Aristotle on How Animals Move: The de Incessu Animalium: Text, Translation, and Interpretative Essays 1108491332, 9781108491334

Table of contents :
01.0_pp_i_iv_Frontmatter
02.0_pp_v_vi_Contents
03.0_pp_vii_vii_List_of_Figures
04.0_pp_viii_viii_List_of_Tables
05.0_pp_ix_xii_Notes_on_Contributors
06.0_pp_xiii_xiii_Preface
07.0_pp_xiv_xvi_List_of_Abbreviations
08.0_pp_1_2_Introduction
08.1_pp_3_18_Explanatory_Strategies_in_the_De_incessu_animalium
08.2_pp_19_32_The_Reception_of_the_De_incessu_animalium
09.0_pp_33_34_Greek_Text_and_Translation
09.1_pp_35_40_Preface_to_the_Greek_Text
09.2_pp_41_41_Sigla_Manuscriptorum
09.3_pp_42_98__
10.0_pp_99_100_Interpretative_Essays
10.1_pp_101_116_De_incessu_animalium_13_The_Theoretical_Framework_and_the_Beginning_of_the_Actual_Investigation
10.2_pp_117_140_De_incessu_animalium_4_Aristotles_Conception_of_Dimension
10.3_pp_141_161_De_incessu_animalium_56_The_Architecture_of_Locomotive_Bodies
10.4_pp_162_193_De_incessu_animalium_78_Number_and_Distribution_of_Feet_in_Animal_Progression
10.5_pp_194_216_De_incessu_animalium_9_Aristotles_Mathematical_Kinesiology_The_Case_of_Bending
10.6_pp_217_232_De_incessu_animalium_1011_Flight_and_Two-Footedness
10.7_pp_233_265_De_incessu_animalium_1213_Limb-Bending_and_Natural_Teleology
10.8_pp_266_281_De_incessu_animalium_1415_Teleology_Across_Kinds
10.9_pp_282_296_De_incessu_animalium_1619_The_Motion_of_Many-Footed_Animals_and_Cases_of_Peculiar_Motion_in_Water
11.0_pp_297_306_References
12.0_pp_307_309_General_Index
13.0_pp_309_315_Index_Locorum

Citation preview

ARISTOTLE ON HOW ANIMALS MOVE

The De incessu animalium forms an integral part of Aristotle’s biological corpus but is one of the least studied Aristotelian works both by ancient and modern interpreters. Yet it is a treatise where we can see, with some clarity and detail, Aristotle’s methodology at work. This volume contains a new critical edition of the Greek text, an English translation, and nine in-depth interpretative essays. A general introduction that focuses on the explanatory strategies adopted by Aristotle in the De incessu animalium plus a historical essay on the reception of this work in antiquity and beyond open the volume. No other work of this kind has been published in any modern language. andrea falcon is affiliated with Concordia University, (Montréal, Canada) and University of Milan, La Statale (Italy). He is the author of several books on Aristotle and the Aristotelian tradition, including Aristotle and the Science of Nature: Unity without Uniformity (Cambridge, 2005), Aristotelianism in the First Century BC: Xenarchus of Seleucia (Cambridge, 2011) and, coedited with David Lefebvre, Aristotle’s Generation of Animals: A Critical Guide (Cambridge, 2017). stasinos stavrianeas is an Assistant Professor in the Department of Philosophy at the University of Patras. He specializes in Aristotle’s natural philosophy, biology and metaphysics and is the author of a Modern Greek translation and commentary of Aristotle’s Parts of Animals (2021) and is currently preparing a similar edition of Aristotle’s Generation of Animals.

ARISTOTLE ON HOW ANIMALS MOVE The De incessu animalium: Text, Translation, and Interpretative Essays edited by ANDREA FALCON Concordia University (Montréal, Canada) and University of Milan (Italy)

STASI NOS S TAVRI AN E AS University of Patras, Greece

greek text prepared by PANTELI S GOLI TSIS Aristotle University, Thessaloniki

University Printing House, Cambridge CB2 8BS, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06–04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108491334 DOI: 10.1017/9781108868228 © Cambridge University Press 2021 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2021 Printed in the United Kingdom by TJ Books Limited, Padstow Cornwall A catalogue record for this publication is available from the British Library. ISBN 978-1-108-49133-4 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents

List of Figures List of Tables Notes on Contributors Preface (Andrea Falcon and Stasinos Stavrianeas) List of Abbreviations

page vii viii ix xiii xiv

part i  introduction

1

1 Explanatory Strategies in the De incessu animalium (Andrea Falcon)

3

2 The Reception of the De incessu animalium (Andrea Falcon)

19

part ii  greek text and translation

33

Preface to the Greek Text (Pantelis Golitsis)

35

Sigla Manuscriptorum

41

ΑΡΙΣΤΟΤΕΛΟΥΣ ΠΕΡΙ ΠΟΡΕΙΑΣ ΖΩΙΩΝ (Pantelis Golitsis)

42

Aristotle, On the Progression of Animals 

43

part iii  interpretative essays

99

3 De incessu animalium 1–3: The Theoretical Framework and the Beginning of the Actual Investigation (Andrea Falcon)

101

v

vi

Contents

4 De incessu animalium 4: Aristotle’s Conception of Dimension (Panos Dimas)

117

5 De incessu animalium 5–6: The Architecture of Locomotive Bodies  (Klaus Corcilius) 141 6 De incessu animalium 7–8: Number and Distribution of Feet in Animal Progression (Stasinos Stavrianeas)

165

7 De incessu animalium 9: Aristotle’s Mathematical Kinesiology: The Case of Bending (Christopher Frey)

194

8 De incessu animalium 10–11: Flight and Two-Footedness (Timothy Clarke)

217

9 De incessu animalium 12–13: Limb-Bending and Natural Teleology (Spyridon Rangos) 233 10 De incessu animalium 14–15: Teleology Across Kinds (Sarah Ruth Jansen)

266

11 De incessu animalium 16–19: The Motion of Many-Footed Animals and Cases of Peculiar Motion in Water (Pantelis Golitsis)

282

References General Index Index Locorum

297 307 309

Figures

3.1 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11 7.12 9.1 10.1

Bending toward the circumference Possible resting points for a bending Rising from a seated position  Bending the trail leg when walking The position of the head when walking The geometrical argument in MA Walking like Aristotle The articulation of elephants Bending at the hip Moving forward with bending Moving forward without bending Diagrammatic representation of worm movement Geometrical properties of undulation Leg-bending possibilities Walk cycle for diagonal walkers

vii

107 197 199 200 201 202 204 208 209 210 211 213 215 263 271

Tables

2.1 Writings on natural philosophy in Pacius’ and Casaubon’s bilingual edition of Aristotle: a comparison 2.2 Writings on natural philosophy in the Aldine edition of Aristotle 2.3 Writings on natural philosophy in the Basel edition of Aristotle 2.4 Writings on natural philosophy in Bekker’s edition of Aristotle 5.1 Possession of bodily articulations according to directions in living things 5.2 Possession of separate bodily articulations in locomotive animals 

viii

25 27 29 30 146 149

Notes on Contributors

timothy clarke is Associate Professor of Philosophy at the University of California, Berkeley. He is the author of Aristotle and the Eleatic One (Oxford University Press 2019). His articles include “The Argument from Relatives” (Oxford Studies in Ancient Philosophy 2012) and “Aristotle and the Ancient Puzzle about Coming to Be” (Oxford Studies in Ancient Philosophy 2015). klaus corcilius is Professor of Philosophy at the University of Tübingen, Germany. He specializes in ancient philosophy. His most recent publications include “De motu animalium 6” in Ch. Rapp and O. Primavesi (eds.), Proceedings of the XIX. Symposium Aristotelicum (Oxford University Press 2020); “Ideal Intellectual Cognition in Tim. 37 A 2–C 5” (Oxford Studies in Ancient Philosophy 2018: 51–106); and, together with Oliver Primavesi, Aristoteles: De motu animalium /Über die Bewegung der Lebewesen (Felix Meiner 2018). panos dimas is Professor of Philosophy at the University of Oslo, Norway. He previously held the position of Director of the Norwegian Institute at Athens and also served as the cultural attaché at the Norwegian Embassy in Athens. His publications in ancient philosophy are on issues of ethics, moral psychology, metaphysics, and epistemology. He is presently working on a monograph with the working title “Plato on Pleasure” and a project on the issue of divisibility of magnitude in ancient philosophy. andrea falcon is affiliated with Concordia University (Montréal, Canada) and University of Milan (Italy). He works on Aristotle and the Aristotelian tradition in antiquity. He is the author of Corpi e Movimenti. La fortuna del De caelo nel mondo antico (Bibliopolis 2001); Aristotle and the Science of Nature: Unity without Uniformity (Cambridge University Press 2005); Aristotelianism in the First Century BCE: Xenarchus of Seleucia ix

x

Notes on Contributors

(Cambridge University Press 2012); Aristotelismo (Einaudi 2017). He is the editor of the Brill’s Companion to the Reception of Aristotle in Antiquity (Brill 2016); and co-editor, together with David Lefebvre, of Aristotle’s Generation of Animals: A Critical Guide (Cambridge University Press 2017) and, with Pierdaniele Giaretta, of Ancient Logic, Language, and Metaphysics: Selected Essays by Mario Mignucci (Routledge 2019). christopher frey is Associate Professor in the Philosophy Department at the University of South Carolina. He received his PhD in Philosophy at the University of Pittsburgh. His work in the history of ancient Greek philosophy focuses primarily on Aristotle’s natural philosophy and he also publishes in contemporary philosophy of mind and action. His work has appeared in Ancient Philosophy, Oxford Studies in Ancient Philosophy, Phronesis, Philosophy and Phenomenological Research, and many other venues. He is currently writing a book about Aristotle’s concept of life entitled The Principle of Life: Aristotelian Souls in an Inanimate World. pantelis golitsis is Assistant Professor in the Philosophy Department at Aristotle University of Thessaloniki. His main research interests lie in the reception of Aristotle in late antiquity and Byzantium. He is the author of Les commentaires de Simplicius et de Jean Philopon à la Physique d’Aristote: Tradition et innovation (Walter de Gruyter 2008; Prix Zographos de l’Association pour l’encouragement des Études Grecques en France). He has recently completed a new critical edition of Alexander of Aphrodisias’ commentary on Aristotle’s Metaphysics I–III (Walter de Gruyter 2021). sarah ruth jansen teaches philosophy at Pima College in Tucson, Arizona. She is also a Visiting Scholar at the University of Arizona. She received her PhD in philosophy at the University of California, Los Angeles. Her work on Plato’s aesthetics, moral psychology, and ethics has appeared in Ancient Philosophy, Epoché, and the Journal of Aesthetics and Art Criticism. Sarah received institutional support for this project from Carleton College, the University of British Columbia, and Northern Arizona University. spyridon rangos is Professor of Greek Literature and Philosophy at the University of Patras, Greece. His research interests focus on the interrelationships between philosophy and religion in classical and late antiquity. His articles include “Empedocles on Divine Nature” (Revue de métaphysique et de morale 2012) and “Plato on the Nature of the Sudden

Notes on Contributors

xi

Moment, and the Asymmetry of the Second Part of the Parmenides” (Dialogue 2014). stasinos stavrianeas is Assistant Professor in the Philosophy Department at the University of Patras. He has published papers on Aristotelian natural science and biology. His articles include “Spontaneous Generation in Aristotle’s Generation of Animals” (Rhizai 2009) and “Nature as Principle of Change” in M. Leunissen (ed.), Aristotle’s Physics: A Critical Guide (Cambridge University Press 2015). He is also the author of a Modern Greek translation with commentary of Aristotle’s Parts of Animals.

Preface

It was during the winter of 2015 that Sean Kelsey and Stasinos Stavrianeas conceived the idea of organizing a workshop on Aristotle’s De incessu animalium. This writing forms an integral part of Aristotle’s biological corpus but it is one of the least studied Aristotelian works both by ancient and modern commentators. And yet, it is a treatise where we can see, with some clarity and detail, Aristotle’s methodology at work. Moreover, it is an avowedly causal treatise and so of special interest for that reason, too. In organizing the workshop, Sean Kelsey and Stasinos Stavrianeas secured the participation of a number of scholars from both sides of the Atlantic Ocean: Timothy Clarke, Klaus Corcilius, Panos Dimas, Andrea Falcon, Christopher Frey, Jessica Gelber, Pantelis Golitsis, Sarah Ruth Jansen, and Spyridon Rangos. The workshop took place at the University of Patras, Greece, during the week of July 11–15, 2016. The conference was made possible through funding by the University of Notre Dame and the University of Patras. We, the editors of the volume, take this opportunity to thank both institutions for their generous support. The participants at the conference were able to work on the Greek text established by Pantelis Golitsis. During the workshop the idea of publishing the contributions as a collective commentary with an English translation based on a new edition of the Greek text was proposed by Andrea Falcon, and it was clear to everyone that such a project would be a valuable addition to Aristotelian studies. Although not all participants were able to contribute an essay to the volume, their contributions at the workshop helped to conceive and see through this collective project. We would like to express our gratitude to Sean Kelsey, without whom this project would not have gotten off the ground. Last but not least, special thanks go to Kay (Katherine) Rollans, Sierra Billingslea, and Malcom J. Todd, our outstanding copy-editor, who made valuable suggestions on the style and substance of the book. xiii

Abbreviations

Alexander of Aphrodisias DA De anima

On the Soul Aristotle

APo APr DA DC GA GC HA IA

Analytica posteriora Analytica priora De anima De caelo De generatione animalium De generatione et corruptione De historia animalium De incessu animalium

Posterior Analytics Prior Analytics On the Soul On the Heavens On the Generation of Animals On Generation and Corruption On the History of Animals On the Progression of Animals

Juv. De iuventute et senectute On Youth and Old Age Long. De longitudine et brevitate vitae On the Length and Shortness of Life MA De motu animalium On the Motion of Animals Mem. De memoria On Memory Meta. Metaphysica Metaphysics Mete. Meteorologica Meteorology PA De partibus animalium On the Parts of Animals Phys. Physica Physics Poet. De arte poetica On Poetics Pol. Politica Politics xiv

List of Abbreviations Resp. De respiratione Sens. De sensu et sensatu Somn. De somno et vigilia Top. Topica

On Respiration On Sense-perception On Sleep and Waking Topics [Aristotle]

De spir. De spiritu Mech. Mechanica

On Breath Mechanical Problems Plato

Tim. Timaeus

Timaeus Theophrastus

CP

De causis plantarum

On the Causes of Plants

Further Abbreviations DK H. Diels and W. Kranz, Die Fragmente der Vorsokratiker. Weidmann Verlag, Zurich 19516

xv

part i

Introduction

chapter 1

Explanatory Strategies in the De incessu animalium Andrea Falcon

The Initial Agenda of the IA In the IA, Aristotle is concerned with many complex questions (some interrelated and some independent or at least outlying) about how animals move themselves by using parts of their bodies as instruments of locomotion. The key questions are recalled at the outset of the work, which begins with an agenda of eleven questions that Aristotle promises to answer in the course of his investigation by employing three explanatory principles explicitly introduced for that purpose in IA 2. These are not all the questions that Aristotle will try to answer in the IA. It is quite telling that, at the end of his agenda, Aristotle adds that we have to look for the causes of the facts that have been singled out (sc. in the agenda), or for other similar facts.1 Evidently, his initial list does not contain all the questions that are relevant to the topic of animal progression. In the course of his investigation, Aristotle is able to consider issues that are not in his initial purview. A good example is offered in IA 14 and 15, where two questions that are not announced at the outset of the work are discussed. Sarah Ruth Jansen argues that these new questions naturally arise from the need to deal with prima facie exceptions to the account of animal progression developed up to that point.2 Evidently, the agenda does not impose a rigid structure on Aristotle’s inquiry but rather constitutes a convenient platform from which to embark on a study of how animals move that must account for the complexity of the zoological data. My interest in the IA goes back to a reading group devoted to this work held at the University of Pittsburgh in the Winter Semester of 2009. The reading group consisted of four people: Allan Gotthelf, Keith Bemer, Peter Distelzweig, and myself. I am happy to acknowledge that my ideas on the initial agenda of the IA were first developed at those meetings as a result of our teamwork. I remember those meetings with a mixture of fondness and sadness as both Allan Gotthelf and Keith Bemer are no longer with us. This introductory essay as well as the essay on IA 1–3 (ch. 3) are dedicated to the memory of Keith Bemer, whose intelligence, curiosity, and above all kindness, are greatly missed. 1 See IA 1, 704b8. 2 See below, “Teleology Across Kinds” (ch. 10).

3



Part I Introduction

With the caveat introduced in the above paragraph, it remains true that the initial agenda provides not only a theoretical framework to the entire investigation but also unity to the treatise as a whole. Moreover, Aristotle’s methodological explicitness gives us a way to assess the outcome of his investigation. In other words, explanatory success (or failure) in the IA can be measured against the initial agenda. If Aristotle is able to offer an adequate answer to the eleven questions on his agenda, he has completed his task as stated at the outset of the work; if he is not, he has fallen short of the task that he has set for himself. Of course, it remains to be seen if by answering those questions Aristotle has also offered an adequate explanation of the phenomenon of animal progression. For example, one might acknowledge that at the end of the IA Aristotle has answered all the questions on his agenda (as well as a few additional questions) and at the same time object that Aristotle has not given us a completely adequate explanation of animal progression. To counter this objection we would need to be able to establish, among other things, how Aristotle has arrived at the formulation of his initial questions. This would not be an easy task, especially since Aristotle does not elaborate on the origin of his agenda. He does not tell us how the agenda is organized, or why the questions that we have to answer are exactly those eleven questions. This last observation makes it even more pressing for us to try to explain why Aristotle does not defend the adequacy of his initial agenda. In the absence of an explicit statement on the part of Aristotle, we can only venture an educated guess. His position may be that answering the initial eleven questions is also a vindication of the list itself and its merits as an initial agenda for the inquiry conducted in the IA. Put differently, when we follow Aristotle in his attempt to answer the questions on his agenda, we are also expected to come to see their relevance for the investigation, and see that an answer to the eleven questions on the agenda amounts to an adequate account of animal progression. Here are the eleven questions in the order in which Aristotle lists them in IA 1: [Q1] What are the fewest points at which animals move? [Q2] Why do blooded animals move by means of four points while bloodless animals move by means of more than four points? [Q3] Why, in general, are some animals footless, some two-footed, some four-footed, and some many-footed? [Q4] Why do all the animals that have feet have an even number of feet?



1 Explanatory Strategies in the De incessu animalium



[Q5] Why, in general, are the points by means of which animals move even in number? [Q6] Why are human beings and birds two-footed whereas fishes are footless? [Q7] Why do human beings and birds, being both two-footed, bend their legs in opposite directions (human beings convexly while birds concavely)? [Q8] Why do human beings bend their legs and arms in opposite directions? [Q9] Why do the four-footed animals that are live-bearing bend their legs in the opposite way to human beings and also in opposite ways with respect to themselves (front legs convexly while hind legs concavely)? [Q10] Why do the four-footed that are egg-laying bend their legs in a way unique to them, namely laterally away from their body? [Q11] Why do four-footed animals move their legs diagonally? I refer the reader to the first interpretative essay for additional reflection on the organization and structure of this agenda.3 For the time being, I am content to make the following observation: the above questions are not established independently of the investigation conducted in the IA. In other words, Aristotle is not recalling an agenda that was generated at a pre-explanatory stage of inquiry by merely collecting and organizing the relevant empirical data and is now available to us as we are about to approach the explanatory stage of inquiry. On the contrary, his agenda reflects (at least in part) the results reached in the course of the explanatory stage of inquiry. Consider, in particular, [Q1], [Q2], and [Q5]. These questions are concerned with “points of motion.” We do not see points of motion when we observe an animal moving from one place to another. Rather, the points of motion by means of which the animal moves with respect to place are an important explanatory device that Aristotle introduces in the course of his attempt to develop a conceptual model to deal with animal locomotion.4 The significance of this observation can hardly be overstated. At the very least, we are able to infer that our initial agenda 3

See “The Theoretical Framework and the Beginning of the Actual Investigation” (ch. 3). HA I 5, 490a26 ff. refers to points of motion, but this stretch of text seems to presuppose the ­­­­discus­­­­­­­sion offered in IA (rather than vice versa). We should bear in mind that HA may have been the last work that Aristotle wrote on the topic of animals, entailing some rethinking of the results achieved in the rest of the zoological corpus. David Balme was the first to suggest this possibility. The best, and indeed clearest, presentation of this suggestion – with a wealth of evidence – is Lennox 1996: 229–248.

4



Part I Introduction

is not entirely innocent with respect to the explanatory work done in the IA. This last point can be restated as follows: while it is true that Aristotle refers to a collection of the relevant facts (ἱστορία) right after he has given us his initial agenda, this collection alone does not deliver the questions on our initial agenda.5

Points of Motion and Number of Feet One important distinction, which also marks the beginning of Aristotle’s actual investigation in IA 3, is that bodily displacement takes place by means of either jumping or progression (πορεία). In calling the second mode of bodily displacement progression, I am following William of Moerbeke (the first translator of the IA), who rendered πορεία with progressus.6 While in jumping the body is displaced all at once, in progression the body is displaced part by part (κατὰ µέρον). This second mode of bodily displacement is Aristotle’s primary focus in the IA. The parts of the body involved in animal progression are called instrumental parts.7 By adopting this expression, Aristotle indicates that these parts are used by the animal as a tool to produce motion. Hence, my decision to say that Aristotle is concerned with how animals move themselves by using parts of their bodies as instruments of locomotion. As Aristotle explains toward the end of IA 3, the animal that moves itself displays a minimal level of complexity: one part is acted upon by the other by being compressed, the other acts on it by pressing. Hence, nothing that is partless can move itself.8 Aristotle begins his investigation with a study of how animals progress on land by means of feet. Moreover, he defines the foot as the part that is in contact with the ground and as such is productive of locomotion.9 We can restate this first move by saying that Aristotle begins his study by focusing on the case of footed animals. If, however, we want to appreciate Aristotle’s overall explanatory strategy, we need to bear in mind another important decision implicitly made at the outset of the IA. The study of how animals move from place to place begins with a study of locomotive parts in blooded animals. The focus on blooded animals is already in place 5

The language that Aristotle adopts in referring to the collection of facts is quite interesting. For more on this point, see the first interpretative essay. 6 See below, “The Reception of the De incessu animalium” (ch. 2). 7 IA 3, 705a20. 8 IA 3, 705a20–24. Full discussion in “The Theoretical Framework and the Beginning of the Actual Investigation” (ch. 3). 9 IA 5, 706a31–32.



