelesteros, less advanced, 78,23; 80,9; 81,11; 99,1-2
16,20; 17,5; 28,16.20.22; 94,11.16 boulesthai, intend, wish, 2,2.18; 42,2
athanasia, immortality, 36,8 boultikos, for intention, 99,1
athroizein, assemble, 23,24 bouleuesthai, deliberate, 73,11; 74,12;
athroos, a body of, 47,7.10; 80,6
accumulated, 86,6; all at once, 91,18 bouleutikos, for deliberation,
athruptos, undispersed, 50,14 deliberative, 82,16.19; to bouleutikon,
atmidds, vaporous, 3,18 the part for deliberation, 98,28;
atopia, absurdity, 20,18 99,2.23. For bouleutik dunamis, see
atopos, absurd, 18,8; 19,24; 20,8.15; under dunamis; bouleutikos nous, see
27,16 under nous; for bouleutik orexis, see
aulos, oboe, 49,7 under orexis
aulein, play oboe, 23,20 bous, ox, 78,28
aults, oboist, 23,20
aultikos, for playing oboe, 23,20 deiknunai, show, 11,14; 12,18; 13,9;
austros, bitter, 51,24; 55,11 16,2.5; 17,9; 19,1.24; 24,4.13; 38,13;
auto kath hauto, see under kath hauto 39,22; 58,22; 59,1; 66,1.4; 79,20; 81,5;
automatos, spontaneous. For automat 89,17; 91,10; 92,1; 96,10; 97,14.27;
genesis, see under genesis 98,1; 100,13; prove, 80,8-11
auxein, grow, cause to grow, 32,10; 34,6; deiktikos, probative, 100,9
35,24; auxesthai, grow, 9,13; 12,12; deixis, proof, 99,31
16,15; 31,9; 32,21; 34,4.14.22; dekhesthai, receive, 3,12;
35,8-12.18; 36,13.22.26; 38,1 20,11.12.13.16; 31,14-20; 38,6; 42,21;
auxsis, growing, growth, expansion, 43,2; 44,2.24; 50,15; 51,20; 53,6;
9,3; 29,10; 30,6; 36,21.24; 77,1; 78,6; 61,30; 62,2-3; 64,5; 81,14; 82,4;
92,18; 94,25. For arkh ts auxses, 91,14.17; 95,2
see under arkh dektikos, able to acquire or receive,
auxtikos, able to cause growth or receptive, 37,15; 44,3.20.23; 53,4;
enable growing, 34,6; 36,1. For 60,4; 81,24; 82,4; 83,1; 88,20; 91,13;
auxktik dunamis, see under 92,22. For dektik dunamis, see
dunamis under dunamis
dektos, to be received, 39,1
badizein, walk, 23,8; 79,16 dlos, evident, 10,21; 60,24
baktria, stick, 69,21 dloun, be evident, 12,24; make clear,
barus, heavy, 8,22; 79,4; low-pitched, 27,2; reveal, 30,18
34,2; 55,20 dlsis, making something clear, 40,18
baruts, heaviness, 9,18-19; 14,2; 20,21; dltikos, able to make clear or explicit,
22,7-10.18-20; 23,29-30; 55,23; 16,18; 28,19
low-pitched, 34,2; 55,28 Dmokritos, Democritus, 27,6
bathron, bench, 25,17-18 dendron, tree, 8,25; 9,7
batrakhos, frog, 27,16 deuteros, second, subsequent; deuters
belon, needle, 47,6 (contrasted with kuris), in a
bttein, cough, 50,6 derivative or secondary sense, 8,22;
bia, force, 15,17; 48,3; 79,11 31,19; 89,8. For prtos/deuteros, see
biaios, forced, 21,15; forceful, 60,13; under prtos
biais, forcibly, 47,13 dexios, on the right, 97,25
bios, life; ho kata phusin bios, the life in diaballein, be opposed to, 71,20
accordance with nature, 1,9; 2,2 diadidonai, relay (v.), 39,21; 42,18;
blab, harm, 98,20 43,20; 48,16; 64,8; 77,6
blastsis, sprouting, 94,30 diadosis, relay (n.), 41,5; 48,15.20; 63,16
blepein, look, 62,10 diakoneisthai, serve, 39,20; transmit,
blepharon, eyelid, 52,25 41,20; 42,9; 52,20; 62,5.12; 95,1
botheia, assistance, 87,28 diakontikos, able to transmit, 59,14
bothein, help, 98,20-1 diakonikos, for transmitting, 100,16
boul, planning, 74,7.12; 82,17 diakonos, transmitter, 41,3; 94,28
diakrinein, discriminate, 92,26; 93,7
194 Greek-English Index
diakrisis, discriminating, dierkhesthai, extend through; sma
discrimination, 49,14.18 dia smatos dierkhesthai, body
dialambanein, analyse, 2,3.26; 40,18 extending through body, 20,7-19
dialektos, speech, 49,6.15 dithsis, percolation, 54,11
diamenein, continue, 100,10 dihairesis, division, 18,19.24-5; 30,8;
diametros, diagonal, 72,23 31,3-6; 47,20
diangellein, report, 64,2.20 dihairein, divide, distinguish, 11,16;
dianoeisthai, think, 16,15; 78,15; 99,5 13,15; 31,1-4; 38,8; 62,21; 63,8; 73,23;
dianotikos, able to think, 76,11-12. 91,11; 94,3; 97,28; cut, 98,12
For dianotik dunamis, see under dihamartia, error, 41,12; 70,21
dunamis to di hou, that by means of which, 35,4
diaphainesthai, appear through, 43,5 dihugrainein, soak, 53,3
diaphaneia, transparency, 44,21; diienai, extend through, 14,8
45,5.10 dinein, spin, 21,28
diaphans, transparent, 42,1.7; dioikein, govern, 30,16
42,20-43,9; 43,15; 44,1.5-7.15-19.23-5; diktos, to be pursued, 77,16-17
45,1.7-8.11-17.20-1; 46,1.5-9.12.17; Din, Dion, 13,19
52,19; 57,11-13; ahoriston diaphanes, diosmos, transodorant, 51,20; 52,19;
transparent [material] without fixed 53,5.15
boundaries, 44,5.7.15.23; 45,7; see dittos, in two ways, 16,8; 33,13; twofold,
also hrismenos, under hrizein, and 80,24; 81,23; to ditton, duality, 3,14;
horistos; kat energeian diaphanes, 33,15
actively transparent, 42,6.10; dogma, theory, 2,6; 20,18
43,4-6.12 doxa, belief, view, 2,5; 14,20; 20,8;
diapherein, differ, 8,8.24; 9,1.5.7; 27,18; 66,12.15.23; 67,13.16-18.21-6;
10,10; 11,9; 16,11; 19,34; 21,19; 26,18; 71,3; 72.2
27,17.18; 55,23; 56,4; 57,18; doxastikos, for or capable of believing,
60,19-23.26; 61,8.10.13.16-18.21.24; 81,9-10. For doxastik dunamis, see
63,11.21.28; 64,1.19; 65,2; 69,6; 74,14; under dunamis; for doxastikos nous
75,3.11-12; 76,7; 81,1-2.5; 83,15; 92,19 see under nous
diapheronts, distinctively, 52,1 doxastos, can be believed, 81,7
diaphora, difference, differentiation, doxazein, believe, have a belief, 67,15;
differentiating feature, 1,4; 3,22; 72,2-3; 78,15.20; 99,4
6,26-7; 7,10.13; 8,7.14.18.20; 9,10; drastikos, active, 75,24
10,14.17.19.20.21.25; 11,8; 14,22; drimus, acrid, 51,24; 52,10; 55,11
27,20; 28,1; 30,7.22; 45,10-14; 46,6; dunamis, power, 1,3; 2,16.18; 5,13-15;
51,25; 52,8-9; 56,1; 58,13.25; 9,15-26; 11,1.4; 16,8-10.12.21; 17,1;
60,17.20-4.27; 61,2-4.6-10.28; 62,21; 19,33; 20,1-6; 22,2.6; 23,26-4,2;
63,6.25.28; 64,3; 69,20; 71,7.10; 24,23.28; 25,2.8; 26,11.21.27.29;
76,9.13; 77,9.14; 78,3.21; 81,27; 27,5.8-11.17.19; 28,3.5-6.11-14.21.28;
85,17-18; 87,11; 88,23; 94,2.9.14 29,8-9.15.23-6; 30,1-4.10-11.19.22;
diaphoros, distinctive, 8,16; 10,29; 31,2.8; 32,2-8.25.28; 33,8; 34,25;
63,7.17; diaphors, distinctively, 28,1 35,2.4.15.21.25; 36,9.12.19;
diaphuein, extend through, 94,29 37,1.3.10.14; 38,4-6.10.14.18.21;
diaphtheiresthai, perish, 93,8 39,20; 46,2; 50,10; 53,27.30; 59,1-2;
diapseudesthai, be in error, 70,10 60,4-7.16; 63,17.22; 64,9.19-20; 65,5;
diarthroun, differentiate, 95,7 66,18; 68,23-4; 73,27; 74,13.20.25;
diasmainein, mention, 17,4; 28,18 75,2.25; 76,5-9.17; 78,18;
diaspan, sequester, 96,25; 99,16 80,16.19-21.24; 81,1.14; 82,7-9;
diastasis, direction, 9,3 85,6.11; 86,20; 89,17; 90,15-16; 91,3;
diastma, extension, 14,19; distance, 92,10.15.18-20; 94,2-4.9-10.16.23;
41,17-18; 50,19.22.25; 51,2 95,25-6; 96,11.15-17.26; 97,6;
diateinein, extend, 50,17; 100,15 99,10-12.18.25-7; aisthtik dunamis,
diatithenai, modify, 86,8 power for perceiving, 19,2;
diathesis, modification, 80,2 29,7.11.14-17; 38,18; 39,4; 40,10.16;
didaskalia, instruction, 81,25; 83,1 57,5.17; 60,10; 62,18; 63,15.20; 64,18;
dikhs, transonant, 52,19 66,8; 68,7.22; 69,1.6; 73,15; 74,25;
Greek-English Index 195
96,16.20-3; 97,6-8; 98,2; antilptik 30,16; 31,9; 32,4-8.21.27; 33,1-2;
dunamis, power for awareness, for 35,15.26; 36,19; 37,8; 38,14.18; 39,2;
being aware, 53,30; 55,13; 56,21; 74,23.27; 94,16-17.26; 95,8;
auxtik dunamis, power for growing, 96,1.6.14.20-4; 97,9. For dunamei
29,3; 31,9; 35,16; 38,15; bouleutik nous, see under nous
dunamis, power for deliberating, duskhereia, unpleasantness, 95, 19
29,24; dektik dunamis, power for duspaths, not easily disturbed, 95,14
receiving, 66,13; deutera dunamis,
second power, 30,5; cf. 30,4; edaphos, ground, 42,16
dianotik dunamis, power for eg, I (first person pronoun), 61,1
thinking, 73,15; doxastik dunamis, egrgorthai, be awake, 66,27
power for believing, 29,24; dunamei, eidenai, know, 60,25
as a power, by (its) power, 16,12; eidos, form, 3,24.27;
39,6-8; 55,4; 84,22; 85,10; 88,19; 4,1-2.10.16.19-21.26-7;
epistmonik dunamis, power for 5,1.4-5.10-11.22; 6,3.6.18-19; 6,22-8;
knowing, 29,24; genntik dunamis, 7,2-4.7-13.17.24-5; 8,7-10.14-15.24;
power for reproducing, 29,3; 30,16; 9,5.11.15-16.19.24-5;
32,20; 36,6; 38,15; gnstik dunamis, 10,3-4.8.10-14.16; 11,1.6-8.12.20;
power for apprehension, 76,8; 87,19; 12,7-8.24; 13,9.12-15.24; 14,11-17;
haptik dunamis, power for touch, 15,11.23-6.28-9; 16,2.4.5.9;
55,16; 56,15-17; horatik dunamis, 17,11-18,10; 18,14.18;
power for seeing, 44,10-11; hormtik 19,4.5.21.23.24; 20,1.6; 20,19-21,10;
dunamis, power for striving, 21,18.22.23.24; 22,6.9.14.16; 23,18.30;
29,11.18; 73,18; 74,13; 77,23; 97,17; 24,2.13.19; 25,2; 26,9.22;
98,1; kata dunamin, as a power, 31,10.12.17.23.25; 35,1; 36,12; 38,13;
51,18; 54,18; 86,1.4; kata topon 52,19; 60,8.11-12; 64,1.19.22; 65,2;
kintik dunamis, power for changing 79,20; 81,25; 82,1; 83,15.22-3;
place, 30,14; khris dunames, 84,7.14-15.23.25; 85,11-19; 86,19.29;
separate from power, 89,17; kintik 87,2.5-7.10-14.21.24-6;
dunamis, power to change, 8,24-6; 88,6.14-15.21.25; 89,17; 90,3;
power for movement, 98,3; koin 91,12-14.17; 92,10; type, 29,22; 49,3;
dunamis, a common power, 95.25-6; 51,14-15; 55,5; 61,25; ahulon eidos,
kritik dunamis, power for having immaterial form, 66,18; 88,2; 89,19;
cognition, 54,1; 55,13; 65,16; 66,13; eidos khris (ts) huls, form separate
73,14-16; 76,8; 80,22; 99,27; kuritat from (the) matter, 39,13; 87,21; 92,22;
dunamis, highest power, 30,11.15; enhulon eidos, enmattered form, 16,2;
92,12; lptik dunamis, power to 87,5.29; 88,4.11; 89,13.17; 91,8; 90,2;
receive, 66,18; logik dunamis, power kuritaton eidos, highest form,
to reason, rational power, 30,3; 39,23-40,1; meta eidous, conjoined
66,7-8; 73,15; 80,21; 81,5; 94,3; 99,7; with a form, 4,6; oikeion eidos, proper
notik dunamis, power for form, 43,6; phusikon eidos, natural
understanding, 29,24; orektik form, 6,3; sun eidei, joined with a
dunamis, power for desiring, form, 4,9; to epi pasin koinon eidos
29,11.16-17; 74,14; 97,17; 98,1; autois, the form over them all and
phantastik dunamis, power for common to them, 8,11; eidos eidn,
representation, 29,12.16-19; 66,9; form of forms, 8,12; topos eidn, place
68,13.30-1; 69,5.8; 70,15; 73,15.20; of forms, 85,6-9. For aisthton eidos,
praktik dunamis, power for action, see under aisthtos; for noton eidos,
99,27; prt dunamis, first power, see under notos; for eidos
9,12; 28,21-4; 29,3; 30,5-6; 32,4.6; nooumenon, see under noein
38,12; 80,19. (See also prt psukh, eidopoiein, informs, 20,23
under psukh); psukhik dunamis, eikn, portrait, 68,9
soul power, 12,13; 16,11; 17,7; 29,22; eiktikos, yielding, 98,19
30,2.7; 32,4.19; 36,27; 38,9.17; 49,10; einai, to be; m on, that which is not,
62,22; 66,6; smatik dunamis, bodily 70,25; on, (a) being, 14,21; (pl.)
power, 2,18; threptik dunamis, beings, 7,10; 61,5; 84,18-21.27;
power for nourishing, 29,2.6.12-15; 91,10-16.20-2.26; 92,2.5.10; to einai,
196 Greek-English Index
being, 5,22; 6,21; 10,3.7; 13,24; 14,21; 55,19-22.24-7; 56,1.8; 58,27; 59,5-6.9;
15,3; 18,3; 21,9; 26,28; 35,15; 36,20; 93,15
49,22; 59,22; 61,5; 73,3; 75,31; 84,21; enapolambanein, enclose, 50,12
87,25-7; 88,13.16; 89,8; 90,4.9; enapoplunein, steep, 53,12.14.16
91,12.18; 93,16.18-19; to (ti) X einai, enargs, evident, 2.25; 62,4; 81,15;
what it is for (an) X to be, what it is enargesteron, more distinctly 50,25.
to be (an) X, Xs being, 4,13.17; 7,9; For enargs phantasia, see under
10,8; 15,4.24-6; 23,15; 26,24-5; 39,23; phantasia
40,1; 75,5; 87,8-9; see also to tide endeia, shortage, shortfall, 13,6; 59,20
einai, under tode; (touto) ho esti, just endeiktikos, can make reference, 17,7
what it is, 6,21.23.25.29; 7,5; endon, inside, 57,5
15,25.29; to ti n einai, essence, 87,24; eneilein, confine, 43,13
88,11. For arkh tou einai, see under energeia, activity, 9,15-26; 10,29; 12,21;
arkh; for to eu einai, see under eu 16,9; 21,1.3; 23,22.23.25; 24,6.8;
eisangellein, report, 76,13 27,6.10-14.20; 31,12; 32,25;
eiseinai, enter, 15,18-21 33,3-5.9-10; 34,27; 35,2.18.22-7;
eishodos, entrance, 15,21 36,2.6; 37,3; 38,10; 39,2-4.11-12;
eisokh, depth, 71,18; 72,6 40,16-17; 43,8; 44,10-12.15; 50,9;
khein, make a sound, 26,1.2; echo, 51,18; 54,18; 65,8; 68,5.13.20.30;
47,14 69,11-13.15.18.23; 70,4.6.20; 73,4-6;
kh, echo, 47,25 74,9.25; 75,4.6.16-17.28; 76,4.17;
ekklinein, avoid, 73,29 77,22; 78,2.4.7; 80,17; 81,1; 82,8-10;
ekphuein, stem, 94,30 83,1.6; 85,3.22.26; 86,10.14; 89,17;
ekpnein, exhale, 50,1-4 91,7; 97,12; 100,12; aisthtik
ektasis, stretching, 74,21; 77,5 energeia, perceptive activity, 69,3. (cf.
ektithenai, set out, 2,8; 92,14 39,5); energeiai, in actuality, 37,2;
ektos, external, 13,1.4; 39,9-10; 44,25; 46,18; 55,4.12; 84,21-2; 85,5; 90,16;
50,21; 52,18; 58,7.11; 69,9; 79,7. For kat energeian, as an activity,
ektos meros, see under meros; for actively, in actuality, 16,13, 43,5;
ektos aisthton, see under aisthton 46,20-1; 51,21; 53,7; 73,5; 85,8; 86,23.
ektrop, fork, 10,22 For kat energeian aisthsis, see
elleipein, fall short, 59,6 under aisthsis; for kat energeian
elenkhein, refute, 65,19 diaphanes, see under diaphans; kat
empeiria, experience, 83,8 energeian nous, see under nous; for
empiptein, strike, 79,11 kat energeian phantasia, see under
emphainesthai, exhibit, 19,32; reflect, phantasia; kath hormn energeia,
62,14; 71,19 activity due to striving, 76,17; 77,7;
emphasis, reflected image, 44,2 cf. 49,5; 50,8; kata psukhn energeia,
emplastos, ointment, 25,1 activity pertaining to soul, 75,28;
empodizein, impede, 45,9 praktik energeia, practical activity,
empoiein, produce, 42,22; 77,1 71,22; psukhik energeia, soul
empsuxis, cooling: 49,12.21; 98,18 activity, 12,9; 35,11; to peri ho h
empsukhos, animate, 12,3; 14,7.21; energeia, the object towards which
15,19; 16,16; 19,8; 20,7; 21,4.10; the activity is directed, 33,10
23,22; 24,11; 27,20; 29,1.3.6; 30,8; energein, be active, engage in, actively
32,3.5.16.17; 34,26; 49,7; 63,5; 74,23; exercise, 12,11.24; 16,17; 21,4;
77,16; 79,3; 80,2; 92,16; 94,21; 23,11.14.22-3.27-8; 24,4; 27,10.14.15;
96,14-16; prts empsukhos, 28,1.6; 31,11.12; 32,2.3.5.26; 36,27;
pre-eminently animate, 24,5. For 37,10; 39,3.6-7; 68,14.23; 69,9; 70,24;
empsukhon sma, see under sma; to 74,20.24; 75,7; 78,1; 83,3; 85,26;
tn psukhn ekhon, see under psukh 86,2-4.25; 89,23; 100,13
enallattein, cut across, 71,10 engraphein, inscribe, 85,1
enantios, contrary, 5,7; 15,1; engus, close, 50,21.26-7; 51,2; 55,9
33,15.21.24-8; 34,1.5-7; 46,17; 54,8; enhaimos, containing blood, 97,6-7
59,11; 61,21-4.30-1; 64,3-6.11-20; en hmin, in us, 57, 9; 68,5; 72,9.12;
65,2; enantis, just the opposite, 71,8 77,20; 79,19; 99,9
enantisis, contrariety, 5,20-1; enhudros, aquatic, 13,14; 52,20
Greek-English Index 197
enhulos, enmattered, 18,10; 83,19; epiphaneia, surface, 18,11-12; 45,21-2;
87,23; 88,11; 90,5; cf. hs en huli 46,15-16
under hul. For enhulon eidos, see epipolaios, superficial, 13,3; lack detail,
under eidos 71,6
enkatoikodomein, build in, 49,1; 50,13 episkeuastos, reinvigorated, 36,8
enkataleimma, trace, 38,10; 63,3; epistasthai, know, 78,15; 99,5-6;
68,7.10.27; 69,1-2.16; epistmn, knower, person who
70,4-5.9.12.15.19.23; 72,8.11; 97,13 knows, 39,8; 86,1-4
enkatamignunai, mix in, 54,16 epistm, knowledge, (pl.) branches of
enkephalos, brain, 99,31; 100,5 knowledge, 39,7; 66,12.16.23;
enkhumos, flavoured, 53,7.11.25; 54,10. 67,9-11.25; 72,25; 82,3.6.9; 86,2-3;
For enkhumos xrots, see under 89,23; 90,1
xrots; for enkhumos hugrots, see epistmonikos, for or capable of
under hugrots knowing, 82,19. For epistmonik
enkrats, one who is in control of dunamis, see under dunamis; for
himself, 27,8 epistmonikos nous, see under nous
ennoein, conceive, 80,12.14 episttos, what is known, 66,17; 81,8
ennoia, conception, 77,3; 85,23 epitdeios, apt, suitable, fit, 13,5; 36,23;
enokhlein, disturb, 42,2 41,20; 98,14
entelekheia, completion, 16,1.6-10; epitdeiots, aptitude, 12,25; suitable
17,12; 24,1-2; 52,2; 78,25-6; 81,26; condition, 37,1; fitness, 81,14;
entelekheiai, in actuality, 54,20; prt 84,24.28; 85,3
entelekheia, first completion, 16,7-10; epiteinein, increase, 26,10
21,5 epitelein, achieve, 99,29
entos, internal, inside, 45,5; 49,20; epithumia, appetite, 12,16; 13,2; 74,2-3;
56,16; 69,1. For entos meros, see 78,23
under meros; for entos aisthton, see epitithesthai, place directly upon,
under aisthton 57,16.19; 58,1.10
epakolouthein, follow, 80,6 epiztein, enquire further, 60,22
par, liver, 95,2; 97,7 epizeugnunai, join, 63,9
epeisienai, enter, 50,13 eran, love (v.); ermenos, object of love,
epharmozein, harmonise, 2,17; fit, 79,6
28,19; coincide, 63,10-12.24; apply, erasts, lover, 79,6
86,26 remaios, gentle, 50,22-3
ephesis, longing, 36,7; 79,7 remein, be at rest, 22,25-6; 23,2-7; 86,6
ephetos, longed for, 79,6 remia, rest, 22,24; 23,1; 41,5; 65,13;
eph hmin, up to us, 67,1; 73,11.12; 83,21
ouk eph hmin, not up to us, 66,28; remizein, bring to rest, 22,26
73,8 erets, rower, 79,20
ephiesthai, long for, 32,15 ergon, function, task, 2,26; 22,27; 23,3;
ephistanai, reflect, 6,7 32,12; 36,21; 49,12.18; 65,17; 83,12;
epidektikos, receptive, can receive, 5,8; 97,12; (in pl.) workings, activities,
10,26; 15,1; 81,19 work, 2,14.16; 32,10; 69,3
epidosis, development, 30,6; 94,15; erion, wool, 47,5
increase, 35,17; 36,20 rtsthai, depend, 12,18
epigignesthai, supervene, 25,2; 26,22; eruthma, blushing, 77,14
79,26; gignesthai + epi, 10,25; 24,19 eskhatos, last, ultimate, 29,22; 94,4;
epikaluptein, cover up, 52,24; (perf. 94,28; 99,12; (pl.) extremities, 95,11.
pass.) be foggy, 71,25 For eskhaton aisthstrion, see under
Epikouros, Epicurus, 26,17 aisthstrion
epikouros, assistant, 98,16 ethizesthai, become used to, 83,5
epikrisis, discriminating, 71,27 ethos, habit, 81,25
epinoia, thought, 6,17.20; conception, eu, well; to eu einai, well-being, 49,19;
14,9 81,18-19; 93,20; to eu, well-being,
epikttos, acquired. For epikttos nous, 49,23; wellness, 81,20
see under nous euaisthsia, greater sensitivity, 52,6
epipedon, surface, 50,27 eukhroia, lustre, 31,4
198 Greek-English Index
eukintos, easily moved, 48,8 35,17.24; 36,16; to gennsan, parent,
eukrasia, good blend, 52,5 37,2
eukratos, well-blended, 52,1; gennasthai epi or ek, emerge above or
temperate, 77,2 from, 5,6; 24,23-4.28; 26,27-9
eulogos, sensible, makes sense, 8,22; gennsis, reproduction, 36,4
10,25; 11,11; 28,7; 39,23; 94,20; 95,25; genntikos, able to reproduce, 32,19.22;
96,5; 97,19; 99,15; eulogs, sensibly, able to produce, 54,14; genntikos
it makes sense to, 2,19; 7,2.13.24; homoiou, able to produce something
8,16; 10,21; 21,17; 25,4; 39,16; 45,25; similar, 36,2. For genntik dunamis,
49,2; 59,8; 86,27; 89,5; 95,24; 97,18 see under dunamis; genntik psukh,
eunouxos, eunuch, 32,12 see under psukh
eudia, fragrance, 31,4 genos, genus, 13,12-15; 14,11-17;
euphuesteros, naturally more gifted, 19,18-20; 88,18-19
81,27 geds, earthy, 54,10. For geds
euphuia, good natural endowment, 52,6 xrots, see under xrots
euporia, abundance, 37,9 gras, old age, 23,4
eurus, wide, 95,4 graskein, age (v.), 35,13
euskhmosun, grace, 2,22 geuesthai, taste (v.), 55,13
eutheia, straight line; ex eutheias, geusis, taste (n.), 40,11-12.23; 49,17;
directly, 96,28; ep eutheias, directly 51,18.25-6; 54,3.18; 55,5.21; 56,7;
in front, 62,9 59,21; 93,3-4; 96,27
euthuria, straight line; kat geustikos, able to taste; geustik
euthurian tetagmenon, aligned, 43,19 aisthsis, sense of taste, 53,30; 92,25;
euthus, immediately, straight off, to geustikon, that which can taste,
56,18-21; 57,1; 74,19; 81,13; 82,8.20 54,3.12
exergasia, cultivation, 82,15 geustos, tastable, 51,8.14; 54,1-6.19;
exhairein, remove, 100,10 55,11
exhairetos, alone, 80,22 ginskein, know, 2,3.12; apprehend,
exhomoiousthai, become assimilated 84,12; gnthi sauton, Know Thyself,
to, 33,18; 35,1 1,6
exokh, protrusion, 71,18; 72,6 gltta, tongue, 49,14.17; 54,23; 55,5
ex, apart from, 88,14; outside, 97,19.22 glukus, sweet, 33,23-4; 51,22; 52,10;
exthen, from outside, external, 55,8-9.21; 60,21; 61,11-15
externally, 33,13; 34,28; 46,10; 50,13; gnrimos, discernible, 10,6; 23,12; 51,7;
56,19; 80,11; 91,2; 93,1; 97,18.21 92,15; 97,10; intelligible, 32,27;
exousia, power, 22,12 33,6.9; 78,3; gnrims, intelligibly,
30,18.24; 40,7
gastr, womb, 36,27; 38,4.8; 74,19.22 gnrizein, grasp, 41,2; 62,21; 63,6.27;
g, earth, 6,23-4; 8,21; 9,16-18; 22,7; 68,4; 74,15; discern, 51,8; 60,27;
23,29.30; 45,9; 53,8; 54,14.17 61,1.10-11.27
genesis, coming-to-be, coming into gnsis, knowing, knowledge, 2,1; 80,8;
being, becoming, 1,2; 6,11-20; 12,5; apprehension, apprehending, 32,23;
19,27-9; 24,3; production, 46,6; 51,15; 66,17; 72,1; 75,22; 81,4.13; 82,13;
53,13.23; 54,10.15; 68,19; automat h(e)autou gnsis, knowing oneself,
genesis, spontaneous coming-to-be, 1,8-9
32,12; en genesei (te kai phthorai), gnstikos, for apprehension, 73,23;
subject to coming-to-be (and 74,29; 82,18. For gnstik dunamis,
perishing), 1,1; 28,26; 29,2; 81,6; tou see under dunamis
autoi homoiou genesis, generation of grammateion, writing tablet, 84,26-8;
something like itself, 32,17; 36,7 85,1-2
genet, birth, 94,5 gramm, line, 18,12-13
gennan, reproduce, generate, 16,15; graph, drawing, 71,18
26,14; 32,21; 35,25; 36,13; 37,10; 83,2; graphein, write, 85,2
92,20; gennan hoion auto or homoion grapheus, painter, 50,27
(hauti), generate something like
itself, 9,4; 12,12; 32,11 (cf. 32,13.17); haima, blood, 40,2; 94,28; 95,2; 97,5.8
hairesis, choosing, 72,1.4; 82,17
Greek-English Index 199
hairetos, worthy of choice, 73,1 heterots, difference, 75,4
halmuros, salty, 51,23; 55,10 hetoimos, prepared, 86,14
halourgs, murex dye, 42,15 heursis, ascertaining, 31,3
hama, simultaneously, at the same hexis, disposition, 9,20-2; 15,11-15; 16,8;
time, 35,10; 50,26; 57,12-14; 22,2.5; 23,19; 26,21; 31,12.17;
61,14.19.22.28-31; 62,12; 63,6.21.26; 33,3-4.9; 37,1; 81,13; 82,1.10.21;
64,1.5-6.10.14-18.22; 65,1; 67,6; 85,11-12.20.25-6; 86,5.19; 88,24;
68,15; 80,10; 83,19 90,16; 91,3. For kath hexin nous and
hamaxa, cart, 78,28 en hexei nous, see under nous
haph, contact, 21,27; (sense of) touch, hidrusthai, be located in, 39,21
40,11-14.24; 51,26; 52,1.4-6; 54,2.4-5; hikanos, sufficient, 27,7; 56,17.23
55,15-18.26; 56,7.15; 57,13-15.20; himation, cloak, 80,14-15
58,3.6.12.16; 59,4.12.18.20; 93,3-4.14; historein, investigate, 2,21
96,22.27 hodgein, guide, 82,14
haplous, simple, 3,28; 4,7.10; hodos, sequence, 35,7; path, 96,29
7,15-20.27; 11,3; 17,6; 28,21; 37,7; holoklros, total, 19,34
55,7-8; 73,8; 88,8; hapls, simply as holoklrs, fully, 7,2
such, 6,14.17.19; 18,25; 26,26; 38,7; holos, whole, 13,17; to holon, the whole,
69,18; 84,2; 91,14-16. For hapls 11,16; 13,16-18; 14,10-11; 37,9; 38,11
hul, see under hul; for hapl homoeids, same in form, 16,19; 87,17
kinsis, see under kinsis; for homoiomers, homogeneous, 37,13;
haploun sma, simple body, see 98,3-5
under sma homoios, like, alike, similar, 33,16.25;
haptesthai, touch, 56,18-19; 57,1.10; 34,15-21; 35,17; 36,7.25; 37,9;
58,14.17; 93,12 39,12.17-18; 41,13; 48,13; 54,22;
haptikos, able to touch, tactile, 59,5; 55,4.12; 59,4.19; 83,12; 90,9. For
haptik aisthsis, sense of touch, 57,2; gennan homoion, see under gennan
93,10. For haptik dunamis, see homoisis, likening, becoming like,
under dunamis 39,12.15; 40,19; 44,10; 53,28; 54,20-1;
haptos, tangible, 5,20; 40,24; 51,8; 61,29; 77,3; 91,8-9
54,1.5; 55,15.18.21; 56,4.8-10.15-18; homoiots, likeness, 54,25; 55,14; 81,3;
57,6.16; 58,1.5.25-7; 59,5-6; 93,10.16 83,11; pros to theion homoiots,
harmonia, harmony, 24,18.22.27-9; likeness to the divine, 32,13
25,5.9-11.14-23; 26,2-7.11.21.25.28; homoiousthai, become like, 34,17; 39,1;
60,13; kath harmonian, 44,3; 55,17; 59,3; 61,22.29; 89,22;
harmoniously, 25,17-22 (cf. 25,14); 90,17; 91,19
26,15 homokhroos, of the same colour, 42,15
harmozein, be in harmony, 25,11.16.23; homologein, agree, 9,26; 11,14; 13,10
26,1.12; 100,4 homnums, homonymously, 28,27
hdesthai, feel pleasure, 23,15 horan, see, 35,4; 42,1.5; 43,16-18;
hdus, pleasant, 51,11-13; 74,2; 77,19 44,3-4.9-11.14; 46,3-4; 50,26;
hgemonikos, governing; to 51,2-3.11-12; 56,14; 62,4.9-11; 65,3-7;
hgemonikon, the governing part, 72,7; 77,20; 78,9; 83,3; 88,21;
24,5; 39,22; 68,12; 70,4; 76,10.14-15; 97,20.24; to horan, (the activity of)
94,7; 98,7.25; 99,26-8.31; 100,13 seeing, 65,7
hlios, sun, 72,19 horasis, seeing, 40,11; 78,9
helkein, pull, 21,28 horatikos, visual. For horatik
hensis, unity, 21,21 dunamis, see under dunamis
henoun, unify, 38,11; 48,2 horatos, visible, 43,12.19; 44,10-14;
hepesthai, follow, is entailed by, is a 45,24; 46,3; 51,8; 52,12; 56,9.11-12;
consequence of, 10,25; 26,29; 53,23; 65,7; 89,1-2
67,15.24-7; 71,15.22-4; 72,13.17; horismos, definition, 17,2.5; 28,26; 42,6
73,9.20; 75,17; 77,12; come next, horistos, with a definite shape, 46,17
35,18; 38,15; 93,25; ta hepomena, horizein, bound, delimit, give definition
consequences, 27,1 to, 3,2; 46,12; define, 54,16; 58,27;
hermneia, expressing oneself, 49,19 68,11; 92,18; hrismenos,
hsukhia, calm, 74,26 determinate, 26,7-8; with a definite
200 Greek-English Index
shape, 44,19; 45,20; 46,1.5. For simply as such, 4,5-7.24; hs (ontn)
hrismen kinsis, see under kinsis en huli, as being in matter, 83,17.22;
horman, strive, 72,21.26; 73,2.29; 79,14 87,3.12 (cf. enhulos); hs
horm, striving, 12,19; 49,5; 50,8; kekhrismenon huls, as separated
72,14-16.21-3; 73,1.21; 75,22-3; 77,16; from matter, 88,13; hs hul, as
78,23; 79,9.23.26; 80,5; 97,15. For matter, 83,23; 84,6; see oukh/m hs
kath hormn energeia, see under hul below; hs meta huls, as
energeia accompanied by matter, 84,7; khris
hormtikos, able to strive or enable (ts) huls, separate from (the)
striving; to hormtikon, the part for matter, 60,5; 66,13; 78,8; 83,15;
striving, 73,27; 75,10; 76,2-3.6.15; 85,14.19; 86,29; 87,21.26; 88,15.25;
99,8.22; to koinon hormtikon te kai 89,12.17.21; 90,5; 91,8; 92,22; cf.
orektikon, the common part for khris tn hulikn peristasen,
striving and desiring, 78,22. For separate from material material
hormtik dunamis, see under conditions, 85,15; see also aneu (ts)
dunamis; for hormtik psukh, see huls and apo ts huls khrismos,
under psukh above; also khrizein ts huls, under
horos, limit, 35,7 khrizein; kuris hul, matter in the
hothen, that from which (= the principle fundamental sense, 4,4.7.16.22; 7,15;
of change or efficient cause), 73,25 m memigmenos huli, not mixed
hou heneka, that for which (= the end with matter, 89,16; meta (ts) huls,
or final cause), 24,14 (cf. 24,16, 28,8); in conjunction with, 8,9; 18,6;
73,26; see also telos 91,14-15; oukh/m hs hul, not as
hudr, water, 3,17; 6,24; 8,21; 11,18; matter, 62,2-3.13; 83,16.22; 84,6;
34,11; 42,7; 44,5; 47,2; 51,17; 52,18; 87,4; see hs hul above; oukh hs en
53,5.8.12-14.17-18; 54,6-7; 57,3; huli, not as in matter, 83,22;
58,20-2; 59,15; 62,14 oukh/m hs meth huls, not as
hugiainein, be healthy, 16,16; conjoined with matter, 83,23; 84,7;
31,15-20; 41,15 91,15; prosekhs hul, proximate
hugiazein, cure, 100,7 matter, 4,23; 10,24; 44,21 (cf.