1 Explanatory Strategies in the De incessu animalium



at the outset of IA 6. It is in this chapter that Aristotle lays down the foundation for the claim that opens IA 7: motion with respect to place belongs, either only or above all, to those animals that make change with respect to place either by means of two or four points.10 The animals in question are blooded animals, which are used as a paradigmatic case. It is telling that it is only when a study of the locomotive parts in blooded animals is firmly in place that Aristotle ventures into the study of bloodless animals, which are lower on the scala naturae. The central claim made in IA 6 is that there has to be a common origin of motion that is equally disposed with respect to the locomotive parts. In other words, the locomotive system described in IA 6 is envisioned as a centralized system with a single source of locomotion. Aristotle makes this source the cause of motion with respect to place.11 Given that Aristotle is firmly committed to cardiocentrism, there is no doubt that he thinks that this source of motion is a perceptual soul, and that in the case of blooded animals this soul is located in the heart. Still, it is significant that Aristotle refrains from talking about the soul and the heart in IA 6. It is not immediately clear why Aristotle does not elaborate on the soul and the heart in the IA. In the absence of clear hints in the text, we can only offer an educated guess. In all probability, a discussion of the soul and its relation to the body pertains to the study of what is common to the body and the soul rather than to a study of the locomotive parts and their role in the explanation of animal locomotion. In his interpretative essay, Klaus Corcilius expands on the reasons why Aristotle remains silent on the topic of the soul and the heart in IA 6.12 Here I am content to stress that if we accept Corcilius’ reading of IA 6, we have a very good reason to resist the temptation to assimilate the IA to the short essays on natural philosophy that are collectively known as Parva naturalia. This conclusion is important because the place of the IA in the larger project that is known as Aristotle’s natural philosophy is far from being obvious. Evidence of this is that the position of the IA in the Aristotelian corpus is notoriously unstable.13 In IA 6 Aristotle develops a highly abstract account of what a certain kind of locomotive system requires in order to perform its primary function, which is to move from one place to another. Although Aristotle generates this account for the explanation of how blooded animals move, 10

IA 7, 707a16–19. IA 6, 707a12. 12 See “The Architecture of Locomotive Bodies” (ch. 5). 13 See below, “The IA in the Early Printed Editions of Aristotle” (ch. 2). 11



Part I Introduction

he subsequently adopts it to explain the motion with respect to place of bloodless animals. Aristotle’s explanatory strategy becomes clear as soon as we realize that Aristotle conceives of bloodless animals as locomotive systems to which additional points of motion are attached to the original four. Moreover, because additional points are attached, these systems do not display the same level of integration and unity as the one studied in IA 6. One concrete example may help to illustrate Aristotle’s strategy and overall approach to the study of bloodless animals. An ant moves by means of six feet. By Aristotle’s lights, an ant is neither a two-footed nor a four-footed animal: rather, it is a many-footed animal. Aristotle conceives of such an animal as a locomotive system consisting of four + two points of motion. We may think that having more than four feet allows an animal to perform its function – moving with respect to place – in a better, quicker, and indeed more efficient way. But Aristotle never makes this observation. His first and most important concern is to stress that a unit consisting of more than four points of motion is a less unified, and indeed less well-integrated, locomotive system. It is a less unified, and less well-integrated, locomotive system precisely because Aristotle thinks of this unit as having four + two points of motion. Of course, Aristotle will have to posit that there is still a controlling center in this system if he wants to account for the locomotive behavior of the animal. By his lights, however, this controlling principle, which is situated in the analogue to the heart,14 is not equally well disposed with respect to all the six points of motion. It is along these lines that we must understand Aristotle’s claim that, while blooded animals cannot continue to live and move if they are severely mutilated, some many-footed bloodless animals can. They can because they are compounded out of several animals.15 In this context, Aristotle mentions the centipede. By his lights, all animals that have an elongated body and are like the centipede enjoy a weaker type of unity compared to blooded animals. In a few cases, the unity is so weak that the animals consist of relatively independent locomotive units. As a result, each unit can survive and move even if separated.16 The significance of the decision to study bloodless animals by analogy with blooded animals can hardly be overestimated. This decision allows 14

Recall that, according to Aristotle, bloodless animals do not have a heart but they do have something that plays an analogous role in their physiology. 15 IA 7, 707a27–31. 16 For full discussion of this claim, see “Number and Distribution of Feet in Animal Progression” (ch. 6).



1 Explanatory Strategies in the De incessu animalium



Aristotle to arrive at a unified account of animal progression – namely, an account that applies to both blooded and bloodless animals without reducing or eliminating the differences between the two groups of animals.17 But this decision is not without costs. One thing that is obvious, even from a cursory glance at the IA, is how selective the treatment of progression in bloodless animals is. It is hard to resist the conclusion that Aristotle’s approach to the progression of bloodless animals would have been significantly different if it did not depend on the assumption that their progression ought to be understood in light of the results achieved in the study of blooded animals. In IA 6 Aristotle lays the theoretical foundations to answer the first two questions on his initial agenda. Both are questions dealing with points of motion. While the first is about the fewest number of points by which an animal moves, namely [Q1], the second is concerned with those points of motion in blooded and bloodless animals, [Q2]. It is worth noting, however, that Aristotle does not announce IA 6 as an attempt to answer those questions; rather, our answering them is a by-product of Aristotle’s study of what it takes for a locomotive system of the kind he envisions in IA 6 to move from place to place. This observation confirms a point I made earlier in dealing with the agenda of the IA – namely, our agenda cannot be established independently of the investigation conducted in the IA. With this observation in place, we can turn to how Aristotle organizes his discussion in IA 7–8. A close look at the explanatory strategy adopted in these two chapters helps us appreciate how Aristotle deals with the complexity, and indeed the variety, of animal progression. First, Aristotle begins his investigation by (re)stating the main result reached in IA 6: if x is a blooded animal, then x moves by means of four points of motion; moreover, if x moves by means of four points of motion, x is a blooded animal.18 Then, he turns to footless animals in order to show that, despite appearances to the contrary, they too move by means of four points. Dealing with footless animals, and dealing with them at this stage of the argument, is crucial because these animals are a potential counterexample to what Aristotle has just established. Footless animals with an elongated body such as snakes or eels move on the ground or in water by means of two bends. Aristotle argues that we can distinguish two points of motion 17

Recall that the transition from blooded to bloodless animals is made by invoking analogy. Bloodless animals have a bodily part that is analogous to the heart. In the Peripatos, analogy is an explanatory tool specifically designed to deal with reality without reducing or eliminating its complexity. 18 IA 7, 707a20–23.



Part I Introduction

at each of the two bends. Hence, despite the appearance to the contrary, footless animals such as snakes and eels move by means of four points of motion like all blooded animals. By dealing with an apparent challenge to his basic account – footless animals – Aristotle is able to vindicate the statement made at the outset of IA 7 – namely that all blooded animals move by means of four points of motion. Moreover, he is able to give us a better understanding of the sort of locomotive systems he has in mind. It is only at this point that Aristotle turns to the case of snakes in order to explain why they are footless. This happens in the first part of IA 8, which is also one of the bestknown sections of the IA. Here Aristotle applies the explanatory principle outlined at the outset of the IA – namely, that nature does nothing in vain but always what is best with the possibilities available for each kind of animal.19 Scholars have often returned to Aristotle’s explanation of why snakes are footless in the attempt to shed light on how Aristotle understands this teleological principle. The reader will find in the essay by Stasinos Stavrianeas a careful discussion of how Aristotle applies this principle in IA 8, and a reflection on what we can learn from Aristotle’s discussion of the absence of feet in snakes.20 Once Aristotle has dealt with the absence of feet in snakes, he is able to return to footed animals in order to show why all footed animals must have an even number of feet (second part of IA 8). At this point, we also have an answer to two other questions on our agenda. While the first has to do with the even distribution of feet, namely [Q4], the second restates the same point with respect to points of motion, [Q5]. One final point is in order. The discussion conducted in IA 7–8 alone does not confirm that all blooded animals move by means of exactly four points of motion. More directly, humans and birds may be thought to move by means of two rather than four points of motion. Aristotle has a great deal more to say on the topic of birds and humans in the course of IA 10 and 11. It is only when this discussion is finally in place that we can be confident that all blooded animals move by means of exactly four points of motion. Until then, a perceptive reader may wonder about possible exceptions to the rule of the four points of motion. For a lucid discussion of this aspect of Aristotle’s overall strategy, I refer the reader to the introductory remarks Timothy Clarke offers in his interpretative essay.21 19

IA 2, 704b15–17. “Number and Distribution of Feet in Animal Progression” (ch. 6). 21 “Flight and Two-Footedness” (ch. 8). 20



1 Explanatory Strategies in the De incessu animalium



Bending A quick look at the initial agenda of the IA suffices for one to realize that the eleven questions are divided into two blocks: while the first block is concerned with number of points of motion and distribution of feet, the second is about bending and its role in animal progression. Aristotle is far from having exhausted the first block of questions at the end of IA 8, but he is nonetheless ready to approach the second. Aristotle turns to bending at the outset of IA 9. His main claim in this chapter is that without bending (and straightening) there would be no bodily displacement. He proves this claim by focusing on the case of footed animals. Once more, footless animals can be seen as a potential counterexample. As in his discussion of points of motion, Aristotle shows that they are not. Their bodily displacement and progression on land (or in water) takes place by bending. Once established that all progression takes place by bending, Aristotle adds that also jumping requires bending in the supporting part of the body. This is a hardly surprising result, but it is required to show that all modes of bodily displacement require bending (and straightening). By the end of IA 9, Aristotle has given us the promised explanation of why bending (and straightening) is necessary for all animals to move from place to place. However, this does not exhaust the topic of bending. His study of bending continues all the way to the end of IA 15 and remains a main focus in the discussion of bloodless animals offered in IA 16 and 17. Evidence of this is that answers to all the questions that are concerned with bending on our initial agenda can be located in this stretch of text. 1. We find an answer to the question why humans and birds bend their legs in opposite directions, namely [Q7], in IA 15. 2. We are offered the resources to explain why humans bend their arms and legs in opposite directions, [Q8] on the agenda, in IA 12. 3. We find an answer to the question why four-footed animals that are live-bearing bend their front and rear legs in opposite directions, namely front legs convexly and rear legs concavely, [Q9] on the agenda, in IA 12. 4. We can also locate an answer to the question why the four-footed animals that lay eggs move their legs obliquely and away from their body, namely [Q10], in IA 15. 5. Last but not least, the question why four-footed animals move their legs diagonally, namely [Q11], takes up a large part of IA 14.



Part I Introduction

Still, we may wonder whether locating an answer to the questions on the initial agenda helps us advance our understanding of the overall strategy that Aristotle adopts in this stretch of text. I don’t think so. Clearly, there is a great deal more going on than an attempt to answer those questions in this section of the IA. Let us return to the end of IA 9, which consists of a few remarks on swimming and flying animals and how bending is necessary in their case as well. This discussion is carried over into IA 10. As soon as we realize that these animals are a focus of the chapter, we see why Aristotle finds it convenient to deal with topics such as the role of the tail in flyers, or the slow and not so efficient flight of certain insects that do not have a tail, in the course of this chapter. I refer the reader to the interpretative essay by Timothy Clarke for a detailed discussion of the various statements that Aristotle makes in the course of his analysis of flying animals.22 What matters here is that the occasional discussion of bloodless animals and their locomotion – as in the case of the flyers that bump around because they have no tail – is not necessarily a violation of the overall strategy adopted in the IA. Aristotle has programmatically postponed the study of bloodless animals after that of blooded animals, but he deals with particular aspects of their motion in the course of the study of blooded animals, and under a common rubric, if he thinks that this strategy results in an optimal treatment of certain topics. In IA 11 we are back to the study of footed animals – with a concentration on two-footed animals. Aristotle’s main claim is that humans alone are, strictly speaking, two-footed. This chapter contributes to the investigation launched in IA 5. There, Aristotle has provided an initial description of the distribution of feet in footed animals by using the concepts of dimensions (διαστάσεις): a two-footed animal is an animal in which the front part and the upper part are clearly demarcated, whereas a fourfooted or a many-footed animal does not display this distinction.23 This distinction is meant to be at most an initial characterization of what it is to be two-footed (or, for that matter, what it is to be a four-footed or a many-footed animal). It is only in IA 11 that Aristotle completes his discussion with a close analysis of the anatomy of a bird. From this point of view, IA 11 does not only complete the discussion started in IA 5 but also continues, quite naturally, the discussion of flying animals started at the end of IA 9. 22

“Flight and Two-Footedness” (ch. 8). Of course, we can still draw this distinction in four-footed and many-footed animals by invoking the different functions associated with those parts: while their upper part is the entry point of nourishment, their front part is where their organs of perceptions are implanted.

23



1 Explanatory Strategies in the De incessu animalium



Two-footed animals are the focus of Aristotle’s discussion in the first part of IA 12. Aristotle appears to be concerned with the question of how these animals progress on land. Clearly, one of their two legs is standing while the other is leading. The standing leg supports all the weight whereas the leading leg is the one that has no weight. Aristotle explains why it is better, and indeed more efficient, for the leading leg to bend forward (convexly). As soon as this explanation is in place, Aristotle turns to four-footed animals, with a concentration on the live-bearing animals. He explains why it is better for this group of animals to bend the front and rear legs in opposite directions (they bend the front legs forward or convexly, and the rear legs backward or concavely). IA 13 completes the discussion of bending by schematizing the results reached in live-bearing animals. The chapter provides additional remarks on the bending of human limbs. Aristotle concentrates on four-footed, live-bearing animals in IA 14. A central goal of this chapter is to explain why these animals move their legs diagonally (κατὰ διάµετρον). They move their legs in this way because the temporal order in which the legs are moved goes as follows: “after the right front leg, four-footed animals move the left back leg, then the left front leg, and after it the right back leg.”24 Sarah Ruth Jansen offers a full discussion of Aristotle’s explanation of this diagonal movement.25 In the context of this general introduction, I am content to elaborate on how Aristotle is able, and indeed willing, to adapt this explanation to the case of bloodless animals. Recall that some (but not all) bloodless animals are many-footed animals. An obvious question is how these animals move their legs. Aristotle answers this question by extending the explanation developed for the four-footed, live-bearing animals to the bloodless animals that are manyfooted. In the latter, any consecutive set of four legs moves in the way described for the former: the rear legs move diagonally with the front legs. Aristotle conceives of many-footed animals as locomotive units in which the original module of four points of motion and four feet is repeated into a single, complex system. At this point, there is no need to insist on the claim that such a locomotive system does not display the same level of unity as the original module. What matters is that Aristotle can deal with many-footed animals here because he has already developed the conceptual resources to do so. Their treatment in IA 14 feels 24

IA 14, 712a26–28. “Teleology Across Kinds” (ch. 10).

25



Part I Introduction

appropriate even if it is, strictly speaking, a digression from the topic of discussion. Proceeding in another way would have been possible, but it would have resulted in needless repetitions. IA 14 ends with a discussion of the progression of crabs, which are the only animals that progress diagonally rather than forward. Aristotle can explain away this apparent anomaly by pointing out that the eyes of crabs are implanted on their two sides. As a result, we can say that crabs move obliquely with respect to us but forward relative to themselves. They move forward relative to themselves because their eyes are implanted on parts that functionally operate as the front part of the animal. Aristotle returns to his main focus – blooded animals – in IA 15. He is able to explain why birds bend their legs backward (concavely), like the rear legs of the four-footed, live-bearing animals. His reasoning depends on accepting that blooded animals move by means of four points, and accepting that in birds two of the four points of motion are located in their wings while the remaining two are in their legs. In this sense the legs of birds are functionally equivalent to the rear legs of the fourfooted, live-bearing animals. The reader will find an attempt to evaluate this argument in the essay by Sarah Ruth Jansen.26 What matters here is that Aristotle applies the results achieved in IA 6 in order to generate explanations that crucially depend on comparing not only different kinds of animals but also different modes of locomotion. Once Aristotle has completed his account of bending, he turns to the following question: Why is it better for certain locomotive parts (most notably wings and fins) to be attached obliquely? Generally speaking, this is not true of feet. There is, however, one notable exception: some four-footed animals such as lizards, crocodiles, and turtles have their feet attached obliquely. As a result, when those animals move their feet, they move them laterally and away from the body. This is a unique phenomenon that calls for an explanation.27 All the animals that display this peculiar arrangement of their four feet have at least two features in common: they are egg-layers and live in holes. Aristotle takes both their mode of reproduction and their habitat as basic explanatory features. It is time to try to sum up the explanatory strategy adopted in the stretch of text that begins at IA 9. Aristotle focuses on bending and answers the questions on his agenda by concentrating first on two-footed animals and then on four-footed animals. In other words, the order of 26

“Teleology Across Kinds” (ch. 10). [Q10] in the initial agenda.

27



1 Explanatory Strategies in the De incessu animalium



explanation that Aristotle adopts is first two-footed animals and then four-footed animals. It is only with the help of the additional information that Aristotle provides while dealing with two-footed animals that we see that birds and humans are no exceptions to the rule, and that they too move at exactly four points of motion like all blooded animals. In dealing with four-footed animals, Aristotle deals first with live-bearing animals and then with egg-laying animals. Four-footed, live-bearing animals and four-footed, egg-laying animals are two large kinds (µέγιστα γένη) of blooded animals. Aristotle defends the existence of these two distinct groups in HA I 6.28 In the IA we see, in a clear way, how this division is adopted in the course of his explanatory work to generate the relevant explanations.

Blooded and Bloodless Animals; Footed and Footless Animals; Imperfectly Developed Animals Aristotle has already touched upon the progression of bloodless animals in the course of his discussion of footed animals. Whenever it has been not only possible, but also better, to deal with certain aspects of their locomotion in the course of his discussion of blooded animals, Aristotle has not hesitated to do so. Still, it is safe to say that, from the beginning of his actual investigation in IA 3 to the end of IA 15, Aristotle has concentrated his attention on blooded animals. The shift of focus to bloodless animals occurs at the outset of IA 16. Given his previous strategy, it should be no surprise to discover that Aristotle is able, and indeed willing, to return to certain aspects of the progression of blooded animals in his study of bloodless animals. This does not mean that the division of animals into blooded and bloodless is not important, or that it does not play a significant role in the overall explanatory strategy adopted in the IA. It only means that Aristotle is not imposing a too rigid structure on his discussion but remains flexible in order to adapt to the complexity of his task, which is to provide as complete a study as possible of animal progression. One may get the impression that the end of the IA is a somewhat random discussion of exceptions or anomalies that challenge the basic account: crabs, crayfish (κάραβοι), web-footed birds, and the like. I would like to resist this impression and argue that even here Aristotle has a method in dealing with these cases. He first searches for an explanation 28

For a perceptive discussion of HA I 6, I refer the reader to Gotthelf 2012c: 293–306.



Part I Introduction

of why animals have feet, and have them in the particular number or manner they do. In other words, his study of bloodless animals begins with the discussion of footed animals. However, Aristotle has already explained why bloodless animals equipped with feet must be many-footed animals. As a result, questions about number of feet and points of motion are no longer addressed. Instead, Aristotle concentrates on the explanation of the particular manner in which the feet are attached to the body of bloodless animals or why they bend in the particular way they do. Since Aristotle has already dealt with these questions in connection with blooded animals (under the rubric of bending), we can only expect him to try to do the same for bloodless animals. I refer the reader to the commentary on these chapters for a detailed discussion of the explanation given. Here suffice it to say that Aristotle discusses the case of crabs and crayfish. Both are hard-skinned animals and this is relevant to the explanation of the particular way they bend their legs. While crabs use their legs for progression on the ground, crayfish employ them for swimming. The discussion of crayfish leads to a discussion of other swimmers, including web-footed birds. Aristotle must have felt that the discussion of these blooded animals was more appropriate here rather than in the context of his discussion of two-footed animals. His explanation invokes the fact that these birds live in water as well as other considerations about their mode of life and habitat. The discussion of the absence of feet in fishes is a natural development of the discussion of footed animals whose habitat is water. Note that, at this point, the investigation turns from footed to footless animals. This is far from being surprising. On the contrary, a precise explanatory pattern has been observed at several junctures in the work. Aristotle always begins his inquiry by explaining the presence of feet in certain kinds of animals and then explains their absence in other kinds of animals. Fishes are no exception to the rule. In their case, the absence of feet is explained by invoking their habitat and mode of life, as well as the fact that fishes are blooded animals. Fishes are blooded animals, so they can move only at four points of motion. Since they live in water, they have fins rather than feet. The attachment of the fins and the presence of a tailfin are explained by exploiting the analogy between swimmers and flyers. It is only at this point that we have a full answer to the question why human beings and birds are two-footed whereas fishes are footless, namely [Q6]. Of course, Aristotle did not forget or overlook this question. His explanatory strategy is to deal first with progression on land and only then with progression in air by means of flying or in water by means of swimming.



1 Explanatory Strategies in the De incessu animalium



The previous remarks go some way toward mitigating what seems to be a too brief and cursory discussion of progression in water (swimming). However, we should also take the following rule of inquiry into account as we evaluate the merits of the overall explanatory strategy adopted in the IA: any investigation ought to begin with the study of what is more easily observable, and as such more familiar to us. There is no doubt that this rule of inquiry, which is stated not only in the Posterior Analytics but also at the outset of Physics I, is at work throughout the whole IA. The adoption of this rule explains why Aristotle begins his investigation with a study of progression on land. It also explains why flying is studied before swimming. Aristotle must have relied on the assumption that swimming and flying are analogous in order to extend the results achieved in the study of the first mode of progression to the second. This explains why, at the end of IA 16, he is confident that he has adequately dealt with swimming despite the fact that he has barely touched on it. Aristotle’s study of progression is not over yet. He still must deal with hard-shelled animals (ὀστρακόδερµα). These animals appear to move, but it is far from clear how they move, and whether they are engaged in the form of locomotion that we have called progression (πορεία). It is telling that Aristotle approaches this topic by announcing that the motion of these animals is a real puzzle. They do move but it is unclear whether they have a right and a left in their body. His solution to the puzzle consists in considering hard-shelled animals a maimed or mutilated kind of animal (ἀνάπηρον). Aristotle elaborates on this idea by adding that these animals move as a footed animal whose legs are cut would move.29 This analogy with an animal that has a principle of motion but is maimed implies that the hard-shelled animals, too, have a principle of motion in their body. As a result, we should be able to distinguish a right and a left in their body. Unfortunately, what Aristotle says seems to contradict this. He tells us that the hard-shelled animals move but they do so against nature, because they are not naturally able to move (they are not κινητικά). Perhaps we can shed some further light on this issue with the help of what Aristotle says on the topic of the imperfectly developed animals in the context of his discussion of the locomotive soul (DA III 11). His claim is that imperfectly developed animals move, but they do so in an indefinite way (ἀορίστως).30 As a result, the capacities of the soul needed for 29

Or as a seal does: Aristotle regards the seal as the equivalent of a maimed four-footed animal. DA III 11, 434a4.

30



Part I Introduction

locomotion, as well as the relevant bodily distinctions, are present in them but in an equally indefinite way (ἀορίστως).31 If we are on the right track, Aristotle would say that the distinction between a right and left is present also in the hard-shelled animals, but he would add that this distinction is present in them in an indefinite way. Be this as it may, the discussion of hard-shelled animals is the last topic dealt with in the IA. It is a sort of coda to the study of animal locomotion for at least two reasons. First, it is no longer a case of progression, which is the focus of the treatise. Second, the whole kind is regarded as an intermediate group between stationary and non-stationary animals.32 Clearly, the IA is a relatively well-organized treatise. But is there an overall argument in the treatise as a whole? The answer to this question appears to be negative. There does not seem to be a grand argument in the IA. Instead, Aristotle conceives of his main task as that of answering a set of complex and difficult questions: the eleven question that constitute his initial agenda plus a few other questions that naturally arise in the course of his study of animal progression. His main concern is to cover as many modes of progression and as many animal kinds as possible. This explains why dealing with exceptions is an important part of the inquiry rather than an afterthought. Still, Aristotle seems to have a set of rules of inquiry that he consistently adopts throughout the IA. He begins his investigation with a study of blooded animals, and within blooded animals with a study of footed animals. He also deals with the more familiar (and easier to study) case of progression on land by means of feet. It is around this case that he develops his basic account of animal progression. He deals with anomalies or exceptions only when his basic account is in place. The discussion of peculiar cases of locomotion is typically postponed. The discussion of the oblique motion of crabs at the end of IA 14 is a good case in point.33 Crabs are a prima facie exception to forward progression because they appear to move obliquely. By showing how prima facie exceptions are not really exceptions to his general principles, Aristotle also strengthens the case for his principles. In this sense, his discussion of an exception is not a digression but rather an integral part of his task, which is to account for the complexity of the zoological data.

31

Aristotle appears to ascribe indeterminate phantasia, desire and perception to these animals in DA III 11, 434a5. 32 Hard-shelled animals are also regarded as an intermediate kind between animals and plants (GA I 23, 731b8–13). 33 IA 14, 712b15–21.

chapter 2

The Reception of the De incessu animalium Andrea Falcon

The Greek and Latin Reception of the IA With the notable exception of the medical tradition, and in particular of Galen, the biological writings remained largely at the margin of the postHellenistic engagement with Aristotle. While there is plenty of evidence of an interest in claims made in the course of the study of animals, this interest did not result in a systematic engagement with Aristotle’s biology.1 Quite the opposite: the biological works were conspicuously absent from the philosophical curriculum of late antiquity. Given this situation, it is far from surprising to discover that no ancient Greek commentary was written on the three great biological treatises (HA, PA, and GA), the MA, the IA, and the short essays that are collectively known as Parva naturalia. The De sensu is the only exception to the rule.2 This selective engagement with Aristotle was already noted by Karl Praechter, who also highlighted the Byzantine attempt to fill the gaps left by the ancient commentators.3 We now know that this attempt can be dated to the first half of the twelfth century, and that it can be traced back to the circle of intellectuals gravitating around Anna Comnena.4 Michael of Ephesus was a key figure in this circle. He wrote commentaries on the PA, GA, IA, and MA. He also produced commentaries on all the essays collected in the Parva naturalia with the exception of the De sensu. This chapter is dedicated to the memory of Pieter De Leemans. His innovative approach to the study of the codicological tradition of the MA and the IA has been very important to my own research. Pieter was a brilliant philologist and an outstanding historian. He will be greatly missed beyond the boundaries of the Aristoteles Latinus. 1 For a succinct presentation of the extant evidence, see Falcon 2021: 246–260. Falcon 2017b: 104– 108 evaluates this selective engagement in the larger context of the reception of Aristotle’s philosophy in antiquity. 2 Alexander wrote a commentary on this essay, which is available in English translation (Towey 2000). Alexander may have also written on Mem. but his commentary is not extant. 3 Praechter 1906: 861–862; Praechter 1909: 535 [1990: 51–53]. 4 Browning 1962: 1–12 [1990: 393–406].





Part I Introduction

His production in this area almost perfectly fills the gaps left by the ancient commentators in the field of biology.5 Michael has been described as a sober and intelligent interpreter of Aristotle, and a commentator who tried to remain close to the spirit of Aristotle’s text.6 Although his exegesis feels at times rather dull and uninspiring, we should not underestimate how difficult it was to write a commentary without having the luxury of relying on a previous exegetical tradition. On the contrary, we must recognize that Michael is often working with rather limited exegetical means. Consider, for example, his commentaries on the MA and the Parva naturalia. In both cases, Michael found in Alexander of Aphrodisias and his treatise on the soul a useful source of information. To understand why, we need to realize that Alexander goes well beyond what Aristotle achieved in his own treatise on the soul (the DA). Aristotle’s DA is about the soul as a form of life. As such, it is a distinct project from the one attempted in the so-called Parva naturalia. At the outset of the De sensu, Aristotle introduces the reader to the project of the Parva naturalia by saying that he has completed a study of the soul and is about to embark on a study of “what is common to the soul and the body” (1, 436a7–8). Aristotle could not be clearer on the fact that the DA and the Parva naturalia are different kinds of projects. One obvious way in which Alexander goes beyond Aristotle is that he concerns himself not only with questions that pertain to the soul as a form of life, but also with questions that by Aristotle’s lights fall under the rubric of what is common to the body and the soul. One of them is the question of the location of the soul in the body. Toward the end of his own treatise on the soul, Alexander engages in a sustained defense of Aristotelian cardiocentrism (DA 94.7–100.17). In this section, Michael found a rich and authoritative source of information, which he used in the interpretation not only of the MA but also of the Parva naturalia.7 As Michael turned to the IA, he could not avail himself of Alexander (or any other authoritative source of information). In this case, his exegetical resources are stretched to the limit. In the complete absence of an exegetical 5

I say “almost perfectly” because we have no extant commentary on the HA. It is still debated whether Michael wrote a commentary on this work. One should not overlook, however, that the HA is not on a par with the other biological works. Since the HA is not concerned with causes, one may take the view that it contributes to the science of animals at most indirectly – by providing the facts to be explained. If one takes this view, one may see no need to write a commentary on this work. 6 Praechter 1909: 536 [1990: 53]. 7 For a full discussion of how Michael uses Alexander and quotes from the last part of his work on the soul, see Donini 1968: 316–323. Cf. Wenland 1903: XII, suppl. II (“De Alexandri usu”).



2 The Reception of the De incessu animalium



tradition on the IA, Michael could only interpret Aristoteles ex Aristotele, which amounts to making contact (when possible) with the PA. This explains why his commentary on the IA remains one of his most elementary and least helpful. As a rule, Michael shies away from the more technical questions, so the reader should not approach it with the expectation that she will find new insights on the text. The value of the commentary for the reconstruction of the text is also limited.8 I will not engage in a detailed discussion of the exegetical choices or solutions that Michael adopts in the course of his explication of this Aristotelian text. Here, suffice it to say that Michael regarded the IA as an appendix to the PA. This is clear from what we are told at the end of his exegesis of the Parva naturalia, which is one of the few places where Michael – rather uncharacteristically – talks about his own work. There, in looking back to what he has achieved as a commentator, he considers the PA and the IA a single project.9 Michael of Ephesus is only a chapter in the long and complex history of the gradual re-appropriation of Aristotle’s biology. This process started before him in Arabic philosophy and continued after him in the Latin tradition. The IA exemplifies this long and tortuous process very well. This Aristotelian text does not appear to have been known in the Arabic tradition.10 It was literally rediscovered by William of Moerbeke, who produced the first Latin translation of this text around 1260. This translation was part of Moerbeke’s larger project of translating Aristotle’s biology into Latin, which itself was part of his larger project of translation of Greek science into Latin.11 The challenge of translating a Greek text into another language for the first time is comparable to the task of writing a commentary in a vacuum. We can think of Moerbeke not only as a gifted translator but also as an interpreter who is doing his best to render the sense of Aristotle’s text with little or no help.12 Here I am content to note two things. First, as we have already seen, Moerbeke opted to render the Greek πορεία with the Latin progressus. Hence, his chosen title for our work was De progressu animalium. Second, his translation of the IA (along with that of the MA) generated an interesting debate on the place of this work in Aristotle’s project of natural investigation. 8

For more on this point, I refer the reader to the preface to the Greek text by Pantelis Golitsis. Michael of Ephesus, In PN 149.1. 10 The Arabic tradition seems to have known the MA only indirectly via Nicolaus of Damascus and his compendium of Aristotle’s philosophy. For more on this topic and in general about the reception of Aristotle’s biology in the Arabic world, I refer the reader to Cerami–Falcon 2014: 35–56. 11 For a clear and authoritative introduction to Moerbeke and his activity as a translator of Aristotle, see Brams 2003: 105–130. 12 The Book on Animals (Liber de animalibus) was a translation from the Arabic of the HA, PA, and GA. It did not contain the IA and the MA, which do not seem to have circulated in the Arabic world. 9



Part I Introduction

We can reconstruct this debate by looking at how Moerbeke’s translation of the IA circulated in Paris and beyond. 13 This translation is ­sometimes copied along with that of the MA in exemplaria that contain the HA, GA, and PA; other times it is found along with the Parva naturalia. In the first case, the assumption is that the IA contributes directly and immediately to Aristotle’s biology (presumably by being an appendix to the PA). In the second case, the situation is more complex. By placing the IA in the group of writings dealing with what is common to the soul and the body, the link with the biological corpus is not lost but the obvious and important connection with the PA is severed. I have already touched on the reasons why we do not want to consider the IA a contribution to the project attempted in the Parva naturalia.14 However, the question of its precise place in Aristotle’s larger explanatory project remains to be discussed. In this chapter, I will offer additional data by looking at the order in which the treatises on natural philosophy are printed in the early editions of Aristotle. Those data will prepare the groundwork for a more in-depth discussion of the place of the IA in Aristotle’s explanatory work, which will be offered in the course of the analysis of the opening lines of the treatise.15 William of Moerbeke was a pioneer but his translations were soon felt to be inadequate precisely for the reason we like them: they were considered too literal and at times even clumsy. His translation of the IA was no exception to the rule. Not surprisingly, there were a few attempts to produce a better translation of this work in the Renaissance. By far the most successful translation of the IA was the one produced by Niccolò Leonico Tomeo (Nicolaus Leonicus Thomaeus). Leonico Tomeo was an interesting scholar in his own right. Appointed professor of philosophy at Padua University in 1497 (where he taught Aristotle in the original language for about ten years), Leonico Tomeo concentrated his exegetical efforts (if not his teaching) on texts that had been left at the margin of scholarly interest. In addition to a translation and a commentary of the Parva naturalia that included the IA and the MA,16 he produced a translation of the 13

The relevant codicological data can be found in De Leemans 2011: lxi–lxxix. For a discussion of their significance, I refer the reader to Falcon 2012. On the formation, organization, and structure of the medieval corpus aristotelicum see also Beullens–De Leemans 2007: 87–135. 14 See Introduction, p. 7. But I refer the reader to the interpretative essay by Klaus Corcilius (ch. 5) for a fuller discussion of the reasons why the IA does not belong to the study of what is common to the body and the soul. 15 See “The Theoretical Framework and the Beginning of the Actual Investigation” (ch. 3). 16 The full title of this work, published in 1523, is: Parva naturalia explicata. De sensu et sensibili, De memoria et reminiscientia, De somno et vigilia, De insomniis, De divinatione per somnia, De animalium motione, De animalium incessu, De extensione et brevitate vitae, De iuventute et senectute, morte



2 The Reception of the De incessu animalium



Mechanica.17 It should not be overlooked that Leonico Tomeo translated and commented on the IA and the MA as part of his exegetical project on the Parva naturalia. At this point, we are entitled to see in this decision a deliberate editorial choice that reveals how the MA and IA are understood to contribute to Aristotle’s natural philosophy. We know of two other attempts to replace the translation of the IA produced by William of Moerbeke. The first, by Pietro Alcionio (Petrus Alcyonius), was published in 1521. It was not well received. Quite the opposite: Jan Ginés de Sepúlveda (Johannes Genesius Sepulveda) offered his own alternative translation of the IA in 1522, as well as a list (lost) of the errors made by Alcionio in his translation of Aristotle (Errata Petri Alcyonii in interpretatione Aristotelis). What was at stake in this humanist debate was how Aristotle ought to be translated in order to meet the scholarly standards and literary tastes of the time. I will not elaborate further on this debate. Here suffice it to say that both attempts were eclipsed by the publication in 1523 of the translation offered by Niccolò Leonico Tomeo. It is telling that Francesco Patrizi in his learned Discussiones peripateticae feels no need to recall any other interpreter of the Parva naturalia and says so explicitly.18 But even more impressive is the fact that his translations of the MA and the IA imposed themselves quite independently of his translation and commentary of the Parva naturalia. For example, both translations were adopted in the first bilingual edition (Greek and Latin) of Aristotle edited by Isaac Casaubon in 1590.19 I will return to the significance of this important edition of Aristotle below. For the time being, I am content to stress that Leonico Tomeo’s translations of the IA and the MA remained in use well beyond the Renaissance, as is documented by Bekker’s decision to reprint them in the third volume of his Berlin edition of Aristotle.20 It is thanks to the influence of Leonico et via et de spiratione. Omnia in latinum conversa et antiquorum more explicata a N. Leonico Thomaeo. Venetiis (apud Bernardinus et Mattheus, fratres Vitales). His translations of the MA and IA were reprinted together with the Mechanica in the Opuscula in 1525. This editorial project was quite successful. The Opuscula circulated widely in the Renaissance and beyond. Renaissance knowledge and use of the Mechanica is largely due to this edition. 18 Patrizi 1581 [1999]: talis est etiam Nicolai Leonici expositio in parvos naturales ac fortasse alterius cuius­piam nec enim numerare aut vacat aut est necesse (I, XXI, 149). 19 The full title of this edition, which was printed in Lyons, is Operum Aristotelis nova editio, graece et latine. Graecus contextus quam emendatissime praeter omnes omnium editiones est editus, adscriptis ad oram libri et interpretum veterum recentiorumque et aliorum doctorum virorum emendationibus: in quibus plurimae nuc primum in lucem prodeunt, ex bibliotheca Isaaci Casauboni. Lugduni, apud Guillelmum Laemarium. For the Parva naturalia, Casaubon adopted the translations produced by François Vatable. 20 Bekker 1831: vol. 3. 17



Part I Introduction

Tomeo’s translation that the title De incessu animalium eventually replaced the earlier title De ingressu animalium.21

The IA in the Early Printed Editions of Aristotle At this point, it should be abundantly clear that the position of the IA within the biological works of Aristotle remained quite unstable well beyond the rediscovery of the treatise, which was due to Moerbeke’s translation. In this section I would like to look at the position of the IA in the early printed editions of Aristotle. The editors and printers who concerned themselves with the Aristotelian corpus had to answer questions that are immediately relevant to the debate on the order and organization of the writings on natural philosophy. We can reconstruct their answers to those questions by looking at their editorial decisions. For a first idea of the variability and latitude in the organization of the writings on natural philosophy, we can compare the first two bilingual editions (Greek and Latin) of Aristotle: the one published by Isaac Casaubon in 1590 and the other put together by Julius Pacius (Giulio Pace) in 1597.22 Both editions were printed by Guillelmus Laemarius (Guillaume de Laimarie) in Lyons.23 John Glucker has discussed the complex relation between these two editions (and the two men), so I will begin my brief discussion of these two editions by recalling the main conclusions reached in his article.24 Pacius undertook the project of a new edition of Aristotle for Laemarius in 1585 (or even in 1584). At the time, he had just completed his bilingual edition of Aristotle’s Organon.25 In all probability, the early success of this edition prompted him to engage in the more ambitious project of a bilingual edition of the complete works by Aristotle. Pacius must have worked on this editorial project before 21

The reader will find further information on the reception of the IA in the Latin world in a series of learned articles by Pieter De Leemans (De Leemans 1999: 199–218; De Leemans 2004: 165–185; and De Leemans 2006: 525–541). 22 The full title of this second bilingual edition is: Operis Aristotelis nova editio, graece et latine. Latinae interpretationes Graeco contextui convenientiores et emendatiores, quam antehac editae sunt. Lugduni, apud Guillelmum Laemarium. It was reprinted in 1605 and 1607. 23 In fact, Guillaume de Laimarie (1533–1598) printed his books in Geneva. However, it was the rule at the time for the printers in Geneva either to print their books with a false address or to omit this information altogether in order to circumvent the French law which prohibited circulation in France of books published in reformed territories. Laimaire used Lyons for his editions of Aristotle. 24 Glucker 1964: 274–296. 25 Pacius’ bilingual edition of the Organon was published by Laemarius in 1584 (Aristotelis Organum, hoc est, libri omnes ad logicam pertinentes, Graece et Latine. Iulius Pacius recensuit, e lingua Graeca in Latinam linguam convertit, necnon notis illustravit).



2 The Reception of the De incessu animalium



leaving Geneva for Heidelberg in 1585. Unable to complete his project, and under pressure from Laemarius, Pacius asked his former student and friend Isaac Casaubon to take over his project. The result was the first complete bilingual edition of Aristotle published by Casaubon in 1590. Pacius was able to return to his original project when Laemarius wanted to reprint this edition in a smaller (and more affordable) edition. Pacius adopted both the text and the bilingual format of Casaubon’s edition, but he did away with the marginalia to the Greek text, which were one of the most interesting features of Casaubon’s edition (and arguably his only contribution to the text). More importantly, however, Pacius rearranged the writings on natural philosophy. His reorganization of the biological writings was quite extensive, as we can infer from Table 2.1.26 Table 2.1  Writings on natural philosophy in Pacius’ and Casaubon’s bilingual edition of Aristotle: a comparison

26

Operum Aristotelis nova editio. Graece et Latine. Ex bibliotheca Isaaci Casauboni Lugduni (= Lyons) 1590 Guillelmus Laemarius

Operis Aristotelis nova editio. Graece et Latine. Edidit Julius Pacius Lugduni (= Lyons) 1597 Guillelmus Laemarius

Physics DC GC Meteorologica De mundo

Physics DC GC Meteorologica

DA PN1 MA PN2 IA De spiritu HA PA GA

HA PA IA DA PN1 MA GA PN2

Hereafter I adopt the following, additional, abbreviations: PN1 = De sensu et sensibili/De memoria et reminiscentia/De somno et vigilia/De insomniis/De divinatione per somnum. PN2 = De longitudine et brevitate vitae/De iuventute et senectute, De vita et morte, De respiratione.



Part I Introduction

While Casaubon does not address the topic of the organization of the Aristotelian corpus (either in general or with specific reference to the writings on natural philosophy), Pacius is forced to do so because he not only departs from the organization adopted by his predecessor but also corrects him in more than one way. Here is what he says, in his “Letter to the Reader,” on the topic of the arrangement of the biological writings: And here is their order: the books on the history of animals are to be placed first because they say the hoti, while other books deal with the dioti. As for what pertains to the latter: the matter of animals is considered in the books on the parts of animals and the progression of animals, which is to say in the books that are concerned with the parts that pertain to the progression of animals. Concerning form, it is discussed in the books on the soul. The proper accidents of animate bodies are: sense-perception, memory and recollection, sleep and waking, dreams and divination in sleep, motion, generation, length and brevity of life, youth and old age, death, health and illness.

Pacius relies on the ideas that animals are hylomorphic compounds and that their study amounts to a study of their matter and their form as well as their per se accidents. The study of the matter and form has explanatory priority: it is only when we know matter and form that we know the nature and essence of something. It is only at that point that we can engage in a study of the proper accidents (accidentia propria). Among the latter, we find all the activities (operationes) discussed in the context of the Parva naturalia as well as the topic of generation (GA). Pacius makes it clear that the PA combined with the IA deal with the matter/body of animals, whereas the DA is concerned with their form/soul. Taken together, these three works give us a complete study of the nature of animals. What Pacius is recalling in this passage is a view of the Aristotelian corpus that is far from being original. It was already adopted in the Aldine edition, the Greek editio princeps of Aristotle (see Table 2.2). Aldus Manutius, the printer of the Aldine edition, does not explain, let alone defend, his arrangement of the biological writings. However, in the prefatory letter to his third volume, he tells us that he sought the help and advice of Francesco Cavalli, whom he describes as a man of deep learning, a great expert in philosophy, and an outstanding doctor in Venice.27 He also mentions the essay that the latter wrote on the number and order of Aristotle’s writings on natural philosophy (De ordine et numero partium ac librorum physicae doctrinae Aristotelis, published by 27

His letter can now be found in an English translation in Wilson 2016: 47–50.