hugieia, health, 25,4-9; 31,15-20 prosekhs hupokeimenon,
hugrainein, moisten, make moist, proximately underlying subject,
54,20; 55,2 7,21); sun (ti) huli, conjoined with
hugros, moist, moisture, 26,8; 40,1-3; matter, 18,1; 89,21; 91,15
44,5; 53,2; 54,4-10.14.18-20; hulikos, material; hulik peristasis,
54,23-55,1; to hugron, moisture, material condition, 84,8; 85,15. For
3,18-19; 51,19; 53,25; 54,4 hulikos nous, see under nous
hugrots, moisture, 8,19; 19,35; 40,25; humn, membrane, 56,19-22
54,7.16.24; 55,22; 58,26; 94,19; huparxis, inherence, 90,4
enkhumos hugrots, 53,19 huperballein, to exceed, overflow, 13,4;
hul, matter, 2,26-5.1; 5,6.19-22; 59,7
6,2.5.17.22.28; 7,7; 8,1.18; 9,6; huperbol, excess, 59,19; 60,9
10,3.4.12; 11,20; 12,7; 13,24; 14,3; hupresia, task, 77,23; apparatus, 86,8;
15,12.23.26; 17,16-18,10; 18,14.18; 99,28
19,4.5.23.24; 26,28; 31,13.18.24; hupretein, subserve, 16,12; 75,29;
36,23; 54,6; 66,14; 81,25; 84,1-3.20; 76,17
85,17; 87,7.10; 88,13.18-22; 89,15-16; hupretikos, subservient, 76,9.14;
94,19; akhristos ts hul, 99,27; 100,14
inseparable from matter, 4,28; 6,1; huphanein, weave, 23,21
aneu (ts) huls, without (the) matter, huphants, weaver, 23,21
83,14; 91,15; see also khris (ts) huphantikos, for weaving, 23,21
huls below; apo ts huls khrizein, huphistasthai, exist, 4,17.26; 5,1.21;
separate from the matter, 85,13 (apo 18,4.17; 19,7; 29,13; kath hauto
ts huls khrismos, separation from huphistasthai, existing on its own,
the matter, 18,23; 89,14); en (ti) 4,8.20.24-6 (cf. 23); 17,11.14; 19,7.20
huli, in matter, 4,21; 88,16; see hs (cf. 19)
en huli below; hapls hul, matter hupnos, sleep, 71,1.26; 74,26-7
Greek-English Index 201
hupobathra, basis, 71,27; 85,24 katalptikos, secure. For katalptik
hupodokh, receptacle, 84,25 phantasia, see under phantasia
hupokeisthai, underlie, 7,16; kataphanai, make an affirmation,
correspond, 68,31; hupokeimenon 79,27
(adj.), underlying, 2,27; kataphatikos, affirmative, 67,21
3,4.5.10.16.23.25.27.28; 4,28; 5,6; katapsukhein, cool down, 77,2
6,5.22.28; 9,6; 10,11.17; 11,4.11; katapsuxis, cooling down, 49,21
31,13; 60,5; 66,14; 87,10.26; to katarithmsis, ennumeration, 31,2
hupokeimonon, underlying subject, kataskeuazein, construct, make, 2,25;
4,5-7; 6,5.28; 7,15.21; 80,15
8,1-3.6.9-11.14-17; 10,27-8; 14,25; kataskeu, construction, 2,19; 11,9;
15,2-3; 19,6.22.25; 31,18; 57,6; 75,4.8; structure, 98,8
83,21; 85,4; 89,15; 94,2.8; 99,9; kata sumbebkos, see under
corresponding object, 41,11; 56,5-9; sumbainein
60,14-16; 61,26; 69,6; 81,5 katgorein, attribute, 17,3; 29,4;
hupokhrein, withdraw, 48,11 characterise, 66,24; 82,11
hupolambanein, assume, suppose, kateirgnunai, digest, 33,14
take, 11,10; 23,21; 24,18; 63,12; katekhein, hold in, 21,13; 50,2
78,15.20; 99,4 katergasia, digestion, 95,3; 96,5
hupolpsis, judgement, 66,15 katharos, pure, 45,15
hupomenein, persist, remain, 47,17; kathekaston, particular, 83,7; 85,16.19;
68,8; 72,12 87,14.16; 90,5. For kathekaston
hupopiptein, come in view, 41,3; fall aisthtrion, see under aisthtrion;
under, 42,5 for kathekaston aisthton, see under
huposmos, guided by smell, 52,21 aisthtos
hupostasis, existence, 18,6; 88,7; 90,4; kathidrusthai, have ones seat, 100,14
hupostasis kath hauto, existence on kath hauto, on its own, in its own
its own, 19,19 right, intrinsically (contrasted with
hupotassesthai, submit, 74,10 kata sumbebkos, see under
sumbainein), 4,23; 5,13.15; 12,23;
iatrikos, medical, 24,23 17,10; 21,25; 22,4-12.23; 23,7.23;
idion, exclusive to, exclusive feature of, 24,6; 38,7; 53,8; 54,7.24; 78,25; 79,18;
43,20; 71,9; 73,16; 78,6; 98,27; idiai, 86,17.21; 90,22; 93,13; 96,15-17; auto
individually, 30,19; 40,8-9; idis, kath hauto, just itself on its own,
exclusively, 55,25; 98,25; in a sense 4,20; 5,12; 12,8; 20,28; 54,23; 79,17;
exclusive to X, 66,20; kat idian, 84,8; 89,13; 96,13; 99,7; to kath
taken individually, 54,12. For idios hauto, something in its own right,
aisthtos, see under aisthetos above 87,26. For kath hauto aisthton, see
ikhnos, vestige, 63,1; 72,11 under aisthtos; for kath hauto
iskhuros, strong, 13,1; 98,10-11 huphistasthai, see under
huphistasthai; for hupostasis kath
kakos, bad, 93,8 hauto, see under hupostasis
kardia, heart, 77,13; 94,24.30; 95,7.12; katholou, universal, 14,14; 82,13;
96,27-9; 97,1.7-9.14.23; 98,2.6.23-5; 83,8.12; 85,12.15; 87,16; 90,4-6;
100,10; peri tn kardian (topos), generally, in general, 21,17; 27,2;
neighbourhood of or region around 38,21; 40,20; 60,3; 78,11; 92,14; 99,6
the heart, 39,22; 40,2; 49,13; 77,10; kat, downwards, 7,20; 9,16-18; 22,7.19;
94,18-19; 95,4.9.16.22; 96,8.28; 23,29-30; lower, 97,28
97,9.17.24.29; 98,1.8; 99,15.24-5; katoptron, mirror, 43,3; 48,21; 62,14;
100,13 71,19
karpos, fruit, 53,13 kaulos, stem, 94,22
kataballein, sow, 36,22 keisthai, grant, 9,26; be located,
katakhrsthai, use, 49,13 62,9-10; 98,9
katalambanein, attain a secure grasp, kenos, empty, 48,4; groundless, 100,1; to
78,20 kenon, empty space, 20,9; 46,26-47,1
kataleipein, leave behind, 35,17 kentron, centre, 63,9-10
katalpsis, securing, 71,12
202 Greek-English Index
kephal, head, 13,17; 18,19; 96,3.29; 18,12-13; 38,9; 61,10.19.21; 62,21;
99,31; 100,6.11 63,27; 75,16; 76,7; 94,6; 100,10
kephalaios, main, 92,14 khrusos, gold, 42,14
keramion, jug, 13,20 khrusoeids, golden, 42,14
kerannunai, blend, 11,17; 13,22; 15,5; khumos, humour, 13,6; flavour, 31,5;
68,2 33,24; 49,14.17; 51,14-16.22; 52,9-10;
kros, wax, 3,6; 4,10-15; 6,18 53,2.10-13.16-20.23-4;
khalkeus, smith, 96,7 54,1-6.10.13-14.17.19; 54,25-55,1;
khalkeutikos, for smithing, 96,8 55,5.8; 56,7
khalkos, bronze, 3,5.17-19; 11,21; 14,2; khusis, exiting, 47,16
18,18; 87,8-11 kinein, to produce change, move, set in
khamailen, chameleon, 100,11 motion, 5,12.15-18; 7,21; 12,15.19;
kheir, hand, 13,16; 24,10; 96,4; 98,4.7; 13,2; 15,17; 21,14.25-8;
99,30 22,1-4.8-13.17-26; 23,3.6-7; 42,8;
khelon, tortoise, 100,10 50,13-15; 50,29; 61,21.30; 62,7.13;
khitn, membrane, 98,11 63,1; 65,12.15; 69,21-6; 70,2.15;
khlra, foliage, 42,15 71,17; 73,19.29; 74,23; 77,6.20;
khol, bile, 13,7 78,24-9; 79,2.5-18.22-7; 85,6; move,
khord, string, 60,13 78,28; 97,26; kata topon kinein, 22,22;
khorgein, provide, 36,9; 95,21 29,20
khorgia, resources, 94,25-7 kinsis, change, motion, movement,
khorgos, provider, 96,25 5,16; 7,18; 10,23; 12,13; 21,15.25;
khrein, pass, 14,9 22,2-3.10.15.18.23.27; 23,1; 24,12;
khrismos, separation. For apo ts 41,5; 43,2; 46,24; 46,26; 47,21;
huls khrismos, see under hul 58,11.17; 60,13-14; 61,21.24; 62,1.16;
khristos, separable, 5,15; 6,18.19; 63,2-5.14.17.24-6; 65,14-15; 66,21;
20,27; 22,5; 89,11-14; 91,1; separate, 68,8; 69,24; 70,3.22; 75,15; 77,12.16;
12,23 79,3.12.21.24-5; 83,20; 97,18-19.22-3;
khrizein, separate (v.), distinguish, 100,16-17; hapl kinsis, simple
6,18; 9,7; 14,8; 15,18; 17,14; 21,12; change, 7,19; 9,1; hrismen kinsis,
22,5; 28,24; 29,4; 40,12; 74,15; 75,3; determinate change, 35,6; kata topon
78,27; 84,8; 87,25; 90,7.13; 96,17; kinsis, change in place, 46,24;
kekhrismenos, separate (adj.), 73,19.27; 80,2; 93,9; 97,16; smatik
distinct, 21,18.26; 24,9; 31,1; 58,7-9; kinsis, bodily change, 12,9.21;
60,23.26; 61,8; 69,7; 79,11; 88,13; tetagmen kinsis, ordered change,
93,1; 94,10; 96,12.23; 99,9; khorizein 35,6. For arkh ts kinses, see
ts huls, separate (v.) from matter, under arkh
84,20; 85,13; 88,13; cf. khrizein apo kintikos, able to produce change, 22,8;
pass huliks peristases, 84,8; 35,5; 42,6; 43,12.14; 44,18;
khorizesthai ti nosei (moni), be 58,8-10.13; 72,24; 79,8; kata topon
separated in thought (alone), 4,10 kintikon, for changing in place,
khrsimos, useful, 10,29; 39,9 98,29; metabatiks kintikon, capable
khrsthai, use, employ, involve, 16,13; of spatial movement, 29,22. For
23,27; 24,6.8; 42,8; 49,25; 66,22; 95,24 kintik dunamis, see under dunamis
khrizein, colour (v.), 45,25 kithara, cithara, 26,1; 49,7; see also lura
khrma, colour (n.), 33,21; 41,22; koilos, hollow (adj.), 47,8.11; 48,2.8.11
42,4-43,8; 43,12-17; 44,2-8.16-24; koilots, hollow (n.), 47,12; 48,1
45,1.6.10-14.18-25; 46,3.6-7.16-18; koimasthai, be asleep, 66,27; 70,17;
50,27-9; 55,6; 56,1.7.10.12; 59,15; 74,24; 75,8
60,17; 61,26; 62,6.10.14; 65,13-18; koinos, common, 8,11; 9,5; 14,14; 17,6;
70,22; 72,10; 83,10.19.22; oikeion 78,5.17; 95,25-6; 96,3; general, 40,4;
khrma, colour of its own, 42,8; 44,4; combined, 32,22; mutual, 100,9; to
45,18; 46,9.13 koinon, what is in common, 85,14-20;
khrmatizein, colour (v.), 44,25 90,4-5; koinoteron, more general,
khrnnunai, tinge, 42,17.23; 45,16; 30,8; in a broader sense (contrasted
46,11 with kuris), 34,11; 72,5; 78,28;
khronos, time, period of time, 9,21; koins, in common, 17,7; as a group,
Greek-English Index 203
29,23. For koin aisthsis, see under metaphoran); 72,6 (contrasted with
aisthsis; for koinon aisthton, see koinoteron); 78,26; 80,4; 81,19;
under aisthtos; for koinon 88,25-6; 89,7-8 (contrasted with
hormtikon te kai orektikon, under deuters); 97,8; kuriteron, more
hormtikos; for koinon kritikon, properly, 82,14; malista kai kuris,
under kritikos; for koinos logos, see (being a certain sort of thing) to the
under logos; for koinos nous, under highest degree and in the primary
nous; for koin tn zin psukh, see sense, 88,25-6. For kuris hul, see
under psukh under hul; for kuris noton, see
koinnein, share in (common), have a under noton
share of, 29,9; 30,12; 44,20; 98,27
koinnia, community, 49,25 lakhanon, vegetable, 9,8
kluein, prevent, 20,11.13.14; 27,15; lambanein, receive, acquire, get, 18,8;
48,5.16; 62,8.11; 80,11; 84,16 36,23; 37,16; 38,5; 39,10; 64,22; 65,2;
kosmos, universe, 6,14 67,2; 72,20; 73,29; 77,7; 81,15;
kor, eyes interior, 44,7; 59,14 83,3.9.16.21-3; 84,1.7-9.14;
kouphos, light, 8,22; 79,4 85,9.14-15.19; 86,17; 87,3-5.28;
kouphots, lightness, 5,6.11-15; 11,7; 88,11-13; 89,22; 93,3; 95,17; 97,22;
20,22.25; 55,23; 79,1 conceive, consider, 4,11-13.24; 12,22;
krama, blend, 13,22 15,11; 19,1-3; 88,18; to
krasis, blend, 8,23; 9,7; 10,18; 11,19; lambanomenon, assumption, 80,8
12,6; 15,8; 20,2; 24,3.19-23; 25,3.4; legein, speak, say, state, 1,8; 2,10.15-19;
26,22.27-8; 68,2; 92,8; see also mixis 4,2-4; 5,16; 6,13.25; 7,3.14;
krinein, to have cognition of, 39,2; 11,6.10.15-17.21-2; 12,6;
44,13; 52,13; 60,16.20-3; 61,5-8; 13,11-12.17-23; 14,3.16; 16,1.6.13.16;
62,20; 63,25; 64,1-3.11-13.16-17.21; 17,15.21; 18,4.8-10.27;
65,16; 78,11.21; 84,6; 87,6; (pass.) be 19,10-12.18.21.26-7.35;
chosen, 74,11; 81,11; 89,4 20,16.19.23-4.27-9; 21,1.17; 22,24-5;
krisis, cognition, acts of cognition, 23,7-8; 24,18; 25,5.9-11;
43,15; 50,18; 51,6; 55,16; 60,3; 26,3.11-12.16.20-1.29; 27,12.16-18;
64,12-14; 65,1.17; 67,18-20; 72,4; 28,2.5.28; 30,8.18.27; 31,11.16.21.28;
74,7; 75,22; 78,13-17; 99,29 33,5.14-16; 34,11.20.23; 36,15;
kritikos, for cognition, able to have 38,7.16; 39,16; 40,9-10.15; 41,7-8;
cognition, 39,5; 46,21; 50,10; 44,16; 45,25; 48,7; 49,2.6-8; 52,10;
52,12-16; 53,29; 55,19.25.27; 59,10; 55,6; 57,2; 58,5; 59,9; 60,25-7;
60,5.17; 61,25; 66,9; 73,21-3; 74,28; 61,3.15-18; 64,15; 65,9.19;
75,13-14; 76,8-11; 80,22; 87,12; to 66,10-12.20-1; 67,6-7; 68,17-18.21;
kritikon, the part for cognition, 69,20; 70,3; 71,11.14; 73,13; 76,5;
73,22-3; 76,2-4.14; 78,17-19; 78,26-9; 79,2.5; 80,3-5; 81,28; 82,10;
99,17-19.23; to koinon kritikon, the 83,18; 84,28; 85,11; 86,2.28; 87,2;
common part for cognition, 78,21. For 91,2.23-5; 98,22; 99,10.30
kritik dunamis, see under dunamis; leios, smooth, 44,1; 47,8-9
for kritik psukh, see under psukh leiots, smoothness, 40,25; 55,23.29;
kubernts, pilot, 14,4; 15,9-26; 56,2
20,26-21,16; 96,7 leipein, be missing, 66,3
kuberntik, of or for piloting, 15,11.23 lpsis, acquiring, acquisition, receiving,
kuklos, circle, 63,8-10 78,8; 79,10; 83,12.14; 84,11.16; 91,7-9
kurios, principal, authoritative; lptikos, capable of receiving, 92,10. For
kuritatos, highest, 30, 11; 40,1; lptik dunamis, see under dunamis
73,4.6; 85,6. For kuritat dunamis, leukos, white, 6,9; 13,21; 33,20.23; 34,3;
see under dunamis; for kuritaton 41,8; 55,7.20; 60,21; 61,11-15.27.31;
eidos, see under eidos; for kurion 62,4.7-11; 64,16; 72,10.21; 79,4;
onoma, see under onoma; for kuria 83,8.10; 84,9; bright, 45,11.15; 63,3
arkh, see under arkh liparos, savoury, 51,23
kuris, in the fundamental sense, 7,20; lithos, stone, 3,5; 25,18; 69,22
34,12 (contrasted with koinoteron); logikos, rational, 29,23; 36,15; 67,17;
43,6; 66,20 (contrasted with kata 74,8; 80,3.24; logik praxis, rational
204 Greek-English Index
action, 80,2. For logik dunamis, see melikraton, honey water, 11,17-20
under dunamis; for logik orexis, see melos, melody, 26,6; song, 49,6
under orexis; for logik psukh, see mlos, apple, 31,4-6
under psukh; for logikon zion, see menein, stay, remain, persist, 2,15;
under zion 9,17; 15,8; 18,17; 20,5; 26,10; 28,20;
logistikos, capable of reasoning, able to 34,25-6; 48,1-3.13; 50,14; 56,3; 86,9;
reason; to logistikon (morion), the 90,13; 93,22; 100,12
part for reasoning, 76,11; 98,24-8; mnuein, communicate, 71,7; 76,13
99,2. For logistik orexis, see under mnusis, communication, 49,24
orexis; for logistik psukh, see under mnutikos, able to communicate, 59,14
psukh meristos, divisible into parts, 31,1
logizesthai, reason (v.), 78,15 meros, part, 10,28; 11,16; 13,16-18;
logos, characterisation, 2,11; 16,7.18; 14,5-11; 15,15; 16,11; 18,10-27; 19,23;
28,19; 30,19; 69,6; 70,3; definition, 27,3; 30,5; 30,26-31,1; 37,4.13-15;
4,2; 18,2; 75,2-6; 86,26; 87,10.15-16; 38,5-10; 40,6.10; 47,12; 49,22; 56,20;
94,9; account, 6,17.20; 28,25; 29,1; 63,19.21; 64,5.10; 74,21-3; 75,26-8;
30,21; 32,9.23.30; 33,10.10; 36,13; 95,7.20; 96,4; 97,24-8; ektos meros,
38,13.15.20; 40,4.8; 55,15; 89,14; external part (of the body), 2,21; en
92,12; essence, 60,8; argument, 6,24; merei, in turn, 61,14; entos meros, the
18,6.27; 21,16; 60,19; 97,27; inner part (of the body), 2,20; kata
proportion, 24,25.28-9; 25,1-4.12.21-3; meros, serially, 91,19; sumphuton
26,4-8.12-13.19.28; discussion, 39,22; meros, congenital part, 24,8. See also
phrase, 56,12; language, 49,23; under morion
reason, 30,12; 67,12; 71,24; 73,10; mesos, in the middle, in the midst,
74,10-13; 80,3; koinos logos, general intervening, 52,18; 57,3; 58,5; 59,9;
characterisation, 17,2.6; 28,16-18 (cf. 95,15-17.20; 97,28
20); 30,18; 40,4.7; general account, mesots, balance, 59,8.18; 93,14-16
92,12; oikeios logos, own definition or mestos, filled, 20,8-10
account, 3,12; 32,29; prton ti logi metaballein, transform, 19,13-17;
(or kata ton logon), first in account, 33,24-7; 34,1.5.11.23-8; 53,22; 54,22;
32,28-30; to logon ekhon, what has 59,11
reason, 74,13 metabasis, transition, 83,8; 85,21;
lupeisthai, be upset, 13,7; feel pain, change in position, 100,17
23,15 metabatiks, spatially, 29,20.22
lupros, painful, 51,10-13; 77,19 metabltikos, able to trigger, 51,18
lura, lyre, 49,8; see also kithara metabol, change, transformation, 5,9;
8,4; 19,12; 27,5; 33,18-28; 34,4.6;
makh, conflict, 27,8 35,2; 53,24; 54,22-3; 59,11
malakots, softness, 55,23 metadikein, pursue, 80,10
manthanein, learn, 2,12; 23,16; metalambanein, have a share of, 82,12
appreciate, understand, 6,7; 12,8; metalleutos, metal, 3,18
82,2; 98,6 metapherein, transfer, 35,4; extend,
marturion, evidence, 83,18.21 72,12; 85,7; translate, 80,15
mastix, whip, 47,21 metaphora, transfer; kata metaphoran,
matn, in vain, superfluously, 12,23; in an extended sense, by extension,
27,9.13; 28,5; 98,9. For mden matn 49,6; 55,6; 66,22
poiein tn phusin, see under phusis metastasis, shift, 77,10; changing place,
mathmatikos, mathematical, 72,22; 93,5
90,10 metaxu, intermediate, (in) between,
megethos, extension, something intervening, 25,11; 33,21; 47,16;
extended, 30,27-9; 35,14; 41,5; 60,8-9; 55,10.20; 58,9; 62,12; 94,22; 98,12; to
65,13; 70,21; 83,20; largeness, 55,28; metaxu, the medium, 41,19; 54,2-4;
56,2; 84,9; 92,4 57,11.20; 58,2.5.17; 93,13.24
meiousthai, shrink, 35,13 metekhein, have a share in, 32,14; 36,8;
melas, black, 33,21.23; 34,4; 45,16; 67,2; 72,18; 74,3; 93,20
55,7.20; 61,27.31; 62,4.7-11; 64,17; 79,4 metempsukhsis, metempsychosis,
meli, honey, 11,18; 13,22; 26,5 27,18
Greek-English Index 205
methistanai, convert, 19,17; noma, idea, 49,24; concept, 85,23; 86,6;
methistasthai, change place, shift, 90,11.20; 91,4
20,17; 77,9; cease from, 95,12 nosis, act of understanding, 68,24.30;
methuein, to be drunk, 23,5 83,12.14; 84,10; 89,5; thought,
metrein, measure, 50,19 thinking, 77,12.18; 79,22. For
mignunai, mix, co-mingle, 19,24.33; khrizesthai ti nosei, see under
24,29; 25,1.12-14.22; 45,5.9.12.18; khrizein
46,5; 53,16-18; 89,16 notikos, able to understand, 76,11. For
mikrots, softness, 55,28; 56,2 notik dunamis, see under dunamis
miktos, can be mixed, 55,8; 58,23 notos, understandable, can be
mimeisthai, imitate, 3,15; 49,7 understood, 61,4.7; 68,24-5.28;
misein, hate, 23,16 84,19-20; 86,7.27; 87,25; 89,7; 90,4;
mixis, mixture, 8,16.23; 9,6; 10,18; 91,11-13; kat energeian noton,
19,28-34; 24,3.19-28; 25,14-15.23; understandable in activity, 87,29;
26,9-14.22.27; 45,8.14; 54,13-14; 88,1; kuris noton, understandable
55,9-10; 92,6-8; 93,16; see also krasis in the primary sense, 87,27; 88,25;
mnm, memory, 68,9.20; 69,19; 83,6 noton eidos, understandable form,
mnmoneuein, remember, 23,16; 83,5 86,17.22; 88,25; 91,7; see also eidos
morion, part, 24,10; 28,7; 32,22; 37,3.8; nooumenon, under noein; sphodra
59,2; 64,7-9.19; 74,29; 75,9; 77,12.20; noton, vividly understandable,
78,2; 81,2; 94,7; 96,15; 97,3.10.13.25; 86,12; ti hautn phusei noton, what
99,16; 100,2.8-9.16; see also under is understandable through their own
meros. For organikon morion, see nature, 61,7; 87,28; 89,4-6 (cf.
under organikos; for smatikon nooumenon kata tn hautou phusin,
morion, see under smatikos; for under noein)
morion ts psukhs, see under nosein, be ill, 16,16; 98,23; 100,5
psukh; for logistikon morion, see nosos, illness, 71,25; 98,23; kardiak
under logistikos nosos, lit. cardiac illness, probably
morph, contour, 3,9; 18,18 heartburn, 98,23
muein, close shut, 52,24 nous, understanding, 61,4.8;
mus, muscle, 100,16 66,13.18.23; 67,9-11.25; 68,24-5.29;
71,25; 76,4; 81,9.28; 82,9-11;
naus, ship, 14,4; 15,9; 20,28-1,10; 79,20 83,13.22; 84,4-6.14.18-19;
nes, temple, 25,19 85,4.6.21.25; 86,7.12.23;
neots, youth, 23,4 87,6.14.17.20-1.25;
nphein, to be sober, 23,4 88,1-6.9-11.14-15.19.21; 89,11;
neuron, sinew, 24,9; 76,16; 77,6; 90,3.6-7.10.13.20-2; 91,7.12.24;
98,12.15; 100,16 bouleutikos nous, understanding for
noein, understand, 23,16; 61,5.8; 68,4; deliberation, 81,10; doxastikos nous,
78,16; 84,9.18-19.22-3; the understanding for belief,
86,13.16-24.28-9; 87,1; 88,3-15; 81,10.20; dunamei nous, the
89.21-2; 90,2-9.12-13.17-22; understanding in power, 81,23; 82,19;
91,4-6.21-2; 92,11; think, 79,10.13-16; 91,3; en hmin nous, the
80,12-14; to noein, (activity of) understanding in us, 88,5; 89,19;
understanding, 23,2.4; to nooumenon, 90,19; 90,23-91,2; en hexei nous,
what is understood, 84,11-12.23-4; dispositional understanding, 85,11;
86,24; 88,2.10-12.14; 89,11.22; 86,16; cf. hs hexis kaloumenos nous,
90,5.11.17.22-3; 91,5; nooumenon 85,26-86,1; see kath hexin nous
eidos, the form understood, 86,15.18; below; epikttos nous, acquired
88,14; 89,22; to nooumenon kata tn understanding, 82,1; epistmonikos
hautou phusin, that which is nous, understanding for knowing,
understood in virtue of its own 81,12.22; hulikos nous, the material
nature, 90,11 (see ti hautn phusei understanding, 81,24.26; 82,20;
noton, under notos); to nooun, who 84,17.21.24.28; 85,10; 88,23-4; 89,20;
or what understands, 86,24; 87,28; 90,14; hupokeimenos nous,
88,2.10 underlying understanding, 84,17;
90,14; kat energeian nous,
206 Greek-English Index
understanding in activity, 82,6.21; activity, 73,5; opsis theretik,
86,4.15.27-8; 88,2; cf. 90,16; kath theoretical vision, 85,22
hexin nous, dispositional oregesthai, desire, 16,15; 73,28; 74,11;
understanding, 82,1 (title), 86,5; 88,6; 79,14
see en hexei nous above; koinos nous, orektikos, able to desire or enable
common understanding, 82,14; kuris desiring; to orektikon, the part for
nous, understanding in the primary desiring, 73,27; 74,10; 75,11.14-16;
sense, 88,2; 89,18-19; phusikos nous, 76,6.16; 98,29; 99,22. For orektik
natural understanding, 81,26; 82,2; psukh, see under psukh; for orektik
poitikos nous, productive dunamis, see under dunamis; for
understanding, 88,17 (title); 88,24; koinon hormtikon te kai orektikon,
89,6.10; praktikos nous, the under hormtikos
understanding for action, 81,10.20; orektos, desirable, 79,6
82,16; thertikos nous, orexis, desire, 12,14; 73,1; 74,1-6.11;
understanding for study, 81, 12.22; 75,18-20.23-4; 76,18; 77,7; 78,23;
82,18; 83,1; 87,17.21; 99,6; thurathen 79,9.23-6; 80,5-7.11; 81,11; 97,16;
nous, the understanding from 99,1-2; bouleutik orexis, deliberative
outside, 90,19-21; 91,2 desire, 80,7; 99,2; logik orexis,
nun, now, 20,13; 61,10-11.17-18 rational desire, 74,8; logistik orexis,
reasoning desire, 99,3
oiax, rudder, 77,9; 96,7 organ, be excited, 13,4
oikeios, of Xs own, proper to X, 2,9; organikos, organic, 11,1-5; 15,4;
3,12; 4.19; 6,27; 12,23; 18,5.7; 16,4.11-13; 75,29; 99,28; organikon
22,16.27; 23,1.3; 30,24; 31,12; meros, organic part, 24,6; organikon
32,2.10.18; 35,27; 36,2; 38,20; 40,17; morion, organic part, 77,22. For
46,8-9; 47,20; 52,8.20; 56,5.11; 59,15; organikon sma, see under sma
60,2.27; 65,12; 74,22; 78,4; 84,14-15; organon, organ, 23,25; 24,7;
86,14; 87,27; 88,7; 93,12; 96,7; 100,12; 27,12.15.19; 37,7.17; 38,3; 39,19;
characteristic of, 35,7; oikeis, 49,1.10.15; 52,24; 54,19; 55,5;
appropriately, 40,9. For oikeion 57,14.18; 58,1; 62,20; 65,12; 75,29;
khrma, see under khrma 84,10; 95,17.23; 96,7; 97,23; 99,29;
oikeiots, conformity, 81,4 100,3; tool, 35,2; 96,8; smatikon
oikia, house, 11,17 organon, bodily organ, 84,10. For
oikonomia, regulation, 2,21; aisthtikon organon, see under
management, 38,2 aisthtikos; for phntikon organon,
oinomeli, honeyed wine, 13,22; 15,7 see under phntikos
oinos, wine, 13,20.23; 15,7; 26,5 org, anger, rage, 12,16; 13,1; 97,2
okhein, carry, 22,1 orgizesthai, become angry, 13,6; 23,15
khrots, blanching, 77,14 orkhsts, dancer, 79,18
oligrein, make a slight, 74,4 orkhstikos, for dancing; orkhstik, art
omma, eye, 44,10 of dancing, 79,18-19
on, (a) being, see under einai osmasthai, perceive smell, 51,10
oneirgmos, wet dream, 77,21 osm, odour, 40,23; 51,7.11; 52,21;
onoma, name (n.), term, 56,6-12; 72,13; 53,4.9.14.17-19; 56,7; 59,17; 72,11
73,3.7; kurion onoma, proper term, osmds, odorous, 53,11
72,11 osphrantikos, capable of smelling, 53,27
onomazein, name (v.), call by name, osphrantos, smellable,
4,2; 36,14; 52,9 51,6-7.13-14.18-22.25;
onux, nail, 75,2; 97,5 53,7-9.20.24.27-8; 55,6
ophthalmos, eye, 52,25; 53,1; osphrsis, smell, 40,11.23; 51,5-6;
57,4.10.13; 58,4 52,15-17.23-4; 53,21; 56,7; 59,16
opis, backwards, 48,6 osphrtikos, for smelling, 52,26
opsis, sight, 40,11.22; 41,17; 42,4.9; osphrtos, smellable, 52,8.15; 53,23
50,24.27; 52,12-14.17; 55,20; 56,6; osteon, bone, 75,2; 97,5; 98,10
57,14; 58,4; 60,18; 61,26-7; 62,6; 63,3; ostreds, like oysters, 67,3
65,13.18; 73,4; 83,20; eye, 43,14.20; ostreon, oyster, 74,6
62,4; kat energeian opsis, sight as an thein, push, 21,28
Greek-English Index 207
ous, ear, 49,1; 50,13.17.21.24; 57,4 89,15.20; 98,21; be affected, 13,2;
ousia, essence, 1,3; 2,11; 4,14; 15,22; experience, 71,18
16,18; 19,12.15-17; 22,27; 27,1; 28,16; pathtiks, by being modified, 62,13
30,20; 32,7; 35,15; 61,7; 92,6; pathos, modification, 39,18.21; 41,13;
substance, 5,2-3.10; 6,2-6; 12,8; 14,25; 43,20; 44,8; 51,17-19.21; 53,25; 54,17;
15,1; 20,27; 26,19; 59,13; 59,13; 57,12-14; 62,2-3.17; 63,22; 64,2.5-8;
aistht ousia, perceptible substance, 76,18; 77,1.12.21; 84,1-3; 86,11;
2,27; meros ts ousias, part of 100,5; effect, 50,27; affect, 13,2-5;
substance, 6,3; smatik ousia, bodily 71,25; 97,2; state, 98,24; smatikon
substance, 2,27 pathos, bodily affection, 84,5
oxus, high-pitched, 34,2; 55,28; acidic, pauein, cease, 4,3; 7,1-2; 86,25
51,24; 55,21 pdsis, palpitation, 98,18
oxuts, acidity, 53,22 pg, source, 40,2
peithein, persuade, 2,13; peithesthai,
pakhuts, density, 44,8 obey, 1,5; 2,2; 74,10
palaiein, wrestle, 23,18-19 pepsis, cooking, 33,19.25-6; 35,1; 51,18;
palaists, wrestler, 23,18-19 53,17; 54,10-11; 95,3; see pessein
palastikos, for wrestling, 23,19 below
palindromein, return back again, 95,11 perainein, limit; peperasmenon, limited
paradekhesthai, admit, 25,10 (in number), 27,4; 28,7.14
paradoxos, incredible, 2,14.23; peras, boundary, 14,20; 45,20-46,1;
11,6.10-11; paradoxs, incredibly, 2.24 46,8-13.18; limit, 17,13-14; endpoint,
parakeisthai, put together, 25,17 63,9-12.15-17.23-4; 64,1; 97,28
parakolouthein, examine in detail, 2,10 peratoun, bound (v.), 46,10
parakoptein, be insane, 70,18 periekhein, include, contain, enclose,
parallag, diversity, 8,14.16; disparity, 14,20; 18,7; 32,29; 47,12; 48,2; 50,15;
10,22 80,18; 98,10
parallattein, deviate, 75,12 perikardios, pericardium, 98,11
paramonos, lingering, 86,11 perikeisthai, surround, 98,19
paramuthia, explanation, 2,9 perilambanein, comprehend, 16,19;
parapempein, deliver, 50,16 class, 56,12; surround, 98,11
paraphrosun, delirium, 100,6 perilpsis, comprehension,
parapomp, delivery, 95,3 comprehending, 82,13; 83,11; 85,12;
paraskeuazein, prime, 13,5; enable, 92,14
36,21; arrange, 50,28; facilitate, 51,1 peripatein, take a walk, 80,12-13;
paraspeirein, distribute, 37,12 82,5-7
parathesis, juxtaposition, 12,2; peripiptein, encounter, 93,5; incur,
combination, 13,6 98,20
pareinai, be present, 40,14; 43,1; 44,4; periphereia, circumference, 63,8
45,6; 58,13-18; 63,2-4; 66,25; peristasis, circumstance, 41,12. For
69,8.11-13; 70,16; 71,1-2; 77,4; ou/m hulik peristasis, see under hulikos
pareinai, not present, 39,5; 62,16; periteinein, stretch over, 56,19
66,26-8; 67,1; 68,8; 69,10.15.26; perittma, end product, 36,4; 92,20;
70,16; 71,1; 73,1. See also parousia 95,18
paremphainein, manifest in addition, perittos, magnificent, 2,12; peritts,
84,15 magnificently, 2,24
parhuphistanai, exist in conjunction prma, maimed, 32,12
with, 65,18 proun, maim, 32,18; 35,20; 81,26; 82,11
parodos, passing, 62,18 pessein, cook, 33,27; 53,12; 95,2
parousia, presence, 29,5.9; 31,7.28; pxis, solidity, 3,19
35,27; 36,26; 38,1; 43,2.9; 45,2.11; pezos, footed, 13,14
92,17. See also pareinai phainesthai, appear, 44,25; 45,24;
paskhein, be modified, undergo, 50,26; 67,7; to phainomenon, what is
7,10-14; 16,15; 39,17.20-1; 41,9.11; represented, 71,3
42,18; 43,13.20; 54,7.24; 57,7-11.15; phaneros, obvious, clear, 27,8; 35,25;
59,3.7; 61,30; 63,22; 64,6-11.19-21; 99,30; apparent, 45,23-4; phaners,
66,12; 75,23; 84,2.6; 85,2-4; 86,7; plainly, 8,14; 12,15
208 Greek-English Index
phantasia, representation, 12,17.20; phtheirein, destroy, 60,9-12;
49,5.11; 50,7; 63,2; 66,11.19-27; phtheiresthai, perish, 4,19; 12,5;
67,1-13.18-23.27; 68,4.11-21.26.29; 90,8.15; 96,22-3
69,15-17.20; 70,2.5-8.11-13.23; phthora, perishing, destruction, 1,3;
71,3.11-12.17.22-4; 19,33-4; 28,26; 29,2; 47,23; 89,14
72,4-5.13.16-19.24-5; 73,3.7-9.12.20; phuein, in perf., be by nature (such as
75,17; 76,10; 77,2.17-18; 79,22; 83,3; to), 3,11; 43,10; 45,2.14; 81,3
97,14; 98,19; appearance, 45,22; phug, avoiding, 72,1.4; 79,10
akatalptos phantasia, insecure phugein, avoid, flee, 39,10; 74,1; 80,10
representation, 71,12; amudra phulassein, keep, maintain, 11,20;
phantasia, obscure representation, 41,14; 48,8; phulassesthai, protect
71,5-7.13-15; enargs phantasia, themselves against, 93,23
evident representation, 71,13; kat phusikos, natural, 3,14; 12,10. For
energeian phantasia, active phusikon eidos, see under eidos; for
representation, 68,18; katalptik phusikos nous, see under nous; for
phantasia, secure representation, phusikon sma, see under sma
71,11.15; sphodra phantasia, vivid phusis, nature, 2,12-13; 3,15.21; 4,1.4;
representation, 71,5.8.11.14-16.19-21 5,3-4.7.10-11; 6,4; 7,17-21; 7,23;
phantasiousthai, represent, 78,14.19 8,9-10; 9,17; 11,7; 15,6.22; 18,5-7;
phantasma, representation, 66,21 19,17-18.24; 21,14.19; 22,9; 27,15;
phantastikos, able to represent, 29,20; 28,10.18; 30,20-1.25; 31,26-7; 32,8.19;
to phantastikon, the part for 36,9; 39,19; 40,7; 44,21-2; 46,4; 52,6;
representing, 29,21; 68,28; 99,8.23. 53,4; 71,26; 82,14; 84,15; 90,11.23;
For phantastik dunamis, see under 95,14; 98,9; en ti oikeiai phusei, in
dunamis; for phantastik psukh, see its own nature, 87,27; kata phusin,
under psukh natural(ly), in accordance with
phantaston, representable, 68,26.30; nature, 1,9; 2,2; 5,3; 7,5; 8,4; 23,2.6;
69,9; 71,7 24,16; 32,15-17.22; 41,15; 81,2; 88,17;
phantazesthai, present, 51,3 93,9; 95,14; kata tn hautou phusin,
pharmakon, drug, 24,24-5.28 in its own nature, 85,8; katakhrtai
pherein, move, carry, shift, drive, h phusis epi duo erga, nature uses
9,16-18; 23,5.29-30; 47,13.23; for two functions, 49,12; mden
48,1-8.11 matn poiein tn phusin, nature does
pheuktos, to be avoided, 77,16-17 nothing superfluously, 27,9.13; 28,4;
philein, love, 23,16 98,9; panta h phusis heneka tou
phleps, blood vessel, 94,27 poiei, nature makes all things for
phobeisthai, be afraid, 13,7; 23,15 something, 28,8.12; para phusin,
phobos, fear, 12,17; 13,1; 97,2 contrary to nature, 21,14-15; phusei,
phn, voice, 49,3-5.8-9.14.23; 50,3-8; by nature, natural, 3,15.21; 28,10;
55,27; 56,3 61,7; 81,3; 82,4-10; see also phuein.