2 The Reception of the De incessu animalium



Table 2.2  Writings on natural philosophy in the Aldine edition of Aristotle Aristotelis Opera Graece Venetiis 1495–1498 Aldus Manutius

Aristotelis Opera Graece Venetiis 1495–1498 Aldus Manutius

Vol 2

Vol 3

[…]

HA PA IA DA PN1 MA GA PN2

Physics DC GC Meteorologica De mundo Philoni de mundo Theophrasti de igne, de ventis, de signis, de lapidibus

Capcasa in Venice between 1490 and 1495 [GW 5832]). In this essay, Cavalli concerns himself with the via et ordo of Aristotle’s natural philosophy. In so doing, he provides us with the reasoning behind the arrangement adopted in the Aldine edition. His reasoning is similar to that of Pacius at the outset of his own edition of Aristotle, so we do not really need to elaborate on it. Suffice it to note that Cavalli stresses the close connection between the PA and the IA.28 In particular, he regards the latter work as an appendix to the PA. He finds the textual basis for this view in the closing words of the IA. There, the investigation conducted in the whole work is described as a study of the parts involved in animal locomotion, completing what is said on the topic of the parts of animals in the PA. The final words of the IA are a notoriously difficult text. Reference is made to the research into the soul conducted in the DA, which is presented as the next instalment in Aristotle’s explanatory project. This contradicts what we read at the end of the PA, where reference is made to the GA. Cavalli severs the connection between the PA and the GA. He condemns the passage at the end of the PA, where Aristotle announces a study of the generation of animals. By his lights, this passage 28

Cavalli speaks of the IA as liber de progressu vel processu seu incessu. By so doing, he acknowledges the existence of a plurality of translations of the Greek title of the work.



Part I Introduction

is either a later editorial addition or a corruption of the text. His treatment of the whole issue turns out to be quite naïve: it is far from obvious that the explicit of the GA is an editorial addition while the one at the end of the IA is not. For all we know, neither one of the two passages may be by Aristotle. We need, at the very least, a good reason if we are to retain one and reject the other. The organization of the biological corpus recommended by Cavalli, and adopted first by Aldus and then by Pacius, was quite common in the Renaissance. It goes all the way back to the Arabic tradition and is endorsed (among others) by Averroes. As a result, it is not surprising to see this order adopted in the Juntine edition of Aristotle–Averroes (1562– 1574), which represents the culmination of the exegetical tradition of reading and interpreting Aristotle solely in Latin.29 Still, it is far from being the only possible arrangement of the Aristotelian writings, as the bilingual edition of Casaubon clearly shows. At least in this case, however, Casaubon was not an innovator; rather, he relied on an alternative tradition that we find documented in the Basel edition of Aristotle. This is the second Greek edition of Aristotle. It was printed by Johann Bebel (Bebelius) in 1531 and 1539; it was printed for a third time by Johann Bebel and Michael Isengrin (Isingrinus) in 1550. The best known and most important innovation adopted in this edition (but introduced only in the 1550 reprint) was the division of the Aristotelian text into chapters and the addition of short informative headings to improve the overall readability of the Greek text. The Aristotelian writings on natural philosophy are arranged as shown in Table 2.3. This alternative arrangement of the biological works largely depends on a different understanding of the place of the DA in Aristotle’s explanatory work. Aristotle was the first to write on the soul as such. Arguably, the ultimate reason for this innovation is to be found in Aristotle’s interest in life, which was an interest in all forms of life. An essential part of this project was to gain clarity about the phenomenon of life. If one takes this view, one may be led to the conclusion that the study of the soul conducted in the DA plays a foundational role with respect to the study of life. At a minimum, it supplies the conceptual apparatus for the study of life, which is a prerequisite for embarking on an optimal study of life in general and of animal life in particular.30 While we do not have any 29

For a recent study of this editorial project and its significance for the history of Aristotelianism, I refer the reader to Schmitt 1984: 121–142 and Burnett 2012: 55–64. 30 The thought that the DA, because of its foundational role for the study of life, is to be placed before the study of animals is fairly common in the Aristotelian tradition. In the fifteenth and



2 The Reception of the De incessu animalium



Table 2.3  Writings on natural philosophy in the Basel edition of Aristotle Aristotelis opera Per Des(iderium) Eras(mum) Roterodamum (et Sim. Grynaeum) Basileae, Bebelius 1531, 1539 Basileae, Bebelius et Isingrinus 1550 Vol 1 Physics DC GC Meteorologica DA PN1 MA PN2 IA De spiritu GA PA HA

explicit statement on this topic by Erasmus in his preface to the Basel edition, we can imagine that this conclusion played an important role in the decision to place the DA and the Parva naturalia augmented by the MA and the IA (as well as the pseudo-Aristotelian De spiritu) before the study of animals. Of course, the price to pay for this editorial decision is the severance of the link between the PA and the IA. I would like to end this brief overview of the early printed editions of Aristotle by recalling the arrangement adopted by Immanuel Bekker in his Berlin edition of Aristotle. This is not only the first modern edition of Aristotle but also the most common entry point into the Aristotelian corpus. As such, it is still an immensely influential point of reference. Interestingly enough, Bekker prints the PA and the IA together, while at the same time recognizing the foundational role of the DA and the link between this work and the Parva naturalia (Table 2.4). sixteenth centuries, it was common to trace this idea back to Avicenna, who placed the study of the soul as the sixth instalment in his study of nature before the study of plants and the study of animals.



Part I Introduction

Table 2.4  Writings on natural philosophy in Bekker’s edition of Aristotle Aristotelis opera Ex recensione Immanuelis Bekkeri Edidit Academia Regia Borussica Berolini 1831 Vol 2 Physics DC GC Meteorologica De mundo DA PN1 PN2 De spiritu HA PA MA IA GA

The arrangement that Bekker offers should be read in light of what Aristotle says at the outset of the MA.31 There, Aristotle envisions the integration of two accounts of animal motion: the first is a detailed account of the different modes of locomotion, and the second explains animal motion in general by reference to a common cause. While the first is traditionally taken to be the detailed account of animal locomotion offered in the IA, the second is the account that Aristotle offers in the MA. Hence, there is a clear textual basis for the proximity that Aristotle establishes between the IA and the MA. However, the most obvious downside to this arrangement is that the MA is isolated from the DA and the Parva naturalia. It is easy to see that all the early editors of Aristotle have tried to save this link. In this respect, Bekker is an exception to the rule.

31

MA 1, 698a1–7. For a translation and discussion of this text, I refer the reader to the first interpretative essay (ch. 3). See also Falcon 2017a: 215–235.



2 The Reception of the De incessu animalium



This brief review of the early printed editions of Aristotle suggests that there was no consensus on the position in the Aristotelian corpus of a few works beyond the IA. For instance, the position of the DA is at least as variable as the position of the IA. At this point, it should be clear that the different position reflects a different understanding of how the work contributes to the overall project of natural investigation. A discussion of the place of the DA in the Aristotelian corpus remains outside the scope of this volume. I will attempt a fuller discussion of how Aristotle conceives of the position of the IA in discussing its opening statement.32 32

See “The Theoretical Framework and the Beginning of the Actual Investigation” (ch. 3). But the reader should also see what Klaus Corcilius says on this topic in connection with his reading of IA 6, as well as the discussion of the epilogue of the IA by Pantelis Golitsis.

part ii

Greek Text and Translation

Preface to the Greek Text Pantelis Golitsis

Werner Jaeger published in 1913 what became the standard edition of the IA.1 During a trip in Italy in 1911–1912, which he chiefly undertook for his edition of Gregory of Nyssa’s works,2 Jaeger also made fresh collations in situ of four of the five Byzantine codices that Immanuel Bekker collated for the editio bekkeriana. 3 The five codices are the following (arranged in alphabetical order): (1) Laurentianus plut. 81,1 [S], to be dated to the last quarter of the thirteenth century (dated by Jaeger to the twelfth/thirteenth century);4 (2) Oxoniensis Corpus Christi 108 [Z], possibly the oldest extant Aristotelian manuscript,5 from the ninth century (inspected by Jaeger on photographs and dated to the twelfth century); (3) Vaticanus gr. 260 [U], from the late eleventh century (dated by Jaeger to the twelfth century); (4) Vaticanus gr. 261 [Y], copied by the Byzantine philosopher George Pachymeres and his collaborators6 in the beginning of the fourteenth century (dated by Jaeger to the thirteenth century); (5) Vaticanus gr. 1339 [P], copied by the monk Ioasaph around the middle of the fourteenth century (dated by Jaeger, following Eduard Schwartz and Giovanni Mercati, to the twelfth century).7

1

Jaeger 1913. See the praefatio to his edition. 3 Bekker 1831: vol. 1, 704a4–714b23. 4 On this manuscript, see Golitsis 2018: 463–469. 5 Wilson 2011: MS 108. 6 The part containing the IA (ff. 84v–97v) was copied by Anonymus x; see Golitsis 2010: 168. 7 See Gamillscheg et al. 1997: n. 343. 2





Part II Greek Text and Translation

Jaeger corrected the few errors committed by Bekker,8 while he introduced some further mistakes of his own.9 However, Jaeger did not reproduce in his apparatus criticus all the variants collected by Bekker but made a selection from them. The rationale behind his selective reading of Bekker’s apparatus is left unexplained in Jaeger’s preface. Both editors did not record in their apparatuses the careful corrections made to Z by a scholar from the twelfth century (Z2).10 Jaeger separated the five manuscripts selected by Bekker into two classes, (i) Z Y and (ii) P S U, to which he added the testimony of Michael of Ephesus’ commentary (ca. 1120), published some years earlier by Michael Hayduck.11 Chiefly due to the incorrect dating of the manuscripts, Jaeger considered Michael’s indirect testimony to be older than the direct testimonies (whereas both Z and U are older than Michael’s commentary) and thus privileged in his edition the convergence of Michael’s commentary with one of his two classes of manuscripts. This, however, is an erroneous editorial principle, since, as is also the case with the rest of his commentaries, Michael had access to more than one manuscript belonging to different branches of the textual tradition. For the present edition I checked anew the five manuscripts, I distinguished as carefully as I could between Z1, Z2, and Z3 (a further corrector of Z from the fifteenth century),12 and I added to these testimonies three new witnesses: (6) Laurentianus plut. 87,4 [Ca], copied by Ioannikios toward the middle of the twelfth century;13 (7) Vaticanus gr. 258 [N], copied by Ioannes Proasteiotes Vardales in the first quarter of the fourteenth century;14 705a13 τῷ ἄνω S; 705a16 πλοῖον Z; 706a4 κινήσεται Z; 707b10 τὸ om. Y; 709b30 γε habet Y; 710a14 ἀποτείναντα S; 711a27 προσίεται Z; 712b34 ἔδει Y; 713a14 τὰ om. Y; 713b21 βλαισοῦται Z; 713b25 τε habent U S P. Further errors in Bekker reproduced by Jaeger: 708b1–2 ὅλως ουδέν U; 709b30 ἡ om. Y; 713b18 χελῶναι Z. 9 705a16 οἶ (sic) ἔχοντες Z; 705b12 αὑτὸ S; 707a10 ταῦτ᾽ Y; 713b25 τε om. P; 714b23 ζωῆς Z2. The following readings are wrong in both Bekker and Jaeger: 707a31 γὰρ om. Z; 708a6 ἐν τῇ γῇ οm. Z; 709a21 κάμψαντα δ᾽ Z; 709b7 συγκαμπτόν Z; 710a4–5 τοῖς σχιζoπτέτέροις S; 710a8 πλοῖον ἀπήδαλον P; 710b32 τὸ σῶμα habet Z; 713b5 ἔχειν Z; 713b31 μέρος U; 714a9 μέντοι Z. Note that the words συγκαμπτόν and ἀπήδαλον are recorded in LSJ according to their sole occurrence in the IA (as edited by Bekker and Jaeger). 10 On this scholar, see Golitsis 2014: 33–52. 11 Hayduck 1904. 12 I have some hesitations as to whether a fourth corrector, contemporary to Z2, intervened on the manuscript. 13 Among several studies dealing with this copyist, see Wilson 1991: 447–455 and Degni 2008: 179– 248. 14 See Gamillscheg et al. 1997: n. 296. 8



Preface to the Greek Text



(8) Vaticanus gr. 266 [V], from the first quarter of the fourteenth century. I selected these manuscripts by taking into account Frederike Berger’s very useful study of the manuscript tradition of the IA.15 The stemma codicum that I arrived at, however, differs from hers in some of its details (especially with regard to the stemmatic position of S and the so-called “contaminations”):  

 Z

 Z2 Ca



U

*

o Y

S V

N

P Z3

As happens with other Aristotelian treatises, most copies of the IA turn out to be copies of some manuscript collated with some other.16 From the point of view of textual criticism, the most important observation is that, at several places, γ appears to preserve the authentic reading against the common reading of Z and U. This is due either to corrections ex ingenio (which is less probable) or, more plausibly, to marginal readings in α that were not taken into account by the amanuensis copyist who is responsible for Z. Thus, the present edition is a new critical edition of the IA. It records in its critical apparatus all divergences between the independent witnesses of Aristotle’s text,17 while it also provides the choices of the previous editors (Bekker, Jaeger) whenever they assessed the authenticity of the 15

Berger 1993: 23–42. Berger thinks that there are fourteen independent manuscripts of the IA. I think that this is too high a number. But even if it were true, the six manuscripts that have not been taken into account in the present edition are unlikely, due to their stemmatic position, to yield genuine readings of the IA. It should also be noted that the stemma codicum proposed by Jaeger for the five manuscripts (see his praefatio: xvi) is not very remote from the present one. (The manuscript Berolinensis Phillips 1507, which is of particular importance for the constitutio ­textus of the MA, has no value for the constitutio textus of the IA; it is quasi-identical to P.) 16 In Golitsis 2015: 1–23, I argue that it is preferable to speak, not of contaminated manuscripts, but of collated manuscripts. 17 The manuscripts are arranged in the apparatus in chronological order. Note, however, that N and V are roughly contemporary.



Part II Greek Text and Translation

variant readings differently from the present editor.18 Contrary to the editorial practice of the nineteenth and early twentieth centuries, the present apparatus is positive (except for cases in which a manuscript transmits an isolated reading). Here is a list of the differences between the readings adopted in the present edition and those preferred by Jaeger: 704a23 τὰ τετράποδα καὶ ζῳοτόκα] τὰ τετράποδα τὰ ζώοτοκα Jaeger 704b8 τούτων ἁπάντων] πάντων τούτων Jaeger 705a16 πλεῖον] πλέον Jaeger 705a28 τὸ ἔμπροσθεν καὶ ὄπισθεν] τὸ ἔμπροσθεν καὶ τὸ ὄπισθεν Jaeger 705a28 τὸ δεξιὸν καὶ ἀριστερόν] τὸ δεξιὸν καὶ τὸ ἀριστερόν Jaeger 705b20 ἑκάστῳ] ἑκάστων Jaeger 706a7 ἐὰν] ἂν Jaeger 706a29–30 τό τε πρόσθεν] τὸ πρόσθεν Jaeger 706b21 τι κοινόν] τὸ κοινόν Jaeger 706b23 δῆλον ὅτι] δηλονότι Jaeger 706b27 ἀριστερῶν] τῶν ἀριστερῶν Jaeger 706b28 ὄπισθεν (pr.)] τῶν ὄπισθεν Jaeger 706b28 ὄπισθεν (alt.)] τὸ ὄπισθεν Jaeger 707a8 δεῖ γ’] δεῖ δ’ Jaeger 707a9 ἐφ’ ἑκάστου] ἀφ’ ἑκάστου Jaeger 707a14 ἄνω τε] ἄνω Jaeger 707b5–6 σημείοις πέφυκε κινεῖσθαι] κινεῖσθαι πέφυκε σημείοις Jaeger 707b7 ὅσα τῶν ἐναίμων] τῶν ἐναίμων ὅσα Jaeger 707b10 καὶ τὸ πρόσθιον] τὸ πρόσθιον Jaeger 707b27 δεξιὸν ἐφ’ οὗ Γ, ἀριστερὸν ἐφ’ οὗ Δ] ἀριστερὸν ἐφ᾽ οὗ Γ, δεξιὸν ἐφ᾽ οὗ Δ Jaeger 708a4 ἐγχέλεις] ἐγχέλυες Jaeger 708b10 ἀλλ’ οὐ βάδισις] del. Jaeger 708b17 πορευόμενα] πορευόμενον Jaeger 709a5 ἔχων … ἁπτόμενον addidi] lacunam indicavit Jaeger 18

I also took into account Pieter De Leemans’ very informative study of William of Moerbeke’s Latin translation (De Leemans 2011) of a lost Greek manuscript (Γ) somehow related, as also occurs with other translations by Moerbeke, to Ioannikios’ copy (Ca). However, I do not share his conviction that Moerbeke’s translation may help us reconstruct the genuine Greek text at 708b14 and 709a4. I cannot explain why Aristotle would introduce a variation between 708a8 (ἀνισάζειν) and 708b14 (ἰσάζειν), while the reading ἐγγὺς (instead of ἐν γῇ) at 709a13 appears to be redundant with regard to παρά (which denotes per se closeness).



Preface to the Greek Text 709a20 τὸ προϊόν] τό τε προϊόν Jaeger 709b7 συγκάμπτον] συγκαμπτὸν Jaeger 710a8 ἀπηδάλιον] ἀπήδαλον Jaeger 710a29 οἰκείαν κίνησιν] ὠκεῖαν κίνησιν Jaeger 711a7 οὐθὲν] οὐδὲν Jaeger 711a11 μηθενὸς] μηδενὸς Jaeger 711a18 οὐθὲν] οὐδὲν Jaeger 711a28 τοὔμπροσθεν] τὸ ἔμπροσθεν Jaeger 711b12 ζῳοτόκα] τὰ ζώοτοκα Jaeger 712a14 ὀλέκρανον] ὠλέκρανον Jaeger 712b13 γὰρ] δὲ Jaeger 713a12 διαστέλλοντα] διαστέλλοντα Jaeger 713a16 τρωγλοδυτικὰ] τρωγλόδυτα Jaeger 713b18 τρωγλοδυτικὰ] τρωγλόδυτα Jaeger 713b32 ἡγουμένους] ἡγεμόνας Jaeger



Sigla Manuscriptorum

Ca = Laurentianus plut. 87,4 (secundo quarto saeculi XII ab Ioannicio exaratus) N = Vaticanus gr. 258 (saeculo XIII exeunti exaratus) P = Vaticanus gr. 1339 (saeculo XIV exaratus) S = Laurentianus plut. 81,1 (ca. 1280–1290 exaratus) U = Vaticanus gr. 260 (saeculo XI exeunti exaratus) V = Vaticanus gr. 266 (saeculo XIV ineunti exaratus) Y = Vaticanus gr. 261 (ca. 1310 a Georgio Pachymere et collaboratoribus exaratus) Z = Oxoniensis Corp. Christi 108 (saeculo IX exeunti exaratus)

Conventions : angle brackets enclose additions to the transmitted Greek text deriving from parallel sources or editorial conjectures. In the translation, they are used to mark amplifications. [...]: square brackets enclose words or phrases that may have been added to the original Greek text.