phnein, have a voice, 49,5.8; 50,1 For ti hauton phusei noton see
phntikos, vocal; phntikon organon, under notos; for ho kata phusin bios,
vocal organ, 49,10.15; 50,7 see under bios
phora, motion, 22,7.19 phutikos, for vegetation, vegetative,
phran, detect, 30,20; 71,16 74,15. For phutik psukh, see under
phs, light, 42,4; 42,11-43,1; 43,7-12; psukh
44,13-18; 44,24-45,7; 45,19; 46,2-3; phuton, plant, 8,25-9,26; 10,5.13.16;
73,5-7; 89,1 29,6.14; 30,9.16; 31,7-8.10; 32,18-20;
phtizein, illuminate, 42,1.10.18-20; 37,4.13.18; 74,16; 75,9; 78,5; 92,22.25;
43,1.10-11.13; 45,2.13; 46,18-19; 62,5 94,20-1; 96,14
phtoeids, luminous, 45,15 pikros, spicy, 51,23; 55,9-10.21
phrittein, shudder, 77,4 pinakis, slate, 84,25-6
phrixis, shudder, 77,15 piptein, fall; hupo tn aisthsin piptein,
phronein, think, regard, 2,6; 12,20 fall within the scope of perception,
phronsis, thought, 67,11 91,22
phthartikos, destructive, 93,6-7.23 pisteuein, be convinced, hold, 1,7; 2,14;
phthartos, perishable, 21,24; 90,10 95,13
Greek-English Index 209
pistis, conviction, 67,15-20 pous, foot, 24,10; 96,4; 98,4; 99,30
pithanots, plausibility, 100,1 pragma, object, 16,6; 27,11; 28,18;
planan, to be confused, 18,14 68,24; 72,1; 77,3; 88,12; 91,15-18.25;
plan, error, 100,2 92,6; subject, 40,7
Platn, Plato, 26,18; Timaeus, 26,20 pragmateia, treatise, 30,24; 38,20
plg, impact, 46,21-47,1; prakteos, ought to be done, 73,10; 80,6
47,4-5.14-15.18-21.24; 48,12-14.17.20; praktikos, for action, concerned with
49,9; 50,6.14.21.24; 60,14 action, 75,15; 80,1; 81,6; to praktikon,
pleonakhs, in many ways, 13,10 the part for action, 73,22-6; 76,2-3.15;
pleonasmos, excess, 98,16 80,1; 98,26-7; 99,16.19; praktikon
pleonazein, be in excess, 13,7; 49,13; morion, part for action, 98,4. For
multiply, 85,23 praktik dunamis, see under
pleonektein, surpass, 86,4 dunamis; for praktik energeia, see
plrs, full, 20,10-18; 48,4 under energeia; for praktikos nous,
plroun, fulfil, 2,7; fill, 20,17 see under nous; for praktik psukh,
plssein, make an impact, hit, see under psukh
46,22-47,19; 47,25; 48,8-12.16-18; praktos, can be done, 77,16; 80,9; 82,16
49,16; 50,2-3 prattein, do, act, 16,16; 32,15-16; 73,2;
plthos, number, 8,7.15; 10,18; 70,23; 76,18; 80,4; 82,17
abundance, 13,6; 95,6 praxis, action, 72,15-16; 73,24; 75,23;
pleumon, lungs, 49,13; 98,19 78,1; 80,5.9; 81,12; 99,29-30. For
pleura, side, 72,23 arkh praxes, see under arkh
plinthos, brick, 11,17 presbeuein, have great reverence, 2,4
pluntikos, able to wash, 51,20; 53,5 proagoreuein, declare, 1,8
pneuma, breath, 19,10-20; 26,16; 40,3; prodiaballein, predispose against,
49,20; 77,11; 98,17; to sumphuton 71,21
pneuma, congenital breath, 77,5 progignesthai, precede, 79,21
podaios, a foot long, 72,19 progignskein, foreknow, 1,7
poiein, act, make, produce, do, 2,26; progumnazein, train, 86,13
7,10-14; 16,14; 19,9.12.15.22-4; 23,8; prohairesis, choice, 80,6
27,9.13; 28,5.8.12; 29,1; 30,19.27; prohgeisthai, occur first, lead, 79,24;
31,4; 32,24; 35,3-4; 40,8; 41,5; 80,4; prohgoumens, primarily,
45,14.18; 46,3; 47,3-9.14.17-18; 22,16; 35,6.16; 86,21
53,17-18; 54,22; 55,12; 59,3; 61,28; prohiesthai, emit, 36,25; 49,8
62,7.14; 65,1; 66,12; 71,23; 74,22; prohodos, progression, 28,13; 48,15
77,5.14; 80,13; 83,17; 84,10.19; 85,25; prohuparkhein, be there before, 94,13
87,11-13.25; 89,20; 91,25; 98,9; 100,9; prohuphistasthai, exist prior, 9,21.24
poiein heteron hoion auto, producing proienai, progress, 35,7; 82,8; 95,5
another like oneself, 32,13 prokruttein, decree, 1,6
poitikos, productive, able to produce or prokinein, change first, 97,3
make 24,12; 35,17; 46,2-4; 74,28; prokrinai, prefer, 73,11
75,13-14.24; 88,20; 89,10. For prolegein, proclaim, 1,8
poitikos nous, see under nous propaskhein, be affected first, 97,3
poikilos, complex, 7,23; 8,10; 11,2.5; prra, bow, 77,10
85,23 prosanatupoun, mould further, 70,13
poion, character, 12,3; 14,7; 18,22 prosangellia, message, 97,21
poiots, quality, 5,20; 6,6; 17,17; 18,3-7; prosekhs, proximate; prosekhs,
19,10-16; 54,6; 70,22 proximately, 7,21; in detail, 28,17;
polutropos, diversified, 85,24 40,9. For prosekhs hul, see under
pods, greenish, 42,15 hul
poreutikos, capable of walking, 93,4.21 proskonts, appropriately, 51,4
poros, passage, 50,17; 52,26 prosginesthai, be added, 19,8; 37,18;
porphuroeids, purplish, 42,15 apply, 54,9; accrue, 81,19; 82,9
porr, far away, 50,22.26.29; 52,21 proshiesthai, approach, 93,7
ps ekhein, be disposed in a certain prosienai, approach, 72,25
way, 13,8 proskaiein, set on fire, 53,21
poson, size, 12,4; 14,7; 18,16.22; 35,17 proskeisthai, qualify, 16,17
210 Greek-English Index
proskhrsthai, use, make use of, 74,13; 75,25.28; 76,10; 78,25;
23,25.29; 84,11 79,2.8.16-19; 80,21; 81,2; 84,2.27;
proskrinein, incorporate, 33,15; 85,6-7; 90,15; 91,10.14.19-20.23-6;
34,15.17.25; 35,1 92,2.8-17; 94,7-10.16-18; 96,13; 97,17;
proslambanein, add, take on in 98,7; 99,11-14.16-18.25-6; 100,13;
addition, acquire, 16,22; 18,9; 53,9; aisthtik psukh, soul for perceiving,
54,11 10,9.14.17; 38,21; 39,19; 50,18;
pros, forwards, 48,5.16 69,13-15.25; 71,27; 74,20; 78,7; 93,25;
prospheresthai, ingest, 33,13; 34,28; 96,11; akoustik psukh, auditory
92,26 soul, 50,9; alogos psukh,
prosphuesthai, be naturally attached or non-rational soul, 73,14; 80,16;
appended to, 58,6; 92,24; 95,24; 96,1 auxtik psukh, soul for growing,
prospiptein, collide with, impinge on, 36,18; 96,9; deutera psukh, second
47,17; 48,4; 93,22 soul, 29,11; 38,16; en hmin psukh,
prostassein, command, 1,5 soul in us, 91,3; genntik psukh,
prosthk, addition, 19,10.11.13; soul for reproducing, 32,9; 36,17;
80,19-20; supplementation, 94,12.15 96,9; hormtik psukh, soul for
pros ti, relative, 86,25-6 striving, 78,24; 94,1; koin tn zin
prostithesthai, be added, 19,14; 30,5; psukh, soul common to animals,
34,23; supplement, 94,13 78,2; kritik psukh, soul for
protasis, premiss, 80,13 discerning, 99,23; logik psukh,
prothesis, aim, 1,2; 2,7 rational soul, 74,8; 80,18; 94,3;
prtos, first, 9,20; 48,8.12; 75,25; basic, logistik psukh, soul for reasoning,
5,19; prts, initially 24,5; 99,15; meros ts psukhs, part of the
pre-eminently, 89,2; prtos/deuteros, soul, 18,26-19,3 (cf. 27,3); 30,26;
first vs second, earlier vs subsequent, 73,22; morion ts psukhs, part of the
28,8-9.15-16 (cf. 21); 31,11-15; 75,30; soul, 81,2; cf. 98,24; orektik psukh,
cf. 30,4; prton/husteron, first vs soul for desiring, 78,24; 94,1; peri
subsequent, 40,6. For prta smata, psukhs legn, someone who speaks
see under sma; for prt psukh, see about the soul, 28,28; 30,18;
under psukh phantastik psukh, soul for
pslaphoun, probe, 100,4 representation, 69,24; 71,26; 93,25;
pseuds, false, 66,11; 67,4.11.13-14; 94,1; 97,11; phutik psukh, soul for
70,14; 70,25-71,4; 71,9-10.13.17; to vegetation, 35,24; 38,13; 75,31; 96,9;
pseudos, falsehood, 67,22; 70,7.10-13 praktik psukh, the soul for action,
pseudesthai, possess falsehood, 66,10 99,21; prt psukh, first soul, 16,20;
psophein, emit a sound, 47,4 35,24-6; 36,11.16; see also prt
psophtikos, able to emit a sound, 47,6 dunamis, under dunamis; threptik
psophos, sound, 26,3; 40,23; 42,2; psukh, soul for nourishing oneself,
46,20-1; 47,2-9.15.17-18.22; 31,25; 32,9-10.20; 33,11; 34,12.18.28;
48,7.15.20; 49,3-4; 50,4-6; 50,18; 35,21; 36,11.18-19; 37,6; 92,15;
56,6.10; 59,16; 60,18; 65,19 94,18.22-4; 95,9; 96,9; to tn psukhn
psukh, soul, 1,1-2; ekhon, that which has the soul, 2,20;
2,1-3.6-7.10.15.20.23; 9,12-14.25; 15,16; 16,14; 22,21; 23,22; 24,2.15;
10,2.9-13.16; 11,2-6.22; 12,2-4.7.21; 28,7; 30,8; 31,25; 32,20; 38,14;
13,9.12; 14,5.10-18.22.25; 39,1.19; 50,10; 84,27; tn then
15,1.4-7.9.12.16-29; 16,1-10.14.18-19; psukh, soul of the gods, 28,27;
17,5.9.14; 18,28; 19,2-6.9-15.18-21; trephousa psukh, soul that
20,6-7.23-4.27; 21,3-8.10- 11.16-17.22; nourishes, 34,24; ztik psukh, soul
22,6.10.20; 23,7.14.17.23-5; for life, 38,12; 95,9
24,1.4-5.8.11.15.18.22; 25,2-5; psukhikos, soul (adj.). For psukhik
26,9-10.14-15.18-28; 27,10.14.17.19; energeia, see under energeia; for
28,1-3.7-10.13.21.25-8; 29,5.20; psukhik dunamis, see under
30,7.17.25.29; 31,7-10.21-8; dunamis
32,1.6-7.16.25; 33,8; 35,8; 36,1; 37,13; psukhros, cold, 6,9; 26,8; 33,22-4; 79,4
38,12.21; 39,22; 40,5; 49,15.24; 52,3; psukhrots, coldness, 8,19; 19,35-6;
53,27; 60,4-6; 66,21; 68,7.12; 73,23; 40,25; 55,22; 58,26
Greek-English Index 211
psuxis, cooling, 77,15.18-19 68,6; 69,10; 75,28; 76,17; 77,1.4.14.21;
ptnos, winged, 13,14 92,4.19; 93,6; 94,7.25.29; 95,3.22.26;
pterux, wing, 96,4 96,1; 98,5.10-13; 100,2-3.12-14;
Puthios, the Pythian god (= Apollo), 1,6 apsukhon sma, inanimate body,
pur, fire, 5,4-12; 6,21.23; 8,21; 11,7; 63,5; empsukhon sma, animate body,
16,4; 19,22.35; 20,3; 26,17; 34,12; 11,8; 14,21; 63,5; 75,27-9; haploun vs
35,8; 43,9; 45,2.12; 46,1; 53,8; 79,2 suntheton sma, simple vs compound
body, 3,21-4,9; haploun sma, simple
rhein, flow, 54,9 body, 3,22.25.28; 4,10.16.19.27;
rhis, nose, 57,5 5,4.8.19; 7,16; 8,3-8; 8,26; 9,14;
rhiza, root, 37,18; 94,22 10,13.16; 11,9; 16,3; 19,29-20,6 (cf.
rhop, tendency, 12,10 19,21-3); 20,20.23-6; 26,22; 31,27;
rhuptikos, able to cleanse, 51,20; 53,6 34,9; 54,12; 58,20; 93,11.14; meros tou
rhuthmos, rhythm, 26,6 smatos, part of the body, 18,15-27;
24,7-9; 94,29 (cf. 95,7); 100,14;
salpinx, trumpet, 49,8 organikon sma, organic body,
saphs, perspicuous, 17,2; clear, 16,4.10-14; 24,11; 27,17; phusikon
32,24-5; saphs, clearly, 2,8; 3,3 sma, natural body, 3,14.16.22-4;
sarx, flesh, 56,15-23; 4,16; 5,4; 7,15; 8,2.5.13.26; 9,8; 10,1;
57,2-4.8-9.12-14.17; 58,3.7.24; 93,17; 16,3.10; 19,16; prta smata, basic
98,6.10.13-14; 100,15 bodies, 7,16; 10,14; 24,4; 26,27; 58,27;
smainein, signify, 13,11; 40,7; suntheton sma, compound body,
smainomenon, sense, 13,11 3,22-5; 4,5.23.27; 7,22-8,8; 8,26;
smeion, indication, evidence, 46,7; 9,9.23; 16,4; tekhnikon sma, 16,3;
52,5; 56,17.23; 94,27; 95,7-9; 96,22 theion sma, divine body, 43,9; 45,3.
spesthai, rot, 53,21 For sma dia smatos dierkhesthai,
sidros, blade, 98,11 see under dierkhesthai
skelos, leg, 18,20 smatikos, bodily; 21,15.27; 86,8;
skepasma, clothing, 80,14-15 smatikon morion, bodily part, 78,1.
skhma, shape (n.), 3,8; 4,11-15; 6,19; For smatik dunamis, see under
11,21; 14,1-2; 18,21; 20,22; 21,2.8; dunamis; for smatik kinsis, see
31,4; 41,5; 50,16; 65,13; 70,21; under kinsis; for smatik ousia,
72,7-10; 83,20; 84,9; 98,5-6 bodily substance, see under ousia; for
skhmatizein, shape (v.), 3,1.6; 19,25-6; smatikon organon, see under
34,3; 46,10; 48,13.16.19; 50,14 organon; for smatikon pathos, see
skhesis, relation, 43,3.10; 45,3; 56,9; 71,5 under pathos
skhizein, divide, 95,5 sros, pile, 47,22
sklrophthalmos, hard-eyed, 53,1 sstikos, able to preserve, sustaining,
sklrots, hardness, 55,22 35,26; 93,6.8
sklx, worm, 67,3; 74,5 steria, preservation, preserving, 35,14;
skopos, function, 35,10.13 36,5; 75,31; 93,4; 95,15
skotos, dark, darkness, 42,5; 43,17; szein, preserve, retain, 10,15;
52,14; 56,14 11,13.16.19; 15,6.7; 20,19; 37,9; 63,1;
Skrats, Socrates, 72,25 67,27; 68,9-10.16.20; 70,6.12; (pass.)
sma, body, 1,3, 2,20.24; 3,11.13; 4,5.8; persist, survive, 18,23-4; 19,8; 98,23
5,9.14-22; 6,7.12-21; 7,3.11-14.17; speirein, to sow seed, 32,11
8,16.24; 10,2-3.11.27-8; sperma, seed (n.), 32,2-4; 36,3.23;
11,2-4.7-9.12.22; 12,2-3.7-25; 37,10.14; 38,4; 92,20
13,5.8-12.16.18.21; 14,5-19.25; spermatikos, seed (adj.), 37,17
15,4-5.8-9.12.15-20.27; 16,2; 17,10-11; sphaira, ball, 48,18
17,15-19,8; 19,20-2; 20,2; sphodros, strong, intense, vivid,
21,11.14-18.22-4; 22,6.10-12.15-19.26; 50,20.23; 60,13; 63,3; 71,9. For
23,5.26; 24,2.20-2; 25,3.9; 26,18; sphodra aisthton, see under
28,1.7; 31,6.16-18.21-4; 34,7; 35,1; aisthtos; for sphodra phantasia, see
36,1.12.20; 37,6; 43,11; 44,20; 45,9.24; under phantasia; for sphodra noton,
46,15; 47,10; 48,8; 52,2.5; 54,2; 57,5; see under notos
58,9.19.23-7; 59,5; 60,7.10.15; 62,18; sphodrots, strength, 50,19
212 Greek-English Index
sphragis, signet ring, 72,8 sunamphoteros, composite, 17,11.19;
splankhnon, inner organs, 98,8 18,20; 20,24; 87,9
spoudaios, accomplished, 82,11 sunanapherein, carry along with,
stegein, retain, 44,7 42,14; 65,11.15
stenots, narrowing, 95,5 sunaperkhesthai, depart or leave
sterein, lose, 29,8; 59,24 along with, 42,22; 45,4
stereos, solid, 42,8; 45,5; 46,26; sunauxein, enlarge, 34,8-10
47,3-5.19; 48,1.17-18; 58,23 sunduo, pairing, 8,19
stereots, solidity, 98,13 suneinai, be compresent with, 22,14;
stersis, lack, 4,3; 30,15; absence, 45,6.18.22
52,13-17 sunekheia, continuity, 48,14; 98,17
stilbein, shine, 56,14 sunekhein, hold together, 21,18; 48,5
Stoa, the Stoa, 17,16 (cf. 18,2); 26,16 sunekhs, continuous, 18,25; 35,11;
stoikheion, element, 7,16; 26,9; 34,10 38,10; 48,13; 62,22; 63,27; 83,6;
stokhazesthai, aim, 95,15 85,21.25
stomakhos, oesophagus, 98,23 sunergein, contribute, 93,5.20
stratgos, general, 97,21 sungens, of a similar kind, 25,10; 32,2;
stratopedon, army, 97,20 58,10
struphnos, sour, 51,24; 55,11 sunhaptein, coincide, 46,13-14;
su, you, 61,1 conflate, 52,9
sullambanein, to give a synopsis, 60,3 sunhistanai, form, 3,13.15.21; 7,4;
sullegein, bring together, 27,2 32,14; 36,24-5; 37,18; 38,1.3; 58,20;
sullogismos, argument, 66,16 88,17; 92,3
sumbainein, follow, 26,12; happen, sunhodeuein, accompany, 22,14; 80,11;
41,11; sumbebkenai, is a 81,11
characteristic of, 41,6-7; 86,22; to sunhodos, coming together, 21,21
sumbebkos, extrinsic characteristic, sunhoran, spot (v.), 85,19
13,20; 14,24-15,5; 19,16; kata sunhuparkhein, be together, 64,11.16
sumbebkos, extrinsically, 22,14.19-21; sunkatathesis, endorsement,
86,27. For kata sumbebkos aisthton, endorsing, 67,16-20; 71,12.16.22-3;
see under aisthtos 72,13-26; 73,7-12.21; 97,15; 99,5
sumbebkos, see under sumbainein sunkatathetikos, for endorsing; to
summenein, remain, persist, 15,21; 21,20 sunkatathetikon, 99,3
summetapherein, carry along with, sunkatatithenai, endorse, 67,16;
14,24 72,19-21.25; 73,12; 78,14.19
summetria, balance, 25,6-9; 60,10-12; sunkeisthai, be composed,
commensurateness, 41,17 compounded, 10,2; 12,6; 18,20;
summetros, appropriate balance, 98,18 25,10.21; 26,12-13.16.19.23.26; 30,28;
sumpatheia, responsiveness, 100,1.4 31,2.16; 68,1; to sunkeimenon,
sumpaths, responsive, 15,17; 100,3 composite, 26,24-5; 67,27; 68,3
sumpauesthai, cease, 62,15 sunkinein, change jointly, 22,13.17
sumperasma, conclusion, 80,8-9.13-14 sunteleia, contribution, adding, 35,14;
sumphnein, agree with, 71,3 89,4
sumphnia, agreement, 2,22 suntelein, contribute, make a
sumphtheiresthai, perish along with, contribution, 7,24; 8,11.18; 12,4;
21,23; 90,15 14,7.21; 18,16.22; 36,6; 49,22.25;
sumphus, congenital, 56,22-3; 58,12 59,22; 75,26
sumphutos, congenital, 59,17. For suntetrainein, connect by passages,
sumphuton meros, see under meros; 96,29
for sumphuton pneuma, see under sunthesis, composition, 24,25.29;
pneuma 25,4-6.10-11; 26,5.14-15.19.24; 70,23;
sumplekein, interweave, 83,18 82,13; compound (n.), 31,11; 34,19.21;
sunagein, join, 21,18; 95,5 67,21.25; 68,1; 92,6; 98,13;
sunaisthanesthai, is conjointly aware, compounding, 83,12
50,18.22 sunthetos, compound (adj.), 2,27-3,2;
sunaisthsis, accompanying awareness, 3,22-7; 4,4-6.9; 7,7; 10,3.28; 26,17;
15,17; 65,9
Greek-English Index 213
31,10; 87,7.23-4; 92,3. For suntheton telos, end, 16,6; 24,14-16; 35,23;
sma, see under sma 36,14.16; 73,21.24-6; ta pros tou
suntithesthai, come to a resolution, telous, things leading up to the end,
2,11; accept, 72,3 24,15; tou telous kharin, for the sake
suntoms, in summary, 92,14 of the end, 24,15; see also hou heneka
suntrekhein, concur, 73,2 temnein, cut, carve, 18,15-17
sustasis, consistency, 3,19; formation trein, keep, preserve, 37,8; 50,27
10,19; 36,19; 44,9; 54,11.15; h prt trsis, preservation, 83,5
sustasis, the initial formation, 74,18 trtikos, able to preserve, 35,15
sustellein, contract, 77,2.4 thalassios, sea creatures, 67,3
sustol, contracting, contraction, 12,17; thalatta, sea, 53,9
74,21; 77,5 thanatos, death, 74,27; 95,10
suzeugnunai, link, 29,2.17; 32,8; 33,9; thauma, marvel, 2,14
73,27 thaumazein, be surprised, 8,13
theios, divine, 2,17; 32,13; 91,5. For
takhos, speed, 48,12 theion sma, see under sma
takhus, quick, 47,15.21-3; takhes, therein, engage in study, 84,9.21; to
quickly, 48,7 theroumenon, what is studied,
takhuts, speed, 47,19 89,23-90,1
tass, order, array, 43,19. For tetagmen thertikos, able to study, pertaining to
kinsis, see under kinsis study, theoretical, 87,15.18.22; to
taxis, order, sequence, 16,19; 28,10; thertikon, the part capable of study,
35,7; 68,4; 72,16; 73,20; 75,16; 92,15 98,27; 99,24. For thertikos nous, see
tekhn, art, 2.18-3.21; 5,2; 7,4; 15,11; under nous; for thertik opsis, see
24,9; 79,1; 88,21; to kata tekhnn under opsis
ginomenon, artefact, what is thertos, can be studied, object of
produced by art, 2,13; 3,2-21; 4,25.28; study, 80,7; 83,2
5,1; 6,5; 20,21; 88,21; cf. 7,4 theoria, study, 2,22
tekhnikos, artificial. For tekhnikon theos, god, 1,5.7; 2,2; 28,27
sma, see under sma therapeuein, treat, 100,6-7
tekhnits, artisan, 3,7; 78,29 thermainein, heat (v.), 84,3
tktos, can be melted, 3,18 thermos, hot, 6,8-11; 26,8; 33,22.24;
teleioun, develop, mature, 7,1; 37,5; 35,2-9; 40,1; 79,4; 84,3; to thermon,
66,7; 81,19; 82,8 heat (n.), 35,3-6; 49,13.21; 96,5-6; to
teleios, advanced, highly or fully entos thermon, internal heat, 49,20
developed, mature, 16,14; 32,11; thermots, heat (n.), 5,5; 8,19; 20,3.25;
35,21-2; 36,7; 66,2.6; 74,6; 80,20.23; 40,24; 51,17; 53,12; 54,9-11.15-16;
91,23; 92,13; 93,21; 94,4.12.15-16; 55,22; 58,26; 94,19; 98,16; heating,
teleioteros, more advanced, more 77,1.14.18-19
developed, 7,23; 8,10; 10,12; 11,4; thesis, position, placing, 41,16; 50,29;
16,21; 17,4; 28,16.17; 36,4; 52,4; arrangement, 68,3
58,13; 66,7; 74,6; 80,18-20.23; 93,21; thntos, mortal, 36,8
94,11.17; teleiotatos, most advanced, thrax, trunk, 13,18; 18,19; 49,22;
29,25; 66,6; most developed, 35,21; 95,16.22; 96,1-2.8
36,14 threptikos, capable of being nourished,
teleisis, development, 36,6 29,21; to threptikon, the part for
teleiots, culmination, 6,30-7,3; 7,6.8; nourishing oneself, 74,18.28-9; 75,5-12;
9,11.18-22; 10,8.26; 11,1; 16,1.5-10; 78,5; 99,8.19-21. For threptik psukh,
17,12; 19,26-8; 20,29-30; 23,18; see under psukh; for threptik
24,1.14; 37,11; 43,6-8; 52,2; 79,20; dunamis, see under dunamis
82,1; 85,11; 90,16; 92,20; 99,13-14; thrix, hair, 75,1; 97,5
prt teleiots, first culmination, 9,26 thrupsis, dispersion, 47,15.20.22-3
(cf. 20); teleiots tis teleiottn, a kind thruptein, disperse, 48,2.6
of culmination of culminations, 8,12 thumos, passion, 12,15; 74,2-4; 78,23
teleutaios, highest, 28,11; last, 48,15; thurathen, from outside, 90,20. For
95,10 thurathen nous, see under nous
teleutan, terminate, 97,11; 97,18
214 Greek-English Index
timiteros, superior, 29,25; 80,23; tupsis, impression, 68,11-12.16; 69,4;
89,19-20 70,4
timreisthai, take revenge, 74,3 tupoun, make impression, 72,7; 83,4
tode, a this; 83,7; 86,29; 87,3.15.18-20; tuptein, strike, 47,9-10.23
(to) tode ti, a specific thing, 4,24;
6,8.27; 7,12; 19,7.36; 87,7.13-15; to xros, dry, 26,8; 51,17; 53,3-4.18.25;
tide einai, what it is to be this, 54,11.17
87,1.8.18-20; see also to ti X einai xrots, dryness, 5,5; 8,19; 9,35; 19,35;
under einai 20,3.26; 40,25; 53,10.12; 54,8.17;
toikhos, wall, 42,16 55,22; 58,26; enkhumos xrots,
toionde, a such, 83,7; 87,15 flavoured dryness, 51,19-20;
topos, place, region, 10,23; 13,19; 53,6.15.19; geds xrots, earthy
14,17-24; 15,16; 20,15; 48,11; 70,22; dryness, 51,17; 53,11.15
95,2; kata topon, in location, 94,9. For xulos, wood, 3,5; 25,17; 53,11
kata topon kintik dunamis, see
under dunamis; for topos eidn, see zn, live, be alive, 2,3; 9,12-13; 16,13;
under eidos; for kata topon kinein, see 29,4-5.10; 31,7-8; 31,20-32,1; 37,5;
under kinein; for peri tn kardian 38,7-8.19; 59,24; 74,20; 78,4.7; 92,
topos, see under kardia 17-19; 92,24; 94,18-19; 95,12-13;
to ti n einai, see under einai 96,18.22-5; 100,10-12; to zn, living
trakhlos, neck, 13,17 thing, 9,12; 29,8; 31,7; 32,11;
trakhuts, roughness, 40,25; 55,23.29; 92,17.21-3. For arkh tou zn,
56,2 principle of life, see under arkh
trans, distinctly, 71,7 ztein, to query, 30,23; investigate, 81,12
trephein, nourish, 34,1.13.16.17.19.24; ztsis, inquiry, 33,8
35,3.13.24; 36,10-11; 93,1; z, life, 16,12-13; 78,5
trephesthai, be nourished, 16,15; 31,9; zion, animal, 10,1.4.8-10; 11,22;
32,21; 32,26-33,7; 33,15-19; 12,4.7.14-16; 13,10.13-14; 14,17;
34,4.8.12-13.18.20-2; 34,27; 15,8.29; 18,13; 19,22; 20,1.5-7.20.27;
35,5.7-14.19; 36,10-13.25; 38,3-5; 21,4-9; 22,10.21; 23,8.14; 24,7;
40,3; 74,19.24; 75,1; 92,24; 93,1-2; 27,16.19; 28,6; 29,14-15; 30,9; 31,15;
95,8; 96,18-21; nourishment, 9,2; 32,18; 35,11; 36,20; 37,5.12;
12,12; di hautou trephesthai, nourish 38,2.7-8.19; 39,23; 40,1.13-14; 49,3-5;
oneself, 9,13; 36,22 50,4; 51,10; 52,3; 53,1; 58,14.19.22-3;
tromos, shakes, 77,15 59,21-3; 66,3-5.12; 67,2; 71,26;
trop, variation, 77,13 72,15-17; 73,18.23.28; 74,18-19.29;
troph, nourishment, 33,6-7; 75,15.22.31; 76,5; 77,13; 78,5-6.24;
33,12-34,26; 34,27.28; 35,27; 79,2.8.16.27; 80,23; 92,23;
36,1.3.13.24.26; 37,9.16; 38,1-5; 93,5.10.14.18-19; 94,4.23.27; 95,23-5;
39,3.16; food, 52,21; 92,20.24; 94,25-8; 96,3.15.18.23-4; 97,4.10; 98,22;
95,17.21; 96,5; di hautou troph, 99,11.26; 100,2.8-11.17; alogon zion,
self-nourishment, 29,10; 78,5; 92,17; non-rational animal, 30,10.12-15;
cf. trephesthai di hautou under 67,19; 71,22; 98,26; hsper zion ti on
trephein; eskhat troph, last stage of en zii, like an animal within an
nourishment, 36,4; 96,28 animal, 97,4; logikon zion, rational
trophimos, nourishing, 92,26 animal, 30,9-12. For koin tn zin
tupos, imprint, 68,5.11-16.19; 69,4; psukh, see under psukh
72,5-6.13; 83,5; tupi, in outline, 60,3 ztikos, vital, 23,23; 24,12. For ztik
psukh, see under psukh
Index of Passages Cited

The passages of ancient authors cited are listed by the abbreviations used above,
together with the full title, with passages from these sources cited in bold italics. If the
passage is mentioned in a note to the translation, the reference is to the page and line
numbers of Alexanders text to which the note corresponds (e.g. 5,5-6); if the passage is
mentioned in the introduction, the reference is to the page and note numbers of the
introduction (with an i appended to the page number of the present volume (e.g. 12i or
15i n.40)). Passages that only partially overlap are listed as separate entries, so in
searching for a passage, the reader is advised to glance at neighbouring entries as well.

AESCHYLUS 3,9-11: 3,19-21; 5,2-3; 3,13-31: 5i;


PV (= Prometheus Vinctus) 942: 41,1-3 3,19-21: 3,9-11; 3,22-8: 5i; 3,23-4:
AETIUS 4,8-9; 3,23-5: 10,3; 17,18-19; 3,24-7:
Dox. Gr. (= Doxographi Graeci, Diels) 7,22; 3,26-7: 3,28-4,4; 5,19; 3,27:
1.3.4, 278.12-15: 19,21; 1.3.5, 3,22; 3,28-4,4: 5,20-1; 17,17-18;
279.12-17: 33,15-16; 1.9.4-5, 18,2-7; 3,28-4,20: 5i; 4,7-9: 5,21-2;
308.4-9: 3,28-4,4; 1.11.5, 310.6-7: 4,8-9: 6i; 17,11-12; 4,8-18: 5i;
7,9-13; 1.15.3, 313.8-20: 42,19-22; 4,9-15: 6,19; 4,9-20: 3,28-4,4;
4.3.10, 388.16.20: 21,13-16; 4.3.11, 4,12-17: 15,14-15; 4,20-1: 5i;
388.21-389.8: 26,17-18; 4.4.4, 4,20-5,1: 5,22-6,2; 4,20-5,2: 6i;
390.5-13: 32,1-4; 4.7.3, 393.1-7: 4,22-25: 17,11-12; 4,23-5: 4,8-9;
17,12-15; 4.8.6, 395.1-12: 41,3-5; 5,21-2; 4,24: 6,8; 5,1-2: 6,2-3; 5,2:
4.8.6, 395.12-15: 41,1-3; 4.13.7, 6,4-6; 5,4-6: 10,17-19; 5,4-18: 20i
403.26-8: 42,11-19; 4.21.1, n.14; 5,5: 7,18-20; 9,14-22; 5,5-6:
410.25-9: 24,5; 27,6-7; 4.21.1-4, 11i; 5,5; 9,4-7; 10,24-6; 22,5-12;
410.25-411.24: 27,6-7; 30,26-31,2; 24,18-20; 24,22-3; 25,2; 26,26-30;
4.21.2-3, 410.30-411.13: 18,25-19,1; 36,19-21; 5,7: 5,5; 5,11: 5,5; 5,11-15:
32,1-4; 40,4-5; 5.15, 425.14-426.14: 11,7; 20,19-22; 22,5-12; 5,18-19:
38,4-8; 5.15.2, 425.18-22: 38,4-8; 6,6-7; 6,6-20; 5,18-6,2: 6i;18,2-7;
5.27.1, 440.4-10: 33,15-16 5,19-6,6: 18,25-7; 5,20: 15,14-15;
ALCINOUS 5,21: 17,11-12; 5,21-2: 6i; 5,22-6,2:
Didask. (= Didaskalikos, Whittaker) 14,12-14; 17,11-12; 6,2-3: 5,1-2;
8, 162.32-39: 3,28-4,4; 8, 163.4-10: 6,2-4: 12,8; 18,10-14; 6,6-7: 6,14-15;
3,28-4,4; 11, passim: 17,15-17; 24, 6,6-12: 14,5-7; 18,10-14; 6,6-20:
176.41-177.3: 27,4-8; 25, passim: 18,2-7; 6,8: 4,24; 6,17-19: 5i; 6,20:
12,8; 25, 178.33-9: 27,15-28,2 4,9-15; 6,21-3: 6i; 31,10-14; 6,21-9:
ALEXANDER OF APHRODISIAS 15,28-9; 6,24-7,8: 10,7-8; 18,25-7;
in An. Pr. (= in Aristotelis Analytica 6,29-7,3: 16,5; 6,30: 7,23; 16,5-6;
Priora) 46,2-16: 7,6 7,4-5: 6,21-2; 7,9-11: 14,20-2; 24,13;
DA ( = de Anima) 1,1-3: 16,18; 24,5; 7,9-13: 6i; 7,9-14: 6i;10,17-19;
27,1-4; 1,2: 20i n.13; 1,4-2,4; 31,10-14; 7,12: 4,24; 7,13: 24,13;
28,25-9; 1,2: 5,1-2; 1,3-2,4: 21,24; 7,14-20: 7,18-20; 7,14-21: 5,5;
1,6-9: 39,4-5; 2,1: 1,8; 6,21-2; 39,4-5; 7,21-8,25: 11,11-13; 7,21-11,13:
2,4-9: 2i; 2,10-25: 4i; 11,5-6; 23,29-24,4; 7,23: 29,22-30,2; 7,24:
2,10-26,30: 1,2-4; 2,17: 3i; 2,17-18: 7,24-8,1; 7,24-8,13: 14,20-2; 8,5-13:
11,11-13; 2,25: 17,9; 2,25-3,13: 5i; 8,17-18; 8,5-17: 4i; 8,8-10: 8,8-10:
11,20-1; 19,24; 2,26-3,2: 10,3; 2,27: 11i; 8,10-12: 10,17-19; 8,10-13: 12i;
3,22; 3,2-15: 20,19-22; 3,2-21: 8,10-13.15-17: 23,29-24,4; 8,11:
2,12-15; 3,6: 3,7-9; 3,7-9: 11,20-1; 7,24-8,1; 8,13-15: 7,24-8,1; 8,13-25:
216 Index of Passages Cited
11i; 8,13-11,13: 2,18-22; 8,15-17: 20,26-30; 15,10-13: 20,26-30;
8,10-12; 10,17-19; 8,16: 19,31-6; 15,10-26: 8i; 15,11: 15,9-10;
8,17-11,13: 4i; 11i; 8,22: 5,5; 15,13-26: 20,26-30; 15,14: 15,9-10;
31,16-20; 8,22-5: 10,17-19; 15,14-15: 14,12-14; 15,15-17:
8,25-11,13: 20,22-6; 9,4-7: 11,11-13; 12,2-3; 21,10-11; 15,17: 21,13-16;
36,19-21; 9,11-12: 17,9; 23,29-24,4; 15,17-21: 21,11-13; 21,13-16;
9,11-14: 9,1-4; 16,1-2; 31,7-10; 15,26-8: 13,10-14,3; 15,26-9:
9,14-18: 14,2-3; 9,14-25: 20i n.14; 14,12-14; 16,4-10: 9,14-22; 16,5-7:
9,14-26: 20,19-22; 9,15-26: 16,8-10; 6,30; 16,7: 21,4-5; 16,7-18: 39,6-8;
9,17-19: 19,24-8; 9,23-6: 31,7-10; 16,10: 28,25-9; 16,11-12: 11,1;
10,4-14: 16,1-2; 10,7-10: 15,28-9; 16,18: 1,2-4; 17,1-8; 27,1-4;
10,8-10: 29,14-15; 38,18-19; 16,18-20: 28,25-9; 16,18-17,1: 13i;
10,12-17: 4i; 10,14-19: 10i; 28,15; 35,24; 40,5-6; 16,18-17,8:
10,14-26: 31,10-14; 10,15: 20,19-22; 28,15; 17,1: 17,1-8; 17,1-8: 30,17-20;
10,17-19: 12i; 5,5-6; 8,10-12; 40,6-9; 17,2-5: 28,17-20; 17,9: 21,22;
8,15-17; 8,16; 10,12-14; 10,24-6; 22,5-6; 31,20-5; 17,9-10: 21,22;
11,11-13; 23,29-24,4; 24,18-20; 25,2; 17,9-15: 6i; 15,10-13; 17,10-11:
26,26-30; 36,19-21; 10,24: 4,23; 17,5; 17,11-12: 6i; 4,8-9; 5,21-2;
10,24-6: 10i; 5,5-6; 8,10-12; 8,15-17; 17,11-12; 17,11-15: 17,9-10; 21,22;
10,17-19; 11,11-13; 23,29-24,4; 27,1-4; 17,14-15: 17,5; 17,15-17:
24,18-20; 25,2; 26,26-30; 36,19-21; 18,10-14; 19,3-5; 19,21; 31,2-6;
10,25: 20i n.20; 10,26-11,5: 16,2-4; 17,15-18,10: 6i; 17,15-20,26:
10,28-9: 9,4-7; 10,28-11,7: 16,1-2; 17,9-10; 21,22; 27,1-4; 17,17-18:
11,1: 16,11-12; 23,26-8; 11,5-13: 4i; 6i;18,10-14; 19,21; 17,18: 19,3-5;
20i n.14; 2,12-15; 2,17-18; 11,11-13: 17,19: 24,13; 18,2-3: 3,28-4,4;
11i; 8,10-12; 8,15-17; 10,17-19; 18,2-7; 18,2-7: 5i; 18,6: 18,1;
11,5-6; 23,29-24,4; 11,12: 20,19-22; 18,10-13: 18,10-14; 18,10-14:
11,14-12,7: 8i; 11,14-13,8: 17,9; 18,25-19,1; 18,15-17: 18,24;
11,17: 13,22; 15,7; 11,17-18: 14,8-9; 18,10-27: 6,6-7; 14,5-7; 18,26-19,3:
15,5-9; 11,18-20: 11,17-18; 12,5-6; 22,5-12; 19,2-3: 40,4-5; 19,6-9:
15,5-9; 19,31-6; 12,2: 11,15; 26,15-17; 19,21-3: 20,22-6;
11,15-17; 12,2-3: 13,10-14,3; 19,14-16: 19,9-16; 19,21: 19,21;
15,15-17; 20,6-7; 21,10-11; 12,3-4: 19,26-8: 19,21; 19,26-9: 20,22-6;
14,5-7; 17,17-18; 18,21-3; 12,5-6: 19,31-6: 20,1-6; 20,1.5: 20,1-6;
11,18-20; 12,6-8: 16,1-2; 12,7-13,8: 20,1-4: 20,1-6; 20,4-5: 20,1-6; 20,6:
9i; 11,15; 16,2; 12,9-11: 12,9-13,8; 20,1-6; 20,6-7: 12,2-3; 21,10-11;
12,9-13,8: 12,21-4; 21,25-6; 20,6-19: 20,8-19; 20,7-19: 14,8-9;
12,12-13: 12,9-13,8; 12,13-19: 20,6-7; 20,18-19: 20,8-19; 20,19:
12,9-13,8; 12,14: 12,9-13,8; 12,17: 21,16-21; 20,19-22: 20i n.14;
29,12-14; 12,17-18: 12,9-13,8; 12,19: 22,5-12; 20,23: 19,21; 20,25-6: 5,5;
29,12-14; 29,18-22; 12, 21: 20,26-21,21: 8i; 15,9-10;
12,9-13,8; 12,20-21: 12,9-13,8; 20,28-21,10: 14,4-5; 21,10-11:
23,4-5; 37,1-3; 12,21-2: 12,9-13,8; 12,2-3; 15,15-17; 21,11-13: 15,17-22;
13,1: 13,1-8; 13,4: 13,1-8; 13,5: 21,13-16; 21,13-16: 15,17-22; 21,16:
13,1-8; 13,8: 13,1-8; 13,9: 14,12-14; 21,13-16; 21,16-21; 21,16-21: 8i;
13,9-17,8: 17,9; 13,10: 14,12-14; 21,17: 21,16-21; 21,22-4: 21,22;
13,22: 11,17-18; 15,7; 26,3-7; 21,22-22,23: 7i; 21,22-24,17:
13,22-14,2: 20i n.14; 14,4-5: 8i; 5,11-15; 12,9-13,8; 14,24-15,1;
15,9-10; 20,26-30; 14,5-7: 18,10-14; 17,9-10; 27,1-4; 21,24: 28,6-8;
14,6: 14,10-11; 18,10-14; 14,6-7: 37,1-3; 21,24-5: 21,22; 22,2-6:
18,21-3; 14,8: 14,12-14; 15,9-10; 23,6-24; 22,4-12.15-23: 20i n.14;
14,8-9: 20,8-19; 14,12-14: 14,14-16; 22,5-6: 24,18-20; 22,5-12: 22,13-15;
14,18; 15,5-9; 15,14-15; 14,24-15,5: 22,15-23; 23,18-22; 23,29-24,4;
8i; 10i; 15,2-5: 17,17-18; 21,1-10; 22,7-10: 5,11-15; 22,7-12: 20,19-22;
15,3-5: 15,2-5; 17,11-12; 15,5-8: 22,14-15: 5,11-15; 22,18-20:
11,17-18; 15,5-9: 14,12-14; 15,7: 5,11-15; 22,15-23; 22,18-21:
13,22; 26,3-7; 15,9-26: 14,4-5; 20,19-22; 22,18-22: 22,5-12; 22,27:
Index of Passages Cited 217
22,24-23,6; 23,6-24: 7i; 23,15-17: 38,16-18; 29,14-15: 10,8-10;
23,11-13; 23,24-8: 11,1; 23,6-24; 29,16-18: 29,18-22; 29,18: 29,18-22;
23,24-24,4: 7i; 20i n.14; 23,26-8: 29,16-22: 29,1-3; 29,17: 29,16-18;
23,9-11; 23,26-24,3: 12,9-13,8; 29,22-30,6: 12i; 29,26-30,6: 13i;
23,29-24,1: 22,5-12; 23,29-24,3: 38,16-18; 30,2-3: 27,1-4; 30,3-4:
9,14-22; 14,2-3; 23,29-24,4: 8,10-12; 28,23-5; 30,5: 30,2-6; 30,6: 30,2-6;
8,15-17; 10,17-19; 36,19-21; 24,3-4: 30,6-17: 13i; 30,13: 10,8-10;
20i n.20; 24,9-11: 24,6-7; 24,16-17: 31,16-20; 30,13-15: 29,18-22;
28,8; 32,11-15; 24,18-20: 11,17-18; 30,15-17: 31,8-9; 30,17-20: 17,1-8;
25,2; 25,2-9; 25,23-26,3; 26,3-7; 28,15; 40,6-9; 30,18: 17,1-8; 31,7-25:
24,18-23: 11i; 36,19-21; 38,13-15; 31,7-37,3: 14i; 31,7-38,11:
24,18-26,30: 11,17-18; 27,1-4; 9,1-4; 31,7-92,11: 1,2-4; 31,10-25:
24,19: 20i n.20; 24,19-23: 26,21-2; 6i; 31,15: 31,16-20; 31,18-20:
24,22-3: 26,26-30; 24,23-4: 5,5-6; 31,16-20; 31,20-5: 17,9; 31,25-32,1:
24,23-25,2: 10i; 24,28: 5,5-6; 11,14; 32,1-4; 38,4-8; 31,27-8:
24,22-3; 26,26-30; 25,2: 20i n.20; 31,25-32,1; 31,28: 31,7; 32,2: 32,1-4;
5,5-6; 10,17-19; 26,7-13; 28,14; 32,4-5: 32,1-4; 32,6-7: 32,15-22;
25,2-3: 10i; 25,2-4: 26,15-17; 32,6-23: 38,13-15; 32,9-23: 40,5-6;
26,21-2; 25,2-9: 25,9-15; 25,3-9: 11i; 32,11-15: 14i; 36,6-9; 32,11-16:
25,4-7: 24,18-20; 26,3-7; 31,16-20; 35,22-3; 36,16; 32,11-13: 35,19-20;
25,7-9: 24,22-3; 25,2; 26,21-2; 32,14-15: 32,11-15; 32,15: 32,11-15;
26,26-30; 25,9-23: 25,2-9; 25,16-19: 32,15-22: 13i; 32,6-7; 35,19-20;
24,18-20; 26,3-7; 25,24-26,2: 32,17: 36,7; 32,18-19: 35,19-20;
24,18-20; 26,3ff.: 22,24-23,6; 32,22: 32,15-22; 32,23-33,12:
26,3-13: 25,2-9; 26,9-11: 26,7-13; 34,27-35,2; 35,24-5; 32,25-7:
26,10: 26,7-13; 26,11-13: 26,7-13; 40,15-19; 32,27-9: 32,30-33,5;
26,13-20: 26,15-17; 26,13-30: 32,29-30: 32,29-30; 32,30-33,5:
25,2-9; 26,15-17: 19,6-9; 26,21-2: 32,29-30; 33,1: 32,30-33,5; 33,7-10:
10i; 26,22: 20i n.20; 10,17-19; 25,2; 39,2-4; 40,15-19; 33,13-15: 36,3-4;
26,26-30: 36,19-21; 26,27-9: 5,5-6; 33,13-28: 14i; 33,13-34,26:
24,22-3; 26,29: 10,24-6; 27,1ff.: 2i; 34,27-35,2; 33,15-16: 33,13-15;
27,1-2: 1,2-4; 27,1-4: 40,4-5; 27,1-8: 33,15-34,18: 39,10-18; 33,18-19:
13i; 27,3: 27,1-4; 27,3-4: 28,3-13; 34,27-35,2; 33,25-6: 34,27-35,2;
27,3-28,13: 1,2-4; 27,4-8: 27,1-4; 33,28-34,26: 14i; 34,5-9: 35,8-9;
27,8-15; 28,4-6; 27,4-28,3: 27,1-4; 34,8-14: 34,18-26; 34,13-14:
27,5: 27,1-4; 27,7-8: 27,6-7; 27,8-15: 34,8-14; 34,14: 34,18-19; 34,16-17:
37,1-3; 37,11-18; 27,8-16: 27,1-4; 36,3-4; 34,18-19: 34,18-26;
27,8-28,2: 13i; 27,9: 12,21-4; 34,27-35,2: 35,24-5; 34,27-35,8:
27,9-11; 28,4-6; 27,12-28,2: 27,4-8; 34,27-35,2; 34,27-35,25: 38,13-15;
27,13: 12,21-4; 27,15-28,2: 20i n.23; 35,2: 11,1; 35,2-3: 36,9-13; 35,8-17:
28,3: 27,1-4; 28,3-5: 28,25-9; 34,27-35,2; 35,9-22: 13i; 35,10-13:
28,3-13: 13i; 28,4: 12,21-4; 28,4-6: 35,9-17; 35,11: 35,10-11; 35,13-15:
27,9; 28,4-13: 27,1-4; 28.26-8: 20i 35,9-17; 35,13-16: 14i; 35,15-17:
n.13; 28,8: 24,16-17; 32,11-15; 35,9-17; 35,17-25: 34,27-35,2;
28,13-29,1: 13i; 28,13-30,17: 27,1-4; 35,23-36,9: 40,5-6; 35,24: 38,12-13;
28,13-31,6: 1,2-4; 28,14-20: 40,6-9; 35,24-36,9: 35,24-5; 35,25-36,9:
28,14-29,1: 17,1-8; 28,8-13; 34,27-35,2; 35,26-36,9: 38,13-15;
30,17-20; 28,20-5: 40,5-6; 28,23-5: 35,27: 36,2-3; 36,2: 36,2-3; 36,3-4:
29,12-14; 29,22-30,2; 32,15-22; 36,13; 36,6-9: 14i; 32,11-15; 36,8:
40,11-15; 28,25-9: 29,1-3; 29,3-10; 36,6-9; 36,9-13: 34,18-26; 35,2-3;
29,22-30,2; 38,16-18; 28,25-29,2: 36,13-15: 30,10-12; 36,22-37: 14i;
1,2; 28,26: 28,25-9; 28,26-9: 28,25-9; 37,1: 37,1-3; 37,1-3: 37,11-18;
29,1-3: 13i; 29,16-18; 32,6-7; 32,8; 37,4-5: 38,8-11; 37,4-38,11: 13i;
32,15-22; 35,9-17; 35,19-20; 29,1-6: 37,12-15: 37,11-15; 37,18-38,4:
35,24; 29,1-10: 12i; 9,1-4; 31,7; 37,5-11; 38,2: 2,18-22; 38,4-8: 14i;
38,12-13; 29,2: 35,26; 29,10: 38,7-8: 36,22-3; 38,8-11: 37,4-5;
29,3-10; 29,10-22: 12i; 29,12-22: 38,12-13: 35,24; 38,12-92,14: 14i;
218 Index of Passages Cited
38,21-39, 2: 39,10-18; 42,19-22; 44,15; 43,8-9: 42,22-43,4; 43,8-11:
42,22-43,4; 44,2-3; 44,9-13; 17i; 45,2-3; 43,8-46,19: 40,9-10;
38,21-39,6: 14i; 38,21-39,8: 43,9-10: 42,22-43,4; 43,11: 17i;
38,21-39,2; 39,1: 39,10-18; 39,1-2: 42,19-22; 42,22-43,4; 43,11-16:
15i; 39,10-18; 39,2: 38,21-39,2; 39,5: 42,8-11; 42,19-22; 44,21-3;
38,21-39,2; 39,8: 39,6-8; 39,9-10: 43,11-44,1: 43,8; 43,11-44,2: 17i;
14i; 39,10-18: 38,21-39,2; 44,2-3; 43,12: 43,11-16; 44,21-3; 43,13-15:
39,12: 38,21-39,2; 39,10-18; 43,11-16; 44,9-13; 43,14: 43,11-16;
39,12-18: 34,14-18; 39,13: 43,16-17: 42,5-7; 43,16-18: 16i;
38,21-39,2; 39,13-14: 15i; 39,14-18: 43,11-16; 43,16-18; 44,13-14;
14i; 39,16: 39,10-18; 39,18: 43,18-20: 44,23-5; 43,18-44,2:
39,18-21; 39,18-19: 39,10-18; 42,19-22; 42,22-43,4; 43,16-18;
39,18-21: 15i; 43,11-16; 39,20: 44,21-3; 43,19: 43,18-44,1; 43,20:
41,1-3; 39,20-21: 39,18-21; 39,21-2: 43,11-16; 44,1-9: 43,8; 44,2-3:
24,5; 39,21-40,3; 39,21-40,3: 15i; 44,9-13; 44,3-4: 42,7-8; 46,6-11;
44,6; 40,2-3: 36,3-4; 40,4-10: 17,1-8; 44,3-9: 44,9-13; 44,5: 39,21-40,3;
28,15; 40,9-10; 40,5: 40,5-6; 40,6-9: 44,5-45,10: 17i; 44,5-46,19: 42,4;
40,5-6; 40,10-15: 28,23-5; 40,11: 42,5-7; 44,7: 44,1-2; 44,6; 44,8-9:
40,11-15; 40,11-15: 40,5-6; 44,6-9; 44,9-11: 43,11-16; 44,9-13;
40,14-15: 10,8-10; 40,20-42,3: 44,9-13: 43,8; 44,10: 44,2-3;
40,9-10; 40,21-41,5: 40,20-1; 44,11-13: 44,9-13; 44,12: 38,21-39,2;
41,10-13; 43,18-44,1; 40,26: 44,13-14: 46,1-6; 44,13-15: 43,7-8;
40,21-41,5; 41,1-3: 40,21-41,5; 43,8; 44,14-15: 44,13-14; 44,16-19;
41,3-5; 42,8-11; 41,1-5: 16i; 44,16: 17i; 44,13-14; 45,1; 45,5-6;
39,18-21; 41,3: 41,1-3; 41,4-5: 16i; 46,1-6; 46,6-11; 44,16-46,19: 43,8;
41,1-3; 41,5-10: 40,20-1; 41,6-10: 44,17-18: 44,16-19; 44,17-19:
16i ; 41,7: 41,8-10; 41,9: 15i; 46,1-6; 44,18-19: 44,13-14; 45,20;
41,8-10; 41,13-15: 16i; 40,21-41,5; 44,18-23: 17i; 44,4-5; 44,20-1: 17i;
41,13-42,3: 16i; 41,14: 41,13-42,3; 42,7-8; 44,13-14; 44,21-3; 44,23-5;
41,20: 41,1-3; 42,4: 43,8; 42,4-7: 16i; 44,20-3: 44,21-3; 44,21: 4,23;
43,16-18; 44,13-14; 42,4-43,4: 42,4; 44,21-3: 44,20-1; 44,23-5: 17i;
42,4-43,8: 40,9-10; 42,5-7: 44,21-3; 43,5-8; 44,13-14; 44,20-1;
42,5-11: 17i; 42,6: 42,5-7; 42,6-7: 44,23-45,1: 44,4-5; 46,6-11;
43,11-16; 42,7: 44,5; 42,7-8: 44,16; 44,23-45,5: 43,8; 45,1: 17i; 44,16;
45,5-6; 42,7-12: 43,11-16; 42,8: 45,1-4: 42,22-43,4; 43,8-11; 45,2-5:
42,7-8; 42,19-22; 42,8-11: 42,5-7; 17i; 46,1-6; 45,3: 43,8-11; 45,5-11:
43,7-8; 44,21-3; 42,9: 41,1-3; 45,10-19; 45,5-46,6: 44,4-5;
42,9-10: 43,4-5; 43,11-16; 42,11: 44,16-19; 44,21-3; 45,7: 44,4-5;
42,19-22; 42,11-12: 43,11-16; 45,7-10: 42,7-8; 45,7-11: 44,13-14;
42,11-19: 17i; 44,23-5; 42,12: 45,7-19: 46,1-6; 45,9-10: 42,7-8;
42,11-19; 42,19-22; 42,12-19: 45,10-19: 44,13-14; 45,17-19: 17i;
43,18-44,1; 42,14: 42,11-19; 45,5-6; 45,20: 45,24-46,1; 46,1-6;
42,18-19: 42,11-19; 42,19: 39,18-21; 45,21-2: 46,12-17; 45,25-46,1:
42,22-43,4; 42,19-22: 17i; 42,11-19; 45,20; 46,1-4: 44,13-14; 46,3: 44,16;
42,22-43,4; 43,11; 43,11-16; 46,4-6: 45,7-10; 45,10-19; 46,6-7:
43,18-44,1; 44,21-3; 42,19-43,11: 46,1-6; 46,6-12: 44,4-5; 46,7-9:
43,8; 42,21: 42,19-22; 42,21-2: 42,7-8; 42,7-8; 46,7-11: 17i;
44,21-3; 42,22-3: 42,22-43,4; 46,12-14: 17i; 46,12-17: 45,21-2;
42,22-43,1: 17i; 42,22-43,4: 17i; 46,12-19: 44,4-5; 44,16-19; 46,19:
42,19-22; 43,8-11; 43,11; 45,3-5; 45,10-19; 46,20-1: 39,4-5;
45,6-7; 43,1: 42,22-43,4; 43,1-2: 46,20-47,24: 40,9-10; 47,25-51,4:
42,19-22; 42,22-43,4; 43,2: 40,9-10; 50,9-11: 39,4-5; 50,11:
42,22-43,4; 43,8-11; 45,2-3; 43,3: 38,21-39,2; 50,15-18: 39,18-21;
42,22-43,4; 43,4-5: 42,8-11; 50,18-51,6: 41,5; 51,5-53,29:
43,4-45,5: 42,4; 43,5-8: 22i n.48; 40,9-10; 51,26: 40,11-15; 52,18-19:
42,5-7; 44,20-1; 44,23-5; 43,7-8: 44,5; 52,20: 41,1-3; 53,26-9: 39,4-5;
42,19-22; 43,5-8; 43,11-16; 44,13-14; 53,28: 38,21-39,2; 53,30-54,2:
Index of Passages Cited 219
39,4-5; 53,30-55,14: 40,9-10; 54,1: 39,13-14; 83,17-22: 16i; 41,3-5;
40,11-15; 54,1-5: 40,11-15; 54,21: 83,19: 16,2; 83,23-84,6: 42,19-22;
38,21-39,2; 55,12-14: 39,4-5; 55,14: 84,4-6: 38,21-39,2; 84,10-24: 21,24;
38,21-39,2; 55,15-59,24: 40,9-10; 28,6-8; 84,10-85,5: 44,3-4; 84,12:
55,26-56,3: 40,21-41,5; 56,1-3: 41,5; 39,4-5; 84,12-13: 38,21-39,2; 87,5:
56,5-9: 41,10-13; 56,8-10: 16,2; 88,4: 16,2; 88,11: 16,2;
42,22-43,4; 57,13-14: 43,11-16; 88,12-13: 18,1; 88,23-89,11:
59,1-20: 44,3-4; 59,8: 38,21-39,2; 44,13-14; 88,26-89,2: 44,13-14;
59,11-12: 33,18-24; 59,14: 41,1-3; 89,9-15: 28,25-9; 89,12: 44,13-14;
44,7; 59,14-15: 42,7-8; 59,19-20: 89,13: 16,2; 90,2: 16,2; 90,11-91,6:
44,3-4; 59,20: 40,11-15; 60,1-66,8: 21, 24; 90,13-16: 21, 24; 90,19-21:
40,9-10; 60,3-6: 39,13-14; 60,6: 37,1-3; 90,23-91,4: 21, 24; 91,2:
39,18-21; 60,14-16: 41,10-13; 37,1-3; 92,12-15: 1,2-4; 92,15-94,6:
60,14-17: 39,21-40,3; 61,24-7: 1,2-4; 28,23-5; 40,11-15; 92,17: 31,7;
39,4-5; 61,26: 41,10-13; 62,3-7: 92,18-20: 9,12-14; 92,19-21: 36,3-4;
42,11-19; 42,19-22; 43,18-44,1; 62,4: 92,21-3: 10,8-10; 39,13-14;
43,11-16; 62,5: 41,1-3; 62,5-7: 92,23-93,24: 39,9-10; 40,11-15;
44,21-3; 62,11-13: 42,19-22; 62,12: 92,28: 41,1-3; 93,4: 40,11-15;
41,1-3; 62,16-20: 39,21-40,3; 93,17-18: 10,8-10; 94,7-100,17: 2i;
62,22-63,4: 38,8-11; 63,8-64,11: 1,2-4; 24,5; 39,21-40,3; 94,17-20:
39,18-21; 63,15: 39,18-21; 63,16: 39,21-40,3; 94,26-95,8: 36,3-4; 95,1:
39,18-21; 63,18: 39,18-21; 41,1-3; 96,5-8: 35,2-3; 96,8: 11,1;
63,28-65,2: 39,21-40,3; 64,2: 96,22-3: 40,11-15; 96,25-97,14:
39,21-40,3; 64,8: 39,18-21; 64,12-20: 39,21-40,3; 97,4-6: 12,2-3; 15,15-17;
39,4-5; 64,17-65,2: 39,18-21; 64,18: 97,6: 39,18-21; 97,26-7: 36,3-4;
39,18-21; 64,20: 39,21-40,3; 65,8: 98,1-7: 39,21-40,3; 98,4-5: 9,4-7;
39,18-21; 65,10.16-21: 16i; 98,8-9: 27,9; 98,9: 12,21-4; 98,24-5:
65,10-21: 41,1-3; 41,3-5; 65,11-16: 24,5; 99,17-24: 39,4-5; 99,19-30:
16i; 41,3-5; 65,13-14: 41,5; 66,5: 39,21-40,3; 100,1-8: 15,15-17;
10,8-10; 66,9-19: 39,4-5; 66,9-73,13: 100,9-13: 37,4-5; 100,16: 41,1-3
12,17; 29,16-18; 66,13-14: 39,13-14; Eth. Prob. (= Ethica Problemata) 9,
66,15: 41,13-42,3; 66,17: 39,4-5; 129,22: 24,6-7; 12, 133,12: 24,6-7;
66,24-8: 39,5-6; 67,2: 10,8-10; 30, 162,7-9: 33,18-24; 25, 148,15:
68,4-10: 12,17-18; 68,6: 39,18-21; 3,15-16
69,6: 41,10-13; 69,11-14: 39,5-6; Fat. (= de Fato) 1,164,1-3.13-15: 19i
69,14-70,5: 12,17-18; 69,22: n.4; 4, 167,16-168,24: 24,16-17; 4,
39,18-21; 70,5-71,5: 41,10-13; 168,3-7: 35,5-8; 4, 168,14-15:
70,10-12: 41,10-13; 70,20-3: 35,5-8; 6, 170,9-11: 35,5-8; 6,
41,10-13; 71,26-72,5: 39,9-10; 170,9-171,7: 1,6; 6, 171,7-11: 1,6;
72,3-5: 39,4-5; 73,14-80,15: 14i; 175,17: 20,4; 176,10: 20,4; 11,
12,14-15; 12,19; 29,18-22; 73,20-6: 178,11-12: 27,9; 11, 179,25-6: 27,9;
39,4-5; 73,26-8: 29,18-22; 74,1: 17, 185,11-33: 1,6; 17, 188,11-17:
12,14-15; 29,18-22; 74,17-23: 37,1-3; 1,6; 22, 192,12: 2,18-22; 22, 192,28:
38,4-8; 74,23-5: 35,10-11; 2,18-22; 30, 200,24-201,3: 1,6; 30,
74,29-75,2: 12,2-3; 15,15-17; 201,13-21: 1,6; 31 passim: 1,6; 31,
75,13-15: 39,4-5; 75,31-76,6: 202,9: 1,6; 31, 203,1-24: 1,6
28,8-13; 76,1-3: 39,4-5; 76,8-12: Fragmenta (Giannakis) fr. 16: 43,11
39,4-5; 76,15-78,2: 12,14-15; 78,5-6: Fragmenta (Moraux 1942) fr. IIIa:
9,12-14; 78,6-7: 10,8-10; 78,7-8: 41,1-3
39,13-14; 78,10-13: 39,18-21; Mant. (Mantissa) 101,3-18: 1,3;
78,13-21: 39,4-5; 78,24-5: 21,25-6; 101,18-21: 4,24; 101,28-102,1:
78,24-79,21: 12,9-13,8; 79,10-11: 3,15-16; 102,1-7: 3,7-9; 103,9-10:
21,13-16; 79,19-20: 8i; 14,4-5; 31,7; 103,10-20: 9,14-22; 103,11-20:
20,26-30; 80,8: 39,4-5; 81,5: 16,8-10; 103,29: 5,1-2; 103,32-3:
41,10-13; 81,17-18: 38,21-39,2; 3,22; 103,36-8: 9,1-4; 103,38-104,5:
81,24-5: 44,21-3; 82,13: 39,4-5; 37,11-18; 104: 21i n.24; 104,5-8:
82,18: 39,4-5; 83,13-84,10: 37,11-18; 104,8: 41,1-3; 104,10: 1,2;
220 Index of Passages Cited
104,19: 3,28-4,4; 104,21-2: 20,6-7; 142,3-8: 42,22-43,4; 142,5: 43,8-11;
21,10-11; 104,27-34: 9i ; 104,28-34: 142,8-10: 43,7-8; 142,10-16:
5,5-6; 10,17-19; 10,24-6; 105,4-12: 42,19-22; 142,13-16: 43,11; 142,17:
31,25-32,1; 105,6-12: 11,14; 44,7; 142,21: 44,7; 142,21-31:
105,10-12: 9,1-4; 105,12: 31,7; 44,1-2; 142,26-32: 44,6-9; 142,31:
105,12-13: 10,8-10; 106,24-9: 44,7; 142,31-143,1: 41,1-3; 142,32:
44,3-4; 106,30-107,1: 42,7-8; 39,18-21; 143,4-10: 42,19-22;
106,30-107,7: 44,3-4; 107,6-7: 143,6-7: 43,11; 143,11-29:
39,4-5; 107,7-20: 44,3-4; 108,3: 16,2; 42,22-43,4; 143,17-18: 42,19-22;
108,21: 16,2; 113,25-118,4: 143,19-23: 43,11; 143,21:
14,12-14; 17,9-10; 113,26-8: 42,22-43,4; 143,23-8: 43,11;
14,12-14; 115,6-12: 19,6-9; 143,26-8: 43,11; 143,29-33: 43,11;
115,12-14: 18,2-7; 115,28-116,1: 143,31-4: 43,11; 143,35-144,6:
13,10-14,3; 116,1-2: 18,7-10; 42,8-11; 144,5-6: 44,16; 44,23-5;
116,5-6: 11,17-18; 116, 5-13: 144,16: 44,16; 144,18-19: 43,8-11;
11,18-20; 116,7: 11,17-18; 116,12: 144,19: 42,22-43,4; 144,19-23:
11,17-18; 116,13-15: 18,7-10; 42,4-5; 43,8-11; 144,23-4: 43,8-11;
116,15-18: 15,28-9; 116,18-31: 144,29: 43,7-8; 144,31-2: 42,19-22;
7,9-13; 117,20-1: 17,11-12; 118,6-7: 144,31-145,7: 42,19-22; 42,22-43,4;
27,1-4; 118,6-9: 27,4-8; 144,34-154,2: 42,19-22; 145,2-3:
118,6-119,20: 27,1-4; 118,12-14: 43,8-11; 145,2-5: 43,11; 145,2-7:
11,14; 118,12-16: 31,25-32,1; 42,8-11; 42,19-22; 145,25-30:
118,19: 31,7; 118,19-21: 11,14; 42,11-19; 145,26-7: 41,1-3; 145,27:
31,7-10; 118,19-25: 31,25-32,1; 41,1-3; 146,17-18: 43,18-44,1;
118,25-6: 32,1-4; 118,31-5: 27,4-8; 146,19: 44,7; 146,20-8: 43,18-44,1;
28,4-6; 119,21-122,15: 14,24-15,1; 44,1-2; 146,21-5: 42,8-11; 146,25:
119,31-120,17: 15,2-5; 120,2: 4,8; 44,7; 146,31-3: 44,7; 147,16-17:
120,5-7: 6,4-6; 120,13-17: 15,2-5; 42,19-22; 147,27-9: 44,5; 147,27-30:
17,11-12; 121,17-19: 5,1-2; 6,2-3; 44,20-1; 147,30-148,2: 44,21-3;
6,4-6; 121,23-5: 6,3-4; 121,25-7: 148,1-2: 44,20-1; 148,2-6: 44,16-19;
14,24-15,1; 121,29-30: 6,12-13; 148,4-5: 45,22-4; 148,7-9: 43,5-8;
121,31-122,4: 6,6-20; 121,32-4: 44,23-5; 148,7-10: 44,13-14;
4,9-15; 122,4-15: 6,3-4; 148,9-12: 44,20-1; 148,11-14:
122,16-125,4: 17,15-17; 45,7-10; 148,14-16: 43,7-8; 148,23:
122,22-123,3: 21,29-22,2; 124,9-10: 44,13-14; 148,25-7: 3,11-13;
17,15-17; 42,19-22; 43,11; 126,25-8: 148,29-149,3: 44,21-3; 149,1-3:
5,5; 128,5: 44,7; 128,5-7: 42,4-5; 44,20-1; 149,11-13: 34,8-14;
128,14-15: 42,19-22; 129,21-4: 149,15-32: 45,7-10; 149,24-6:
43,18-44,1; 129,2: 44,7; 129,5: 44,7; 44,13-14; 149,26-9: 42,7-8;
129,10: 44,7; 129,21: 44,7; 149,32-5: 44,13-14; 149,34-5:
130,10-12: 42,19-22; 130,15: 44,7; 44,13-14; 149,36-150,6: 45,7-10;
132,37: 43,11; 133,18-19: 42,7-8; 150,2: 44,16; 150,7-9: 45,10-19;
134,27-136,28: 42,19-22; 135,6-10: 150,13-18: 44,16-19; 150,13-19:
44,7; 136,14: 44,7; 136,16: 44,7; 45,24-46,1; 157,15-18: 27,9;
136,18: 44,7; 136,29-138,2: 162,32-163,1: 1,9; 163,1-18: 1,9;
42,19-22; 138,3-139,28: 43,11; 163,21-31: 1,9; 163,24-30: 27,9;
138,14-23: 42,19-22; 138,15-16: 164,25-7: 27,9; 165,29-30: 27,9;
42,4-5; 138,16-18: 42,22-43,4; 166,11: 27,9; 167,17-168,1: 1,9;
138,22-3: 42,4-5; 139,17-19: 178,26-9: 27,9; 181,21-7: 1,2;
42,22-43,4; 43,11; 139,23: 43,11; 182,32: 1,2; 183,25-6: 27,9;
139,29-141,28: 20,8-19; 140,22-5: 185,14-15: 35,5-8; 185,33-186,4:
20,8-19; 141,9-22: 20,8-19; 1,6
141,29-147,25: 43,16-18; 141,33: in Metaph. (= in Aristotelis
42,19-22; 141,35-6: 41,1-3; Metaphysica) 20,5-6: 6,21-2; 32,1-5:
141,36-142,16: 42,8-11; 142,1-2: 35,5-8; 72,34-6: 12,10-11; 77,24-6:
43,7-8; 142,2: 42,19-22; 142,2-4: 1,2; 101,22-6: 2,25-3,13; 101,28-30:
42,19-22; 142,2-5: 42,22-43,4; 35,5-8; 103,8: 6,14-15;
Index of Passages Cited 221
103,31-104,18: 35,5-8; 110,4: 19,21; 177,20-178,10: 3,17-19; 186,21-6:
112,11-16: 19,21; 117,25: 1,3; 33,13-15; 186,24-6: 33,15-16;
121,16-18: 24,16-17; 121,17-19: 186,35-187,3: 35,2-3; 197,3:
5,1-2; 127,3-5: 5,5; 127,12-17: 5,5; 3,15-16; 200,25-6: 42,7-8;
142,13-16: 44,13-14; 158,7-13: 212,24-33: 3,19; 223,25-224,11:
6,14-15; 158,9-14: 3,28-4,4; 159,3: 15,2-5; 224,14-19: 3,28-4,4;
3,28-4,4; 159,10-26: 6,14-15; 224,24-225,16: 9,4-7; 225,3-33:
159,11-17: 3,28-4,4; 159,27-160,27: 15,2-5; 226,5-6: 3,19; 227,11-14:
28,8-13; 160,17-18: 24,16-17; 3,19; 227,15-22: 9,4-7
161,8-9: 10,8-10; 178,15-19: Mixt (= de Mixtione) 1, 213,2-13:
17,12-15; 185,24-198,3: 33,13-15; 11,18-20; 2, 214,28-215,8: 26,17-18;
198,29-30: 10,8-10; 212,35-213,4: 3, 216,4: 24,23-25,5; 3, 216,4-14:
4,23; 213,2-4: 3,28-4,4; 225,3-5: 11,17-18; 3, 216,14-17: 15,15-17;
17,15-17; 251,36-7: 1,2; 259,17: 5,5; 19,6-9; 20,8-19; 3, 216,17-22:
229,31-230,1: 17,12-15; 263,25-33: 11,15-17; 3, 216,22-5: 11,18-20; 3,
28,8-13; 310,23-33: 1,2; 311,28-9: 216,25-217,2: 11,18-20; 4, 217,27-9:
41,13-42,3; 311,31-2: 41,13-42,3; 20,8-19; 4, 217,32-218,1: 11,17-18;
312,1-2: 39,5-6; 312,2-5: 12,17-18; 4, 217,33-6: 20,8-19; 4, 217,36:
312,8-9: 39,5-6; 312,24-31: 12,2-3; 4, 218,1-2: 20,8-19; 4,
41,13-42,3; 313,20-1: 41,13-42,3; 218,2-5: 5,5; 4, 218,6: 24,23-25,5; 5,
313,20-32: 41,10-13; 313,32-314,1: 218,17-19: 20,8-19; 5, 218,21-4:
41,13-42,3; 313,36: 41,13-42,3; 20,8-19; 6, 219,9-10: 20,8-19; 6,
315,33-4: 10,8-10; 316,18-19: 220,13-16: 20,8-19; 7,
10,8-10; 321,24-30: 15,15-17; 220,23-221,25: 11,18-20; 7,
323,26: 38,21-39,2; 324,33-325,1: 221,11-15: 11,18-20; 7, 221,20-5:
38,21-39,2; 329,18-21: 33,18-24; 11,18-20; 9, 222,31-2: 15,10-13; 10,
334,31-4: 11,17-18; 349,14-16: 223,7-9: 15,15-17; 10, 223,30-3:
6,30-7,2; 24,16-17; 357,13-18: 15,15-17; 10, 224,14-15: 26,15-17;
3,28-4,4; 357,16-18: 4,23; 357,25-9: 11, 225,3-5: 20i n.17; 3,28-4,4; 11,
12,10-11; 358,36-359,4: 4,23; 225,5-8: 26,15-17; 11, 226,15:
359,10-11: 3,28-4,4; 4,23; 359,21-2: 3,28-4,4; 12, 227,1-10: 20,8-19; 12,
7,2; 359,30-2: 7,2; 363,21-4: 227,8-9: 15,15-17; 13, 228,17-20:
15,15-17; 373,5-8: 1,2; 15,10-13; 13, 228,20-1: 4,20-2; 13,
374,37-375,9: 1,3; 386,33-5: 228,22: 4,8; 13, 228,27-36: 8,16; 13,
32,29-30; 393,2-17: 32,11-15; 228,28-34: 11,15; 13, 228,34-6:
399,1-6: 12,3-4; 400,28-31: 12,3-4; 11,18-20; 13, 229,30-230,2: 6,28;
402,8-13: 38,21-39,2; 13, 229,30-230,5: 5,5; 14,
406,25-407,15: 38,21-39,2; 230,17-19: 11,18-20; 19,31-6; 14,
409,26-36: 38,21-39,2; 417,14-19: 230,26-34: 19,31-6; 14, 231,1-4:
30,26-31,2; 421,33-4: 3,17-19; 19,31-6; 14, 231,10-12: 11,18-20;
421,34-6: 3,28-4,4; 422,16-17: 7,2; 19,31-6; 15, 231,15-19: 11,18-20;
423,28: 33,18-24; 423,33-9: 15, 231,25-6: 11,18-20; 15,
30,26-31,2; 424,17-19: 18,17; 232,13-18: 11,18-20; 15, 232,28-31:
424,29-31: 31,2-6; 429,12-14: 11,18-20; 16, 233,23-238,23:
12,3-4; 429,17-19: 12,3-4; 430,3-6: 35,9-17; 16, 233,26-7: 34,18-19; 16,
3,28-4,4; 4,23; 430,8-9: 3,17-19; 233,30-234,11: 34,8-14; 16,
433,10-11: 41,10-13; 433,14-15: 234,11-15: 34,18-26; 16, 236,19-30:
41,10-13 34,18-26; 16, 236,24-6: 35,9-17; 16,
in Meteor. (= in Aristotelis 238,10-20: 34,18-26
Meteorologica) 4,25-5,1: 5,5; 6,5-6: Quaest. (= Quaestiones) 1.1, 3,10-14:
1,2; 6,15-16: 6,14-15; 14,18-19: 5,5; 28,25-9; 1.1, 3,15-19: 28,25-9; 1.2,
58,26-59,1: 5,5; 78,11-12: 6,14-15; 5,2-3: 44,16-19; 1.2, 5,3-8: 45,21-22;
129,3-14: 43,11; 141,11-12: 1.2, 5,11-15: 46,6-11; 1.2, 5,19-23:
43,16-18; 141,12-16: 43,18-44,1; 46,11-13; 1.2, 5,21-2: 44,16-19; 1.2,
141,23-9: 43,18-44,1; 143,13-14: 5,10-19: 44,4-5; 1.2, 5,11-15: 42,7-8;
42,4-5; 148,9-11: 39,9-10; 149,28: 1.2, 5,12-19: 44,16; 1.2, 5,14-16:
44,7; 158,27-159,3: 41,13-42,3; 41,1-3; 1.2, 5,15-19: 44,13-14; 1.2,
222 Index of Passages Cited
5,18-19: 41,1-3; 1.2, 5,22-3: 44,20-1; 5,5-6; 2.27, 77,14-15: 10,8-10;
1.2, 5,30-6,1: 44,21-3; 1.2, 6,1-3: 3.2-3.3 passim: 38,21-39,2;
44,20-1; 1.2, 6,3-7: 45,22-4; 1.2, 3.2, 81,9: 39,6-8; 3.2, 81,27-9:
6,7-8: 44,4-5; 1.2, 6,7-8: 42,7-8; 1.2, 38,21-39,2; 3.3, 83,3: 38,21-39,2;
6,8-10: 42,19-22; 1.2, 6,9: 41,1-3; 3.3, 83,18: 38,21-39,2; 3.3, 84,8-12:
1.2, 6,10-11: 42,22-43,4; 1.2, 6,13: 38,21-39,2; 3.3, 84,18: 38,21-39,2;
41,1-3; 1.2, 6,16-17: 44,16; 1.2, 3.2, 84,33-6: 39,6-8; 3.3, 85,15:
6,17-19: 42,22-43,4; 1.2, 6,20: 39,5-6; 3.3, 85,34-86,2: 38,21-39,2;
41,1-3; 1.2, 6,24-5: 42,7-8; 42,19-22; 3.3, 86,8-10: 39,5-6; 3.3, 86,23-5:
1.2, 6,25-6: 42,22-43,4; 1.2, 6,26-30: 38,21-39,2; 3.3, 86,29-31:
44,13-14; 1.2, 6,26-33: 45,7-10; 1.2, 38,21-39,2; 3.6, 91,13-14: 44,7; 3.7,
6,26-7,7: 45,10-19; 1.2, 6,33-7,1: 92,27-31: 39,13-14; 3.7, 92,35:
44,13-14; 1.2, 7,1: 44,13-14; 1.2, 39,13-14; 3.7, 93,9-11: 39,13-14; 3.8
7,1-4: 43,7-8; 1.2, 7,4-7: 45,7-10; passim: 21i n.41; 40,21-41,5;
1.2, 7,10: 41,1-3; 1.2, 7,12: 41,1-3; 41,6-8; 3.9, 94,32-95,1: 39,4-5; 3.9,
1.2, 7,12-13: 44,3-4; 44,4-5; 1.2, 95,20-1: 39,13-14; 3.9, 96,31:
7,13-17: 45,21-22; 1.8 passim: 39,18-21; 3.9, 96,33: 39,18-21; 3.9,
14,24-15,1; 1.8, 17,19: 4,8; 1,8, 96,35: 38,21-39,2; 3.9, 96,36:
17,34-18,4: 14,24-15,1; 1.8, 18,32-4: 39,18-21; 3.9, 97,5: 39,18-21; 3.9,
4,9-15; 1.8, 19,3-6: 4,9-15; 1.8, 97,10: 39,18-21; 3.9, 97,12:
19,9-15: 14,24-15,1; 1.10 passim: 39,18-21; 3.12, 102,31-103,1:
3,24-5; 1.10, 20,23-30: 28,25-9; 24,16-17
1.10, 20,30: 28,25-9; 1.11a and b, in Sens. (= in Aristotelis de Sensu et
passim: 17,1-8; 28,15; 40,6-9; Sensibilia) 3,19-20: 10,8-10; 5,21-4:
1.11a,21,18: 19i n.9; 1.11b,22,26: 10,8-10; 8,2-3: 10,8-10; 9,1-2:
19i n.9; 1.11, 21,25-6: 10,8-10; 1.11, 10,8-10; 9,5-6: 38,21-39,2; 9,7-11:
22,9-12: 10,8-10; 1.11, 23,27-9: 40,11-15; 10,12-15: 40,11-15;
10,8-10; 1.11, 24,11-14: 10,8-10; 10,15-26: 39,9-10; 10,26-14,6:
1.15 passim: 3,24-5; 1.15, 27,22-3: 39,9-10; 11,12-17: 41,5; 11,14-19:
3,28-4,4; 1.17 passim: 14,24-15,1; 41,3-5; 16,5-6: 38,21-39,2; 17,5-7:
1.17, 29,31-30,22: 15,2-5; 1.17, 42,4-5; 17,13: 38,21-39,2; 18,12-13:
30,2-6: 3,28-4,4; 1.17, 30,3-6: 4,8; 42,4-5; 24,7-9: 42,19-22; 24,14-26,7:
1.17, 30,5-6: 4,20-2; 1.17, 30,14-22: 44,1-2; 25,21: 41,1-3; 25,21-6:
14,24-15,1; 15,2-5; 17,11-12; 1.21, 43,16-18; 43,18-44,1; 44,1-2; 26,10:
35,6: 5,1-4; 6,4-6; 1.21, 35,9-10: 43,11-16; 26,10-1: 44,7; 26,10-27,6:
43,7-8; 1.21, 35,9-16: 43,5-8; 1.21, 44,6; 26,13: 44,5; 26,15-27,6: 44,6-9;
35,12-14: 42,22-43,4; 1.21, 26,19: 44,6-9; 26,21: 44,6-9;
35,14-15: 44,16; 1.22, 35,19-21: 26,21-5: 44,1-2; 26,22: 44,5;
43,11; 1.25, 39,28: 16,2; 1.25, 26,26-27,6: 44,6-9; 26,27-27,1:
40,8-9: 1,2; 28,25-9; 1.25, 40,9-11: 42,7-8; 27,6-7: 44,7; 28,27-8:
28,25-9; 1.25, 40,9-30: 1,2; 1.25, 43,18-44,1; 29,5-7: 42,7-8; 30,1-2:
40,10-23: 28,25-9; 1.25, 40,17-19: 43,11; 31,4: 44,7; 31,11-18:
32,11-15; 1.25, 41,10-11: 1,2; 1.26 42,22-43,4; 31,13-15: 42,22-43,4;
passim: 14,24-15,1; 2.2, 31,13-17: 43,11; 31,15-18:
46,22-47,27: 22,23-7; 2.3, 49,1-4: 42,22-43,4; 31,20-9: 42,19-22; 33,13:
12,10-11; 2.3, 49,30-1: 3,28-4,4; 2.7, 43,11; 34,3: 39,21-40,3; 34,13-15:
52,20-53,30: 3,28-4,4; 2.7, 52,28-30: 44,7; 35,6-9: 42,8-11; 35,6-12:
3,28-4,4; 2,7, 53,25-30: 4,9-15; 2.8, 43,11-16; 35,8-9: 43,7-8; 35,13-15:
54,6-9: 11,1; 2.8, 54,9-10: 11,1; 43,5-8; 43,7-8; 35,17-36,4: 44,7;
16,11-12; 2.8, 54,11: 11,1; 2.10, 35,25-36,4: 44,6-9; 36,2: 44,1-2;
55,9: 16,2; 2.12 passim: 20,8-19; 36,9: 39,21-40,3; 36,11-19:
2.19, 63,8-28: 6,14-15; 2.19, 39,21-40,3; 36,12-17: 15,15-17;
63,15-26: 1,2; 2.19, 63,18-20: 39,20-1: 43,7-8; 39,23: 40,11-15;
28,25-9; 2.19, 63,20-1: 28,25-9; 40,22-3: 39,21-40,3; 41,15-21:
2.23, 74,5-6: 21,29-22,2; 2.24, 40,21-41,5; 42,25: 43,7-8;
75,5-10: 3,28-4,4; 4,23; 2.24, 42,25-43,1: 42,22-43,4; 42,26: 44,16;
75,16-19: 3,28-4,4; 2.24, 75,24-6: 42,26-7: 42,19-22; 43,5-7:
Index of Passages Cited 223
42,22-43,4; 43,8-11; 43,12-17: 44,16; 132,17-133,6: 42,22-43,4;
43,13-15: 44,13-14; 43,20-2: 42,7-8; 133,12-134,19: 42,22-43,4;
44,13-14; 44,1: 42,7-8; 44,4: 44,5; 133,13-14: 42,19-22; 133,18-27:
44,4-9: 44,20-1; 44,18-20: 45,21-2; 42,22-43,4; 133,19: 42,19-22;
44,20-1: 44,16-19; 44,22-45,3: 133,23-6: 42,22-43,4; 134,11:
44,21-3; 44,25-7: 3,11-13; 45,9-11: 42,19-22; 134,11-19: 42,22-43,4;
44,20-1; 45,11-14: 43,5-8; 45,11-16: 134,17-19: 42,22-43,4; 135,7-8:
44,23-5; 45,11-21: 44,13-14; 42,8-11; 135,9-11: 43,11; 135,9-13:
44,16-19; 45,16: 44,13-14; 45,17-18: 42,19-22; 135,9-17: 42,22-43,4;
42,7-8; 45,18-20: 42,7-8; 45,20-1: 135,12-13: 42,22-43,4; 135,13-14:
43,5-8; 45,21-47,20: 44,20-1; 43,11; 135,13-17: 43,11; 135,16-17:
45,23-6: 42,7-8; 45,7-10; 45,26-46,3: 43,11; 135,17-20: 42,19-22;
44,13-14; 46,1-2: 44,13-14; 46,6: 135,21-2: 42,22-43,4; 141,34-5:
45,7-10; 46,6-9: 42,7-8; 46,9-12: 42,19-22; 147,5-8: 40,20-1;
34,8-14; 46,12-16: 42,7-8; 46,12-19: 163,11-14: 15,15-17; 165,26-166,2:
45,7-10; 46,17-18: 44,13-14; 46,19: 31,2-6; 166,3: 39,4-5; 167,6-8:
42,7-8; 46,19-21: 42,7-8; 46,21-47,3: 27,6-7; 168,3: 39,18-21; 168,20-2:
44,13-14; 46,21-47,8: 42,7-8; 47,2-3: 40,11-15
44,16; 47,3-7: 42,19-22; 47,3-8: in Top. (in Aristotelis Topica) 16,23-4:
44,4-5; 47,4-6: 42,22-43,4; 47,8-15: 10,8-10; 17,9-11: 10,8-10; 50,9-11:
42,7-8; 45,7-10; 47,12-13: 44,13-14; 5,5-6; 50,10: 5,5-6; 50,18-29: 5,5-6;
47,15-20: 45,10-19; 47,24-5: 50,20: 5,5-6; 50,23: 5,5-6;
44,16-19; 47,25-48,10: 46,12-17; 50,29-51,4: 5,5-6; 50,30: 5,5-6; 51,3:
48,10-13: 45,24-46,1; 48,17: 44,4-5; 5,5-6; 71,4-7: 38,21-39,2; 89,16-17:
48,22-49,4: 42,7-8; 46,6-11; 49,4-8: 10,8-10; 93,1-3: 10,8-10; 104,2-6:
44,16-19; 46,11-13; 49,8-50,5: 38,21-39,2; 116,25: 40,11-15;
44,23-5; 49,15-21: 46,12-17; 50,4-5: 116,28-9: 39,5-6; 124,1-2: 10,8-10;
42,7-8; 50,4-7: 42,7-8; 50,5-7: 160,20-1: 10,8-10; 167,8-11:
45,5-6; 50,7-13: 44,23-5; 50,8-9: 35,10-11; 171,21-172,3: 24,16-17;
44,5; 50,9-11: 42,11-19; 50,16-18: 215,22-216,10: 4,5; 218,21-2:
42,19-22; 50,16-26: 45,6-7; 50,20-2: 38,21-39,2; 314,22-4: 12,3-4;
42,19-22; 50,22-4: 41,13-42,3; 321,12-16: 22,23-7; 355,21-4:
51,1-3: 41,13-42,3; 51,21-4: 10,8-10; 360,10: 17,15-17;
44,16-19; 45,24-46,1; 51,25-7: 376,11-15: 10,8-10; 376,27-30:
45,5-6; 52,2: 42,19-22; 52,8-12: 35,2-3; 381,17: 10,8-10; 391,6-7:
42,22-43,4; 52,9: 44,16; 52,19-20: 10,8-10; 394,6-7: 10,8-10; 405,2-5:
44,5; 53,5-8: 45,10-19; 54,4-55,8: 10,8-10; 407,4-5: 38,21-39,2;
26,7-13; 56,10-58,22: 42,19-22; 407,23-4: 10,8-10; 432,1: 10,8-10;
59,4-12: 42,8-11; 59,7-15: 42,19-22; 455,17: 40,11-15; 463,11-15:
59,8-12: 42,22-43,4; 59,12-15: 10,8-10; 493,3-4: 10,8-10
39,18-21; 59,25: 43,11; 61,17-21: AMMONIUS
42,19-22; 73,10: 7,9-13; 73,18-21: in An. Pr. (= in Aristotelis Analytica
7,9-13; 73,26-9: 12,10-11; 78,2: Priora) 39,24-5: 42,4-5
40,11-15; 78,6-12: 35,2-3; 83,3-7: in de Int. (= in Aristotelis de
42,19-22; 84,9-10: 41,10-13; Interpretatione) 13,19: 18i n.1;
41,13-42,3; 84,11-18: 41,5; 84,22-3: 39,14: 18i n.1; 100,20: 18i n.1;
41,10-13; 41,13-42,3; 85,16: 41,1-3; 202,4: 18i n.1
85,25-86,1: 41,1-3; 89,2-3: 44,5; ANAXAGORAS
89,9: 44,5; 105,4-5: 40,11-15; 105,9: Fragmenta (in Die Fragmente der
42,8-11; 108,10-14: 33,13-15; Vorsokratiker, Diels-Kranz) 59 A
116,7-122,23: 40,20-1; 116,21-3: 45: 33,15-16; 59 A 46: 33,15-16; 59
40,20-1; 123,6-11: 42,22-43,4; B 6: 33,15-16; 59 B 11: 33,15-16
124,6-7: 42,19-22; 124,9-10: 43,11; ANAXIMENES
127,5-128,6: 38,21-39,2; 127,20-1: Fragmenta (in Die Fragmente der
41,13-42,3; 132,2-16: 42,22-43,4; Vorsokratiker, Diels-Kranz)13 B 2:
132,7-8: 43,8-11; 132,8-16: 43,11; 19,21
132,11-16: 42,22-43,4;
224 Index of Passages Cited
ANTIOCHUS OF ASCALON 13,1-8; 1.1, 403a22: 13,1-8; 1.1,
Fragmenta (Mette) fr. 9a: 1,9 403b3-9: 3,7-9; 1.2, 403b25-7:
APOLLODORUS 29,3-10; 1.2, 404a27-8: 27,6-7; 1.2,
Fragmenta (in Stoicorum Veterum 404b16-17: 26,18-20; 1.2,
Fragmenta, von Arnim) fr. 7 (= 404b27-405a7: 30,26-31,2; 1.2,
SVF vol. 3, p. 260.12-14): 18,10-14 405a8-9: 27,6-7; 1.2, 405a8-13:
ASCLEPIUS 19,21; 1.2, 405a25-6: 19,21; 1.3
in Metaph. = in Aristotelis 405b31-406a4: 21,25-6; 1.3, 406a2:
Metaphysica 93,18: 18i n.1; 22,5-12; 1.3, 406a4-22: 22,13-15;
245,26-7: 18i n.1 1.3, 406a13-14: 21,29-22,2; 1.3,
ARISTOTLE 406a16-22: 21,29-22,2 1.3
An. Post. (= Analytica Posteriora) 1.2, 406a22-7: 15,17-22; 1.3, 406b2-5:
71b33-72a5: 32,23-4; 1.2, 15,17-22; 1.3, 406b4-5: 21,11-13;
72a29-32: 44,13-14; 1.22, 83a1-14: 1.3, 406b11-15: 22,23-7; 1.3,
41,6-8; 2.19, 100a6-9: 2,13; 2.19, 407a2-19: 30,26-31,2; 1.3,
100a16-b1: 41,8-10; 2.19, 100a17: 407a32-4: 23,2-3; 1.3, 407b13-26:
41,6-8 2,18-22; 21,16-21; 1.3, 407b13-27:
An. Pr. (= Analytica Priora) 1.27, 20i n.23; 27,15-28,2; 1.3, 407b23-4:
43a33-6: 41,6-8; 1.32, 47a24-8: 9i; 1.3, 407b25-7: 23,26-8; 1.4,
6,3-4 407b27-408a28: 24,18-20; 1.4,
Cael. (= de Caelo) 1.2, 268b26-7: 3,22; 407b34-408a5: 24,18-20; 1.4,
1.2, 268b26-269a1: 7,18-20; 1.2, 408a1-5: 25,4-7; 1.4, 408a5-9:
269a25-7: 7,18-20; 1.4, 271a33: 25,9-15; 1.4, 408a10-13: 25,2-9; 1.4,
27,9; 1.5, 271b17-18: 3,22; 1.9, 408a13-26: 25,2-9; 1.4,
277b30-278a10: 2,25-3,13; 1.10 408a14-b18: 22,5-12; 1.4,
passim: 6,14-15; 2.1, 408a30-4: 22,13-15; 1.4,
283b26-284a13: 6,14-15; 2.2, 408a34-b18: 7i; 23,6-24; 1.4,
285a29-30: 28,25-9; 2.11, 408b5-30: 23,15-21; 1.4, 408b7-12:
291b13-14: 27,9; 2.12, 292a18-21: 24,5; 1.4, 408b8-12: 23,11-13; 1.4,
28,25-9; 2.14, 296b33: 6,14-15; 3.2, 408b18-25: 21,24; 1.4, 408b19-25:
302a3-7: 6,13-14; 3.6: 5,8-9; 3.3, 27,8-15; 1.4, 408b22-5: 23,4-5; 1.4,
302b6-8: 7,18-20; 3.4, 303b4-5: 408b25-7: 24,5; 1.4,
7,18-20; 3.6, 305a19-20: 20,8-19; 408b30-409a30: 22,5-12; 1.4,
3.7, 305b29-30: 4,9-15; 3.8, 409a1-3: 22,5-12; 1.4, 409a9-10:
306b16-21: 3,28-4,4; 3.8, 306b17: 37,4-5; 1.5, 409b2-4: 20,8-19; 1.5,
3,28-4,4; 4.3-5, passim: 12,10-11; 410b18-20: 29,18-22; 1.5, 411b6-19:
4.3, 310b7-12: 14,19-20; 4.5, 21,16-21; 1.5, 411b19-27: 28,23-5;
312a30-3: 3,28-4,4 37,4-5; 37,11-18; 38,8-11; 1.5,
Categ. (= Categoriae) 1, 1a1-6: 411b27-30: 28,23-5;
28,25-9; 2, 1a23-b3: 13,10-14,3; 3, 2.1, 412a5-6: 17,1-8; 2.1, 412a6-9:
1b10-15: 14,16-17; 5, 2b5-6: 14,24-15,1; 2.1, 412a6-12:
15,10-13; 5, 2b17-22: 14,16-17; 5, 2,25-3,13; 6,2-3; 2.1, 412a13-15:
3b13-21: 12,3-4; 5, 4a10-b19: 29,3-10; 2.1, 412a19-b6: 11,2; 2.1,
14,24-15,1; 7, 6b2-8: 38,21-39,2; 7, 412a19-20: 14,24-15,1; 2.1,
6b33-6: 38,21-39,2; 7, 7b33-8a12: 412a21-8: 9,14-22; 2.1, 412a22-7:
38,21-39,2 16,5-6; 16,8-10; 2.1, 412a22-8:
DA (= de Anima) 1.1, 402a6-7: 9,12; 39,6-8; 2.1, 412a23-6: 16,12-18; 2.1,
1.1, 402a7-8: 1,2-4; 1.1, 402a10-11: 412a27-8: 10i; 16,7; 21,4-5; 2.1,
2,12-15; 1.1, 402b5-9: 17,1-8; 412a28-b6: 11,1; 23,26-8; 2.1,
28,25-9; 1.1, 402b10-14: 32,25-7; 412b4-5: 17,1-8; 2.1, 412b5-6:
1.1, 402b14-16: 33,7-10; 39,2-4; 1.1, 16,10; 2.1, 412b6-8: 4,9-15; 2.1,
403a2-5: 12,9-13,8; 1.1, 403a3-25: 412b6-9: 21,16-21; 2.1, 412b10-17:
12,9-13,8; 16,2; 21,25-6; 1.1, 403a8: 3,7-9; 2.1, 412b17-23: 12,2-3; 2.1,
23,4-5; 1.1, 403a8-10: 12,17-18; 412b17-25: 19,1-3; 2.1,
12,20-1; 1.1, 403a8-12: 21,24; 1.1, 412b17-413a3: 10i; 2.1, 412b18-20:
403a10-12: 12,21-4; 1.1, 43,11-16; 2.1, 412b20-3: 15,2-5; 2.1,
403a19-24: 13,1-8; 1.1, 403a20: 412b20-5: 18,17; 2.1, 412b25-6: 10i;
Index of Passages Cited 225
16,12-13; 2.1, 412b27-413a1: 2.4, 416a33-34: 33,18-24; 2.4,
16,5-6; 16,8-10; 2.1, 412b27-413a3: 416b3-7: 33,13-15; 2.4, 416b9-11:
39,6-8; 2.1, 412b28-413a1: 34,8-14; 2.4, 416b11-15: 35,9-17;
16,12-18; 2.1, 413a4-7: 21,24; 2.4, 416b11-17: 29,1-3; 2.4, 416b13:
37,1-3; 2.1, 413a8-10: 7i; 14,4-5; 35,10-11; 2.4, 416b14-19: 35,26-7;
2.1, 413a9-10: 17,1-8; 2.2, 2.4, 416b20-3: 36,9-13; 2.4,
413a11-12: 32,23-4; 2.2, 416b23-5: 36,14-18; 2.4, 416b25-9:
413a20-b4: 29,3-10; 2.2, 35,2-3; 36,9-13; 2.4, 416b28-9:
413a20-414a3: 27,4-8; 2.2, 35,2-3; 2.5 passim: 38,21-39,2;
413a21-5: 16,12-18; 2.2, 413a22: 39,9-10; 2.5-12: 40,9-10; 2.5-3.2:
28,25-9; 2.2, 413a25-31: 9,1-4; 2.2, 38,19-20; 2.5, 416b34: 38,21-39,2;
413a25-b2: 31,7-10; 2.2, 413a31-2: 2.5, 417a1-2: 39,10-18; 2.5,
28,25-9; 2.2, 413a31-b4: 29,12-14; 417a9-14: 39,5-6; 2.5, 417a14-21:
2.2, 413a31-b7: 28,23-5; 2.2, 38,21-39,2; 2.5, 417a18-20:
413b1-4: 10,8-10; 2.2, 413b2: 34,14-18; 39,10-18; 2.5,
29,14-15; 2.2, 413b2-4: 29,18-22; 417a21-b16: 12,9-13,8; 16,5-6;
29,18-22; 2.2, 413b4-5: 40,11-15; 16,8-10; 2.5, 417a21-b19: 39,6-8;
2.2, 413b8-9: 40,11-15; 2.2, 2.5, 417b2-16: 38,21-39,2; 2.5,
413b4-7: 40,11-15; 2.2, 413b11-13: 417b19-25: 39,9-10; 2.5, 417b29-32:
16,12-18; 2.2, 413b16-21: 37,4-5; 39,6-8; 2.5, 418a3-6: 38,21-39,2;
2.2, 413b22-4: 29,16-18; 2.2, 2.6: 41,10-13; 41,13-42,3; 2.6,
413b24-7: 21,24; 2.2, 414a3-4: 418a8-12: 40,20-1; 2.6, 418a10-11:
40,11-15; 2.2, 414a12-13: 9,12; 2.2, 21i n.42; 41,1-3; 2.6, 418a11-12:
413b27-8: 21,24; 2.2, 414a4-14: 40,21-41,5; 2.6, 418a12: 41,13-42,3;
31,7-10; 31,20-5; 2.2, 414a12-14: 2.6, 418a14-16: 41,13-42,3; 2.6,
9,12-14; 2.2, 414a14-16: 6,2-3; 2.2, 418a17-18: 41,5; 2.6, 418a18-19:
414a22-5: 27,15-28,2; 2.2, 21i n.42; 41,1-3; 2.6, 418a20-3:
414a22-7: 2,18-22; 2.3 passim: 41,6-8; 2.6, 418a23-4: 40,20-1;
16,18-17,1; 2.3, 414a29-b1: 27,4-8; 41,8-10; 2.6, 418a24-5: 16i; 40,20-1;
2.3, 414b1-15: 29,16-18; 2.3, 414b3: 40,21-41,5; 41,8-10; 2.7-11 passim:
40,11-15; 2.3, 414b6-9: 39,9-10; 40,21-41,5; 42,5-7; 2.7, 418a26-8:
40,11-15; 2.3, 414b17: 29,18-22; 2.3, 42,4-5; 44,13-14; 2.7, 418a29-30:
414b18-19: 21,24; 2.3, 414b20-31: 45,5-6; 2.7, 418a31-b2: 42,5-7;
28,8-13; 2.3, 414b20-415a1: 17,1-8; 42,5-7; 2.7, 418b4-6: 44,13-14;
2.3, 414b20-415a12: 16,18-17,1; 44,16; 2.7, 418b4-9: 42,7-8; 2.7,
2.3, 414b32-415a12: 27,4-8; 2.3, 418b5: 44,13-14; 2.7, 418b9-10:
415a3-5: 40,11-15; 2.3, 415a6-7: 43,7-8; 2.7, 418b11: 44,13-14; 44,16;
29,18-22; 2.3, 415a8-12: 13,10-14,3; 2.7, 418b13-16: 42,19-22; 43,11;
28,25-9; 2.3, 415a11-12: 21,24; 2.3, 2.7, 418b13-18: 20,8-19; 2.7,
415a12-13: 28,28-29,1; 2.4, 418b13-20: 42,22-43,4; 2.7,
415a14-23: 40,15-19; 2.4, 418b16-17: 43,11; 2.7, 418b20-3:
415a16-20: 32,25-7; 2.4, 42,19-22; 2.7, 418b20-6: 42,19-22;
415a16-22: 34,27-35,2; 2.4, 43,11; 2.7, 418b26-9: 44,3-4; 2.7,
415a16-23: 40,21-41,5; 2.4, 415a20: 419a1-7: 42,4-5; 42,5-7; 44,13-14;
32,29-30; 2.4, 415a20-2: 33,7-10; 2.7, 419a6-7: 43,16-18; 2.7,
39,2-4; 2.4, 415a26-28: 35,19-20; 419a9-10: 42,5-7; 2.7, 419a11:
2.4, 415a26-b2: 32,11-15; 2.4, 42,8-11; 43,7-8; 2.7, 419a13:
415b8-28: 24,11-15; 2.4, 415b16-17: 43,11-16; 2.7, 419a13-15: 42,8-11;