ΑΡΙΣΤΟΤΕΛΟΥΣ ΠΕΡΙ ΠΟΡΕΙΑΣ ΖΩΙΩΝ

[1] Περὶ δὲ τῶν χρησίμων μορίων τοῖς ζῴοις πρὸς τὴν κίνησιν τὴν κατὰ τόπον ἐπισκεπτέον διὰ τίνα αἰτίαν τοιοῦτόν ἐστιν ἕκαστον αὐτῶν καὶ τίνος ἕνεκεν ὑπάρχει αὐτοῖς, ἔτι δὲ περὶ τῶν διαφορῶν τῶν τε πρὸς ἄλληλα τοῖς τοῦ αὐτοῦ καὶ ἑνὸς ζῴου μορίοις καὶ πρὸς τὰ τῶν ἄλλων τῶν τῷ γένει διαφόρων. πρῶτον δὲ λάβωμεν περὶ ὅσων ἐπισκεπτέον. ἔστι δὲ αὐτῶν ἓν μὲν πόσοις ἐλαχίστοις τὰ ζῷα κινεῖται σημείοις, ἔπειτα διὰ τί τὰ μὲν ἔναιμα τέτταρσι τὰ δ’ ἄναιμα πλείοσι, καὶ καθόλου δὲ διὰ τίν’ αἰτίαν τὰ μὲν ἄποδα τὰ δὲ δίποδα τὰ δὲ τετράποδα τὰ δὲ πολύποδα τῶν ζῴων ἐστί, καὶ διὰ τί πάντα ἀρτίους ἔχει τοὺς πόδας, ὅσαπερ ἔχει πόδας αὐτῶν, ὅλως δ’ οἷς κινεῖται σημείοις, ἄρτια ταῦτ’ ἐστιν· ἔτι δὲ διὰ τίν’ αἰτίαν ἄνθρωπος μὲν καὶ ὄρνις δίπους, οἱ δ’ ἰχθύες ἄποδές εἰσι· καὶ τὰς κάμψεις ὅ τε ἄνθρωπος καὶ ὁ ὄρνις δίποδες ὄντες ἐναντίας ἔχουσι τῶν σκελῶν (ὁ μὲν γὰρ ἄνθρωπος ἐπὶ τὴν περιφέρειαν κάμπτει τὸ σκέλος, ὁ δ’ ὄρνις ἐπὶ τὸ κοῖλον· καὶ ὁ ἄνθρωπος αὐτὸς ἑαυτῷ ἐναν- τίως τὰ σκέλη καὶ τοὺς βραχίονας· τοὺς μὲν γὰρ ἐπὶ τὸ κοῖλον, τὰ δὲ γόνατα ἐπὶ τὴν περιφέρειαν κάμπτει. καὶ τὰ τετράποδα καὶ [τὰ] ζῳοτόκα τοῖς τ’ ἀνθρώποις ἐναντίως κάμ-

704a5

10

15

20

704a3 Ἀριστοτέλους Περὶ πορείας ζῴων Z U S : Ἀριστοτέλους Περὶ ζῴων πορείας Ca P : Περὶ ζῴων πορείας V : om. N Y   4–5 τὴν κίνησιν τὴν κατὰ τόπον ἐπισκεπτέον] τὴν κατὰ τόπον ἐπισκεπτέον κίνησιν U   6 ἕνεκεν] ἕνεκα Z | ὑπάρχει U S V Y : ὑπάρχουσιν Z Ca P : ὑπάρχ´ N | δὲ Zs.l. Ca S Y N V P : τε Z U   7 τοῖς] om. V | τοῦ] οm. N | καὶ ἑνὸς ζῴου] ζῴου καὶ ἑνὸς U   8 τῶν (alt.)] οm. Ca | διαφόρων Z U Y P : διαφορῶν Ca S N V   9 ἐπισκεπτέον Z U Ca S N V : ἐστιν ἐπισκεπτέον Y P   10 αὐτῶν ἓν μὲν Ca U Z2 S N V Jaeger : πρῶτον μὲν Z P Bekker : πρῶτον μὲν αὐτῶν Y   11 ἔπειτα Z U S Y P : ἔπειτα δὲ Ca N V | τέτταρσι Z U Y N V P : τέτρασι Ca S   13 δίποδα τὰ δὲ Y P : om. Z U Ca S N V   14 τοὺς] om. Y   16 δίπους Z Y N V P : δίπουν Ca : διττοὺς U S   16–17 οἱ δ᾽ ἰχθύες Z Y P : ἰχθύες δὲ Ua.c. Ca N V : οἱ δ᾽ ἰχθύες δὲ Up.c. S   18 ὁ ὄρνις Z U N, ex oἱ ὄρνις V : οἱ ὄρνις Y : οἱ ὄρνιθες Ca : ὄρνις P | τῶν] om. N   19 γὰρ] om. Ca | ἐπὶ Z S P : περὶ U Z2 s.l. Ca Y N V   20 αὐτὸς Z U Ca S N V : δ᾽ αὐτὸς P Y | ἑαυτῷ Z U S Jaeger : αὑτῷ Ca N V P Bekker : αὐτῷ Y   21–22 τὸ κοῖλον] τὰ κοῖλα Z   23 τὰ τετράποδα καὶ scripsi (cf. HA 497b13; IA 711b12) : τὰ τετράποδα καὶ τὰ U Ca S N V P : τετράποδα τὰ Z1 : τὰ τετράποδα τὰ Z2 Y Bekker Jaeger | τ᾽ Z U S N Y P : om. Ca V | κάμπτει U Ca S V N : κάμπτουσι Z P Y  



Aristotle On the Progression of Animals

This translation is a joint work. Each author translated her or his chapter(s) of the IA. Andrea Falcon revised the entire translation with the goal of making it as uniform as possible. [1] As for the parts that are useful to animals for motion with respect to 704a4 place, we must investigate why each part is such as it is and for what end it belongs to them, and also the differences in the parts of one and the same animal and compared to the parts of animals different in kind. First of all, however, let us grasp how many questions are to be investi- 704a9 gated. One of these is what the fewest points by means of which animals move are; then, why blooded animals move by means of four points while bloodless animals by means of more than four; and, in general, why some animals are footless, some two-footed, some four-footed, and some many-footed, and why as many as have feet have an even number of them. In general, why the points by means of which animals move are even in number. A further question is why a human being and a bird are two-footed 704a16 whereas fish are footless, and why a human being and a bird, although they are both two-footed, have opposite bendings of their legs. For a human being bends its leg toward the circumference while a bird toward the concave, and a human being bends its legs and arms in opposite directions with respect to itself: arms toward the concave and knees toward the circumference. Furthermore, the four-footed animals that are also live-bearing bend their limbs in the opposite ways with respect to





1, 704b4–2, 705a2

πτει καὶ αὐτὰ αὑτοῖς· τὰ μὲν γὰρ πρόσθια σκέλη ἐπὶ τὸ κυρτὸν τῆς περιφερείας κάμπτει, τὰ δ’ ὀπίσθια ἐπὶ τὸ κοῖλον. ἔτι δὲ τῶν τετραπόδων ὅσα μὴ ζῳοτοκεῖ ἀλλ’ ᾠο- τοκεῖ, ἰδίως καὶ εἰς τὸ πλάγιον κάμπτει)· πρὸς δὲ τούτοις διὰ τίν’ αἰτίαν τὰ τετράποδα κινεῖται κατὰ διάμετρον. περὶ δὴ τούτων ἁπάντων, καὶ ὅσα ἄλλα συγγενῆ τούτοις, τὰς αἰτίας θεωρητέον. ὅτι μὲν οὖν οὕτω ταῦτα συμβαίνει, δῆλον ἐκ τῆς ἱστορίας τῆς φυσικῆς, διότι δέ, νῦν  σκεπτέον. [2] Ἀρχὴ δὲ τῆς σκέψεως ὑποθεμένοις οἷς εἰώθαμεν χρῆσθαι πολλάκις πρὸς τὴν μέθοδον τὴν φυσικήν, λαβόντες τὰ τοῦτον ἔχοντα τὸν τρόπον ἐν πᾶσι τοῖς τῆς φύσεως ἔργοις. τούτων δ’ ἓν μέν ἐστιν ὅτι ἡ φύσις οὐθὲν ποιεῖ μάτην, ἀλλ’  ἀεὶ ἐκ τῶν ἐνδεχομένων τῇ οὐσίᾳ περὶ ἕκαστον γένος ζῴου τὸ ἄριστον· διόπερ εἰ βέλτιον ὡδί, οὕτως καὶ ἔχει κατὰ φύσιν. ἔτι τὰς διαστάσεις τοῦ μεγέθους, πόσαι καὶ ποῖαι ποίοις ὑπάρχουσι, δεῖ λαβεῖν· εἰσὶ γὰρ διαστάσεις μὲν ἕξ, συζυγίαι δὲ τρεῖς, μία μὲν τὸ ἄνω καὶ τὸ κάτω, δευτέρα  δὲ τὸ ἔμπροσθεν καὶ τὸ ὄπισθεν, τρίτη δὲ τὸ δεξιὸν καὶ τὸ ἀριστερόν. πρὸς δὲ τούτοις ὅτι τῶν κινήσεων τῶν κατὰ τόπον ἀρχαὶ ὦσις καὶ ἕλξις. καθ’ αὑτὰς μὲν οὖν αὗται, κατὰ συμβεβηκὸς δὲ κινεῖται τὸ φερόμενον ὑπ’ ἄλλου· οὐ γὰρ αὐτὸ δοκεῖ κινεῖν αὑτό, ἀλλ’ ὑπ’ ἄλλου κινεῖσθαι τὸ  ὑπό τινος φερόμενον.

704b5

10

15

20

705a1

24 αὑτοῖς Z U N Y P, ex αὐτοῖς V : αὐτοῖς Ca S | πρόσθια Z Y P : ἐμπρόσθια U Ca S N V   704b4 κάμπτει U S N V : κάμπτουσι Z Ca Y P   6 ἰδίως] ἰδίᾳ V | καὶ] om. Ca | τὸ πλάγιον U Z2 Ca S V Y P : πλαγίως Z1 : τὰ πλάγια N | κάμπτει U S Y N V : κάμπτουσι Z Ca P   8 δὴ τούτων ἁπάντων Ca N V Y : δὴ πάντων τούτων P Bekker Jaeger : δὲ πάντων τούτων Z : δὲ ἁπάντων τούτων U : τούτων δὴ ἁπάντων S   9 θεωρητέον Z U S Y P : ῥητέον Ca N V | οὖν Z U S Jaeger : γὰρ Ca Y N V P Bekker   12 ὑποθεμένοις Z Ca Y P : προθεμένοις U S N V | οἷς Z S P : ὡς U Ca Y N V   13 τὰ] om. Z, add. Z3  14 τοῦτον] τούτων P | πᾶσι Z U S Y P : ἅπασι Ca N V   15 δ᾽ ἓν μὲν U Ca Y N V P : δὴ ἓν Z : δὲ εἰ μὲν S | ἐστιν Z Y P : om. U Ca S N V | ἡ φύσις οὐθὲν ποιεῖ μάτην Z P : ἡ φύσις οὐθὲν μάτην ποιεῖ U S : ἡ φύσις ποιεῖ οὐθὲν μάτην Ca N V : οὐθὲν ἡ φύσις ποιεῖ μάτην Y   16 ἐκ τῶν ἐνδεχομένων] τὸ ἐνδεχόμενον Z | γένος ζῴου] ζῴου γένος V   17 ἄριστον Z U Ca V Y P : ἀόριστον S N | διόπερ] διὸ Z | εἰ βέλτιον ὡδί Z (ὠδή Z2 s.l.) : βέλτιον ὡδί Y : βέλτιον δὴ U Ca N S P : ὃ βέλτιον ᾖ V | οὕτως] τοῦτο V | κατὰ] τὰ κατὰ Z   18 ἔτι U Ca N V P : ἔτι δὲ Y S : ἔπειτα Z | ποῖαι U Z2 Ca Y N V P : om. Z1 S   19 γὰρ] om. N | μὲν Z Y P : om. U Ca S N V  20 τὸ κάτω Z U S : κάτω Ca Y N V P   21 τὸ ὄπισθεν Ca P : ὄπισθεν Z U S Y N V  21–22 τὸ δεξιὸν καὶ τὸ ἀριστερὸν Z U Y P : δεξιὸν καὶ ἀριστερὸν Ca S N V   23 ὦσις καὶ ἕλξις Z U S P : ὥσεις καὶ ἕλξεις Y : ἕλξις καὶ ὦσις Ca N V | οὖν Z Ca Y N V P : om. U S  24–705a2 ὑπ᾽ ἄλλου … φερόμενον Z2 U2 Ca S Y N V P : om. propter homoioteleuton Z1 U1  1 αὑτὸ Z U Ca N Y P, ex αὐτό V : ἑαυτὸ S   2 ὑπό τινος φερόμενον] φερόμενον ὑπό τινος Y



1, 704b4–2, 705a2



human beings, and also in opposite ways with respect to themselves: front legs toward the convex part of the circumference while hind legs toward the concave part. Moreover, among the four-footed animals, those that are not live-bearing but are egg-laying bend their legs in a way unique to them, namely laterally away from their body. An additional question is why four-footed animals move their legs diagonally. The causes of all these facts, and of other facts similar to these, are to 704b8 be studied: for that things are in this way is evident from natural research; we must now investigate why. [2] We begin our investigation by positing those that we are used to employing often in natural investigation, assuming that things occur in the same manner in all the works of nature. One of these is that nature does nothing in vain but always what is best from among the possibilities for the substance of each kind of animal, which is why if it is best in a certain way, it is also in this way according to nature. Furthermore, we ought to assume how many and what sort of dimensions of magnitude are present in what kinds of animals. There are indeed six dimensions but three pairs: the first is the up and the down, the second is the front and the back, and the third is the right and the left. In addition, that sources of motions in place are pushing and pulling. Now, these are sources of per se motion, whereas what is carried by something else is moved per accidens: the reason is that what is carried by something else does not seem to move itself but rather to be moved by something else.

704b12 704b15

704b18

704b22



3, 705a3–25

[3] Τούτων δὲ διωρισμένων λέγωμεν τὰ τούτων ἐφεξῆς. τῶν δὴ ζῴων ὅσα μεταβάλλει κατὰ τόπον, τὰ μὲν ἀθρόῳ παντὶ τῷ σώματι μεταβάλλει, καθάπερ τὰ ἁλλόμενα,  τὰ δὲ μορίοις, καθάπερ τῶν πορευομένων ἕκαστον. ἐν ἀμφοτέραις δὲ ταῖς μεταβολαῖς ταύταις ἀεὶ μεταβάλλει τὸ κινούμενον ἀποστηριζόμενον πρὸς τὸ ὑποκείμενον αὐτῷ. διόπερ ἐάν τε ὑποφέρηται τοῦτο θᾶττον ἢ ὥστ’ ἔχειν ἀπερείσασθαι τὸ ποιούμενον ἐπ’ αὐτοῦ τὴν κίνησιν, ἐάν θ’ ὅλως μηδεμίαν  ἔχῃ τοῖς κινουμένοις ἀντέρεισιν, οὐθὲν ἐπ’ αὐτοῦ δύναται κινεῖν ἑαυτό. καὶ γὰρ τὸ ἁλλόμενον καὶ πρὸς αὐτὸ ἀπερειδόμενον τὸ ἄνω καὶ πρὸς τὸ ὑπὸ τοὺς πόδας ποιεῖται τὴν ἅλσιν· ἔχει γάρ τινα ἀντέρεισιν πρὸς ἄλληλα τὰ μόρια ἐν ταῖς καμπαῖς, καὶ ὅλως τὸ πιέζον πρὸς τὸ πιεζόμενον.  διὸ καὶ οἱ πένταθλοι ἅλλονται πλεῖον ἔχοντες τοὺς ἁλτῆρας ἢ μὴ ἔχοντες, καὶ οἱ θέοντες θᾶττον θέουσι παρασείοντες τὰς χεῖρας· γίνεται γάρ τις ἀπέρεισις ἐν τῇ διατάσει πρὸς τὰς χεῖρας καὶ τοὺς καρπούς. ἀεὶ δὲ τὸ κινούμενον δυσὶν ἐλαχίστοις χρώμενον ὀργανικοῖς μέρεσι ποιεῖται τὴν με- ταβολήν, τῷ μὲν ὡσπερανεὶ θλίβοντι, τῷ δὲ θλιβομένῳ· τὸ μὲν γὰρ μένον θλίβεται διὰ τὸ φέρειν, τὸ δ’ αἰρόμενον τείνεται τῷ φέροντι τὸ φορτίον. διόπερ ἀμερὲς οὐδὲν οὕτω κινηθῆναι δυνατόν· οὐ γὰρ ἔχει τήν τε τοῦ πεισομένου καὶ τὴν τοῦ ποιήσοντος ἐν αὑτῷ διάληψιν.

5

10

15

20

25

3 δὲ Z Vs.l. Y P : om. U Ca S N | λέγωμεν U Ca S Y N V: λέγομεν Z P   5 καθάπερ τὰ ἀλλόμενα] om. Z, καθάπερ ἀλλόμενα Z3  6 μορίοις U Ca Y N V P Z3 mg : κατὰ μέρος Z : μέρει S Z3 s.l.   8 ἀποστηριζόμενον] om. Z, add. Z2  9 ἐάν τε U Ca Ye corr. N V : ἄν τε Z P : ἐάν του S   11 ἐπ᾽ αὐτοῦ U Ca S Y : ὑπ᾽ αὐτοῦ Z : ἐπ᾽ αὐτὸ N V P   12 αὐτὸ] ἑαυτὸ Z | ἀπερειδόμενον] ἐπερειδόμενον V   13 τὸ ἄνω Y U Ca N V P : πρὸς τὸ ἄνω Y : τῷ ἀνθρώπῳ S | ἅλσιν] ἕλξιν Y   14 τινα] τὴν U   15 πιέζον] πιεζοῦν (sic) Z   16 πλεῖον codd. : πλέον Bekker Jaeger | ἔχοντες] οἱ ἔχοντες Z   17 ἢ] ἢ οἱ Z   18 διατάσει Ca N V P : διαστάσει Z U S Y   19 καρποὺς] καμποὺς P   19–20 δυσὶν ἐλαχίστοις Z Ca Y P : ἐλαχίστοις δυσὶν U S V N   20 ποιεῖται Z Ca Y N V P : ποιεῖ U S   21 τῷ μὲν ὡσπερανεὶ θλίβοντι, τῷ δὲ θλιβομένῳ Y P : τῷ μὲν ὡσπερεὶ θλίβοντι, τῷ δὲ θλιβομένῳ U Z2 Ca S N V : ὡσπερανεὶ τῷ μὲν φέροντι, τῷ δὲ φερομένῳ Z   22 τὸ μὲν γὰρ μένον Z U : τὸ γὰρ μένον Ca N V P : τὸ γὰρ μὲν S : τὸ μένον γὰρ Y | θλίβεται] φαίνεται Z   23 τείνεται] γίνεται Z1, corr. Z2 | ἀμερὲς Z Y N : ἀμελὲς U Ca S P : ἀσκελὲς V | οὕτω] οὔτε N   24 δυνατόν Z U Ca S P : δύναται Y N V | τε Z2 Ca Y N V P Jaeger : om. Z1 U S Bekker | καὶ τὴν Ca Y N V : καὶ Z U S P   25 ποιήσοντος Z U Ca N V : ποιήσαντος S Y : πείσοντος P | ἐν Z Y P : om. U Ca S N V | αὑτῷ Z Jaeger : αὐτῷ U Ca S Y N V P Bekker



3, 705a3–25



[3] Having established these points, let us state what follows from them. 705a3 Some of the animals that change with respect to place change by means of their whole body all at once, like the jumpers, while others by means of their parts, like each of the animals that engage in progression. In both these cases of change, what moves always changes by supporting itself against what is beneath it – which is why whenever this thing gives way too quickly for what makes motion on it to support itself, or whenever it offers no resistance at all to what moves, then nothing can move itself on it. In fact, even the jumper makes its jump by supporting itself both against its own upper part and against what is under its feet. The reason is that its parts offer some resistance to one another at the joints – and, in general, what presses against what is pressed (which is why pentathletes jump farther with weights than without, and runners run faster if they swing their arms. The reason is that some support is generated in the distension toward the hands and the wrists). What moves always makes its change by using at least two parts as its 705a19 tools: the first, as it were, compressing, and the second compressed. The part that remains still is compressed because of the carrying, while the part that is raised is extended by carrying the weight, which is why nothing that is partless can move in this way. The reason is that does not have the distinction between what is to be acted upon and what is to act on it.



4, 705a26–b17

[4] Ἐπεὶ δ’ εἰσὶν αἱ διαστάσεις τὸν ἀριθμὸν ἕξ, αἷς ὁρίζεσθαι πέφυκε τὰ ζῷα, τό τε ἄνω καὶ κάτω καὶ τὸ ἔμπροσθεν καὶ ὄπισθεν, ἔτι δὲ τὸ δεξιὸν καὶ ἀριστερόν, τὸ μὲν ἄνω καὶ κάτω μόριον πάντ’ ἔχει τὰ ζῶντα· οὐ μόνον γὰρ ἐν τοῖς ζῴοις ἐστὶ τὸ ἄνω καὶ κάτω, ἀλλὰ καὶ ἐν τοῖς φυτοῖς·  διείληπται δ’ ἔργῳ, καὶ οὐ θέσει μόνον τῇ πρός τε τὴν γῆν καὶ τὸν οὐρανόν. ὅθεν μὲν γὰρ ἡ τῆς τροφῆς διάδοσις καὶ ἡ αὔξησις ἑκάστοις, ἄνω τοῦτ’ ἐστι· πρὸς ὃ δ’ ἔσχατον αὕτη περαίνει, τοῦτο κάτω. τὸ μὲν γὰρ ἀρχή τις,  τὸ δὲ πέρας· ἀρχὴ δὲ τὸ ἄνω. καίτοι δόξειέ γ’ ἂν τοῖς φυτοῖς οἰκεῖον εἶναι τὸ κάτω μᾶλλον· οὐχ ὁμοίως γὰρ ἔχει τῇ θέσει τὸ ἄνω καὶ κάτω τούτοις καὶ τοῖς ζῴοις. ἔχει δὲ πρὸς μὲν τὸ ὅλον οὐχ ὁμοίως, κατὰ δὲ τὸ ἔργον ὁμοίως·  αἱ γὰρ ῥίζαι εἰσὶ τὸ ἄνω τοῖς φυτοῖς· ἐκεῖθεν γὰρ ἡ τροφὴ διαδίδοται τοῖς φυομένοις, καὶ λαμβάνει ταύταις αὐτήν, καθάπερ τὰ ζῷα τοῖς στόμασιν. ὅσα δὲ μὴ μόνον ζῇ ἀλλὰ καὶ ζῷά ἐστι, τοῖς τοιούτοις ὑπάρχει τό τε ἔμπροσθεν καὶ τὸ ὄπισθεν. αἴσθησιν γὰρ ἔχει ταῦτα πάντα,  ὁρίζεται δὲ κατὰ ταύτην τό τε ὄπισθεν καὶ τὸ ἔμπροσθεν· ἐφ’ ὃ μὲν γὰρ ἡ αἴσθησις πέφυκε καὶ ὅθεν ἐστὶν ἑκάστοις, ἔμπροσθεν ταῦτ’ ἐστι, τὰ δ’ ἀντικείμενα τούτοις ὄπισθεν. ὅσα δὲ τῶν ζῴων μὴ μόνον αἰσθήσεως κοινωνεῖ, ἀλλὰ δύναται ποιεῖσθαι τὴν κατὰ τόπον μεταβολὴν αὐτὰ δι’ αὑτῶν, ἐν  τούτοις δὴ διώρισται πρὸς τοῖς λεχθεῖσι τό τ’ ἀριστερὸν καὶ τὸ δεξιὸν ὁμοίως τοῖς πρότερον εἰρημένοις, ἔργῳ τινὶ καὶ οὐ

30

705b1

5

10

15

27 ζῷα Y Jaeger : ζῶντα U Ca S V P Bekker : ὄντα N   28 ἔμπροσθεν] ἐπίπροσθεν Ca  27–29 καὶ τὸ … κάτω] om. N   28 ὄπισθεν Ca S U Z Bekker : τὸ ὄπισθεν V Y P Jaeger | δὲ Z Ca V Y P : δὲ καὶ U S P | τὸ δεξιὸν καὶ ἀριστερὸν Y : τὸ δεξιὸν καὶ τὸ ἀριστερὸν Z Jaeger : καὶ τὸ δεξιὸν καὶ τὸ ἀριστερὸν U S : δεξιὸν καὶ ἀριστερὸν Ca V P Bekker   28–29 τὸ μὲν … μόριον] τὰ μὲν … μόρια Z  31 πρός τε Z U S P : πρός γε Ca V : τε πρὸς Y : πρὸς N   33 ἡ αὔξησις ἑκάστοις Z U S : αὔξησίς ἐστιν ἑκάστοις Y P : αὔξησις καθ᾽ ἑκάστοις Ca : αὔξησις ἑκάστοις N V | ἄνω τοῦτ᾽] τοῦτ᾽ ἄνω Y   705b1 αὕτη Ca P N : αὐτὴ Z U S V Y | τοῦτο Z Y P V N : τοῦτο δὲ U Ca S | τὸ μὲν γὰρ Ca S Y N V P : ἡ μὲν Z : τὸ μὲν U   2 γ᾽ ἂν U Ca N Jaeger : ἂν Z S N Y P Bekker   5 κατὰ … ὁμοίως] om. Z1, add. Z2  7 ταύταις αὐτὴν] ταύτῃ Z1, corr. Z2  9–10 ἔμπροσθεν καὶ τὸ ὄπισθεν Z U S V Y P : ὄπισθεν καὶ ἔμπροσθεν Ca N   10 ταῦτα πάντα U Ca S N Y : πάντα ταῦτα Z V P   11 δὲ Ca S Y N V P : γὰρ Z U | ὄπισθεν καὶ τὸ ἔμπροσθεν Z U S Jaeger : ἔμπροσθεν καὶ τὸ ὄπισθεν Ca N V P Y Bekker   12 ἐφ᾽ ὃ] ἐφ᾽ ᾧ S | ἐστὶν] ἔσθ᾽ Ca  16 δὴ Y : om. P : δὲ Z U Ca S N V   17 ἔργῳ τινὶ Z1 U : ἔργῳ δέ τινι Z2 Y P : ἔργῳ Ca S N V



4, 705a26–b17



[4] The dimensions by which animals are naturally determined are six in 705a26 number: the up and the down, the front and the back, and also the right and the left. Furthermore, all living beings have the up and the down, for the up and the down are found not only in animals but also in plants. Now, this distinction is one of function, and not merely of position relative to the earth and the heavens. For that from which food and growth is distributed is the up, while that to which it is distributed and in which it ends is the down. The former is some kind of origin; the latter is an end. The up is an origin, although in the case of plants it might seem that the down is more appropriate as origin. But that is because plants do not have the up and down in the same position as animals. With respect to the universe plants do not have the up and down in the same way as animals, but with respect to their function they do. The reason is that roots are the up for plants. For it is from the roots that the nutriment is distributed to the growing parts, and it is by means of them that plants take it, just as animals do by means of their mouths. In those creatures that not only live but are also animals we find both 705b8 the front and the back, for they all have sense-perception, and it is by reference to sense-perception that the front and the back are determined. For the part in which sense-perception is naturally implanted and from which each animal derives it is the front, whereas the opposite parts are the back. Those animals that not only partake in sense-perception but are able to 705b13 make change with respect to place by themselves, in addition to what has already been said, are determined also by the left and the right, each one of which is, like the dimensions we spoke of previously, a determination of function and not of position. That from which a body’s



4, 705b18–706a12

θέσει διωρισμένον ἑκάτερον αὐτῶν· ὅθεν μὲν γάρ ἐστι τοῦ σώματος ἡ τῆς κατὰ τόπον μεταβολῆς ἀρχὴ φύσει, τοῦτο μὲν δεξιὸν ἑκάστῳ, τὸ δ’ ἀντικείμενον καὶ τούτῳ πεφυκὸς  ἀκολουθεῖν ἀριστερόν. τοῦτο δὲ διήρθρωται μᾶλλον ἑτέροις ἑτέρων. ὅσα μὲν γὰρ ὀργανικοῖς μέρεσι χρώμενα (λέγω δ’ οἷον ποσὶν ἢ πτέρυξιν ἤ τινι ἄλλῳ τοιούτῳ) τὴν εἰρημένην μεταβολὴν ποιεῖται, περὶ μὲν τὰ τοιαῦτα μᾶλλον διήρθρωται τὸ λεχθέν· ὅσα δὲ μὴ τοιούτοις μορίοις, αὐτῷ δὲ  τῷ σώματι διαλήψεις ποιούμενα προέρχεται, καθάπερ ἔνια τῶν ἀπόδων, οἷον οἵ τε ὄφεις καὶ τὸ τῶν καμπῶν γένος, καὶ πρὸς τούτοις ἃ καλοῦσι ἔντερα γῆς, ὑπάρχει μὲν καὶ ἐν τούτοις τὸ λεχθέν, οὐ μὴν διασεσάφηταί γ’ ὁμοίως. ὅτι δ’ ἐκ τῶν δεξιῶν ἡ ἀρχὴ τῆς κινήσεώς ἐστι, σημεῖον καὶ  τὸ φέρειν τὰ φορτία πάντας ἐπὶ τοῖς ἀριστεροῖς· οὕτως γὰρ ἐνδέχεται κινεῖσθαι τὸ φέρον, λελυμένου τοῦ κινήσοντος. διὸ καὶ ἀσκωλιάζουσι ῥᾷον ἐπὶ τοῖς ἀριστεροῖς· κινεῖν γὰρ πέφυκε τὸ δεξιόν, κινεῖσθαι δὲ τὸ ἀριστερόν· ὥστε καὶ τὸ  φορτίον οὐκ ἐπὶ τῷ κινήσοντι ἀλλ’ ἐπὶ τῷ κινησομένῳ δεῖ ἐπικεῖσθαι ἐὰν δ’ ἐπὶ τῷ κινοῦντι καὶ τῇ ἀρχῇ τῆς κινήσεως ἐπιτεθῇ, ἤτοι ὅλως οὐ κινήσεται ἢ χαλεπώτερον. σημεῖον δ’ ὅτι ἀπὸ τῶν δεξιῶν ἡ ἀρχὴ τῆς κινήσεως καὶ αἱ προβολαί· πάντες γὰρ τὰ ἀριστερὰ προβάλλονται, καὶ ἑστῶτες προβεβλήκασι τὰ ἀριστερὰ μᾶλλον, ἐὰν μὴ ἀπὸ τύχης συμβῇ· οὐ γὰρ τῷ προβεβηκότι κινοῦνται, ἀλλὰ τῷ ἀποβεβηκότι· καὶ ἀμύνονται τοῖς δεξιοῖς. διὰ ταύτην δὲ τὴν αἰτίαν καὶ τὰ δεξιὰ ταὐτά ἐστι πάντων. ὅθεν μὲν γὰρ ἡ ἀρχὴ τῆς κινήσεως, τὸ αὐτὸ πᾶσι καὶ ἐν τῷ αὐτῷ τὴν θέσιν ἔχει κατὰ φύσιν· δεξιὸν δ’ ἐστὶν ὅθεν ἡ ἀρχὴ τῆς

20

25

30

706a1

5

10

18 αὐτῶν] αὐτῶν ἐστιν Y | ἐστι τοῦ σώματος Z U S Y P : τοῦ σώματος Ca N : τοῦ σώματός ἐστι V   20 ἑκάστῳ Z Ca Y N V P Bekker : ἑκάστων U S Jaeger   24 ποιεῖται] om. Ca1, πoιεῖται μεταβολὴν Ca2 | μᾶλλον] om. Z1, add. Z2  26  σώματι] στόματι Y | διαλήψεις] ex διαλείψεις V   27 οἵ τε Z Y N V P : om. U Ca S   28 ἔντερα γῆς Z U Ca N V Jaeger : γῆς ἔντερα Y P Bekker : ἔντερα τῆς γῆς S   31 τὰ φορτία πάντας Z1 Y : τὰ φορτία πάντα U Ca S V P Z3 : πάντα τὰ φορτία N   32 φέρον λελυμένου Z U Ca S, φερὸν (sic) λελυμένου V : φερόμενον λελυμένου Y P : φερόμενον λελυμένον N   33 ἀσκωλιάζουσι U Z2 S Y : ἀσκωλίζουσι Z1 : ἀσκωλιάζοντα Ca : ἀσκωλιάζονται V P Z3 : ἀσκωλιάζοντες N   706a1 καὶ] οm. V   3 ἐὰν Z Ca N V P Y : ἂν S U   4 κινήσεται] κινηθήσεται Z   7 προβεβλήκασι U Ca S N V P Jaeger : προβεβήκασι Z Y Bekker | ἐὰν Ca Y N V : ἂν P Bekker Jaeger : εἰ Z U : δὲ εἰ S   7–8 ἀπὸ τύχης συμβῇ] συμβῇ ἀπὸ τύχης Y   8 προβεβηκότι] προβεβληκότι Ca  10 ταὐτά] τὰ αὐτὰ P | γὰρ] om. N   11 τὸ αὐτὸ Z U S P : ταυτὸ Ca N V : τοῦτο Y | τῷ αὐτῷ Z U S P : ταυτῷ Ca N V Y



4, 705b18–706a12



change with respect to place naturally originates is in each animal the right, whereas that which is opposite to it, and dependent on it, is by nature the left. The distinction between right and left is more pronounced in some animals than in others. For those that use instrumental parts – I mean feet, wings, or the like – in order to produce the aforesaid change have the mentioned distinction more clearly articulated around these parts; those, however, that do not make use of such parts, but progress by making sections in their own bodies – as do some footless animals such as snakes, the entire class of caterpillars, and also those that people call earthworms – have the mentioned distinction as well, although it is not nearly as clear. That the origin of movement is from the right is shown by the fact that all carry loads on the left shoulder; the reason is that in this way it is possible for the part that carries the load to be set in motion, while the part that causes the movement is free. This is why it is also easier to hop on the left leg; for the nature of the right is to move and that of the left is to be moved. So the load, too, must rest on the side that is to be moved, not on that which is going to cause movement; and if it be set on the side that moves, which is the origin of motion, it will either not be moved at all or be moved with more labor. evidence that the right is the source of motion is the way in which they put their feet forward; all lead with the left, and after standing still prefer to put the left foot forward, unless they do otherwise by chance. For they move not by the leg they put forward but by the leg with which they step off. And they defend themselves with their right. Due to this cause the right is the same in all , for that from which motion begins is the same for all of them, and it has its natural position in the same place. Right, then, is the whence of motion, and for

705b21

705b29

706a5

706a9



4, 706a13–5, 706b2

κινήσεώς ἐστι. καὶ διὰ τοῦτο τὰ στρομβώδη τῶν ὀστρακοδέρμων δεξιὰ πάντ’ ἐστιν· οὐ γὰρ ἐπὶ τὴν ἑλίκην κινεῖται, ἀλλ’ ἐπὶ τὸ καταντικρὺ πάντα προέρχονται, οἷον πορφύραι καὶ  κήρυκες. κινουμένων οὖν πάντων ἀπὸ τῶν δεξιῶν, κἀκείνων ἐπὶ ταὐτὰ κινουμένων ἑαυτοῖς, ἀνάγκη πάντα δεξιὰ εἶναι ὁμοίως. ἀπολελυμένα δ’ ἔχουσι τὰ ἀριστερὰ τῶν ζῴων μάλιστα ἄνθρωποι διὰ τὸ κατὰ φύσιν ἔχειν μάλιστα τῶν ζῴων· φύσει δὲ βέλτιον τὸ δεξιὸν τοῦ ἀριστεροῦ  κεχωρισμένον. διὸ καὶ τὰ δεξιὰ ἐν τοῖς ἀνθρώποις μάλιστα δεξιά ἐστι. διωρισμένων δὲ τῶν δεξιῶν εὐλόγως τὰ ἀριστερὰ ἀκινητότερά ἐστι, καὶ ἀπολελυμένα μάλιστα ἐν τούτοις. καὶ αἱ ἄλλαι δ’ ἀρχαὶ μάλιστα κατὰ φύσιν καὶ διωρισμέναι ἐν τῷ ἀνθρώπῳ ὑπάρχουσι, τό τ’ ἄνω καὶ τὸ  ἔμπροσθεν. [5] Ὅσοις μὲν οὖν τὸ ἄνω καὶ τὸ ἔμπροσθεν διώρισται, καθάπερ τοῖς ἀνθρώποις καὶ τοῖς ὄρνισι, ταῦτα μὲν δίποδα (τῶν δὲ τεττάρων τὰ δύο σημεῖα τοῖς μὲν πτέρυγες τοῖς δὲ χεῖρες καὶ βραχίονές εἰσιν), ὅσα δ’ ἐπὶ τὸ αὐτὸ τό τε πρόσθεν ἔχει καὶ τὸ ἄνω, τετράποδα καὶ πολύποδα καὶ  ἄποδα. καλῶ γὰρ πόδα μέρος ἐπὶ σημείῳ πεζῷ κινητικῷ κατὰ τόπον· καὶ γὰρ τὸ ὄνομα ἐοίκασιν εἰληφέναι ἀπὸ τοῦ πέδου οἱ πόδες. ἔνια δ’ ἐπὶ τὸ αὐτὸ ἔχει τὸ πρόσθιον καὶ τὸ ὀπίσθιον, οἷον τά τε μαλάκια καὶ τὰ στρομβώδη τῶν ὀστρακοδέρμων· εἴρηται δὲ περὶ αὐτῶν πρότερον ἐν ἑτέροις.

15

20

25

30

706b1

13 ἐστι] om. V | καὶ] om. N   14 ἑλίκην] νίκην S | κινεῖται] κινεῖ Z   15 τὸ Z Y P : τοῖς U S N V : τῶν Ca | πάντα προέρχονται U Z2 Ca S N V : πᾶν τὸ προέχον Z1 Y : πάντα προέρχεται P   16 οὖν Z Y N V P : om. U Ca S   20 τῶν ζῴων· φύσει δὲ U Ca Y N V P : τῶν δὲ ζῴων φύσει Z S | βέλτιον Ca S Y N V P : βελτίονα U : τὰ βελτίονα Z | τὸ Z U Ca S N Jaeger : τε τὸ Y P Bekker : τό τε V   21 κεχωρισμένον V Y P : καὶ κεχωρισμένον Z : om. U Ca S N | ἐν Z Ca N Y P : om. U S V   22 δὲ Z U S V Y P : τε C a N | τὰ ἀριστερὰ Z U C a S Y P : τἀριστερὰ N V   23 ἀκινητότερα] ἀκινητότατα Z1, corr. Z2  24 αἱ ἄλλαι Z Y P : ἄλλαι U Ca N V : ἄνω S | δ᾽ ἀρχαὶ] ἀρχαὶ Z | καὶ διωρισμέναι Z Ca Y N V P Jaeger : διωρισμέναι U S Bekker   25 ἐν Z U Ca S Y P : om. N V | τό τ᾽ U Ca S Y P : καὶ τὸ N : τό τε V   25–26 τό τ᾽ ἄνω … oὖν] om. Z, oἷς μὲν οὖν add. Z3    26  ὅσοις U Ca S Y N V Jaeger : οἷς P Z3 Bekker | τὸ ἄνω] καὶ τὸ ἄνω Z | ἔμπροσθεν Z Ca N Y P : πρόσθεν U S V   28 τεττάρων ex τετραπόδων Y   29–30 τό τε πρόσθεν U Z2 : τὸ πρόσθεν Z Bekker Jaeger : τὸ πρόσθιον Ca Y P : τό τε πρόσθιον S N V  32  κατὰ τόπον] om. Z1, add. Z2  33–706b1 καὶ τὸ ὀπίσθιον U Ca S Y N V Z3 : καὶ ὀπίσθιον V : οm. Z1 P   1 καὶ τὰ] τά τε Z  



4, 706a13–5, 706b2



this reason among the hard-shelled animals the stromboid ones have their shells on the right. The reason is that they do not move in the direction of the spiral, but they all move in the opposite direction. Examples are the murex and the ceryx. Then, since all animals start their movement from the right, and the right moves in the same direction as the animals themselves, it is necessary for the right to be in them all alike. Human beings have the left limbs most detached of all animals because 706a18 of all animals they are most in accordance with nature, and by nature it is better that the right side is separate from the left side. And so the right side is especially right-sided in human beings. And since the right side is differentiated, it is reasonable that in human beings the left side should be less movable and most detached. The other origins too, namely the up and the front, are found most naturally and most clearly distinct in the human being. [5] All the animals that, like human beings and birds, have the upper part 706a26 set apart from the front part are two-footed (two of the four points are wings in one group, and hands and arms in the other group), whereas all the animals that have their upper and front parts in the same place are four-footed, many-footed, and footless. I use the term “foot” for a part connected with a point on the ground capable of producing motion; for the feet (πόδες) also appear to have got their name from “ground” (πέδον). Some animals, too, have their front and back parts in the same place, for example soft-bodied animals and stromboid hard-shelled animals. These animals have been discussed earlier elsewhere.



5, 706b3–6, 706b27

τριῶν δ’ ὄντων τόπων, τοῦ ἄνω καὶ μέσου καὶ κάτω, τὰ μὲν δίποδα τὸ ἄνω πρὸς τὸ τοῦ ὅλου ἄνω ἔχει, τὰ δὲ πολύποδα ἢ ἄποδα πρὸς τὸ μέσον, τὰ δὲ φυτὰ πρὸς τὸ  κάτω. αἴτιον δ’ ὅτι τὰ μὲν ἀκίνητα, πρὸς τὴν τροφὴν δὲ τὸ ἄνω, ἡ δὲ τροφὴ ἐκ τῆς γῆς. τὰ δὲ τετράποδα ἐπὶ τὸ μέσον καὶ τὰ πολύποδα καὶ ἄποδα διὰ τὸ μὴ ὀρθὰ εἶναι· τὰ δὲ δίποδα πρὸς τὸ ἄνω διὰ τὸ ὀρθὰ εἶναι, μάλιστα δ’ ὁ ἄνθρωπος· μάλιστα γὰρ κατὰ φύσιν ἐστὶ δίπους.  εὐλόγως δὲ καὶ αἱ ἀρχαί εἰσιν ἀπὸ τούτων τῶν μορίων· ἡ μὲν γὰρ ἀρχὴ τίμιον, τὸ δ’ ἄνω τοῦ κάτω καὶ τὸ πρόσθεν τοῦ ὄπισθεν καὶ τὸ δεξιὸν τοῦ ἀριστεροῦ τιμιώτερον. καλῶς δ’ ἔχει καὶ τὸ ἀνάπαλιν λέγειν περὶ αὐτῶν, ὡς διὰ τὸ τὰς ἀρχὰς ἐν τούτοις εἶναι ταῦτα τιμιώτερα τῶν  ἀντικειμένων μορίων ἐστίν. [6] Ὅτι μὲν οὖν ἐκ τῶν δεξιῶν ἡ τῆς κινήσεώς ἐστιν ἀρχή, φανερὸν ἐκ τῶν εἰρημένων. ἐπεὶ δ’ ἀνάγκη παντὸς συνεχοῦς, οὗ τὸ μὲν κινεῖται τὸ δ’ ἠρεμεῖ, ὅλου δυναμένου κινεῖσθαι ἑστῶτος θατέρου, ᾗ ἄμφω κινεῖται ἐναντίας κινήσεις, εἶναί  τι κοινόν, καθ’ ὃ συνεχῆ ταῦτ’ ἐστιν ἀλλήλοις, κἀνταῦθ’ ὑπάρχειν τὴν ἀρχὴν τῆς ἑκατέρου τῶν μερῶν κινήσεως, ὁμοίως δὲ καὶ τῆς στάσεως, δῆλον ὅτι, καθ’ ὅσας τῶν λεχθεισῶν ἀντιθέσεων ἰδία κίνησις ὑπάρχει τῶν ἀντικειμένων μερῶν ἑκατέρῳ, πάντα ταῦτα κοινὴν ἀρχὴν ἔχει κατὰ τὴν τῶν εἰρημένων μερῶν σύμφυσιν, λέγω δὲ τῶν τε δεξιῶν καὶ ἀριστερῶν καὶ τῶν ἄνω καὶ κάτω καὶ τῶν ἔμπροσθεν καὶ

5

10

15

20

25

3 μέσου καὶ κάτω U Ca S N V P : κάτω καὶ μέσου Z Y   4 τὸ ἄνω Zp.c. Ca Y N V P : τοῦ ἄνω Za.c. : ἄνω U S | ὅλου] ὀμφαλοῦ Z1, corr. Z2 | ἄνω ἔχει] ἔχει ἄνω P   6–707a14 αἴτιον … κίνησίς ἐστιν] οm. Y   8 ἄποδα] τὰ ἄποδα Z   9 τὰ δὲ … εἶναι] om. Z, add. Z2  10 μάλιστα γὰρ Z U S N Y P : om. Ca Va.c. | δίπους] δίπουν Z1, corr. Z2  11 αἱ ἀρχαί] ἀρχαί V   12 πρόσθεν] πρόσθιον V   14 τὸ ἀνάπαλιν P : om. Z U Ca S N V | ὡς Ca N V : om. U Z S P   16 μορίων Z U S N Y P : πάντων μορίων Ca : om. V   17 ἐστιν ἀρχή] ἀρχή ἐστι V   18 ἐκ τῶν εἰρημένων] om. Z   19 ὅλου Z e corr. U Ca Y N V P : ἀλλ᾽ οὐ S   20 ᾗ Z e corr. | τὰς Jaeger cum Mich. : om. codd. Bekker | κινήσεις] ex κινήσεων Ca  21 τι κοινὸν P Bekker : τὸ κοινὸν Ca S N V Jaeger : τέ τι κοινὸν Z : τέ τι τὸ κοινὸν U | καθ᾽ ὃ] καθὼς V | κἀνταῦθ᾽ Z N P : καὶ ἐνταῦθ᾽ U, καὶ ἐνταῦθα Ca S V   23 δῆλον ὅτι Forster cum Leonico Thomaeo : δηλονότι codd. Bekker Jaeger   25 ἑκατέρῳ P : ἑκατέρων U Ca S N V : ἑκατέραι Z | πάντα ταῦτα P : ὡς πάντα ταῦτα Z : ταῦτα πάντα U Ca S N V | κατὰ Z2 P : om. U Ca S N V   26 τε] om. Z



5, 706b3–6, 706b27



Since there are three regions – an upper, an intermediate, and a lower 706b3 region – the two-footed animals have their upper part lined up with the upper part of the universe; the many-footed and the footless animals lined up with the intermediate region ; plants lined up with the lower . The reason is that plants have no power of 706b6 locomotion, and their upper part is determined relative to nutriment, and their nutriment is taken from the earth. Four-footed, many-footed, and footless animals lined up with the intermediate because they are not erect. Two-footed animals, however, have theirs lined up with the upper region of the universe because they are erect, and most of all the human being; the reason is that the human being is the most natural two-footed animal. And it makes good sense for the origins to operate from these parts; for 706b11 the origin is honorable, and the upper part is more honorable than the lower part, the front part more than the back part, and the right side more than the left side. But it is also correct to state the reverse about them and say that these parts are more honorable than their opposites because the origins are present in them. [6] Now, the above discussion has made it clear that the origin of motion 706b17 is from the right side. Since for every continuous whole of which one part is moved while another remains at rest, in order for it to be able to move as a whole while one of its parts stands still, it is necessary, insofar as both parts are moved with opposite motions, that there exists something common, according to which these parts are continuous with each other, and that the origin of the motion of each of the two parts is located there, and likewise of the absence of motion; it is evident that, for each motion of the opposite parts, proper to the aforesaid opposite pairs, there exists a common origin at the natural juncture of the parts just mentioned. I mean the parts on the right and the left side, the upper and lower parts, and the front and back parts.