28,8; 2.4, 415b24: 28,8; 2.4, 2.7, 419a23: 44,13-14; 2.8, 420a14:
2.4, 415b24: 38,21-39,2; 2.4, 44,7; 2.8, 420b16-22: 28,6-8; 2.9,
416a4-9: 34,14-18; 2.4, 416a9-15: 421a19: 40,11-15; 2.9, 421b23-5:
35,3-5; 2.4, 416a13-18: 35,5-8; 2.4, 44,13-14; 2.10, 422a8-9: 40,11-15;
416a14: 35,3-5; 2.4, 416a19: 29,1-3; 2.10, 422a10: 40,11-15; 2.10,
32,15-22; 2.4, 416a21-2: 33,15-16; 422a14-15: 42,19-22; 2.10,
2.4, 416a21-5: 33,28-34,8; 2.4, 422a21-8: 44,13-14; 2.10,
416a25-9: 34,8-14; 2.4, 416a29-34: 422a29-32: 44,13-14; 2.10,
33,15-16; 2.4, 416a29-b9: 39,10-18; 422b7-10: 41,13-42,3; 2.11,
226 Index of Passages Cited
422b21-3: 39,18-21; 2.11, 422b25-7: 3.12, 434a31-2: 27,9; 3.12,
40,21-41,5; 2.11, 422b27-33: 434a32-b1: 39,9-10; 3.12, 434b3-8:
40,21-41,5; 2.11, 422b32-4: 38,16-18; 3.12, 434b4-7: 28,25-9;
41,10-13; 2.11, 423a20-1: 40,11-15; 3.12, 434b7-8: 29,18-22; 3.12,
2.11, 423b22-3: 39,18-21; 2.11, 434b9-14: 40,11-15; 3.12,
423b27-9: 5,5; 5,20; 2.11, 434b13-17: 39,9-10; 3.12,
423b29-31: 39,18-21; 2.11, 434b18-19: 40,11-15; 3.12, 434b18:
423b31-424a10: 44,3-4; 2.11, 40,11-15; 3.12, 434b18-22: 39,9-10;
424a1: 38,21-39,2; 2.11, 424a7-9: 434b21: 40,11-15; 3.12, 434b22-4:
44,3-4; 2.11, 424a10-15: 44,13-14; 40,11-15; 3.12, 434b24-7: 39,9-10;
2.12, 424a17-24: 39,13-14; 2.12, 3.13, 435b4-7: 40,11-15; 3.13,
424a24-5: 39,18-21; 2.12, 435b5-7: 10,8-10; 3.13, 435b16-17:
424a24-8: 19,1-3; 2.12, 424a28-32: 10,8-10; 3.13, 435b7-9: 44,13-14;
44,13-14; 3.13, 435b16-19: 40,11-15; 3.13,
3.1, 425a4: 44,7; 3.1, 425a13-20: 435b19-21: 39,9-10; 3.13, 435b22-3:
40,20-1; 3.1, 425a14-b10: 40,20-1; 39,9-10; 3.13, 435a12-13: 40,11-15;
3.1, 425a14-16: 41,5; 3.1, 3.13, 435b2: 40,11-15; 3.13,
425a21-30: 40,20-1; 3.1, 425a24-7: 436b18-437a3: 39,9-10
41,6-8; 3.1, 425a25-7: 40,20-1; 3.1, Div. (= de Divinatione per Somnum) 2,
425a30-1: 41,6-8; 3.1, 425a30-b3: 464a5-6: 42,19-22; 2, 464a9-11:
40,21-41,5; 3.1, 425b8-9: 41,3-5; 42,19-22
3.2, 425b14: 41,10-13; 3.2, EE (= Ethica Eudemia) 1.5, 1216a3-9:
425b25-426a27: 38,21-39,2; 3.2, 38,4-8; 1.8, 1218a1-15: 17,1-8; 2.1,
425b29-31: 39,5-6; 3.2, 426a20-7: 1219b37: 35,24; 7.9, 1241b17-19:
46,17-19; 3.2, 426a23-4: 39,5-6; 3.2, 23,26-8; 7.9, 1241b22: 23,26-8
426a29-b8: 44,13-14; 3.2, EN (Ethica Nicomachea) 1.2,
426b8-11: 41,10-13; 3.2, 426b15-17: 1094a18-22: 28,8-13; 1.4,
39,18-21; 3.2, 427a14-16: 10,8-10; 1095b2-4: 32,23-4; 1.6, 1096a17-23:
3.3 passim: 12,17; 41,10-13; 3.3, 17,1-8; 1.7, 1098b7: 10,19-20; 1.13,
427a17-19: 29,3-10; 3.3, 427a20-1: 1102a32: 35,24; 1.13, 1102b4-5:
39,4-5; 3.3, 427b12: 41,13-42,3; 3.3, 35,10-11; 1.13, 1102b29: 35,24; 3.1,
427b24-6: 29,22-4; 3.3, 428a3-5: 1110a15-18: 24,6-7; 3.3, 1113a5-7:
29,22-4; 39,4-5; 3.3, 428a9-11: 24,5; 3.10, 1118b1: 40,11-15; 6.1,
29,16-18; 3.3, 428a11: 41,13-42,3; 1139a3-6: 29,22-30,2; 6.1,
3.3, 428b11-14: 12,17-18; 3.3, 1139a6-17: 29,22-4; 29,22-30,2; 6.2,
428b17-25: 41,10-13; 3.3, 1139a20: 10,8-10; 6.4, 1140a6-10:
428b18-19: 16i; 41,13-42,3; 3.3, 22,5-12; 6.4, 1140a20-1: 22,5-12;
428b18-30: 40,20-1; 3.3,428b21-2: 6.5, 1140b25-30: 29,22-30,2; 9.9,
41,13-42,3; 3.4, 429a18-27: 37,1-3; 1170a16: 10,8-10; 10.5,
44,3-4; 3.4, 429a31-b3: 44,13-14; 1176a13-16: 41,13-42,3; 10.7,
3.5, 430a16-17: 46,1-6; 3.5, 1177b33-1178a3: 32,11-15
430a22-5: 21,24; 3.6, 430b29-30: Fragmenta (Rose, 3rd ed.; for the
41,13-42,3; 3.7, 431a4-6: Protrepticus, see separate entry
38,21-39,2; 3.7, 431a13-14: below) frr. 1-3: 1,4-2,4; fr. 3: 1,6-8;
29,16-18; 3.7, 431a16-17: 12,20-1; frr. 37-47: 24,18-20
3.7, 431b2: 12,20-1; 3.8, 432a1-3: GA (= de Generatione Animalium) 1.2,
8,12-13; 3.8, 432a8-10: 12,20-1; 3.9, 716b23-5: 27,8-15; 1.18, 724a23-6:
432a15-16: 29,3-10; 3.9, 3,7-9; 1.18, 725a11-13: 36,3-4; 1.18,
432a23-b7: 28,3-13; 3.9, 726a26-7: 36,3-4; 1.19, 726b1-5:
432b13-14: 29,18-22; 3.9, 36,3-4; 1.19, 726b9-11: 36,3-4; 1.19,
432b19-21: 29,18-22; 3.9, 432b21-6: 726b22-4: 12,2-3; 15,2-5; 18,7; 1.19,
27,9; 3.9, 432b23-5: 35,19-20; 3.9, 727a5-8: 35,20-2; 1.20, 728b21-7:
432b25-6: 24,6-7; 3.10, 433b16-21: 35,20-2; 1.20, 729a32: 3,28-4,4;
12,14-15; 3.10, 433b28-9: 29,16-18; 1.22, 730b11-24: 24,7-9; 1.22,
3.11, 433b31-434a5: 29,16-18; 3.12, 730b19-23: 15,10-13; 1.23,
434a22-6: 29,1-3; 3.12, 434a27-31: 731a30-b4: 10,8-10; 2.1,
10,8-10; 3.12, 434a28-9: 40,11-15; 731b31-732a1: 32,11-15; 2.1,
Index of Passages Cited 227
732a12-13: 10,8-10; 2.1, 733b26: 322a20-8: 35,9-17; 1.5, 322a24-5:
3,28-4,4; 2.1, 734b23-7: 12,2-3; 35,10-11; 1.6, 322b32-323a12:
15,2-5; 18,7; 2.1, 734b36-735a3: 21,29-22,2; 1.6, 323a13-14: 21,25-6;
3,9-11; 2.1, 734b36-735a4: 1.7 passim: 33,25-9; 39,10-18; 1.7,
15,10-13; 2.1, 735a2-4: 24,7-9; 323b3-7: 33,15-16; 1.7, 324a30-4:
32,4-6; 2.1, 735a6-8: 12,2-3; 2.1, 21,25-6; 1.7, 324a34-b6: 22,5-12;
735a7-8: 15,2-5; 18,7; 2.1, 1.8, 324b26-35: 42,19-22; 1.10,
735a9-11: 36,22-3; 2.1, 735a12-14: 327b14-15: 4,9-15; 1.10, 327b22:
36,19-21; 36,23-6; 2.1, 735a15-21: 15,10-13; 1.10, 327b23-31:
32,4-6; 2.1, 735a16-19: 32,15-22; 11,18-20; 1.10, 328a5-15: 11,15;
2.1, 735a17-19: 32,11-15; 36,7; 2.1, 1.10, 328b17-19: 11,18-20; 2.1,
735a14-26: 37,1-3; 2.3, 736a29-30: 329a8-11: 3,28-4,4; 2.1, 329a23:
10,8-10; 2.3, 736a31-6: 36,22-3; 2.3, 3,28-4,4; 2.1, 329a24-7: 4,8; 2.1,
736a35-b1: 36,19-21; 2.3, 736b1: 329a24-35: 3,28-4,4; 2.1, 329a29:
10,8-10; 2.3, 736b1-5: 37,1-3; 2.3, 3,28-4,4; 2.2, 329b6-10: 5,20; 2.2,
736b5-8: 37,1-3; 2.3, 736b8-13: 329b6-17: 5,5; 2.2, 329b13:
38,4-8; 2.3, 736b8-15: 36,22-3; 2.3, 41,10-13; 2.2, 329b18-20:
736b21-9: 21,24; 2.3, 736b22-4: 40,21-41,5; 2.2, 329b32-4: 5,5; 2.2,
37,1-3; 2.3, 736b27-8: 37,1-3; 2.3, 330a24-6: 5,5; 2.3, 330a30-b7: 5,5;
736b29: 37,1-3; 2.3, 737a16-18: 8,17-25; 2.3, 330b21-5: 45,10-19;
36,22-3; 2.3, 737a18-20: 36,3-4; 2.3, 2.4, 331a14-16: 6,8-10; 2.4-5
737a18-22: 36,23-6; 2.4, 737b8-12: passim: 5,8-9; 8,3-5; 2.5, 332a7-8:
35,20-2; 2.4, 740a24-7: 38,4-8; 2.4, 33,18-24; 2.5, 332a17-20: 3,28-4,4;
740b25-9: 3,9-11; 2.4, 740b29-36: 2.5, 332a34-b1: 3,28-4,4; 2.7,
35,2-3; 2.4, 740b29-37: 32,15-22; 334a15-25: 3,28-4,4; 2.7,
2.5, 740b36-7: 29,1-3; 2.5, 334a16-26: 8,3-5; 2.8 passim:
741a9-10: 10,8-10; 2.5, 741b4-5: 45,10-19; 2.8, 334b31-3: 45,10-19;
27,9; 2.5, 741b15-16: 24,5-6; 2.5, 2,8, 334b31-335a9: 8,3-5; 2.8,
741b15-17: 37,1-3; 2.6, 742a28-b6: 335a22-3: 45,10-19; 2.9, 335b7-24:
24,6-7; 2.6, 742b1-2: 24,5-6; 2.6, 7,9-13; 2.10, 336b26-33: 32,11-15;
743b25-7: 39,21-40,3; 2.6, 2.11, 338b6-19: 32,11-15
744a36-7: 27,9; 2.6, 744b16-21: HA (= Historia Animalium) 1.1,
2,18-22; 2.6, 744b32-6: 33,13-15; 487b6-11: 29,18-22; 1.3, 489a17-18:
2.8 passim: 35,19-20; 3.2, 10,8-10; 1.3, 489a17-19: 40,11-15;
753b27-9: 38,4-8; 3.7, 757b15: 1.6, 491a26: 24,6-7; 1.9, 491b20-2:
10,8-10; 4.1, 766a22-3: 27,8-15; 4.1, 44,7; 3.18, 520b3-5: 44,7; 4.1,
766b4-7: 35,20-2; 4.1, 766b7-9: 523b21: 42,4-5; 4.8, 533a8-9: 44,7;
36,3-4; 4.1, 766b12-14: 36,3-4; 5.1, 4.8, 533a17-18: 10,8-10; 4.8,
778b32-3: 10,8-10; 5.1, 535a4-5: 40,11-15; 5.19, 551b23-6:
778b32-779a2: 38,4-8; 5.1, 42,4-5; 7.1 passim: 35,20-2; 7.1,
780a1-5: 44,6-9; 5.1, 780a7-14: 581a12-14: 35,20-2; 7.1,
44,13-14; 5.1, 780a25-36: 44,7; 5.1, 582a16-17: 35,20-2; 7.3, 583b9-25:
780b22-9: 44,7; 5.1, 780b23: 44,7; 38,4-8; 7.3, 583b14-20: 35,20-2; 7.5,
5.1, 781a15-17: 41,10-13; 5.2, 585a34-b2: 35,20-2; 8.1,
781a20-3: 39,18-21; 5.8, 788b20-4: 588b12-17: 29,18-22; 8.1,
27,9; 5.8, 789b9-12: 28,6-8 588b17-18: 10,8-10
GC (= de Generatione et Corruptione) IA (= de Incessu Animalium) 2,
1.3, 317a32-b5: 6,13-14; 1.3, 704b12-18: 27,9; 3, 705a20: 24,6-7;
317b5-7: 4,5; 1.3, 317b10: 15,10-13; 4, 705b22: 24,6-7; 7, 707a27-b5:
1.3, 317b31-3: 15,10-13; 1.4, 37,4-5; 8, 708a9-12: 27,9
319b2-4: 3,28-4,4; 1.4, 320a2-5: Insom. (= de Insomniis) 1, 458b4-6:
3,28-4,4; 5,8; 1.5, 320b12-14: 40,20-1; 41,5; 1, 458b31-3: 41,10-13;
3,28-4,4; 1.5, 320b16-17: 4,8; 1.5, 1, 459a15-17: 29,16-18; 2, 459b4-5:
320b24-5: 15,10-13; 1.5, 321a5-10: 38,21-39,2; 2, 459b7-13: 41,13-42,3;
20,8-19; 1.5, 321a9-29: 34,8-14; 1.5, 2, 459b13-18: 43,18-44,1; 2,
321a29-b10: 34,18-26; 1.5, 460b20-3: 41,1-3; 2, 460b23-5:
321b35-322a3: 33,13-15; 1.5, 41,10-13; 3, 460b28-461b20:
228 Index of Passages Cited
39,18-21 1017a5: 3,28-4,4; 5.8, 1017b21-2:
Int. (= de Interpretatione) 9, 19a33: 1,3; 5.11, 1018b29-34: 32,25-7;
41,10-13 5.12, 1019b15-19: 32,11-15; 5.14,
Iuv. (= de Iuventute et Senectute, de 1020a33-b2: 12,3-4; 5.14,
Vita et Morte) 1, 467b18-27: 24,5; 1, 1020b14-15: 12,3-4; 5.23, 1023a13:
467b24-5: 10,8-10; 2, 468a20-3: 13,10-14,3; 5.23, 1023a23-5:
24,5; 2, 468a23-b15: 37,4-5; 2, 13,10-14,3; 5.24, 1023a28-9:
468b1-9: 37,11-18; 38,8-11; 2, 3,17-19; 5.24, 1023a33: 7,2; 5.24
468b5-9: 37,18-38,4; 2, 468b12-14: passim: 11,15; 5.25 passim: 18,7;
37,11-18; 3, 468b28-31: 24,5-6; 3, 5.25, 1023b12-15: 30,26-31,2; 5.25,
469a4-10: 24,5; 3, 469a5-20: 24,5; 1023b19-22: 18,17-21; 5.26,
39,21-40,3; 3, 469a10-16: 39,18-21; 1024a4-5: 4,9-15; 5.28, 1024b4-6:
3, 469a18-20: 10,8-10; 4, 12,3-4; 5.29, 1024b21-4: 41,10-13;
469a23-34: 24,5; 4, 469a23-7: 6.1, 1026a13-16: 22,5-12; 6.3,
39,21-40,3; 4, 469b3-6: 39,21-40,3; 1027a29-30: 20i n.20; 7.3,
4, 469a33-4: 24,5; 4, 469b1-6: 24,5; 1029a7-30: 3,28-4,4; 7.3,
4, 469b4: 10,8-10; 4, 469b6-20: 1029a27-30: 4,24; 7.3, 1029b3-12:
39,21-40,3; 4, 469b11-12: 35,2-3; 4, 32,23-4; 7.7, 1032b7: 25,4-7; 7.7,
469b11-13: 24,5 1032b18-20: 25,4-7; 7.8, 1033b5-8:
Long. (= de Longetudine et Brevitate 20i n.20; 7.8, 1033b12-19:
Vitae) 5, 466a18-19: 39,21-40,3; 6, 2,25-3,13; 7.9, 1034a24: 3,9-11; 7.9,
467a18-23: 37,4-5; 6, 467a18-30: 1034a33-b1: 36,22-3; 7.10,
37,18-38,4; 6, 467a20-2: 37,11-18; 1035a1-4: 6,2-3; 7.10, 1035a5: 20i
18, 479b1-7: 37,4-5 n.20; 25,2; 7.10, 1035a12: 20i n.20;
MA (= de Motu Animalium) 7, 25,2; 7.10, 1035b4-6: 32,29-30;
701b16-18: 38,21-39,2; 7, 7.10, 1035b14-18: 10,8-10; 7.10,
701b24-32: 24,5; 8, 702a17-19: 1035b25-7: 24,5; 24,5-6; 7.11,
29,16-18; 9, 702b12-28: 24,5; 9, 1036a31: 20i n.20; 7.11,
702b20-1: 39,21-40,3; 10 passim: 1036a31-2: 25,2; 7.11, 1036b6: 20i
12,14-15; 10, 703a14-16: n.20; 25,2; 7.12, 1038a19-20:
39,21-40,3; 10, 703a29-b2: 12,2-3; 12,3-4; 7.13, 1038b2-3: 6,2-3; 7.13,
10, 703a36-7: 12,2-3; 11, 703b23-4: 1039a3-8: 6,3-4; 7.13, 1039a14-19:
24,5; 39,21-40,3 6,3-4; 7.15, 1039b23-7: 20i n.20;
Mem. (= de Memoria et Reminiscentia) 7.16, 1040b5-16: 6,3-4; 7.16,
1, 449b13-15: 39,5-6; 1, 449b27: 1041a4-5: 6,3-4; 7.17, 1041b29-30:
39,5-6; 1, 449b31: 12,20-1; 1, 5,1-2; 8.1, 1042a24-31: 2,25-3,13;
450a12: 39,18-21; 1, 450a27-b11: 8.1, 1042a26-31: 6,2-3; 8.1,
12,17-18; 1, 451a17: 39,18-21; 2, 1042a29-31: 17,11-12; 8.2,
453a14-30: 12,17-18 1042b15-17: 11,17-18; 8.2,
Metaph. (= Metaphysica) 1.1, 980a27: 1043a14-8: 2,25-3,13; 8.2,
10,8-10; 1.1, 981a12-b9: 2,13; 1.9, 1043a26-8: 6,2-3; 8.3, 1043b14-16:
991b3-9: 7,9-13; 1.9, 992a29-31: 20i n.20; 8.3, 1043b21-3: 5,1-2; 8.4,
28,8; 2.1, 993a30-b7: 2,4-9; 2.1, 1044a23: 3,28-4,4; 8.4, 1044a23-5:
993b24-6: 44,13-14; 2.2, 994a8-10: 11,15; 8.4, 1044b8-12: 24,16-17;
28,8-13; 2.2, 994a22-b3: 11,15; 2.2, 8.5, 1044b21-2: 20i n.20; 9.2,
994b9-16: 28,8-13; 3.3, 999a6-11: 1046b2-4: 22,5-12; 9.6, 1048b18-35:
17,1-8; 3.5, 1002a4-8: 17,12-15; 4.5, 12,9-13,8; 9.7, 1049a24-7: 3,28-4,4;
1010a29-30: 1,2; 4.5, 1010b2-3: 9.8, 1049b16-17: 32,25-7; 9.8,
41,13-42,3; 4.5, 1010b5-6: 1050a15-16: 7,2; 9.8, 1050b2-6:
41,13-42,3; 4.5, 1010b14-17: 9,14-22; 9.8, 1050b28-9: 32,11-15;
40,21-41,5; 4.5, 1010b19-21: 9.10, 1051b4-5: 41,10-13; 9.10,
41,13-42,3; 4.5, 1010b21-3: 1051b33-5: 41,10-13; 10.1,
41,13-42,3; 4.7, 1011b34-5: 1053b1-3: 41,10-13; 10.7, 1057a31:
33,18-24; 5.1, 1013a4-6: 24,5; 5.2, 33,18-24; 10.8, 1049b12-17:
1013b6-9: 3,7-9; 5.4, 1015a7-10: 32,29-30; 11.6, 1062b36-1063a3:
3,28-4,4; 4,23; 5.4, 1015a9-10: 41,13-42,3; 11.7, 1064a31-3:
3,17-19; 5.4, 1015a10-11: 7,2; 5.6, 22,5-12; 11.8, 1065a26-7: 28,8;
Index of Passages Cited 229
11.11, 1068a1-4: 33,18-24; 11.12, 2.17, 660a21: 40,11-15; 3.1,
1068a11-13: 42,22-43,4; 12.1, 661b23-4: 27,9; 3.1, 662a16-24:
1069b3: 33,18-24; 12.3, 1070a9-13: 28,6-8; 3.4, 665a10-13: 39,21-40,3;
2,25-3,13; 6,2-3; 12.3, 1070a13-20: 3.4, 665a10-15: 24,5; 3.4, 665b24-6:
5,1-2; 12,3 1070a15-17: 20i n.20; 37,11-18; 3.4, 665b27-9: 12,2-3; 3.4,
12.7, 1072a25: 21,25-6; 12.7, 666a7-8: 39,21-40,3; 3.4,
1073a3-7: 28,25-9 13.2, 1077b3-4: 666a10-11: 24,5-6; 3.4, 666a11-18:
32,29-30; 13.10, 1087a19-20: 24,5; 39,21-40,3; 39,21-40,3; 3.4,
41,6-8; 14.1, 1088a29-35: 666a20-2: 24,5-6; 3.4, 666a34:
42,22-43,4; 14.5, 1092a21-35: 11,14 10,8-10; 3.4, 666a34-5: 24,5; 3.4,
Meteor. (= Meteorologica) 1.2, 666a34-b1: 39,21-40,3; 39,21-40,3;
339a13-14: 5,5; 1.4, 341b6-12: 3.4, 666b14-15: 24,5; 3.5,
3,17-19; 2.2, 355a3-11: 34,8-14; 3.4, 667b19-34: 39,21-40,3; 3.5,
375a22-8: 41,13-42,3; 42,11-19; 3.6, 667b21-31: 39,21-40,3; 3.7,
378a19-20: 3,17-19; 3.6, 670a23-6: 39,21-40,3; 3.10,
378a26-31: 3,17-19; 4.2-3: 33,13-15; 672b14-19: 39,21-40,3; 3.12,
4.2, 379b14-17: 33,13-15; 4.2, 673b26: 25,4-7; 3.14, 674b9-13:
379b18-20: 33,13-15; 4.3, 381a9-12: 33,13-15; 3.14, 674b26-34:
3,15-16; 4.3, 381b3-9: 3,15-16; 4.6-7 33,13-15; 4.5, 678b1-4: 24,5;
passim: 3,19; 4.10, 389a7-9: 39,21-40,3; 4.5, 678b2-4: 10,8-10;
3,17-19; 4.12, 389b31-390b2: 4.5, 681a19-20: 10,8-10; 4.5,
15,2-5; 18,17 682a2-9: 37,11-18; 4.6, 682b30-2:
MM (= Magna Moralia) 2.15, 37,4-5; 37,11-18; 37,18-38,4; 4.6,
1213a13-26: 1,4-2,4 683a20-6: 28,6-8; 4.7, 683b4-11:
Oecon. (= Oeconomia) 1.3, 1343b23-5: 29,18-22; 4.10, 687a10-14: 27,8-15;
32,11-15 4.11, 691b4: 27,9; 4.12, 694b13-14:
PA (= de Partibus Animalium) 1.1, 27,8-15; 4.13, 695b19: 27,9
639b14-23: 2,13;1.1, 639b21-30: 20i Phys. (= Physica) 1.1, 184a16-b14:
n.22; 1.1, 640a31-2: 3,9-11; 1.1, 32,23-4; 1.4, 188a6: 15,10-13; 1.4,
640b34-641a5: 15,2-5; 18,17; 1.1, 188a12-13: 15,10-13; 1.5,
641a17-21: 15,2-5; 18,17; 1.3, 188b19-20: 3,7-9; 1.5, 188b20:
642a1-13: 20i n.22; 1.3, 642b33-4: 3,28-4,4; 1.5, 188b21-6: 6,8-10; 1.6,
42,4-5; 1.5, 644b23-645a26: 189a32-3: 6,3-4; 1.7, 190b5-6: 3,7-9;
2,12-15; 1.5, 645b14-20: 27,8-15; 1.7, 190b15: 3,28-4,4; 1.7, 191a2:
2.1, 646a12-24: 3,22; 2.1, 3,28-4,4; 1.7, 191a10: 3,28-4,4; 1.8,
646a16-20: 5,5; 2.1, 647a2-5: 191b13-16: 6,8-10; 2.1 passim:
24,6-7; 2.1, 647a19-21: 39,18-21; 12,10-11; 2.1, 192b8-23: 22,23-7;
2.1, 647a21: 10,8-10; 2.1, 647a24-6: 2.1, 192b16-23: 5,1-2; 2.1, 193a11:
24,5; 2.1, 647a24-30: 24,6-7; 3,28-4,4; 2.1, 193a29: 3,28-4,4; 2.1,
39,18-21; 2.1, 647a24-31: 24,5; 2.1, 193a31-b8: 3,19-21; 2.2, 194a21-2:
647a24-33: 39,21-40,3; 2.2, 3,15-16; 2.3, 195a3-8: 15,10-13; 2.3,
647b22-5: 24,6-7; 2.2, 648b9-10: 195a32-5: 3,7-9; 2.5, 196b17-33:
5,5; 2.3, 650a2-32: 33,13-15; 2.3, 24,16-17; 2.6, 197b22-30: 27,9; 2.8,
650b2-13: 36,3-4; 2.4, 650b22-3: 199a9-b33: 2,13; 2.8, 199a15-17:
39,21-40,3; 2.4, 651a14-15: 36,3-4; 3,15-16; 2.9: 20i n.22; 3.1,
2.5, 651b3-5: 10,8-10; 2.7, 200b33-201a19: 42,22-43,4; 3.1,
652b9-15: 35,2-3; 2.7, 653b5-8: 201a10-11: 42,22-43,4; 3.1,
39,21-40,3; 2.8, 653b22-3: 10,8-10; 201a27-9: 42,22-43,4; 3.1, 201b4-5:
2.8, 653b22-4: 40,11-15; 2.8, 42,22-43,4; 3.2, 202a5-9:
653b23-30: 39,18-21; 2.10, 656a3-4: 21,29-22,2; 42,22-43,4; 3.6, 207a21:
10,8-10; 2.10, 656a26-33: 39,18-21; 6,30; 4.1, 209a4-7: 20,8-19; 4.1,
2.10, 656a27-9: 24,5; 2.10, 209a26-7: 14,18; 4.2, 209b11-16:
656a27-31: 39,21-40,3; 2.10, 3,28-4,4; 4.2, 209b28-9: 14,23-4;
656b24: 39,21-40,3; 2.10, 4.3, 210a14-24: 13,10-14,3; 4.4,
656b29-31: 43,18-44,1; 2.10, 211a17-23: 14,18; 4.4, 211b14-29:
656b34-6: 39,18-21; 2.13, 658a8-9: 14,19-20; 4.4, 212a3-5: 14,19-20;
27,9; 2.16, 659b34-660a2: 28,6-8; 4.4, 212a14-18: 14,23-4; 4.4,
230 Index of Passages Cited
212a20: 14,19-20; 4.5, 212b7: 42,4-5; 2, 437b5: 43,16-18; 2,
14,18; 4.5, 212b25: 20,8-19; 4.5, 438a4-5: 42,19-22; 2, 438a5-12:
212b28-9: 14,18; 4.6, 213a27-9: 44,1-2; 2, 438a12-15: 42,7-8; 2,
14,19-20; 4.6, 213b5-12: 20,8-19; 438a13: 43,11-16; 2, 438a13-15:
4.7, 213b31: 14,19-20; 4.7, 214a5-6: 44,6; 2, 438a14-15: 44,5; 2,
14,19-20; 4.7, 214a20: 14,19-20; 438a15-16: 44,6-9; 2, 438a16-22:
4.7, 214b6: 20,8-19; 4.9, 217a21-33: 44,7; 2, 438b2-5: 42,8-11; 43,11-16;
3,28-4,4; 5.1, 224b4-7: 22,5-12; 5.1, 2, 438b5-8: 44,6; 2, 438b5-11: 44,7;
224b5: 5,11-15; 5.1, 224b11-12: 2, 438b5-16: 44,7; 2, 438b19: 44,7;
5,11-15; 5.1, 224b11-13: 22,5-12; 2, 438b30-439a1: 40,11-15; 2,
5.1, 224b13: 23,2-3; 5.2, 225b11-17: 439a1-4: 39,21-40,3; 3, 439a17:
42,22-43,4; 5.3, 226b34-227a6: 40,20-1; 3, 439a18-19: 44,16; 3,
25,9-15; 5.6, 229b23-8: 22,23-7; 6.1, 439a19-21: 42,22-43,4; 3,
231a23: 25,9-15; 6.4, 234b10-20: 439a21-5: 44,20-1; 3, 439a26-7:
22,5-12; 7.2, 243a16-244a4: 44,4-5; 3, 439a26-8: 44,21-3; 3,
21,29-22,2; 7.2, 244b10-12: 439a26-30: 44,16-19; 3, 439a27-32:
38,21-39,2; 7.3, 246b3-10: 25,4-7; 45,20; 3, 439a28-9: 42,5-7; 3,
7.3, 247b10-248a6: 23,2-3; 7.3, 439a31-2: 45,5-6; 3, 439a31-b6:
246b11-12: 42,22-43,4; 7.3, 44,23-5; 3, 439a33-b1: 45,5-6; 3,
247b13-248b6: 23,4-5; 7.3, 439b1-2: 42,11-19; 44,13-14; 44,16;
248a6-9: 38,21-39,2; 8.4, 254b7-12: 3, 439b1-5: 44,4-5; 3, 439b1-6:
22,13-15; 8.4, 254b15-20: 22,15-23; 45,6-7; 3, 439b5-6: 41,13-42,3;
8.4, 255a30-b3: 16,8-10; 8.4, 45,22-4; 3, 439b6-10: 44,3-4; 3,
255b13-20: 12,10-11; 8.5, 439b7: 44,21-3; 3, 439b8-10:
256b13-24: 21,25-6; 8.5, 258a1-4: 44,20-1; 3, 439b10-18: 44,16-19; 3,
21,25-6; 8.5, 258b4-9: 21,25-6; 8.6, 439b11-12: 42,5-7; 44,13-14; 3,
258b24-6: 22,5-12; 8.6, 259b1-14: 439b11-18: 45,20; 3,
22,15-23; 8.7, 260a30-2: 33,15-16; 439b25-440a3: 26,7-13; 3,
8.7, 261a32-b1: 33,18-24; 8.7, 440a15-20: 42,19-22; 3, 440a15-16:
261a36: 6,30 42,19-22; 3, 440a18-19: 42,8-11; 3,
Polit. (= Politica) 1.2, 1252b1-5: 440b14-23: 45,10-19; 4, 440b16-18:
28,6-8; 1.2, 1253a20-5: 15,2-5; 18,7; 45,6-7; 4, 441a3: 40,11-15; 4,
1.8, 1256b20-2: 27,9; 3.1, 441b12-15: 7,9-13; 4,
1275a35-b2: 17,1-8; 5.4, 441b24-442a12: 40,11-15; 4,
1303b29-31: 10,19-20; 7.16, 442a12-16: 45,10-19; 4, 442a29-b1:
1335b19-26: 38,4-8 27,6-7; 4, 442b4-7: 21i n.42; 41,5; 4,
Protr. (= Protrepticus, Dring; 442b7: 41,1-3; 4, 442b8-10:
citations also given for Iamblichus 41,13-42,3; 4, 442b8-10: 41,10-13; 5,
Protrepticus below) B13: 3,15-16; 444a25-7: 28,6-8; 6, 445b4-8:
B14: 3,15-16; B59: 23,26-8 40,20-1; 6, 446a20-b2: 42,19-22; 6,
Respir. (= de Respiratione) 3, 446a20-447a11: 42,22-43,4; 43,11;
471b20-3: 37,4-5; 3, 471b23-9: 6, 446a25-8: 42,19-22; 6,
23,11-13; 8, 474a25-8: 35,2-3; 10, 446a25-b2: 42,22-43,4; 6, 446b9-13:
476a11-15: 27,9; 15, 478a24: 42,19-22; 43,11; 6, 446b27-447a3:
39,21-40,3 42,19-22; 6, 446b27-447a11:
Rhet. (= Rhetorica) 1.11, 1370a32-5: 42,22-43,4; 6, 447a8-10: 42,19-22;
39,5-6; 2.21, 1395a20-6: 1,4-2,4 6, 447a9-10: 43,11; 6, 447a10-11:
SE (= de Sophisticis Elenchis) 34, 42,19-22; 6, 447a11: 46,1-6
183b22-6: 10;19-20 Somn. (= de Somno) 1, 454a9-11:
Sens. (= de Sensu et Sensibilia) 1 38,21-39,2; 1, 454b25: 10,8-10; 1,
passim: 39,9-10; 1, 436b6-7: 454b32-455a3: 35,10-11; 2,
12,9-13,8; 1, 436b10-12: 10,8-10; 1, 455a12-b2: 39,18-21; 2, 455a27:
436b12-18: 39,9-10; 1, 436b13-14: 40,11-15; 2, 455b17-18: 28,8; 2,
40,11-15; 1, 437a1-17: 39,9-10; 1, 455b34-456a6: 24,5; 39,21-40,3; 2,
437a3-17: 40,20-1; 1, 437a8-9: 41,5; 456a3-10: 39,21-40,3; 2, 456a6-10:
2-5 passim: 40,21-41,5; 2, 39,21-40,3; 3, 456a34-b6: 36,3-4; 3,
437a31-2: 42,4-5; 2, 437b4-7: 458a28: 39,18-21
Index of Passages Cited 231
Top. (= Topica) 2.4, 111b24-6: Corporum Caelestium, Ziegler)
35,10-11; 2.7, 113a29-30: 22,13-15; 2.164.4: 15,15-17; 8.16-21: 15,15-17
2.11, 115b29-35: 4,5; 4.2, CORPUS PAROEM. GR. (= Corpus
122b16-17: 12,3-4; 4.2, 122b30-1: Paroemigraphorum Graecorum
8,16; 5.2, 129a26: 10,8-10; 5.5, App. prov. 1.41: 10,19-20;
135a16: 10,8-10; 6.2, 139b21: Diogenianus 2.97: 10,19-20
25,4-7; 6.4, 141a26-142a16: DAMASCIUS
32,23-4; 6.6, 145b7-11: 25,4-7; 6.10, in Phaed. (= in Platonis Phaedonem,
148a27-31: 28,25-9 Westerink) I 222: 18i n.1; 235:
de Vita et Morte 1, 479b1-7: 37,4-5 18i n.1; 236: 18i n.1; 243: 18i n.1;
ARIUS DIDYMUS 297: 18i n.1; 405: 18i n.1; II 14:
Fragmenta (in Diels, Doxographi 18i n.1; 25: 18i n.1; 88: 18i n.1; 89:
Graeci) fr.5, 449.12-14: 7,9-13; fr. 18i n.1; 100: 18i n.1; 110: 18i n.1;
17, 456,15: 42,7-8; fr. 20, 128: 18i n.1; 129: 18i n.1; 132: 18i
457.25-458.11: 3,28-4,4; fr. 24, n.1; 145: 18i n.1