6, 706b28–707a16

ὄπισθεν. κατὰ μὲν οὖν τὸ ἔμπροσθεν καὶ ὄπισθεν διάληψις οὐκ ἔστι τοιαύτη περὶ τὸ κινοῦν ἑαυτό, διὰ τὸ μηθενὶ φυσικὴν ὑπάρχειν κίνησιν εἰς τὸ ὄπισθεν, μηδὲ διορι- 30 σμὸν ἔχειν τὸ κινούμενον, καθ’ ὃν τὴν ἐφ’ ἑκάτερα τούτων μεταβολὴν ποιεῖται· κατὰ δὲ τὸ δεξιόν γε καὶ ἀριστερὸν καὶ τὸ ἄνω καὶ τὸ κάτω ἔστι. διὸ τῶν ζῴων ὅσα μέρεσιν ὀργανικοῖς χρώμενα προέρχεται, τῇ μὲν τοῦ ἔμπροσθεν καὶ  707a1 ὄπισθεν διαφορᾷ οὐκ ἔχει διωρισμένα ταῦτα, ταῖς δὲ λοιπαῖς, ἀμφοτέραις μέν, προτέρᾳ δὲ τῇ κατὰ τὸ δεξιὸν καὶ ἀριστερὸν διοριζούσῃ, διὰ τὸ τὴν μὲν ἐν τοῖς δυσὶν εὐθέως ἀναγκαῖον εἶναι ὑπάρχειν, τὴν δ’ ἐν τοῖς τέτταρσι πρώτοις.  5 ἐπεὶ οὖν τό τε ἄνω καὶ τὸ κάτω καὶ τὸ δεξιὸν καὶ ἀριστερὸν τῇ αὐτῇ ἀρχῇ καὶ κοινῇ συνήρτηται πρὸς αὑτά (λέγω δὲ ταύτην τὴν τῆς κινήσεως κυρίαν), δεῖ γ’ ἐν ἅπαντι τῷ μέλλοντι κατὰ τρόπον ποιεῖσθαι τὴν ἐφ’ ἑκάστου κίνησιν ὡρίσθαι πως καὶ τετάχθαι ταῖς ἀποστάσεσι ταῖς πρὸς τὰς ῥηθείσας  10 ἀρχάς (τάς τε ἀντιστοίχους καὶ τὰς συστοίχους τῶν ἐν τοῖς μέρεσι τούτοις) τὸ τῶν λεχθεισῶν κινήσεων ἁπασῶν αἴτιον (αὕτη δ’ ἐστὶν ἀφ’ ἧς ἀρχῆς κοινῆς τῶν ἐν τῷ ζῴω ἥ τε τοῦ δεξιοῦ καὶ ἀριστεροῦ κίνησίς ἐστιν, ὁμοίως δὲ καὶ ἡ τοῦ ἄνω τε καὶ κάτω, ταύτην δ’ ἔχειν ἑκάστῳ ᾗ παραπλησίως πρὸς ἑκάστην τῶν 15 ἐν τοῖς ῥηθεῖσι μέρεσιν ἀρχῶν). 27 ἀριστερῶν Z N V P Bekker : τῶν ἀριστερῶν U Ca S Jaeger | ἄνω καὶ κάτω Z P : ἄνωθεν καὶ κάτωθεν U Ca S N V  27–28 τῶν ἔμπροσθεν καὶ ὄπισθεν Z : τῶν ἔμπροσθεν καὶ τῶν ὄπισθεν U S N V Bekker Jaeger : τῶν ὄπισθεν καὶ τῶν ἔμπροσθεν Ca P   28 κατὰ … ὄπισθεν] οm. Z, κατὰ … διάληψις Ca N V P : om. U S, add. U2 | οὖν] om. V | ὄπισθεν U2 Ca N V : τὸ ὄπισθεν P Bekker Jaeger  29 ἔστι] ἔσται Z | περὶ] παρὰ V | μηθενὶ Z U S N P : μηδενὶ Ca V   30 διορισμὸν Ca N P : διωρισμένον Z U S V   31 καθ᾽ ὃν Bekker : καθὸ Z : κατὰ U Ca S P, ex καὶ V : καὶ N   32 ποιεῖται Z Ca : om. U S N V P | δὲ] om. Z, add. Z2 | γε Z : τε V N P : om. U Ca S   33 ἔστι] om. Z, add. Z2 | διὸ ex διὰ Z2  707a1 προέρχεται Z2 Ca N V P : προσέρχεται Z1 U S | τῇ] ταῖς Z | τοῦ Z : οὖν N : om. U Ca S V P   1-2 καὶ ὄπισθεν Z : om. U Ca S N V P   2 διαφορᾷ οὐκ ἔχει] διαφοραῖς οὐκ ἔστι Z   5 ὑπάρχειν Z : πρῶτον ὑπάρχειν Z2 U Ca S N V : ὑπάρχειν πρῶτον P | τέτταρσι Z U N V P : τέτρασι Ca S   6 οὖν] γοῦν V | τὸ κάτω Z U Ca S N V Jaeger : κάτω P Bekker  7 αὑτά] αὐτά Ca  8 δεῖ γ᾽ V : δεῖ τε Z2 U Ca S, δεῖ τ᾽ N : δεῖ δε Z P, δεῖ δ᾽ (tamquam apodosis) Bekker Jaeger, maluerit Corcilius   9 ἐφ᾽ ἑκάστου Z U S N P : ἐφ᾽ ἑκάστῳ V : ἀφ᾽ ἑκάστου Ca Bekker Jaeger   10  ταῖς πρὸς Z2 Ca N V : πρὸς Z U S   11 ἀντιστοίχους] ἀντιστοιχούσας S | τῶν] τοῖς V   13 ἀρχῆς κοινῆς τῶν ἐν τῷ ζῴῳ] ἀρχη τῶν ἐν τοῖς ζώιοις Z, corr. Z2  14–15 τε καὶ U Ca S N V : καὶ Z Y P Bekker Jaeger   15 δ᾽ U Ca S N V P Z3 : οm. Z1 S Y | ἔχειν U2 Ca S N V P Z3 : ἔχει ἐν Z1 Y : ἔχει U1  16 post ἀρχῶν cum Louis punxi



6, 706b28–707a16



Now, the distinction of front and back is not of a kind that concerns 706b28 that which moves itself, because there is no animal for which backward motion is natural, nor has the moving animal any articulation according to which it can make change in each of these directions; but there exists according to the right and the left, the up and the down. This is the reason why all animals that advance using instrumental parts have them distinguished not by the differentiation of front and back, but rather by that of the remaining two pairs. The distinction of right and left is prior. The reason is that this differentiation must appear as soon as you have a division into two, while the other differentiation appears as soon as there is a division into four. Since, then, the up and the down, and the right and the left are con- 707a6 nected with one another by the same common origin (by which I mean that which controls their motion), it follows that in everything that is going to make motion in each such part, properly, the cause of all the said motions must be arranged in a certain definite position relative to the distances from the origins mentioned before (both those arranged coordinately in pairs as well as those that are arranged in a series); and that the common center is the origin from which the animal’s motions of right and left, and similarly of up and down, originate, and that each has an origin of this kind at a place that is similarly related to each of the origins in the parts described.



7, 707a16–b11

[7] Δῆλον οὖν ὡς ἢ μόνοις ἢ μάλιστα τούτοις ὑπάρχει τῶν ζῴων ἡ κατὰ τόπον κίνησις, ἃ δυσὶν ἢ τέτταρσι ποιεῖται σημείοις τὴν κατὰ τόπον μεταβολήν. ὥστ’ ἐπεὶ σχεδὸν τοῖς ἐναίμοις τοῦτο μάλιστα συμβέβηκε, φανερὸν ὅτι πλείοσί τε σημείοις τεττά- ρων οὐθὲν οἷόν τε κινεῖσθαι τῶν ἐναίμων ζῴων, καὶ εἴ τι τέτταρσι σημείοις κινεῖσθαι πέφυκε μόνον, ἀναγκαῖον τοῦτ’ εἶναι ἔναιμον. ὁμολογεῖ δὲ τοῖς λεχθεῖσι καὶ τὰ συμβαίνοντα περὶ τὰ ζῷα. τῶν μὲν γὰρ ἐναίμων οὐδὲν εἰς πλείω διαιρούμενον δύναται ζῆν οὐθένα χρόνον ὡς εἰπεῖν, τῆς τε κατὰ τόπον κινήσεως, καθ’ ἣν ἐκινεῖτο συνεχὲς ὂν καὶ μὴ διῃρημένον, οὐ δύναται κοινωνεῖν· τῶν δ’ ἀναίμων τε καὶ πολυπόδων ἔνια διαιρούμενα δύναται ζῆν πολὺν χρόνον ἑκάστῳ τῶν μερῶν, καὶ κινεῖσθαι τὴν αὐτὴν ἥνπερ καὶ πρὶν διαιρεθῆναι κίνησιν, οἷον αἵ τε καλούμεναι σκολόπενδραι καὶ  ἄλλα τῶν ἐντόμων καὶ προμήκων· πάντων γὰρ τούτων καὶ τὸ ὄπισθεν μέρος ἐπὶ τὸ αὐτὸ ποιεῖται τὴν πορείαν τῷ ἔμ- προσθεν. αἴτιον δὲ τοῦ διαιρούμενα ζῆν ὅτι, καθάπερ ἂν εἴ τι συνεχὲς ἐκ πολλῶν εἴη ζῴων συγκείμενον, οὕτως ἕκαστον αὐτῶν συνέστηκε. φανερὸν δὲ τοῦτο ἐκ τῶν πρότερον εἰρημένων, διότι τοῦτον ἔχει τὸν τρόπον. δυσὶ γὰρ ἢ τέτταρσι ση- μείοις πέφυκε κινεῖσθαι τὰ μάλιστα συνεστηκότα κατὰ φύσιν, ὁμοίως δὲ καὶ ὅσα τῶν ἐναίμων ἄποδά ἐστι. καὶ γὰρ ταῦτα κινεῖται τέτταρσι σημείοις, δι’ ὧν τὴν κίνησιν ποιεῖται. δυσὶ γὰρ χρώμενα προέρχεται καμπαῖς· τὸ γὰρ δεξιὸν καὶ ἀριστερὸν καὶ τὸ πρόσθιον καὶ ὀπίσθιον ἐν τῷ πλά- τει ἐστὶν ἐν ἑκατέρᾳ τῇ καμπῇ αὐτοῖς, ἐν μὲν τῷ πρὸς τὴν

20

25

30 707b1

5

10

16 oὖν] om. P   17 τούτοις] τούτοις τοῖς ζῴοις S   18 τέτταρσι Z U Y N V P : τέτρασι Ca S  19 τοῖς] om. Z, add. Z2  20 τε Z Ca Y N V P : οm. U S   21 καὶ Z S : κἂν U Ca Y N V P | εἴ τι U Ca Y N V P : ἐπεὶ Z : εἴ τι ἐπεὶ S   22 τέτταρσι Z U Y N V P : τέτρασι Ca S | τοῦτ᾽] τοῦτ᾽ ἦν Z, corr. Z2  25 ζῆν οὐθένα Z U S Y P : οὐδένα ζῆν Ca N V   25–27 ζῆν … δύναται] om. S   26 ἐκινεῖτο Z Y : κινεῖ τὸ U Ca N V P | συνεχὲς ὂν Z U Y : συνέχον Ca N V P   27 τε] ποτε fecit Z2  707b1 ἐπὶ] καὶ Z1, corr. Z2 s.l. | ποιεῖται τὴν πορείαν Z2 Ca N V P : τὴν πορείαν ποιεῖται Y : ποιεῖ τὴν πορείαν Z1 U S | τῷ ἔμπροσθεν] τουπροσθεν (sic) Z1, corr. Z2  3 εἴη Z Ca Y N V : om. U S P | οὕτως U Ca Y N V P : οὕτως ἂν Z : οὕτως καὶ Y   5–6 σημείοις πέφυκε κινεῖσθαι Z Bekker : κινεῖσθαι πέφυκε σημείοις U Ca S N V Jaeger : πέφυκε κινεῖσθαι σημείοις Y P   6 τὰ] om. P   7 ὅσα τῶν ἐναίμων Z Y P Bekker : τῶν ἐναίμων ὅσα U Ca N V Jaeger : τῶν ἐντόμων S   8 τέτταρσι Z U N V P : τέσσαρσι Y : τέτρασι Ca S | τὴν κίνησιν ποιεῖται Z Ca Y N V P : ποιεῖται τὴν κίνησιν U S   9 καμπαῖς] καμπαῖς κατὰ τὸ πλεῖστον Y   10 καὶ τὸ Y P Bekker : καὶ Z U Ca S N V : τὸ Jaeger | πρόσθιον καὶ ὀπίσθιον Z U Ca S N : πρόσθιον καὶ τὸ ὀπίσθιον Y P : πρόσθεν καὶ ὄπισθεν V | πλάτει] πλατεῖ Y   11 αὐτοῖς] om. Z1, add. Z2 | τὴν Ca Y N V P : om. Z U S  



7, 707a16–b11



[7] It is evident, then, that motion with respect to place belongs either 707a16 only or above all to those animals that make change with respect to place either by means of two or four points. So, since perhaps this happens above all in the case of blooded animals, it is clear that none of the blooded animals can move by means of more than four points, and that if an animal moves naturally by means of just four points, it is necessary for it to be blooded. What occurs in animals is in accord with what was just said. None of 707a23 the blooded animals can survive if it is divided into a plurality of parts, not even for a moment so to speak, nor can it partake of the motion with respect to place according to which it was moving while it was continuous and undivided. Some of the non-blooded and many-footed animals, however, when divided can live for quite a while in each of their severed parts, and can move with the same motion with which they were moving before they were divided, like the ones called centipedes and others that belong to the insected animals and have elongated bodies. Indeed, in all these animals even the hind part engages in progression in the same direction as the front part. And the cause of their living when they are divided is that each of them is constituted as if it were something continuous compounded out of many animals. And the reason why they are like this is clear from what was established earlier.1 The animals that are constituted according to nature in the highest 707b5 degree move naturally by means of two or four points, and similarly those among the blooded animals that are footless. For these, too, move by means of four points in virtue of which they make their motion. Indeed, they move forward by using two bends. The reason is that the right and the left are found in the front as well as in the back on the wide side of each of their bends: the front right and the front left in the part 1

IA 6, 707a6–16.

7, 707b12–28



κεφαλὴν μέρει τὸ πρόσθιον σημεῖον δεξιόν τε καὶ ἀριστερόν, ἐν δὲ τῷ πρὸς τὴν οὐρὰν τὰ ὀπίσθια σημεῖα. δοκεῖ δὲ δυοῖν σημείοιν κινεῖσθαι, τῇ τ’ ἔμπροσθεν ἁφῇ καὶ τῇ ὕστερον. αἴτιον δ’ ὅτι στενὸν κατὰ πλάτος ἐστίν, ἐπεὶ καὶ ἐν τούτοις τὸ δεξιὸν ἡγεῖται, καὶ ἀνταποδίδωσι κατὰ τὸ ὄπισθεν, ὥσπερ ἐν τοῖς τετράποσι. τῶν δὲ κάμψεων αἴτιον τὸ μῆκος· ὥσπερ γὰρ οἱ μακροὶ τῶν ἀνθρώπων λορδοὶ βαδίζουσι, καὶ τοῦ δεξιοῦ ὤμου εἰς τὸ πρόσθεν ἡγουμένου (τὸ γὰρ ἀριστερὸν ἰσχίον εἰς τοὔπισθεν μᾶλλον ἀποκλίνει, καὶ τὸ μέσον κοῖλον γίνεται  καὶ λορδόν), οὕτω δεῖ νοεῖν καὶ τοὺς ὄφεις κινουμένους ἐπὶ τῇ γῇ λορδούς. σημεῖον δ’ ὅτι ὁμοίως κινοῦνται τοῖς τετράποσιν· ἐν μέρει γὰρ μεταβάλλουσι τὸ κοῖλον καὶ τὸ κυρτόν· ὅταν γὰρ τὸ ἀριστερὸν πάλιν τῶν προσθίων ἡγήσηται, ἐξ ἐναντίας πάλιν τὸ κοῖλον γίνεται· τὸ γὰρ δεξιὸν ἐντὸς πάλιν γίνεται.  σημεῖον δεξιὸν πρόσθιον ἐφ’ οὗ Α, ἀριστερὸν ἐφ’ οὗ Β, ὀπίσθιον δεξιὸν ἐφ’ οὗ Γ, ἀριστερὸν ἐφ’ οὗ Δ. B





A

B

15

20

25

A





οὕτω δὲ κινοῦνται τῶν μὲν χερσαίων οἱ ὄφεις, τῶν δ’ ἐνύδρων αἱ ἐγχέλεις καὶ οἱ γόγ12 δεξιόν τε καὶ ἀριστερόν] δεξιῶν τε καὶ ἀριστερῶν Z1, corr. Z2 | τε Z U Ca S N : om. Y P V  13–14 δυοῖν σημείοιν U Ca S N V: δυεῖν σημείοις Z, σημείοιν corr. Z2 : δυσὶ σημείοις Y P  14 τ᾽ ἔμπροσθεν ἁφῇ καὶ τῇ ὕστερον] τε ὄπισθεν ἁφῆι καὶ τῆι ἔμπροσθεν Z   15 κατὰ] τὸ Z1, corr. Z2  16 ὄπισθεν Z U Ca S N V : ὀπίσθιον Y P   19 πρόσθεν] ἔμπροσθεν fecit Z2 | γὰρ Y Jaeger : om. Z U Ca S N V P Bekker   20 ἀποκλίνει U Y N V P : ἀποκλῖναι Ca : ἀποκλίνεται Z S | καὶ τὸ Z Y P : τὸ δὲ U Ca S N V   21–22 τῇ γῇ Z U Ca S Y P : τῆς γῆς N V   22 λορδούς. σημεῖον δ᾽ Y : λορδοῖς σημείοις. τὸ δ᾽ Z e corr. U Ca N V P : λορδοῖς σημείοις δ᾽ S   22–26 ὅτι … σημεῖον] om. V   24 τὸ ἀριστερὸν πάλιν Z U Ca S N V Jaeger : πάλιν τὸ ἀριστερὸν Y P Bekker  25 ἐντὸς] ἐν τοῖς ὀπισθίοις Z   26 πρόσθιον ἐφ᾽ οὗ A U Ca S N V P : ἐφ᾽ οὗ A πρόσθιον Z Y   27 δεξιὸν ἐφ᾽ οὗ Γ, ἀριστερὸν ἐφ᾽ οὗ Δ U Z2 Ca S N V P Bekker : ἀριστερὸν ἐφ᾽ οὗ Γ, δεξιὸν ἐφ᾽ οὗ Δ Z Y Jaeger | δεξιὸν] ὀπίσθιον δεξιὸν Z | figuram restitui : non habent codd.