460.14-16: 18,10-14; fr. 39, DAMIAN
471.18-21: 17,12-15 Opt. (= Optica, Schne) 2, 19-21: 44,7;
ATTICUS 12,12-16: 44,7
Fragmenta (des Places) fr. 7.34-43, DAVID [ELIAS]
57-63: 2,17-18; 12,8 in Categ. (= in Aristotelis Categoriae)
AUGUSTINE 128,12-13: 19i n.2
CD (= de Civitate Dei) 19.3: 1,4-2,4 DEMOCRITUS
AVERROES Fragmenta (in Die Fragmente der
Comm. Magn. in DA (= Vorsokratiker, Diels-Kranz) 68 A 1:
Commentarium Magnum in 19,21; 68 A 101: 27,6-7; 68 A 105:
Aristotelis de Anima Libros) 27,6-7; 68 A 113: 27,6-7; 68 A 119:
2.67.39-45, 231-2: 42,11-19 27,6-7; 68 A 135: 42,19-22; 68 B
CALCIDIUS 214: 27,6-7; 68 B 234: 27,6-7; 68 B
in Tim. (= in Timaeum, Waszink) 220, 236: 27,6-7
233.5-6: 41,1-3; 228, 243.13-18: Fragmenta (Luria) fr. 459: 27,6-7
26,18-20; 290, 294.8-9: 3,28-4,4; DESCARTES
292, 295.3-6: 3,28-4,4; 293, Meditations on First Philosophy AT
295.17-18: 3,28-4,4; 311, 311.7-8: 7.30-2: 4,9-15; AT 7.81: 14,4-5
3,28-4,4 DICAEARCHUS
CICERO Fragments (Mirhady) fr. 40: 27,15-28,2
Acad. (= Academica)1.19: 1,9; 1.27: DIO CASSIUS
3,28-4,4; 1.39: 7,9-13 Historia Romana (Boissevain) 72.31:
Fin. (= de Finibus) 2.34: 1,9; 3.73: 19i n.3
1,4-2,4; 4.14: 1,9; 4.15-18: 1,9; DIOGENES LAERTIUS
4.25-6: 1,4-2,4; 1,9; 4.43: 1,9; 4.61: 7.4: 1,9; 7.89: 1,9; 7.105: 1,9; 7.110:
1,9; 5.26-7: 1,4-2,4; 1,9; 5.34: 18,25-19,1; 32,1-4; 40,4-5; 7.134: 20i
1,4-2,4; 5.41: 1,4-2,4; 5.44: 1,4-2,4; n.17; 3,28-4,4; 6,6-7; 17,15-17;
1,9; 5.92: 36,14-18 7.150: 3,28-4,4; 7.157: 18,25-19,1;
Leg. (= de Legibus) 1.58-9: 1,6-8; 1.61: 19,6-9; 26,15-17; 32,1-4; 40,4-5;
1,4-2,4 7.159: 24,5; 8.36: 27,15-28,2
Luc. (= Lucullus) 19: 41,13-42,3 DIOGENES OF OENOANDA
de Nat. Deor. (= de Natura Deorum) Fragmenta (Smith) fr. 42: 27,15-28,2
1.70: 41,1-3 DOX.GR.(= Doxographi Graeci, Diels)
Tusc. Disp. (= Tusculanae 278.12-15: 19,21; 279.12-17:
Disputationes) 1.52: 1,6-8; 5.22: 33,15-16; 308.4-9: 3,28-4,4; 310.6-7:
36,14-18 7,9-13; 313.8-20: 42,19-22;
CLEMENT OF ALEXANDRIA 388.16-20: 21,13-16; 388.21-389.8:
Strom. (= Stromata) 1.14.60.3: 1,6-8; 26,17-18; 390.5-13: 32,1-4; 393.1-7:
2.133.4: 1,9; 7.32.9: 1,9; 8.10.3-4: 17,12-15; 395.1-12: 41,3-5;
38,4-8 395.12-15: 41,1-3; 403.26-8:
CLEOMEDES 42,11-19; 410.25-9: 24,5; 27,6-7;
Motu Circ. (= de Motu Circulari 410.25-411.24: 27,6-7; 30,26-31,2;
232 Index of Passages Cited
410.30-411.13: 18,25-19,1; 32,1-4; GALEN
40,4-5; 425.14-426.14: 38,4-8; Comp. Med. Gen. (= de Compositione
425.18-22: 38,4-8; 440.4-10: Medicamentorum per Genera,
33,15-16; 449.12-14: 7,9-13; 456,15: Khn) 13.593: 24,23-25,2
42,7-8; 457.25-458.11: 3,28-4,4; Comp. Med. Loc. (= de Compositione
460.14-16: 18,10-14; 471.18-21: Medicamentorum secundum Locos
17,12-15; 500.28-9: 42,19-22; libri x, Khn) 12.602: 24,23-25,2;
513.20: 42,19-22 12.604: 24,23-25,2; 12.610:
DK (= Die Fragmente der Vorsokratiker, 24,23-25,2
Diels-Kranz) CAM (= de Constitutione Artis
13 B 2: 19,21; 21 B 7: 27,15-28,2; 22 Medicae, Khn)1.242: 24,23-25,2
A 15: 19,21; 31 A 57: 42,19-22; 31 A CC (= de Causis Continentivus
77: 33,15-16; 31 A 92: 42,19-22; 31 libellus, CMG Suppl. 2) 2.3.2:
B 62: 33,15-16; 31 B 84.10-11: 44,7; 24,23-25,2
31 B 89: 42,19-22; 31 B 90: Foet. Form. (= de Foetuum
33,15-16; 31 B 115: 27,15-28,2; 31 B Formatione, Nickel) 3, 68.9-17:
117: 27,15-28,2; 31 B 126: 31,25-32,1; 3, 68.10-70.21: 38,4-8; 3,
27,15-28,2; 31 B 137: 27,15-28,2; 31 74.11-16: 38,4-8; 6, 104.16-19:
B 147: 27,15-28,2; 59 A 45: 31,25-32,1
33,15-16; 59 A 46: 33,15-16; 59 B 6: Hipp. Elem. (= de Elementis ex
33,15-16; 59 B 11: 33,15-16; 68 A 1: Hippocrate libri ii, Khn)1.452:
19,21; 68 A 101: 27,6-7; 68 A 105: 24,23-25,2; 1.455: 24,23-25,2
27,6-7; 68 A 113: 27,6-7; 68 A 119: Hipp. Epid. (= in Hippocratis
27,6-7; 68 A 135: 42,19-22; 68 B Epidemiarum libri vi, Khn)
214: 27,6-7; 68 B 234: 27,6-7; 68 B 17b.251.1-6: 19,6-9; 31,25-32,1
236: 27,6-7 Hipp. Epid. (= in Hippocratis
EMPEDOCLES Epidemiarum libri vi, CMG
Fragmenta (in Die Fragmente der Wenkebach-Pfaff) 4.11,
Vorsokratiker, Diels-Kranz) 31 A 214.24-215.3: 42,19-22
57: 42,19-22; 31 A 77: 33,15-16; 31 Hipp. Fract. (= in Hippocratis librum
A 92: 42,19-22; 31 B 62: 33,15-16; de fracturis commentarii iii, Khn)
31 B 84.10-11: 44,7; 31 B 89: 18b.598: 24,23-25,2
42,19-22; 31 B 90: 33,15-16; 31 B HNH (= in Hippocratis de Natura
115: 27,15-28,2; 31 B 117: Hominis librum commentarii iii,
27,15-28,2; 31 B 126: 27,15-28,2; 31 Khn) 15.18: 24,23-25,2; 15.32:
B 137: 27,15-28,2; 31 B 147: 24,23-25,2; 15.123.14-124.7:
27,15-28,2 31,25-32,1
EPICTETUS Intr. (= Introductio sive medicus,
1.1.10: 1,4-2,4; 1.6.25: 1,4-2,4; Khn) 14.726.7-11: 11,14
1.18.17: 1,4-2,4; 2.6.14: 1,4-2,4; adv. Iul. (= adversus Iulianum, Khn)
3.1.25: 1,4-2,4 18a.266.9-13: 31,25-32,1
EPICURUS adv. Lyc. (= adversus Lycum libellus,
Ep. ad Hdt. (= Epistula ad Khn) 18a.240: 24,23-25,2;
Herodotum) 63-6: 11,15-17 18a.241: 24,23-25,2
Fragmenta (in Arrighetti, Epicuro MM (= de Methodo Medendi, Khn)
Opere2)158: 26,17-18; 159: 26,17-18 10.281: 24,23-25,2; 10.352:
Fragmenta (in Long & Sedley, The 24,23-25,2; 10.353: 24,23-25,2;
Hellenistic Philosophers) 14C: 10.882: 24,23-25,2; 10.883:
26,17-18 24,23-25,2
Fragmenta (in Usener, Epicurea) 314: MMG (= ad Glauconem de Medendi
26,17-18; 315: 26,17-18 Methodo libri ii, Khn) 11.84, 138:
EUSEBIUS 24,23-25,2
PE (= Praeparatio Evangelica, Mras) Nat. Fac. (= de Naturalibis
15.9.7-8, 2.370.4-9: 2,17-18; Facultatibus, Helmreich) 1.1,
15.9.11, 2.371.1-6: 2,17-18; 15.14.1, 101.1-15: 31,25-32,1; 2.4,
2.378.19-379.2: 7,9-13; 15.20.6, 168.12-14: 17,15-17
2.385.9-12: 17,12-15 PHP (= de Placitis Hippocratis et
Index of Passages Cited 233
Platonis, De Lacy) 2.5.35, 134.23-6: JULIAN
39,18-21; 3.1.25, 172.20-6: 15,15-17; Or. (= Orationes) 6, 183A: 1,6-8; 6,
3.7.4, 212.14-18: 15,15-17; 5.3.8, 185D-186A: 1,4-2,4; 6, 188B: 1,6-8
306,23-7: 26,15-17; 6.3.7, JUVENAL
374.18-19: 31,25-32,1; 7.4.1, Sat. (= Saturae) 11.27: 1,6-8
448.7-9: 41,1-3; 7.5.11, 454.27-9: LUCIAN
42,22-43,4; 7.5.35-8, 460.11-16: Eunuch 2-3 : 19i n.3
41,1-3; 7.6.24, 466.25-8: 41,5; LUCRETIUS
7.6.24, 466.26: 41,1-3; 7.7.1-2, DRN (= de Rerum Natura) 3.425-58:
470.5-11: 42,11-19; 7.7.4-8, 11,15-17; 4.72-89: 42,11-19
470.17-472.2: 41,3-5; 7.7.17, MELETIUS
472.29-33: 39,18-21 de Nat. Hom. (= de Natura Hominis,
QAM (= Quod Animi Mores Corporis Cramer) 101.13: 34,16-17
Temperamenta Sequuntur, Mller) MICHAEL OF EPHESUS
4, 37.5-23: 25,2-4; 4, 44.12-20: in EN V (= in Aristotelis Ethica
25,2-4; 4, 44.20-45.3: 25,2-4; 4, Nicomachea V) 556,37-557,8:
45.5-8: 19,6-9; 4, 45.5-25: 26,15-17 38,21-39,2; 584,3: 18i n.1; 598,30-2:
SMT (= de Simplicium 1,9
Medicamentorum Temperamentis ac in EN IX-X (= in Aristotelis Ethica
Facultatibus libri xi, Khn) 12.328: Nicomachea IX-X) 4,17: 18i n.1
24,23-25,2 in Parva nat. (= in Aristotelis Parva
Subst. Nat. Fac. (= de Substantia Naturalia) 135,22-30: 25,2-4
Facultatum Naturalium, Khn) NEMESIUS
4.759.9-15: 31,25-32,1 Nat. Hom. (= de Natura Hominis,
Caus. Sympt. (= de Symptomatum Morani) 2, 20.14-17: 7,9-13; 2,
Causis, Khn) 7.129.9-10: 21.6-9: 7,9-13; 15,15-17; 2, 22.3-6:
31,25-32,1 17,12-15
UP (= de Usu Partium, Helmreich) OLYMPIODORUS
1.2, 1.13-2.2: 23,26-8; 1.408.5-13: in Meteor. (= in Aristotelis Meteora,
3,15-16 Stve), 263,20-1: 18i n.1 ; 273,27-8:
PS-GALEN 18i n.1
Def. Med. (= Definitiones Medicae, ORIGEN
Khn) 116, 19.379.7: 41,1-3; 445, Orat. (= de Oratione, Koetschau) 6.1,
19.451-2: 38,4-8; 445, 19.452.5-10: 311.16-312.10: 31,25-32,1; 27.8,
38,4-8 368.2-10: 3,28-4,4
Qual. Incorp. (= Quod Qualitates PHILO
Incorporeae Sint, Giusta) passim: Aet. (= de Aeternitate Mundi,
17,15-17; 1, 10-11: 17,17-18; 2, Cohn-Wendland) 9, 75.14-76.3:
19-23: 14,19-20; 19, 273-4: 3,28-4,4 6,14-15; 131, 113.5: 6,14-15
HERACLITUS Conf. Ling. (= de Confusione
Fragmenta (in Die Fragmente der Linguarum, Cohn-Wendland)
Vorsokratiker, Diels-Kranz) 22 A 184-7, 2.264.23-265.15: 11,18-20;
15: 19,21 187, 2.265.11: 24,23-25,2
HERODOTUS Ebr. (= de Ebrietate, Cohn-Wendland)
2.123: 27,15-28,2 90, 2.176.13-14: 3,15-16
HIEROCLES Leg. Alleg. (= Legum Allegoriae,
Elem. Eth. (= Elementa Ethica, Long) Cohn-Wendland) 2.22-3, 1.95.7-20:
1.5-27: 19,6-9; 4.6-8: 12,2-3; 11,14; 27,4-8; 31,25-32,1
4.39-53: 15,15-17 Quod Deus sit Immut. (= Quod Deus
PS-HIPPOCRATES sit Immutabilis, Cohn-Wendland)
de Alim. (= de Alimento) 8: 34,16-17 35-46, 2.64.1-66.15: 27,4-8;
IAMBLICHUS 31,25-32,1
Protr. (= Protrepticus, Pistelli; for PHILOPONUS
Drings fragments, see under de Aet. Mundi (= de Aeternitate
Aristotle Protr.) 49.28-50.1: Mundi, Rabe) 48,10: 18i n.1;
3,15-16; 50.1-14: 3,15-16; 191,19-20: 18i n.1; 311,18: 18i n.1;
41.15-20: 23,26-8 482,22: 18i n.1; 520,25: 18i n.1;
234 Index of Passages Cited
521,3: 18i n.1; 524,14: 18i n.1; 27,15-28,2; 91D-92C: 27,15-28,2
553,9: 18i n.1 PHILOSTRATUS
in An. Pr. (= in Aristotelis Analytica VS (= Vitae Sophistarum, Kayser)
Priora) 126,20-1: 18i n.1 566: 19i n.3
in DA (= in Aristotelis de Anima) PLOTINUS
35,10-16: 27,6-7; 51,13-29: 5,5-6; 1.1.3.21-3: 15,9-10; 4.3.17.21-31:
71,19-34: 27,6-7; 159,18-19: 19i n.9; 15,9-10; 4.3.20: 14,18; 4.3.21.5-21:
224,28-37: 15,9-10; 264,15-22: 15,9-10; 4.7.7.7-11: 39,18-21;
32,29-30; 284,11-12: 34,16-17; 4.7.83.1-3: 19,6-9
288,5-19: 35,2-3; 36,9-13: 18i n.1; PLUTARCH
334,34-354,2: 42,19-22; 519,29: Adv. Colot. (= Adversus Colotum) 20,
41,1-3 1118D-E: 26,17-18
in GC (= in Aristotelis de Generatione Comm. Not. (= de Communibus
et Corruptione) 62,17: 18i n.1; Notitiis) 23, 1069F: 1,9; 30,
82,12-13: 18i n.1; 214,23: 18i n.1; 1073E: 7,9-13; 37, 1078C: 20,8-19;
226,17-18: 18i n.1 37, 1077E-1078E: 20,8-19; 40,
PS-PHILOPONUS 1080F: 7,9-13; 45, 1084CD:
in DA (= in Aristotelis de Anima) 42,19-22; 50, 1085E-1086B:
595,37-596,14: 28,25-9; 38,16-18 17,15-17; 18,2-7
PLATO de Lib et Aegr. (= de Libidine et
Alc. I (= Alcibiades I) 129C-130A: Aegritudine, Ziegler-Pohlenz) 4,
23,26-8; 130E: 1,4-2,4; 133A: 44,7 53.13-54.9: 24,5; 4, 53.22-3: 24,5
Laws 4, 721B-C: 32,11-15; 6, de Pyth. Or. (= de Pythiae Oraculis)
753E6-754A2: 10,19-20; 9, 870D-E: 21, 404B: 23,26-8
27,15-28,2; 10, 903D: 27,15-28,2; Fac. in Orb. Lun. (= de Facie in Orbe
10, 904A-C: 27,15-28,2 Lunae) 5, 922E: 42,19-22; 18, 930F:
Meno 76C-D: 42,19-22; 81A-B: 42,19-22
27,15-28,2 Quaest. Conviv. (= Quaestiones
Phaedo 70C-D: 27,15-28,2; 72A-D: Convivales) 4.1, 663A: 33,15-16
27,15-28,2; 81B-82C: 27,15-28,2; Quaest. Nat. (= Quaestiones
85E-86C: 11i; 24,18-20; 89C-91C: Naturales) 19, 916D: 42,19-22
2,17-18; 91C-D: 3i; 2,17-18; Quaest. Plat. (= Quaestiones
91E-95A: 24,18-20; 94E: 3i; 2,17-18; Platonicum) 8.2, 1006D: 23,26-8
95E-96A: 7,9-13; 100A-105C: Sept. Sap. (= Septem Sapientium
7,9-13; 107D-E: 27,15-28,2; 113A: Convivium) 21, 163D: 23,26-8
27,15-28,2 de Virt. Mor. (= de Virtute Morali) 3,
Phdr. (= Phaedrus) 229E-230A: 441C: 27,6-7; 7, 446F-447A: 27,6-7
1,4-2,4; 245C-E: 22,24-23,6; Stoic. Rep. (= de Stoicorum
248C-249C: 27,15-28,2 Repugnantiis) 34, 1050A: 2,18-22;
Polit. (= Politicus) 270A: 36,6-9 41, 1052F: 19,6-9; 38,4-8; 41,
Rep. (= Republic) 2, 377A: 10,19-20; 4, 1053C-D: 19,6-9
434D-441C: 27,4-8; 4, 436B-437A: POLEMON
27,4-8; 10, 614B-621B: 27,15-28,2 Fragmenta (Gigante) fr. 97: 1,9; fr.
Symp. (= Symposium) 206E-207A: 112: 1,9; fr. 127: 1,9; fr. 129: 1,9
32,11-15; 207D-208B: 32,11-15 PORPHYRY
Tht. (= Theaetetus) 184D: 39,18-21; ad Gaurum (Kalbfleisch) 10.4-6:
185B-C: 41,5; 185D: 41,5; 15,9-10; 47.5-48.8: 15,9-10
185D-186C: 41,5 Vit. Pyth. (= Vita Pythagorae, des
Tim. (= Timaeus) 34C-35B: 26,18-20; Places) 19: 27,15-28,2
41D-42D: 27,15-28,2; 45BC: 44,6-9; POSIDONIUS
45D: 39,18-21; 49A-52B: 3,28-4,4; Fragments (Kidd) fr. 92: 3,28-4,4
50B7-C2: 3,28-4,4; 50B8: 3,28-4,4; PRESOCRATICS,see above under DK (=
50C1: 3,28-4,4; 50D4-E1: 3,28-4,4; Die Fragmente der Vorsokratiker,
50D7: 3,28-4,4; 50E2: 3,28-4,4; Diels-Kranz)
51A7: 3,28-4,4; 52A-B: 14,19-20; PROCLUS
58C: 42,19-22; 67C: 42,19-22; in Tim. (= in Platonis Timaeum,
76D-E: 27,15-28,2; 90E-91A: Diels) 1.322.4-7: 18i n.1
Index of Passages Cited 235
PTOLEMY STOBAEUS
de Judic. (= de Judicandi Facultate et Ecl. (= Eclogae, Wachsmuth) 1.19.5:
Animi Principatu, Lammert) 8, 18,10-14; 1.132.27-133.11, 18-23:
13.5-8: 41,1-3; 8, 13.6: 39,18-21 3,28-4,4; 1.353.12-354.6: 31,2-6;
SENECA 1.138.14-22: 17,15-17;
Cons. ad Marc. (= Consolatio ad 1.138.14-139.4: 7,9-13; 1.141.19-22:
Marciam) 11.1-5: 1,4-2,4 7,9-13; 167.9-12: 18,10-14;
NQ (= Naturales Quaestiones) 3, 1.368.12-20: 27,6-7; 30,26-31,2;
praef. 15: 1,4-2,4 31,2-6; 2.77.16-19: 1,9; 2.78.1-6:
Ep. (= Epistulae) 94.28: 1,4-2,4; 121.3: 1,9; 2.80.3-4: 1,9
1,4-2,4; 113.23: 27,6-7 STOICS
SEXTUS EMPIRICUS LS (= The Hellenistic Philosophers,
Adv. Math. (= Adversus Long & Sedley) 14A 1-5: 11,15-17;
Mathematicos) 7.183: 41,13-42,3; 14C: 26,17-18; 14F 2-3: 11,15-17;
7.253-8: 41,13-42,3; 7.307: 27,6-7; 28K: 17,15-17; 28L 1: 17,15-17;
7.359: 27,6-7; 7.424: 41,13-42,3; 28q: 3,28-4,4; 40L: 41,13-42,3; 44B:
7.78-80: 15,15-17; 8.263: 7,9-13; 20i n.17; 6,6-7; 44B 2: 3,28-4,4; 44B
9.11: 3,28-4,4; 9.75: 3,28-4,4; 9.211: 3: : 20i n.17; 3,28-4,4; 17,15-17; 44C
7,9-13; 10.3-4: 14,19-20 1: 3,28-4,4; 44D 3: 3,28-4,4; 45A:
Pyrr. Hyp. (= Pyrrhonei Hypotyposes) 7,9-13; 45C: 7,9-13; 15,15-17; 45D:
1.120: 41,13-42,3; 2.70: 19,6-9; 17,12-15; 45G: 7,9-13; 47N: 11,14;
26,15-17 47P: 11,14; 27,4-8; 31,25-32,1; 47Q:
SIMPLICIUS 27,4-8; 31,25-32,1; 48C 3-4:
in Cael. (= in Aristotelis de Caelo) 11,18-20; 48C 10: 11,17-18; 12,2-3;
121,12: 18i n.1; 279,5-14: 49B: 14,19-20; 49E: 14,19-20; 53A:
14,24-15,1; 379,18-23: 21,13-16; 31,25-32,1; 53H 3: 18,25-19,1; 55A:
380,29-381,2: 9,14-22; 463,3-6: 7,9-13; 55A 1-3: 17,15-17; 53B 5-9:
28,25-9; 578,32-579,2: 20i n.20; 15,15-17; 53G 7: 41,1-3; 53H:
700,9: 18i n.1 27,6-7; 30,26-31,2; 53H 3: 32,1-4;
in Cat. (= in Aristotelis Categoriae) 40,4-5; 53K: 27,6-7; 30,26-31,2;
78,17-20: 15,2-5; 17,17-18; 31,2-6; 53W: 17,12-15; 55B: 7,9-13;
217,32-3: 17,15-17; 218,1: 26,15-17; 55G: 7,9-13; 55N: 2,18-22; 63A: 1,9;
271,20-2: 17,15-17 65G: 27,6-7; 69E 4: 41,13-42,3
in DA (= in Aristotelis de Anima) SVF (= Stoicorum Veterum
52,28: 18i n.1; 96,3-10: 15,9-10; Fragmenta, von Arnim) 1.10-11:
126,27-37: 41,1-3 1,9; 1.13: 1,9; 1.41: 1,9; 1.86:
in Phys. (= in Aristotelis Physica) 3,28-4,4; 1.87: 3,28-4,4; 1.88:
23,33-24,12: 35,3-5; 80,15-16: 18i 3,28-4,4; 1.89: 7,9-13; 17,15-17;
n.1; 211,15-18: 3,28-4,4; 213,23: 18i 1.90: 7,9-13; 1.98: 7,9-13; 1.106:
n.1; 213,34: 18i n.1; 227,23-6: 6,14-15; 1.135: 19,6-9; 26,15-17;
3,28-4,4; 258,16-17: 18i n.1; 1.151: 39,18-21; 1.183: 1,9; 1.484:
270,26-31: 14,24-15,1; 311.1-19: 19,6-9; 26,15-17; 1.518: 7,9-13;
19,6-9; 312,1: 18i n.1; 313,30: 18i 15,15-17; 1.555: 1,9; 2.299: 20i n.17;
n.1; 381,31-382,3: 33,15-16; 3,28-4,4; 6,6-7; 17,15-17; 2.300:
409,27-32: 42,22-43,4; 436,24-35: 3,28-4,4; 2.301: 3,28-4,4; 2.306:
42,22-43,4; 437,1-6: 42,22-43,4; 17,12-15; 2.310: 20i n.17; 3,28-4,4;
437,18-28: 42,22-43,4; 460,15-20: 17,15-17; 26,15-17; 2.311: 3,28-4,4;
33,15-16; 530,9-30: 20,8-19; 2.316: 3,28-4,4; 2.317: 3,28-4,4;
707,33-4: 18i n.1; 795,34-5: 18i n.1; 2.318: 3,28-4,4; 2.323a: 3,28-4,4;
964,9-23: 22,5-12; 1170,1-2.13: 18i 2.326: 3,28-4,4; 2.336: 7,9-13; 2.363:
n.1; 1175,13: 18i n.1; 1176,32: 18i 7,9-13; 2.340: 7,9-13; 2.341: 7,9-13;
n.1 2.377: 17,17-18; 2.379: 17,15-17;
SOPHOCLES 2.380: 17,15-17; 2.383: 17,15-17;
Phil. (= Philoctetes) 497: 41,1-3 2.389: 17,15-17; 26,15-17; 2.394:
SPEUSIPPUS 17,15-17; 2.410: 17,15-17; 2.432:
Fragmenta (Isnardi Parente) fr. 101: 17,15-17; 42,19-22; 2.433a and b:
1,9 42,19-22; 2.441: 15,15-17; 2.442:
236 Index of Passages Cited
26,15-17; 2.458: 11,14; 27,4-8; 3.282: 1,4-2,4; 3.301: 3,15-16; 3.339:
31,25-32,1; 2.472: 11,18-20; 1,4-2,4; 3.459: 27,6-7
24,23-25,5; 24,23-25,2; 2.473: STRATO
11,17-18; 11,18-20; 12,2-3; 15,15-17; Fragmenta (Wehrli) fr. 111: 24,5
2.475: 15,15-17; 2.502: 14,19-20; SYRIANUS
2.505: 14,19-20; 2.525: 7,9-13; in Metaph = in Aristotelis Metaphysica
2.534: 15,15-17; 2.570: 42,19-22; 53,12: 18i n.1; 54,12-13: 18i n.1;
2.620: 6,14-15; 2.665: 42,19-22; 100,6: 19i n.2; 100,6-7: 18i n.1;
2.710: 31,25-32,1; 2.711: 11,14; 122,11-12: 18i n.1; 195,12: 18i n.1
2.712: 31,25-32,1; 2.715: 19,6-9; THEMISTIUS
31,25-32,1; 2.716: 11,14; 2.756: in DA (= in Aristotelis de Anima
38,4-8; 2.757: 38,4-8; 2.774: paraphrasis) 43,30-5: 15,9-10;
26,15-17; 2.785: 19,6-9; 2.786: 44,22-4: 28,25-9; 53,26-33: 36,9-13;
19,6-9; 2.787: 19,6-9; 26,15-17; 57,36-58,5: 41,1-3; 61,21-4:
2.790: 17,12-15; 2.793: 18,10-14; 43,16-18; 61,22-32: 42,4-5; 120,18:
18,25-19,1; 2.804: 19,6-9; 2.806: 18i n.1; 123,24-9: 28,25-9;
19,6-9; 2.806.1: 38,4-8; 2.809: 125,21-30: 28,25-9
17,12-15; 2.810: 17,12-15; 2.823: in Phys. (= in Aristotelis Physica
27,4-8; 2.826: 27,6-7; 30,26-31,2; paraphrasis) 26,20-4: 3,28-4,4
31,2-6; 2.827: 32,1-4; 2.828: THEOPHRASTUS
18,25-19,1; 32,1-4; 40,4-5; 2.836: Sens. (= de Sensibus, in Diels, Dox.
18,25-19,1; 24,5; 27,6-7; 30,26-31,2; Gr.) 7, 500.28-9: 42,19-22; 50,
32,1-4; 40,4-5; 2.837: 24,5; 2.839: 513.20: 42,19-22
24,5; 2.841: 26,15-17; 2.849: 27,6-7; XENOCRATES
2.858: 39,18-21; 2.879: 41,1-3; Fragmenta (Heinze) fr. 79: 1,9
2.882: 39,18-21; 2.886: 15,15-17; XENOPHANES
2.900: 15,15-17; 2.945: 2,18-22; Fragments (Diels-Kranz, Die
2.937: 2,18-22; 2.989: 31,25-32,1; Fragmente der Vorsokratiker) 21 B
2.1013: 15,15-17; 3.16: 1,9; 3.17: 7: 27,15-28,2
1,9; 3.45: 1,9; 3.126: 1,9; 3.136: 1,9;
Subject Index

In the index below, the numerals generally indicate page and line numbers in Alexan-
ders text (which appear in the margins of the translation). References to the Introduc-
tion are indicated by appending an i to the page number of the present volume (e.g. 12i
or 15i n.40). References to the translators notes are indicated by appending n to the
page and line numbers of the passage to which the note belongs. If a citation is given
under a more specific sub-entry within an entry, or is cross-listed as being under
another entry, the citations are given only under that specific sub-entry or cross-listed
entry. Discussions of ancient authors by Alexander are included in the subject index
below, as well as any I have mentioned in the notes (apart from Alexander and
Aristotle). Citations from ancient works are listed in the Index of Passages.

abortion, 38,4-8n; cf. 14i 30,12-13; 38,18-19; initial formation


Academy (Old), 1,9n; 1,4-2,4n; 3,28-4,4n of, 14i, 36,19-37,3; non-rational,
accident, see extrinsic characteristic 30,9-15; rational, 30,9-15; sense of
accompanying awareness (sunaisthsis), touch is necessary, 40,14-15; what it
15,15-17; Stoic view, 15,15-17n is to be an animal is located in what
acting (poiein): in virtue of form, 7,9-13 is hot and moist, 40,1-2
activity (energeia): vs change (kinsis), animate (empsukhon), 16,16; 20,7;
12,9-13,8n; vs power, 8i; 9,20-6; 21,4.10; 23,22; 24,11; 27,20; 29,3-10;
16,8-10.16-18; 32,25-7; 32,30-33,5 body, 12,2-3; 14,7-8.21-2; 15,19;
Aetius, 3,28-4,4n 19,6-9; distinguished from inanimate
affect (pathos), 13,2-5; see also by power for nourishing oneself,
modification 29,3-10
afterimages, 43,18-44,1n Antiochus of Ascalon, 1,9n; 1,4-2,4n;
air, 6,24; 8,21; 19,21.27.35; 26,17; 42,7; 3,28-4,4n; 41,13-42,3n
transparent, moist, and without a appetite (epithumia), 12,15; 13,2;
definite shape, 44,4-5 involves bodily changes, 12,15
akrasia, see under loss of control Apollo, see under Pythian god
Alcinous, 3,28-4,4n; 27,15-28,2n Apollodorus, 18,10-14n; 18,25-19,1n
Alexander: The Commentator, 18-19i Arcesilaus, 20,8-19n
n.1; How We See, 43,16-18n Aristotle of Mytilene, 19,21n
alive, 29,4-10; see also under life Aristotle of Stagira, 2,4-9; 16,5; 19,21n;
alteration (alloisis), 12,17; vs growth, his authority, 2i; 2,4-9n; How We See,
33,28-34,8n; illumination and 43,16-18n; On the Soul, 2,6
tingeing are an alteration only in a Arius Didymus, 7,9-13n
way (ps), 42,19-22n; 42,22-43,4n; arkh, see under origin and principle
perception occurs through a kind of art (tekhn), 2,13n; 3,8-13; 23,26-8n;
alteration, 14i, 39,1-2; perception not 24,8-9; analogy with nature, see
an alteration, though it involves under nature; imitates nature, not
them, 15i, 38,21-39,2n; 44,9-13n vice versa, 3,15-16.
analogy, 10,12-14; 20,19-22; 24,26-7 artefacts, 5i; 3,2-13; 4,25-5,1; 16,3;
Andronicus, 25,2-4n 20,21; forms not substantial, 5,1-2;
anger, 12,16; 13,1.6-8; 23,15 qualities in virtue of their form, 6,4-6;
animal: a natural, compound body, substances in virtue of their matter,
10,1-2; compound of matter & form, 6,4-6
10,1-8; 12,6-7; compound of soul and assimilation, see under likening
body, 10,1-8; 11,14-12,7; defined by Atticus, 2,17-18n
perception, 10,8-10; 29,14-15; aulos, see under oboe
238 Subject Index
awareness (antilpsis, antilambesthai), Boethus, 3,28-4,4n; 15,2-5n; 17,17-18n
see esp. 39,4-5n; also 24,11; 39,4-5n; boundary: bodies lacking a boundary of
44,6; occurs in the heart, 39,21-40,3n; their own also lack a colour of their
see also accompanying awareness own, 46,7-9; of body is surface,
45,20-1, 46,12-17; of body coincides
balance (summetria), 25,6-7 with boundary of transparent, but
becoming like (homoisis, homoiousthai), not the same, 46,9-10.12-17; of
see under likening transparent is colour, 45,20-1,
bench, 25,16-18 46,12-17; that in virtue of which a
bile, 13,7 body is visible, 45,24-46,1
black or dark (melas), 33,21.23; 34,4; breath (pneuma), 19.6-20
45,16 bricks, 11,17
blend (krasis), 9-11i; 9,7; 10,18; bright, see white
11,17-20; 11,18-20n; 12,5-6; 13,21-3;
15,5-9; 19,26-20,6; 24,3-4; 26,17-18n; Carneades, 41,13-42,3n
26,21-2.26-7; always a blend of Cambridge changes (mere), 17i;
bodies, 15,8; powers of blend/mixture 38,21-39,2n; illumination is a mere
in relation to powers of ingredients, Cambridge change, 42,22-43,4n;
19,32-3; 20,1-6; Stoic view, 11,18-20n; perceiving is not a mere Cambridge
12,5-6n; 15,5-9n; see also mixture and change, 38,21-39,2n
diversity of form due to mixture and category mistake, 7i; 22,5-12n; 23,6-24n
blend of underlying matter under cause, see under responsible factors
form chameleon, 37,4-5n
body: act or are modified in virtue of change (kinsis): vs activity, see under
non-bodily items (i.e. forms), 7,9-13; activity; all change is in time, 43,11n;
always exists (as a class), 6,13-14 (cf. bodily changes always accompany
19-20); animate, 12,2-3; 14,7-8.21-2; soul activities, 12,8-13,8; powers and
15,19; 19,6-9; artefacts, see under dispositions do not undergo change,
artefacts; boundary is a surface, 7i; 22,2-10; simple, 7,17-21; what is
45,20-1; can produce growth, 34,6-7; inseparable on its own cannot
cannot pass through body, 14,8-9; produce change, 22,4-5; what
20,8-19; complexity due to variety of produces change does not necessarily
underlying subjects, 7,21-8,13; undergo change, 21,25; what
compound of other bodies, 3,22-5; produces change by contact
4,5.23.27; 7,22-8,8; 8,26; 9,9.23; 16,4; undergoes change, 21,26-22,2; see
compound of matter & form, 2,26-3,2; also Cambridge changes (mere); rest
10,2-8; 11,20-1; 17,11; composed characterisation (logos): general (koinos)
ultimately of non-bodily items, characterisation of perception,
6,6-17; division into parts, 18,10-27; 40,4-10; general (koinos)
every body has a larger or smaller characterisation of soul, 17,1-8;
share of transparency, 17i; 44,20-1; 28,15-20; 30,17-20; cf. 40,6-9
exists on its own, 4,8-9; 5,21 (cf. 6,1); Chrysippus, 3,28-4,4n; 11,17-18n;
17,11-12; has essence in virtue of 15,15-17n; 18,10-14n; 18,25-19,1n
form and culmination, 6,21-9; 7,4-8; cithara, 25,23-26,3
in place, 14,18; living body is animate Cleanthes, 15,15-17n
throughout, 12,2-3; 14,7-9; natural, clear (saphs): activities clearer than
3,21-6; 4,15-20; 8,26; 10,1-2; powers, 32,25-7; proceed from what is
16,2-5.10; organic, 10,28-11,5; 11,1n; less clear to what is more so, 32,23-4
15,2-5; 16,4.10-14; 23,26-8n; 24,6-11; cognition (krinein), see esp. 39,4-5n; also
not an organ of soul, 23,26-8; parts of 39,2
body are bodies, 14,6; 18,10-14.17; cold 6,9; 26,8; 33,22-4; as a tangible,
simple vs compound, 5i; 3,22-7; 4,4-9; 40,25
5,9.18-19; 7,14-8,13; 9,14-26; colour, 40,22; 42,4-7; able to affect the
10,26-11,13; 16,3-4; simple bodies, see actively transparent (= illuminated)
under Elements; Stoic view, 7,9-13n; medium, 16-17i; 42,5-7; 43,12-14;
14,8-9n; tangible, 5,19-20; 44,18-19; appears on surface, 45,22-4;
transparent, 44,20-1 bright (leukon) vs dark (melan),
Subject Index 239
44,13-14n; boundary of transparent correspondence theory of truth,
material with definite shape, 45,20-2; 41,10-13n
45,25-46,1; 46,6-7; cf. 44,17-19; culmination (teleiots), 6,30n; 6,29-7,8;
coincides with surface, but not the 16,5-6n; 19,24-8; 24,1.14; of body is
same, 46,12-17; contrary colours can its form, 6,29-7,3; 16,4-5; completion,
be present on a single surface, 16,5-6; of culminations, 12i; 8,12-13;
46,16-17; contrast with light, 45,5-7; first, 9,20-2; inseparable from what it
definition, 42,5-7; differences in is of, 17,12-13; and mixture, 19,24-8
colours due to proportion of earth,
transparency, or fire, 44,20-1; dark, see black
45,7-19; 46,4-6; due to presence of definition, 17,2
something bright, like fire, 45,11-12; Delphic command, see under Know
in the fundamental sense (kuris), Thyself
45,17-19; having a colour of its own Democritus, 42,19-22n; 22,5-12n;
(oikeion), 42,8; 45,18-19; cf. 43,7-8; 22,24-23,6n; 27,6-7; 44,1-2n
impermanent in transparent Descartes, 4,9-15n; 14,4-5n
materials without a definite shape, desire (orexis), 12,14-15; 29,11; involves
45,6-7; in transparent material with bodily changes, 12,14-15; see also
a definite shape, 17i; 44,17-19; appetite and striving
45,5-6.20; lacking a colour of its own detached parts, see dismemberment
(oikeion), 44,3-4; 46,7-9; only when Diogenes of Babylon, 24,5n
illuminated, 46,18-19; parallel to dismemberment and survival: of
light, 44,17-19; persistent in animals, 37,5; 37,11-38,11; of plants,
transparent materials with a definite 37.4-11
shape, 45,6-7; tingeing the disposition (hexis), 9,20-2; 15,11-13;
transparent medium, 17i; 42,11-19; 16,8; 31,12.17; does not undergo
46,10-11; that in virtue of which change or engage in activity, 22,2-10;
something is coloured is that in 23,24-24,4; inseparable from what
virtue of which it is visible, 45,24-5; has them, 15,13; 22,4-6; non-bodily,
transmitted through medium 15,13
instantaneously, 43,11n; visible in diversity, see under form under diversity
light, 43,12 of form due to mixture and blend of
common perceptibles, see under underlying matter
perceptibles Divine body, 43,9; 45,3
common sense, 41,1-3n; 41,3-5n; drugs, 24,23-25,2; fourfold drug, 10i;
perceives common perceptibles, 24,23-25,2n
65,10-21 drunkenness, 23,5
completion (entelekheia), 16,5-6n; dry 5,5; 8,19; 9,35; 19,35; 20,3.26; 26,8;
16,5-10; 24,1-2; first completion, as a tangible, 40,25
16,7.17-18; 21,5
complexity, see under forms under more Earth, 9,16-20; 14,2-3; 19,24-8n; 22,7-10;
advanced or complex 23,29-24,1; impedes transparency of
compound body, see under body bodies, 45,7-10; responsible for
composed from (ek): different senses, difference in colours, 45,9-11
11,15-12,1 effluences: light is not an effluence,
composition (sunthesis), 24,25-25,2; 42,19-22
25,4-11; 26,14.19.22-5 ek (out of), senses of, 11,15-21
conditional necessity, 9i (esp. n. 22) elements, 5,4-6,2; 7,14-21; 8,17-22;
contraries, 5,7.20-1; 15,1; transformed 16,3-4; 19,21-20,1; 26,7-9; 34,9-14;
into contraries, 33,19-26; not every capable of a single simple change,
contrary is nourishment for its 7,17-21; transformation of, 5,8-9;
contrary, 34,1-4; only those contraries 8,3-5; 32,11-15n; see also air, earth,
that cause growth are nourishment, fire, water
34,5-14 embryo, initial development of,
control (enkrateia), 27,4-8 36,19-37,3; active solely in virtue of
cooking (pepsis), 33,13-15n; 33,18-19 the soul for nourishing oneself, as
other powers are not yet in activity,
240 Subject Index
36,26-37,3; an animal only in power, 45,2; 46,1-3; produces light, 43,8-9;
38,4-8n; Galens views of, 38,4-8n; 46,1-4; responsible for differences in
part of the mother and not an animal colour, 46,4-6; responsible for other
on its own, 14i; 34,4-8; Platonist things being seen, 46,1-3; visible in
views of, 38,4-8n; Stoic views of, the highest degree, 46,1-3; water is
38,4-8n nourishment for fire only in a broader
emergence (gennasthai epi), 9-12i; sense, 34,11-12
5,5-6n; 10,17-19n; 24,18-20n; flavour, 40,24
24,22-3.28: 25,2-9n; 26,26-30; force, 15,17-21; 21,13-16
36,19-21n; see also non-reductive foreknowledge, 1,7-8
materialism form (eidos), 3,1-21; always a form of
Empedocles, 27,15-28,2n; 42,19-22n something, from which it is
en (in), senses of, 13,10-14,5 inseparable, 6i; 17,12-13; cannot
end (telos), 24,14-17; cannot form an undergo change on its own, 6-7i;
infinite series, 28,8-13n; everything 17,8-11; compounded with matter, see
produced naturally is for the end in under body; contributes to the
it, 24,16-17; everything should be supervening form of the compound,
named after the end and most 10-12i; 7,24-8,13 (esp. 7,24-8,1n and
advanced part, 36,13-15; nature does 8,10-12n); completion, 17,12;
everything for something, 28,8 culmination, 6,29-7,3; 9,18-26; 16,4-5;
Epictetus, 1,4-2,4n 17,12-13; of forms, 8,12; diversity of
Epicurus (and Epicureans): on soul as form due to mixture and blend of
composed from bodies, 26,17-18; on underlying matter, 11-12i; 8,13-25;
soul as juxtaposed with body, 11,15n; 9,4-11; 10,14-26; does not undergo
11,15-17n change, 5,11-13; emerges from
error: and common perceptibles, matter, see under emergence;
41,10-13; 70,10-12 (cf. 20-23); and enmattered, 16,2; highest
exclusive perceptibles, 16i; (kuritaton) form located where what
41,13-42,3; and representations, it is to be an animal is, 39,23-40,1;
70,5-71,5; impossibility of error with inseparable from matter, 6i; 4,20-5,1;
exclusive perceptibles, 16i; 6,1; 15,11-13; 17,8-11; 21,22-3; in
41,13-42,3n matter, 13,20-14,3; vs matter,
essence: of soul, 1,3; 16,18; 22,26-7; of 2,25-3,21; 5,4-8; 31,10-25; more
living thing, 35,15 advanced or complex (teleioteron),
eunuchs, 32,12 7,23n; 7,21-8,13; 10,12-17;
exclusive perceptibles (idia aisthta), see 10,26-11,13; not a body, 5,22; 7,9-13;
under perceptibles 15,11-13; 17,8-11; 17,15-19,20; not
existing on its own (or not), matter or composed from matter,
4,8-9.17.20-5; 5,1; 17,11-14; 18,2-7.17; 17,15-18,10; perishes with body,
19,7.18-20; cf. 5,21-2 21,23-4; power, not activity, 9,14-26;
extrinsic characteristic (sumbebkos), quality of a sort, 17,17-18; a
13,20-1; 14,24-15,5 substance, 6,2-3; supervenes on
extrinsic perceptibles (kata sumbebkos), underlying matter, see under
see under perceptibles supervenience; that in virtue of
eye: affected by the illuminated medium, which a body acts or is modified, 6-7i;
43,14-15; capable of receiving colour, 7,9-13; 31,10-13; that in virtue of
44,2-3; composed of water, 44,5-9; which each thing has its being and
interior of (kor), 44,7n; portion that differences, 6,24-7,8; 10,7-8; 14,21-2;
sees lacks colour, 44,3-4; smooth and 15,24-6.28-9; 24,13; 31,10-25; see also
transparent, 44,1-2 under matter under that in virtue of
which something is what it is and
familiar to us vs by nature, see under engages in activities; that which is
intelligible received, 31,15.17.23-5; undergoes
fear, 12,17; 13,1, 23,15; involves bodily change extrinsically, 22,14-20; see
changes, 12,17 also under soul
feet, 24,11 fourfold drug, 10i; 24,23-25,2n
fire, 5,4-18; 11,7; 16,4; 19,21.27.35; 35,8; frogs, 27,15-17
Subject Index 241
fulfilment (eudaimonia), 1,9n heat: not a source of orderly change,
functionalism, 9i (esp. n. 21) 35,5-6; the souls tool (organon) in
fusion (sunkhusis), on Stoic view, nourishment, 35,2-8; as a tangible,
11,18-20n; 15,5-9n 40,24
heaviness (baruts), 9,16-20; 14,2-3;
Galen: on body as an organ or 20,21; 22,7-10.18-20; 23,29-24,1; see
instrument of the soul, 23,26-8n; on also lightness
breath as a messenger, 41,1-3n; on Heraclitus, 35,3-5n
common perceptibles, 41,1-3n; on Hierocles, 15,15-17n
embryo, 38,4-8n; on the fourfold drug Hippasus of Metapontum, 35,3-5n
(tetrapharmakos), 24,23-25,2n; on Hippocratic authors, 34,16-17n
illumination as instantaneous, homonymy, 10i; 12,2-3n; 15,2-5n;
42,22-43,4n; on nourishment, 16,12-18n; 17,11-12n; 18,17n;
34,16-17n; on responsiveness 21,1-10n; 28,28-9; 29,3-10n
(sumpatheia), 15,15-17n; on soul as a honey water, 11,17-18
blend of bodies, 11,17-18n; 25,2-4n; honey wine, 13,22; 15,7; 26,5
on soul vs nature, 31,25-32,1n; on the house, 11,17
tingeing of the medium, 42,11-19n human, 30,10-12; rational, 36,15
genus, 13,12-15; 14,11-17; 19,18-20; humour, 13,5-8
common and universal, 14,13-14 hupokeimenon: as corresponding object,
glare or sheen (aug), 44,13-14n 41,10-13n; as underlying subject,
glow-worm, 42,4-5n 2,26-3,8
god(s), 1,4-8; 28,26-8; Apollo (= Pythian hylomorphism, 5-9i; 12,2-3n
god), 1,4-8; celestial spheres,
28,25-9n; likeness to the divine, illumination: all at once throughout the
32,13-14; Prime mover, 28,25-9n medium, 42,22-43,4n; alteration only
governing part of soul (hgemonikon), in a way (ps), 42,19-22n;
15,15-17n; 15,22-6n; 24,5-6; 39,22; in instantaneous (akhrons), 17i; 43,11;
the heart, 39,21-2; Stoic view, mere Cambridge change, 17i;
15,15-17n; 24,5n; that which is 42,22-43,4n; presence of/relation to a
initially or pre-eminently animate in source of light, 42,22-43,4; 43,10-11;
its own right, 24,5-6 45,1-5; 46,1-3; see also light
growing, 14i; 9,3.13; 29,3.10; 32,9; immortality, 14i; 21,24n; 22,24-23,6n;
34,1-14; 35,8-17.24; 36,20-6; vs 32,11-15n; 36,7-9; reinvigorated
alteration, 33,28-34,8n; by means of immortality, 36,8
nourishment, 35,8-9; involves bodily incorporeal, see under non-bodily items
changes, 12,12-13; 34,6-8; does not insects, 37,4-5n
extend throughout life, 35,12-13; not instantaneous (akhrons) events, 43,11n
the same as nourishing oneself, instrumentalist conception of soul/body
35,9-17; only in something animate, relation, 23,26-8n
34,8-14; purpose is to increase size of intelligible (gnrimteron), more to us vs
the living thing, 35,13-16 by nature, 32,23-4n
interpenetration of bodies, 14,8-9
hand, 24,11 intrinsic perceptibles (kath hauto), see
harmony, 11i; see esp. 24,18-20n; under perceptibles
25,9-15n; 26,3-7n; and also
24,18-26,30 more generally; not a juxtaposition (parathesis), 11,15n;
sound, 25,23-26,3 11,15-17n; 12,2-4
hating, 23,16
health, 25,4-7; 31,15-20; a balance and kath hauta perceptibles, see under
mixture, 25,4-7 perceptibles under perceptible on
heart, 24,5n; hot and moist, 40,1-2; as their own
seat of governing part, 39,21-2; as kinsis, see under change
seat of soul for perception, 15i; kor, not the pupil, but the interior of
39,21-40,3; as seat of vital activities, the eye, 44,7n; see under eye under
12,2-3n; origin and well-spring of interior of
blood and of breath, 40,2-3 Know Thyself, 1,6-2,1; 1,4-2,4n
242 Subject Index
learning, 23,16 form only if it is the matter of a
life: definition, 29,10; due to both body compound body, not a simple body,
and soul, 31,20-5; due to a soul and 4,7-25; cf. 18,4; vs form, see under
not just a nature, 31,25-32,1; in form; in the fundamental sense
accordance with nature, 1,9; (kuris), see under prime matter
homonymous, 16,12-18n; 29,3-10n; as below; prime matter, 5i; 3,28-4,4n;
a power, 16,12-14; see also under alive 3,28-4,20; 5,20-2; 7,15-17; 8,17-18;
light: activity of transparent, in so far as 17,17-18n; proximate matter, 4,23n;
it is transparent, 17i; 43,7-8; 44,15; simply as such (hapls), 4,4n; see also
colour (or like a colour) of prime matter above; Stoic view,
transparent medium, 17i; 44,16; 45,1 3,28-4,4n; 17,15-18,10; a substance,
(though compare 45,17-19); 6,2-3; a sufficient condition for the
completion of the transparent, form that supervenes upon it, 10-11i;
43,7-8n; contrast with colour, 45,5-7; 10,24-6n; that in virtue of which
culmination of transparent, 43,6-8; something is what it is and engages
does not receive colour in virtue of an in activities, 6,21-3; 31,10-14. See
effluence, 42,19-22; does not receive also under form under that in virtue
colour in the way that matter does, of which each thing has its being and
42,19-22; form of transparent, 43,6-7; differences
illuminated or tinged all at once, melody, 26,6
42,22-43,4; instantaneous (akhrons), memory, 23,17
17i; 43,11; not an alteration or mental causation, 11i; 24,18-20n
modification, 42,19-22n; not a body, metempsychosis, 9i; 27,15-28,2
42,19-22n; 43,11; not an effluence, messengers (diakonoi), see under
42,19-21; not a colour in the transmission and transmitters
fundamental sense, 45,17-19; parallel mirrors, 43,3
to colour, 44,17-19; presence of fire or mixture (mixis), 11-12i; 8,16n; 9,6;
divine body in transparent medium, 10,18; 19,26-20,6; 24,3-4; 24,27-25,2;
43,2.8-9; 45,1-3; received by 26,7-9.21-2.26-7; powers of
transparent material without a blend/mixture in relation to powers of
definite shape, 44,23-4; relational, ingredients, 11-12i; 19,32-3; 20,1-6;
42,22-43,4; 43,10-11; 45,3-5; see also blend and diversity of form
responsible for other things being due to mixture and blend of
seen, 44,13-14.24-5; 46,3-4; Stoic underlying matter under form
view, 42,19-22n; in transparent modification (pathos), 12,9-13,8n;
material, as colour is in coloured 39,18.21; 41,13; 44,8; exclusive
things, 44,16-18; visible in the modification, 43,20; see also affect
highest degree, 44,13-14; see also modified, being (paskhein), 7,10-14; 13,2;
illumination 16,15; 39,17.20-1; 41,9.11; 42,18;
lightness (kouphots), 5,4-18; 11,7; 43,13.20; in virtue of form, 7,9-13
20,22; see also heaviness moist 3,18.19; 8,19; 19,35; 26,8; as a
likeness to the divine, 32,13-14 tangible, 40,25
likening (homoisis, homoiousthai), mortality, 1,4-2,4n; 21,24n
38,21-39,2; 39,1.10-18; 40,18-19; motion: forced motion, 15,17-21;
44,2-3 21,13-16; simple motion, 7,14-20;
limits inseparable from body, 17,12-15 spatial motion, 22,7.18-20; 23,29-24,1
line of sight, see under sight multiple realisation, 21i n.15; 26,7-13n
literalism vs spiritualism, 15i;
41,10-13n; 42,11-19n; 44,2-3n natural tendencies (phusikai rhopai),
logos, see under characterisation 12,10-11
loss of control (akrasia), 27,4-8 nature: everything formed naturally
loving, 23,16 does everything to have a share in
immortality and a likeness to the
man, see under human divine, 32,14-15; everything produced
matter (hul), 2,25-4,20; can exist naturally is for the end in it,
separate from form only in thought, 24,16-17; 28,8; in accordance with
5i; 6,17-19; can exist separate from nature, 1,9; 2,2; 5,3; 7,5; 8,4; analogy
Subject Index 243
with art, 3,19-21; contrary to nature, pain, 23,15
21,14-15; does everything for Panaetius, 32,1-4n
something, 28,8; does not imitate art, parts: of the body, 16,11; 24,6-11; 28,6-8;
but vice versa, 3,15-16; does nothing detached, see under dismemberment
superfluous (matn), 12,21-4n; 27,9n; and survival; and division of body,
28,4-6; incredible magnificence of, 18,10-27; in whole or vice versa,
2,11-25; 11,5-13 13,16-18; 14,5-11; preserved in the
Neoplatonism, 44,13-14n whole, 11,15-16; of the soul, 12-13i;
Nicolaus Damascenus, 31,2-6n 19,1-3; 27,1-4n; 28,6-8; 30,5;
non-bodily items (asmata), 4,18 (cf. 30,26-31,6
5,18-19); Platonic view, 7,9-13n; that paskhein, see under modified, being
in virtue of which bodies act or are passion (thumos), 12,15
modified, 7,9-13 perceptibles, 15-16i; 24,11; 40,20-42,3;
non-reductive materialism, 12i; common perceptibles (koina), 16i;
10,24-6n; 25,2-9n; see also emergence 41,1-5.10-13; 65,10-21; 83,17-22;
nourishing oneself, 14i; 9,2.12; 29,1-10; exclusive perceptibles (idia), 16i;
32,9.20-3; 32,26-33,5; 34,26-35,2; 40,21-6; 43,18-44,1n; external
35,24; by contraries, 33,16-34,1; because perception is directed at
39,15-16; by likes, 34,14-18; obtaining or avoiding externals,
continuous throughout life, 35,10-11; 39,9-10; extreme or intense,
heat is the souls tool in nourishment, 44,13-14n; what is extrinsically
35,2-8; in broader sense, water perceptible (kata sumbebkos),
nourishes fire, 34,11-12; in 15-16i; 41,6-10; what is intrinsically
fundamental sense, only animate perceptible (kath hauto), 40,20-41,5
things are nourished, 14i; perception: activity vs power, 39,5-8;
34,12-14.18-26; involves bodily activity clearer than power, 32,26;
changes, 12,12-13; not all cognition of perceptibles, 39,1-2; due
transformation into a contrary is to a kind of alteration, 15i;
nourishment, 34,1-4; nothing 39,1-2.14-15; due to likening
inanimate is nourished, 34,13; occurs (homoisis), becoming like
in a certain sequence and order, perceptibles, 39,1-2.10-14; 40,18-19;
35,6-8; power for, see under power due to likes in one way, unlikes in
under for nourishing; purpose is to another, 39,14-18; due to receiving
preserve the being and essence of a perceptible forms without the matter,
living thing, 35,13-16; three things 39,13-14; due to a modification in the
involved: that which nourishes, that sense organs that have the power of
which is nourished, that by which it perception, 14-15i; 39,18-20; due to
is nourished, 36,9-13; what is power of perception, 29,7; five parts,
nourished is the animate thing, 40,4-5.10-11; instantaneous
34,18-26 (akhrons), 43,11n; involves bodily
nourishment, 33,5-34,26; is a body, changes, 12,13-14; literalist account,
34,6-8; cooked vs uncooked, 33,13-16; see literalism vs spiritualism; natural
only contraries which cause growth conditions of, 16i; 41,13-42,3n; not an
by their transformation are alteration, though involves them, 15i;
nourishment, 34,5-14; the part of 38,21-39,2n; not a mere Cambridge
nourishment that nourishes, change, 38,21-39,2n; for obtaining or
34,16-17; seed as final stage of, 36,3-5 avoiding external objects, 14i;
39,9-10; as a power, see under power
oboe (aulos) and oboe-playing, 23,20 under for perceiving; as relative to its
odour, 40,23 objects, 38,21-39,2n; parts of,
old age, 23,4 40,4-5.10-11; requires presence of
organ, 11,1n; 24,6-11 perceptibles, 39,4-6; spiritualist
organic body, see under body account, see under literalist vs
origin and principle (arkh), 2,23n; spiritualist accounts of perception;
7,18-19; 9,1-3 Stoic view, 41,13-42,3n; for survival,
ousia, 1,3n 39,9-10n; through the body, 12,13-14;
244 Subject Index
useful, 39,9-10; for well-being, being aware (antilptik), 39,5; for
39,9-10n believing, 29,24; for having cognition
perceptual error, see under error (kritik), 39,5; conflict of, 27,7-8;
perceptual organs, 24,10-11; 39,18-20; corresponds with organs, 27,8-28,2;
lack qualities they are sensitive to, 28,6-8; for deliberating, 29,24; for
44,3-4n; relaying (diadidonai, desiring, 29,11.16-17; does not
diadosis) perceptibles to central undergo change or engage in activity
sense, 15i; 39,18-21; 41,3-5 itself, but is that in virtue of which
perceptual system, 39,18-21n; see also change or activity occurs, 5,15-18;
perceptual organs and relaying 22,2-10; 23,24-24,4; emerging from a
perceptible forms blend, 24,22-3.28; first power,
Philoponus, 15,9-10n 9,11-12; 29,1-3; 38,12; see also first
phosphorescent objects, 42,4-5n; 43,16-18 soul under soul; for growing, 29,1-3;
pilot of a ship, 7-8i; 14,4-5; 15,9-26; 36,1-2; 38,15; hierarchy of powers,
20,26-21,21 28,8-13.20-5; 29,1-30,6; see also
place (topos), 13,19-20; 14,17-24; hierarchy of souls under soul; for
boundary of what contains the knowing, 29,24; not a harmony,
occupant, 14,20; only bodies are in 24,27-9; not an infinite number in
place, 14,18. soul, 28,3-13; not a proportion, 25,1;
plants, 8,25-9,26; 29,6; 30,9.16; 32,18.20; not separable from what has it and
do not need many organs, 37,7; have cannot exist on its own, 5,14-15;
a soul, 9,12-14; have only the soul for 22,4-6; for moving in place, 29,18-22;
nourishing oneself throughout the 30,13-14; for nourishing oneself,
body, 37,5-6; natural, compound 29,1-10.12-15.21; 32,20-3; 35,26-7;
body, 8,25-6; parts of plant can 36,19; 38,14.18; for nourishing
survive dismemberment, 37,4-11; oneself requires nourishment to be
roots are easily grafted, 37,18-38,2; active, 39,2-3; for perceiving, 19,2;
Stoic view, 11,14n 29,7.11-22; 38,18; for perceiving
Plato: on control and lack of control, requires perceptibles to be active,
27,4-8n; on the eyes, 44,6-9n; on 39,4-5; for perceiving is that in virtue
having a share of immortality, of which animals are animals,
32,11-15n; on metempsychosis, 38,18-19; plurality of powers of soul,
27,15-28,2n; on non-bodily items and 27,1-28,2; prior to activity, 9,20-6;
causation, 3-4i; 11i; 7,9-13n; Phaedo, 32,30-33,5; rational powers,
2,17-18n; 7,9-13n; 24,18-20n; 29,22-30,6; for representing, 29,12;
Republic, 13i; 27,4-8; on for reproducing, 9,5; 29,1-3; 32,8-23;
self-knowledge, 1,4-2,4n; on 36,2-9; 38,15; for seeing, 44,10-13;
separability of Form, 17,12-15n; on separability or inseparability of one
soul as composition, 26,18-20; on soul power from another, 28,23-5;
as undergoing change, 22,5-12n; 29,12-22; 29,26-30,6; 38,16-18; for
22,24-23,6n; Timaeus, 42,19-22n; striving, 29,11.18; of soul, 12,13;
26,18-20; on tripartition of soul, 16,11; for understanding, 29,24
27,4-8n presence (parousia): light as presence of
Platonists: on embryos, 38,4-8n; on fire or divine body in transparent
non-bodily items and causation, medium, 43,2.9; 42,22-43,4n; 45,1-3
7,9-13n; on soul using body as an preservation, of the being and essence of
organ, 23,26-8n; on space, 14,19-20n; a living thing is the purpose of the
on substance dualism, 11,15n power for nourishing, 35,13-16
pleasure, 23,15 prime matter, see under matter
Plotinus, 14,18n; 15,9-10n prior: in account/definition, 32,25-7n;
Plutarch, 23,26-8n 32,27-33,1
poiein, see under acting proportion, 10,12-14; 24,25-25,2; 25,21-3;
Polemon, 1,9n 26,3-13
Porphyry, 15,9-10n psychological conflict, 27,4-8
Posidonius, 3,28-4,4n Pythagoras, 9i; 27,15-28,2n
power (dunamis): vs activity, 9,20-6; Pythian god (= Apollo), 1,6-8
16,8-10.16-18; 32,25-7; 32,30-33,5; for
Subject Index 245
qualities, 6-7i; 12,3-4n; 17,15-17n; common sense; as parts of the soul,
17,17-18; not bodies, 17,15-17n 19,1-3; vs sense organs, 19,1-3; serial
order of, 40,5-6.11-16; as transmitters
rage, see under anger (diakonoi), 41,1-3n
rational: animals, 30,6-12; human is separate, separable (khrizein,
called rational, 36,15-16; non-rational khristos): in thought alone, 4,9-10;
living things, 30,12-17; powers, 6,17-20; from body, 12,21-4;
29,22-30,6 separability or inseparability of one
receiving forms: colours, 44,2-3; without power from another, 28,23-5;
matter, 39,13-14; what receives 29,12-22; 29,26-30,6; 38,16-18;
colours must be without a colour of separability or inseparability of one
its own, 44,3-4 sense from another, 40,11-15;
recollecting, 23,16 taxonomic separability, 28,23-5n
reductionism, 11i; 10,24-6n; 25,2-9n serial order: of senses, 40,5-6.11-16; of
reflected image (emphasis): in eyes, souls, 13i; 16,18-17,1; 28,20-5
44,1-2 sight: instantaneous (akhrons), 43,11n;
relation (skhesis): illumination/tingeing line of sight, 43,18-44,1; not by
of medium occurs in virtue of a effluence, 42,19-22; occurs because
relation, 42,22-43,4; 43,10-11; 45,3-5; colour affects the illuminated
see also Cambridge changes (mere) medium, which in turn affects the
relaying perceptible forms, 15i; 39,18-21; eye, 43,12-15; power of, 44,10-13;
41,3-5; 42,18-19; 43,18-44,1; see also occurs because the organ receives
transmitters and transmitting colour and is likened to it, 44,2-3;
representation (phantasia), 12,17n; organ of sight is made of moist,
29,12; depends on perception, 12,18 transparent material, without a
reproducing, 14i; 9,4; 32,8-23; shape of its own, 44,5-6
35,17-23.25; end of every natural simple body, see under element
thing, 35,22-3 (cf. 32,11-16); involves Simplicius, 15,9-10n
bodily changes, 12,12-13; living simply as such (hapls), 4,4n
things which are not maimed capable sinews, 24,9
of, 32,11-12.18-19; 35,19-20; not for smooth: as a tangible, 40,25
preservation or development of what sobriety, 23,4
possesses it, 36,5-6; power to some specific thing (tode ti), 4,24n;
reproduce is the last to develop, 15,2-5n
35,20-2 (cf. 36,7); for producing Sosigenes, 42,4-5n; 43,16-18n
another thing like its possessor, 36,7; soul (psukh), 9,12; activities are
provides a share in immortality and inseparable from bodily change,
likeness to the divine, 32,13; 36,7-9 12,9-13,8; of a body subject to
resistance, 21,15 coming-to-be and perishing, 1,2-3;
responsible factor (aition, aitia), 5,16; 28,25-6; as breath, 19,6-20; cannot
8,15.17.20; 9,10; 10,17.21.23; 13,1; undergo change on its own, 6-7i;
15,2.20 17,10; 21,22-24,17; disposition,
responsiveness (sumpatheia), 15,15-17 26,21-2; divisible into parts or
rest, 22,23-23,6; see also change powers, 31,2-6; does not exit and
roots, 37,18-38,2 re-enter the body, 15,17-21; does not
rough: as a tangible, 40,25 inhere in a subject like an extrinsic
rhythm, 26,6 characteristic, 10i; 14,24-15,5; does
Rylean passage, 7i; 23,6-24n not move the body by force, 15,17-21;
emerges from blend, 10-12i; 24,22-3;
scala naturae, 30,6-17n 26,26-30; end and culmination,
seed, 32,1-6; 37,9-11.17-18; 38,3-4; final 24,14-15; essence of, 1,3; 1,2-4n;
stage of nourishment, 36,3-5 16,18; 26,28-30; essence comes from
self-knowledge, 1,6-2,1; Platos view, the power that emerges from a blend,
1,4-2,4n; Stoic view, 1,4-2,4n not the blend, 26,28-30; extends
Seneca, 1,4-2,4n throughout entire body, 12,2-3;
sense organs, 19,1-3 14,7-9; 15,15-17; 37,6; but note
senses: common sense, see under 37,11-18; first soul, 29,1-10; 35,24;
246 Subject Index
36,11.16-17; see also first power Stoic view, 15,15-17n; substance,
under power; first completion, 16,7; 14,25-15,1; supervenes on mixture
21,5; first culmination, 9,25-6; form and blend of underlying bodies, 9-11i;
(of the body), 9,14; 10,10-12; 11,1-2; 24,19-20 (cf. 3-4.22-3); 25,2-3.7-9;
11,14-17,8; 17,9; 21,3-5.22; 24,2-3; 26,21-2.26-7; that in virtue of which
26,21-2; 31,10-25; 38,13-14; form of the living body has its being and
an organic body, 11,1-2; form of differs from others, 14,21-2;
mixture of elements, 26,7-9; of the 15,3-5.28-9; that in virtue of which
gods, 28,26-8; for growing, 36,18; the animal does things as such,
hierarchy of souls, 16,19-17,1; 23,6-24; 24,13; that in virtue of which
17,1-8n; see also hierarchy of powers we live, 9,12-14; third or last soul,
under power; homogeneous in plants, 29,22-30,6; undergoes change
but not in animals, 37,13; in the extrinsically, 7i; 22,13-23; for
body, 13,9-10; inseparable from the vegetation, 35,24; 38,13
body, 12,21-4; 17,9-10; 17,11-15; sound, 25,23-26,3; 40,23
21,22-3; for life, 38,12; cf. 29,1-10; space (khra), 14,19-20
more advanced (teleion), 16,14; species (eidos), 13,12-15; 14,11-17;
nature, 28,10; vs (mere) nature, common and universal, 14,13-14
31,25-32,1; not a balance, 25,7; not a Speusippus, 1,9n
blend, 15,5-9; 24,21-2; not a body, spiritualism, see under literalism vs
14,14; 15,9.14; 17,10; 17,15-20,26; not spiritualism
divisible in the way extension or statue, 3,7-9; 11,20-1; 14,1-2; 16,3;
number is, 30,26-31,2; not an 18,17-25
extended thing, 30,26-8; not an Stoics: on accompanying awareness
extrinsic characteristic of body, 10i; (sunaisthsis), 15,15-17n; on bodies
14,24-15,5; 19,14-16; not a harmony, and causal powers, 7,9-13n; on bodies
24,18-26,30; not part of body, 14,5-9; and matter, 17,15-18,10; 19,3-5n; on
not the proportion or composition of body passing through body, 14,8-9n;
the blend, 25,3-4; not a substance 20,8-19n; on blending (krasis),
itself on its own (ousia aut kath 11,18-20n; 12,5-6n; 15,5-9n; 20,8-19n;
hautn), 12,7-13,8; not a whole of on breath (pneuma), 19,6-9n; 19,21n;
which body is part, 14,10-11; for 26,13-17; deny plants have a soul,
nourishing oneself, 36,18-37,3; for 11,14n; on embryos, 38,4-8n; first
nourishing oneself extends inference schema, 22,24-23,6n; on
throughout the body of plants, but fourfold drug, 24,23-25,2n; on fusion
not animals, 37,4-18; for nourishing (sunkhusis), 11,18-20n; 15,5-9n; on
oneself is sole power active in the governing part of the soul
foetus, 36,26-37,3; one in number, (hgemonikon), 15,15-17n; 24,5n; on
14,13; only admits of a general life in accordance with nature, 1,9n;
characterisation, 13i; 17,1-8; an on light, 42,19-22n; on matter,
origin and principle of vital activities, 3,28-4,4n; on oikeisis, 15,15-17n; on
24,11-12; parts of the soul, 19,1-3; parts of the soul as qualities, 31,2-6n;
27,1-4n; for perceiving, 10,9; 38,16; on perception, 18,25-19,1n; 40,4-5n;
perishes with body, 21,23-4; a power 41,13-42,3n; on perceptual error and
that emerges from a blend, 10-12i; the conditions necessary for veridical
24,22-3 (cf. 24,28); 26,21-2.26-7; perception, 41,13-42,3n; on
powers of the soul, 16,11-16; principles, 17,12-15n; on post-mortem
27,1-28,2; present already in sown survival, 17,12-15n; on powers of the
seed, 36,22-3; principle of life, 9,12; soul, 27,6-7n; 27,8-15n; on qualities,
principle of ordered development and 17,15-17n; on responsiveness
limit, 35,6-8; receives contraries, (sumpatheia), 15,15-17n; on the scala
15,1; relation to body, 21,16-21; for naturae, 27,6-7n; 27,8-15n; on
reproducing, 36,17; responsible for all self-knowledge, 1,4-2,4n; on the
the changes in the animal as an senses as messengers, 41,1-3n; on
animal, 22,10-12; 24,11-12; 32,16; soul as blended with body, 11,15n;
second soul, 29,11-22; 38,16; serial 11,17-18n; 15,15-17n; on soul as a
order of, 13i; 16,18-17,1; 28,20-5; body, 15,15-17n; 18,25-19,1n;
Subject Index 247
30,26-31,2n; on soul as breath (akhrons) for visibles, not for other
(pneuma) or breath qualified in a perceptibles, 43,11n; senses, 41,3; see
certain way, 19,6-9n; 19,9-16n; also relaying perceptible forms
19,18-20n; 26,13-17; on souls relation transparent (diaphanes), 17i; 42,7-43,18;
to body, 17,12-15n; on soul vs nature, 44,4-5; 44,13-45,5; actively,
31,25-32,1n; 32,1-4n; on space, 42,5-7.9-10; 43,4-5; all bodies have a
14,19-20n; on unity of soul, 27,6-7n larger or smaller share of
Strato, 24,5n transparency, 44,20-1; can be seen
striving (horm), 12,19n; 29,11; through (diopton), 17i; 44,13-14n;
29,18-22n; involves bodily changes, common nature/power, 44,20-1n;
12,19; origin of term, 29,18-22n differences in transparency due to
substance (ousia), 5,1-4; bodily, presence of earth, 45,7-10; does not
perceptible substance compounded of receive light in virtue of an effluence,
matter and form, 2,27; itself on its 42,19-22; does not receive light in the
own (aut kath hautn), 12,8; way that matter does, 42,19-22;
natural, 5,3-4; parts of substance are etymology, 43,5-8n; 44,23-5; every
substances, 6,3-4 body is transparent, 44,20-1; matter
substance dualism, 3i; 8i; 11,15n of colour, 44,21-3; medium coloured
sufficient conditions for supervening by objects, 42,11-19; medium does not
form, see under matter have a colour of its own, 46,7-9;
sumbebkos, see under extrinsic medium has extrinsic colour, 44,16n;
characteristic 46,10-11; medium visible through an
sumpatheia, see under responsiveness extraneous colour, 44,13-14n; 44,16n;
sunaisthsis, see under accompanying 44,25; 46,10-11; proximate matter of
awareness colour, 44,21-3; receives colour,
supervenience (epigignesthai), 9-11i; see 44,20.22-3; receives light, 44,23-4;
esp. 25,2n; also 5,5-6n; 10,17-19n; solids that can be seen through,
10,24-6n; 26,7-13n; 36,19-21n; see 42,7-8n; solids that have a persistent
also emergence, multiple realisation, colour of their own present inside,
and non-reductive materialism 45,5-7; tinged by colours of other
things, 42,11-19; 44,25; with a
tangible, 40,24-5 definite shape has a colour of its own,
taste: a form of touch, 40,12 44,18-23; 46,12-14; without a definite
temple, 25,18-19 shape or a boundary of its own also
tendencies, natural, see under natural lacks a colour of its own, 46,7-11;
tendencies without a definite shape has light,
Themistius, 15,9-10n 44,4-5; 44,23-45,1; without a definite
Theophrastus, 24,5n shape is and is said to be most
thinking (phronein): cannot occur transparent, 44,23-4
without representation, 12,20-1n;
involves bodily changes, 12,20 underlying subject (hupokeimenon),
tingeing, 42,11-19; 43,18-20; alteration 3,23-4,20; 7,14-8,8; see also matter
only in a way (ps), 42,19-22n understanding (noein): most proper
tode ti, see under some specific thing function of the soul, 23,2-3; occurs
tortoise, 37,4-5n when soul is at rest, 23,3; see also
touch: necessary to animals, 40,14-15; thinking
separated from other senses, 40,11-15 understanding (nous): immmortal,
traces (enkataleimmata), 38,10 21,24n; lacks a dedicated bodily
transformation (metabol): always into a organ, 21,24n; 28,6-8n; 44,3-4n
contrary, 33,19-26 unity of soul and body, 21,16-21
transmigration, see under universal teleology, 24,16-17n
metempsychosis universe (kosmos), incapable of coming
transmission (diakoneisthai) and to be or perishing, 6,14-15
transmitters (diakonoi), 15i; 41,1-3n;
cf. 39,18-21n; external medium, vessel, 13,18-20; 14,23-4
41,20; 42,9; 52,20; 59,14; 62,5.12;
94,28; 95,1; 100,16; instantaneous water, 3,17; 6,24; 8,21; 11,18; 42,7; can
248 Subject Index
retain modification due to colour 11,15-16; in parts or vice versa,
because of its density and 13,16-18
consistency, 44,7-9; nourishment for womb, 36,26-37,3; 38,4-8
fire only in a broader sense, 34,11-12; wrestling, 23,18-19
transparent, moist, and without a
definite shape, 44,4-5 Xenarchus, 21,13-16n
wax, 4,9-15 Xenocrates, 1,9n; 22,5-12n
weaving, 23,21 Xenophanes, 27,15-28,2n
white or bright (leukos), 6,9; 13,21;
33,20.23; 34,3; 41,8 youth, 23,4
wine, 13,20
whole: composed of parts present in it, Zeno of Citium, 3,28-4,4n

S-ar putea să vă placă și