7, 707b12–28



that is near the head, while the back points in the part that is toward the tail. They seem, though, to be moving by means of two points by touching in front and behind. The cause is that their body is narrow in breadth, since in these animals too, as in the four-footed animals, the right part directs movement, and it is balanced by the movement of the back. The cause of their bends is the length . Indeed, just as 707b17 people who are tall walk with their body curved, and their right shoulder is leading forward (for their left hip is inclined rather toward the back so that their middle becomes hollow and curved), so too we ought to conceive of snakes as moving on the ground with their body curved. Here is evidence that they move in a similar way to the four-footed animals: they alternate the concave and the convex in a part. For whenever the left of their front part is leading again, the concave comes to be from the opposite bend. The reason is that the right becomes internal again. Let the front point on the right be A, and that on the left B, and the rear point on the right C, and that on the left D. B

D

C

A

B

A

D

C

Among land animals, snakes move in this way, while among aquatic ani- 707b27 mals, eels, conger eels, lampreys, and all the others the shape of which is



7, 707b29–8, 708a24

γροι καὶ αἱ μύραιναι, καὶ τῶν ἄλλων ὅσα ἔχει τὴν μορφὴν ὀφιωδεστέραν. πλὴν ἔνια μὲν τῶν ἐνύδρων τῶν τοιούτων  30 οὐδὲν ἔχει πτερύγιον, οἷον αἱ μύραιναι, ἀλλὰ χρῆται τῇ θαλάττῃ ὥσπερ οἱ ὄφεις τῇ γῇ καὶ τῇ θαλάττῃ (νέουσι  708a1 γὰρ οἱ ὄφεις ὁμοίως καὶ ὅταν κινῶνται ἐπὶ τῆς γῆς)· τὰ δὲ δύ’ ἔχει πτερύγια μόνον, οἷον οἵ τε γόγγροι καὶ αἱ ἐγχέλεις καὶ γένος τι κεστρέων, οἳ γίνονται ἐν τῇ λίμνῃ τῇ ἐν Σιφαῖς. καὶ διὰ τοῦτο ταῖς καμπαῖς ἐλάττοσι κινοῦν- 5 ται ἐν τῷ ὑγρῷ ἢ ἐν τῇ γῇ τὰ ζῆν εἰωθότα ἐν τῇ γῇ, καθάπερ τὸ τῶν ἐγχελύων γένος. οἱ δὲ δύο πτερύγια ἔχοντες τῶν κεστρέων τῇ καμπῇ ἀνισάζουσιν ἐν τῷ ὑγρῷ τὰ τέτταρα σημεῖα. [8] Tοῖς δ’ ὄφεσιν αἴτιον τῆς ἀποδίας τό τε τὴν φύσιν μηθὲν ποιεῖν μάτην, ἀλλὰ πάντα πρὸς τὸ ἄριστον ἀποβλέ- πουσαν ἑκάστῳ τῶν ἐνδεχομένων, διασώζουσαν ἑκάστου τὴν ἰδίαν οὐσίαν καὶ τὸ τί ἦν αὐτῷ εἶναι· ἔτι δὲ καὶ τὸ πρότερον ἡμῖν εἰρημένον, τὸ τῶν ἐναίμων μηθὲν οἷόν τ’ εἶναι πλείοσι κινεῖσθαι σημείοις ἢ τέτταρσιν. ἐκ τούτων γὰρ φανερὸν ὅτι τῶν ἐναίμων ὅσα κατὰ τὸ μῆκος ἀσύμμετρά ἐστι πρὸς τὴν ἄλ- λην τοῦ σώματος φύσιν, καθάπερ οἱ ὄφεις, οὐθὲν αὐτῶν οἷόν θ’ ὑπόπουν εἶναι. πλείους μὲν γὰρ τεττάρων οὐχ οἷόν τε αὐτὰ πόδας ἔχειν (ἄναιμα γὰρ ἂν ἦν), ἔχοντα δὲ δύο πόδας ἢ τέτταρας σχεδὸν ἦν ἂν ἀκίνητα πάμπαν· οὕτω βραδεῖαν ἀναγκαῖον εἶναι καὶ ἀνωφελῆ τὴν κίνησιν.  Ἅπαν δὲ τὸ ὑπόπουν ἐξ ἀνάγκης ἀρτίους ἔχει τοὺς πόδας. ὅσα μὲν γὰρ ἅλσει χρώμενα μόνον ποιεῖται τὴν κατὰ τόπον μεταβολήν, οὐθὲν ποδῶν πρός γε τὴν τοιαύτην δεῖται κίνησιν ὅσα δὲ χρῆται μὲν ἅλσει, μή ἐστι δ’ αὐτοῖς αὐ- 

10

15

20

29 αἱ] om. V | μύραιναι Z U Ca N V P Jaeger : σμύραιναι U S Bekker   29–31 καὶ τῶν … μύραιναι] οm. U   31 χρῆται U Ca S N V P : χρῶνται Z Y   708a1 θαλάττῃ] θαλάττῃ δὲ Y | νέουσι U Ca Y N V P : βαίνουσι Z : μένουσι S   2 οἱ U Ca S N V : καὶ οἱ Z Y : καὶ P | καὶ Z Y : om. U Ca S N V P   3 ἔχει πτερύγια] πτερύγια ἔχει S | οἷον] om. P   3–4 καὶ οἱ ἐγχέλεις Z Y Bekker : καὶ οἱ ἐγχέλυες U S Jaeger : καὶ οἱ ἐγχέλoις Ca : καὶ οἱ ἐγχέλυς N V : om. P   5 σίφαις] σίφναις Z1, corr. Z2  6 τῷ ὑγρῷ ἢ ἐν] om. Z1, add. Z2  7 πτέρυγια Y : πτέρυγας Z U Ca S N V P   11 ἐκ τῶν ἐνδεχομένων Jaeger : τῶν ἐνδεχομένων codd. Bekker | διασώζουσαν Z U S N V : καὶ διασώζουσαν Ca Y P | ἑκάστου U Ca S Y N V : ἑκάστῳ Z P   12 ἰδίαν Z S Y N V P : ἴδιον U Ca | αὐτῷ εἶναι Z1 U S Y P : αὐτῶν εἶναι Z2 : αὐτοῖς εἶναι ex αὐτῷ εἶναι Ca : εἶναι αὐτῷ N V | τὸ πρότερον] διὰ τὸ πρότερον Z2  13 οἷον] om. Z1, add. Z2  14 τέτταρσιν Z U Y N V P, τέταρσιν S : τέτρασιν Ca  15 ἄλλην] ὅλην Z2 s.l.   17 θ᾽ ὑπόπουν Z1 U Y P : θ᾽ ὑπόποδον Z2 : τε ὑπόπουν Ca : καθ᾽ ὑπόπουν S : τε ὑπόποδον N V   18 δὲ] γὰρ Y | ἢ] καὶ Z1, corr. Z2 s.l.   19 ἦν ἂν Z P : ἂν ἦν U S : ἦν Ca Y N V   20 εἶναι] φάναι Z   22 μόνον U Ca S N V : om. Z Y P  



7, 707b29–8, 708a24



snake-like. Except that some of these aquatic animals have no fins, for instance lampreys, but they use the sea in the way snakes use both the land and the sea (indeed snakes swim in the same manner as when they move on land). Others have only two fins, for instance conger eels, eels, and a certain kind of mullets that live in the lake at Siphae. For this reason, too, those that are accustomed to live on land move with fewer bends in water than on land, for instance the class of eels. Those mullets that have two fins obtain the four points in water by means of their bending.

[8] The cause of footlessness in snakes is that nature does nothing in vain, 708a9 but rather it always aims to achieve the best from among the possibilities for each animal, preserving the proper substance and essence of each of them. And, moreover, what we have said earlier, namely that none of the blooded animals can move by means of more than four points. From these considerations, it becomes clear that those blooded animals that, with respect to their length, are disproportionate compared to the nature of the rest of their bodies, as is the case with snakes, cannot have feet. The reason is that they cannot have more than four feet (for then they would be bloodless), while if they had two or four feet they would be almost completely motionless. Either way their motion would be necessarily slow and useless. Every footed animal must of necessity have an even number of feet. 708a21 For, all those animals that make their own change with respect to place by using only jumping do not need feet at all for this type of movement. However, as for those that use jumping, though this kind of motion is



8, 708a25–b16

τάρκης αὕτη ἡ κίνησις ἀλλὰ καὶ πορείας προσδέονται, δῆ- λον ὡς τοῖς μὲν βέλτιον τοῖς δ’ ὅλως ἀδύνατον πορεύεσθαι. [διότι πᾶν ζῷον ἀναγκαῖον ἀρτίους ἔχειν τοὺς πόδας.] οὔσης γὰρ τῆς τοιαύτης μεταβολῆς κατὰ μέρος, ἀλλ’ οὐκ ἀθρόῳ παντὶ τῷ σώματι καθάπερ τῆς ἅλσεως, ἀναγκαῖόν ἐστι τοῖς μὲν μένειν μεταβαλλόντων τῶν ποδῶν τοῖς δὲ κινεῖσθαι,  καὶ τοῖς ἀντικειμένοις τούτων ποιεῖν ἑκάτερον, μεταβάλλον ἀπὸ τῶν κινουμένων ἐπὶ τὰ μένοντα τὸ βάρος. διόπερ οὔτε τρισὶ μὲν οὐθὲν οὔθ’ ἑνὶ χρώμενον βαδίζειν οἷόν τε· τὸ μὲν γὰρ οὐθὲν ὅλως ὑπόστημα ἔχει ἐφ’ ᾧ τὸ τοῦ σώματος ἕξει βάρος, τὸ δὲ κατὰ τὴν ἑτέραν ἀντίθεσιν μόνην, ὥστ’ ἀναγκαῖον αὐτὸ οὕτως ἐπιχειροῦν κινεῖσθαι πίπτειν. ὅσα δὲ πολύποδά ἐστιν, οἷον αἱ σκολόπενδραι, τούτοις δυνατὸν μὲν καὶ ἀπὸ  περιττῶν ποδῶν πορείαν γίνεσθαι, καθάπερ φαίνεται ποιούμενα καὶ νῦν, ἄν τις αὐτῶν ἕνα πηρώσῃ τῶν ποδῶν, διὰ τὸ τὴν τῶν ἀντιστοίχων ποδῶν κολόβωσιν ἰᾶσθαι τῷ λοιπῷ πλήθει τῶν ἐφ’ ἑκάτερα ποδῶν· γίνεται γὰρ τούτοις οἷον ἔφελξις τοῦ πεπηρωμένου μορίου τοῖς ἄλλοις ἀλλ’ οὐ βάδισις.  οὐ μὴν ἀλλὰ φανερόν γε ὅτι βέλτιον ἂν καὶ ταῦτα ποιοῖτο τὴν μεταβολὴν ἀρτίους ἔχοντα τοὺς πόδας, καὶ μηθενὸς ἐλλείποντος, ἀλλ’ ἀντιστοίχους ἔχοντα τοὺς πόδας· οὕτω γὰρ ἂν αὑτῶν ἀνισάζειν τε δύναιντο τὸ βάρος καὶ μὴ ταλαντεύειν ἐπὶ θάτερα μᾶλλον, εἰ ἀντίστοιχα ἐρείσματ’ ἔχοι καὶ  μὴ κενὴν τὴν ἑτέραν χώραν τῶν ἀντικειμένων. προ-

25

30 708b1

5

10

15

25 αὕτη ἡ Z Ca Va.c. Y P : ἡ τοιαύτη U N Vp.c. : τοιαύτη S | προσδέονται Y P : δέονται U Z : προσδεῖται Ca S N V   26 ὡς Bekker : ὅτι Z1, ἔστι add. Z2 : ὡς ἔστι U Ca S Y N V : ὡς εἰ P  27 διότι … πόδας οm. U C a S N V P, ut adnotationem del. Jaeger : habent Z Y Bekker  28 ἀθρόῳ Z U S Y : ἀθρόας Ca N V P   30 μεταβαλλόντων] τὸ μεταβάλλον Y | μεταβαλλόντων τῶν ποδῶν Z Ca Y P : om. U S N V   31 καὶ] om. Y   32 τὸ] ἐπὶ τὸ Z   32–708b1 οὔτε τρισὶ μὲν οὐθὲν οὔθ᾽ ἑνὶ Jaeger cum Mich. : οὐδὲ τρισὶ μὲν οὐδὲν οὐδ᾽ ἑνὶ U Ca S N V : οὐδέ τισι μὲν οὐθὲν οὐδενὶ S : τρεῖς μὲν οὐθ᾽ (sic) ἑνὶ οὐδὲν Z1 : τρισὶ μὲν οὐθ᾽ ἑνὶ οὐδὲν Z2 : τρισὶ μὲν οὐθὲν οὐδενὶ Y P Bekker   1-2 oὐθὲν ὅλως Z U S : ὅλως οὐθέν Ca : ὅλως οὐδέν N V Y P   2 ὑπόστημα] ἀπόστημα V | ἐφ᾽ ᾧ Z2 Ca N V : ἐφ᾽ ὃ U Y P : ἀφ᾽ οὗ Z : om. S | ἕξει] ἔχει V   3 μόνην Z Y e corr. : πόνον U Ca S N V P   5 αἱ] om. S | καὶ] om. Z   6 πορείαν Ca Y N V P : καὶ πορείαν Z U S   7 τις] τι U   8 τὸ Z U S Y P : om. Ca N V   9 τούτοις] τούτων Z1, corr. Z2  9–10 οἷον ἔφελξις U Ca S Y N V : ἔφελξις οἷον Z P   10 τοῦ] τοῦ τε U | ἀλλ᾽ οὐ βάδισις codd. Bekker : del. Jaeger   11 φανερόν γε Y P : φανερὸν Z : καὶ φανερὸν U Ca S N V  12–13 καὶ … πόδας Z2 U Ca S N V P : om. Z Y   13 οὕτω γὰρ Z1 U Y P : οὕτως ἄρα Z2 Ca N V : οὕτως S | ἂν V2 add. Jaeger : om. cett.| αὐτῶν U Ca S N V P : αὐτῶν μᾶλλον Z Y   14 ἀνισάζειν codd. Bekker Jaeger : ἂν ἰσάζειν cum Moerbeke (utique equare) proposuit De Leemans | δύναιντο] δύναται Z | τε] om. Z1, add. Z2  16 τῶν] τὴν τῶν Z  



8, 708a25–b16



not sufficient for them as they also need progression, it is clear that for some of them it is better in this way, while for others it would be altogether impossible to progress otherwise. [That is why every animal necessarily possesses an even number of feet.] The reason is that, since this kind of change with respect to place is made part by part and not by the whole body all at once as in jumping, it is necessary that some of the feet that change remain at rest while others move, and for the animal to effect each of the two by means of the feet that stand in opposition, transferring its weight from the moving feet to the feet at rest. And that is why no animal can walk either by using three feet or one foot. In the latter case the animal has absolutely no support on which it could rest the weight of its body, whereas in the former it can rest only in one of the two opposite pairs, so that in attempting to move in this way it will necessarily fall. But animals that are many-footed, like centipedes, are capable of 708b4 engaging in progression even with an odd number of feet, as they can be seen to do even now if someone mutilates one of their feet; for they can remedy the mutilation of one foot out of any of the opposite feet with the remaining multitude of feet present on each of their two sides. What occurs, then, is a dragging along of the mutilated part by the other feet rather than proper walking. It is clear, nevertheless, that these animals, too, would make the change in a more efficient way if they had an even number of feet and no one foot missing, but rather having all their corresponding feet. The reason is that in this way, namely if they had all their corresponding supports, and if they had no empty space in one of the two opposite rows, they would be able to distribute the weight evenly and not sway more to the one side rather than the other. Animals that



8, 708b17–9, 709a9

βαίνει δ’ ἀφ’ ἑκατέρου τῶν μερῶν ἐναλλὰξ πορευόμενα· οὕτω γὰρ εἰς ταὐτὸ τῷ ἐξ ἀρχῆς σχήματι γίνεται ἡ κατάστασις. | ὅτι μὲν οὖν ἀρτίους ἔχει τοὺς πόδας πάντα, καὶ διὰ 19–20 τίν’ αἰτίαν εἴρηται.  21 [9] Ὅτι δ’ εἰ μηθὲν ἦν ἠρεμοῦν, οὐκ ἂν ἦν  κάμψις οὐδ’ εὔθυνσις, ἐκ τῶνδε δῆλον. ἔστι γὰρ κάμψις μὲν ἡ ἐξ εὐθέος ἢ εἰς περιφερὲς ἢ εἰς γωνίαν μεταβολή, εὔθυνσις δ’ ἡ ἐκ θατέρου τούτων εἰς εὐθύ. ἐν ἁπάσαις δὲ ταῖς εἰρημέναις μεταβολαῖς ἀνάγκη πρὸς ἓν σημεῖον τὴν κάμψιν  ἢ τὴν εὔθυνσιν γίνεσθαι. ἀλλὰ μὴν κάμψεώς γε μὴ οὔσης οὔτ’ ἂν πορεία οὔτε νεῦσις οὔτε πτῆσις ἦν. τὰ μὲν γὰρ ὑπόποδα ἐπειδὴ ἐν ἑκατέρῳ τῶν ἀντικειμένων σκελῶν ἐν μέρει ἵσταται καὶ τὸ βάρος ἴσχει, ἀναγκαῖον θατέρου προβαίνοντος θατέρου ποιεῖσθαι κάμψιν· ἴσα τε γὰρ πέφυκεν ἔχειν τῷ  μήκει τὰ ἀντίστοιχα κῶλα, καὶ ὀρθὸν δεῖ εἶναι τὸ ὑφεστὸς τῷ βάρει, οἷον κάθετον πρὸς τὴν γῆν. ὅταν δὲ προβαίνῃ, γίνεται ἡ ὑποτείνουσα καὶ δυναμένη τὸ μένον μέγεθος καὶ  τὴν μεταξύ. ἐπεὶ δ’ ἴσα τὰ κῶλα, ἀνάγκη κάμψαι τὸ μένον, ἢ ἐν τῷ γόνατι ἢ ἐν τῇ κάμψει, οἷον εἴ τι ἀγόνατον εἴη τῶν βαδιζόντων. σημεῖον δ’ ὅτι οὕτως ἔχει· εἰ γάρ τις ἐν γῇ βαδίζοι παρὰ τοῖχον , ἡ γραφομένη ἔσται οὐκ  εὐθεῖα ἀλλὰ σκολιά, διὰ τὸ ἐλάττω μὲν κάμπτοντος γίνεσθαι τὴν γραφομένην, μείζω δ’ ἱσταμένου καὶ ἐξαίροντος. ἐνδέχεται μέντοι κινεῖσθαι καὶ μὴ ἔχοντος καμπὴν τοῦ σκέλους, ὥσπερ τὰ παιδία ἕρπουσι. καὶ περὶ τῶν ἐλεφάντων ὁ

21

25

30 709a1

5 5

17 ἀφ᾽ ἑκατέρου U Ca Y P : ἀφ᾽ ἑτέρου Z1 : ἀφ᾽ ἑκάτερα Z2 : ἑκατέρου S : ἐφ᾽ ἑκάτερα N V | πορευόμενα S U : πορευόμενον Ca N V Jaeger cum Mich. : τὸ πορευόμενον P Y Z Bekker   20 ἔχει τοὺς πόδας Y P : πόδας ἔχει Z U S : ἔχει πόδας Ca N V   22 οὐδ᾽] οὐδὲ νεῦσις οὐδ᾽ Y | δε δῆλον] om. in fenestra S   23 εὐθέος ἢ] εὐθέος P   24 ἐκ θατέρου Z Y : ἑκατέρου U Ca S N V P   26 γε Z U Ca S Y P : τε N V   27 οὔτε νεῦσις οὔτε πτῆσις Y P : οὔτε πτῆσις οὔτε νεῦσις Ca S N V : οὔτε νεῦσις Z U | μὲν] om. V   28 ἐπειδὴ] om. Z1, ἐπεὶ add. Z2  29 ἴσχει Z Y P : ἔχει U Ca N V S | προβαίνοντος] μεταβαίνοντος Z   30 τε γὰρ Z U C a S N V : γάρ τε Y P   31 τὰ Z Y : καὶ U Ca S N V P   709a1 ἡ] om. S   3 ἐν τῇ] ἐν Y | εἴ τι ] εἰσὶν S   5 ἐν γῇ βαδίζοι παρὰ τοῖχον scripsi : ἐν γῇ (Z e corr. ) βαδίζοι (βαδίζει Y) παρὰ τοῖχον codd. Bekker : ἔχων ἐν τῇ κεφαλῇ κάλαμον μετὰ μέλανος καὶ ἁπτόμενον τοῦ τοίχου βαδίζοι παρὰ τοῖχον Mich.p : ἐν γῇ **** βαδίζοι παρὰ τοῖχον Jaeger : ἐγγὺς (cum Moerbeke vicinus) βαδίζοι παρὰ τοῖχον proposuit De Leemans | ἔσται οὐκ Z U Ca N V P : οὐκ ἔσται Y : εὕραι (sic) οὐκ S   6 μὲν Z Ca S Y P : om. U N V   7 τὴν γραφομένην] τὸ γραφομένον (sic) Z2 | καὶ Z2 Ca S Y N V P : καὶ οὐκ Z1 U | ἐξαίροντος] ἐξαιροῦντος Y   8 μέντοι] μὲν V | μὴ Ca S Y N V P : οὐκ Z U   9 ὁ] om. Y  



8, 708b17–9, 709a9



advance progress from each one of the two parts alternately, for it is in this way that the restoration of the self-same initial configuration is achieved. It has now been established that all animals have an even number of feet and due to what cause. [9] That if nothing were at rest no bending or straightening could occur, 708b21 is evident from what follows. For bending is the change from what is straight to what is curved or angled, straightening is the change from either of these to what is straight. In all such changes, the bending or straightening must necessarily be relative to one point. Moreover, without bending there could not be walking or swimming 708b26 or flying. The reason is that, since footed animals stand and take their weight alternately on one or the other of their opposite legs, as one leg strides forward the other must necessarily be bent. For the opposite legs are naturally of equal length, and the one that is under the weight must be a kind of perpendicular at right angles to the ground. When, then, one leg strides forward, it becomes the hypotenuse of a right-angled triangle. Its square then is equal to the square on the other side together with the square on the base. But since the legs are equal, the one at rest must bend either at the knee or, in any kneeless animal that walks, at some other joint. This is shown by the following fact: if a human being were to walk on the ground along a wall the line described would not be straight but zigzag, because it would go lower when the human being bends and higher when it stands and raises itself. It is, however, possible to move even if the leg has no bend, as when 709a8 children crawl. (This is the old account of the movement of elephants,



9, 709a10–b2

παλαιὸς ἦν λόγος τοιοῦτος, οὐκ ἀληθὴς ὤν. κινεῖται δὲ καὶ τὰ τοιαῦτα κάμψεως γινομένης ἐν ταῖς ὠμοπλάταις ἢ τοῖς ἰσχίοις. ἀλλ’ ὀρθὸν οὐδὲν δύναιτ’ ἂν πορευθῆναι συνεχῶς καὶ ἀσφαλῶς, κινηθείη δ’ ἂν οἷον ἐν ταῖς παλαίστραις οἱ διὰ τῆς κόνεως προϊόντες ἐπὶ τῶν γονάτων πολὺ γὰρ τὸ ἄνω μέρος, ὥστε δεῖ μακρὸν εἶναι τὸ κῶλον· εἰ δὲ τοῦτο, κάμ- ψιν ἀναγκαῖον εἶναι. ἐπεὶ γὰρ ἕστηκε πρὸς ὀρθήν, εἰ ἄκαμπτον ἔσται τὸ κινούμενον εἰς τὸ πρόσθεν, ἢ κατα- πεσεῖται ἐλάττονος τῆς ὀρθῆς γινομένης ἢ οὐ προβήσεται. εἰ γὰρ ὀρθοῦ ὄντος θατέρου σκέλους θάτερον ἔσται προβεβηκός, μεῖζον ἔσται, ἴσον ὄν· δυνήσεται γὰρ τοῦτο τό τ’ ἠρεμοῦν καὶ τὴν ὑποτείνουσαν. ἀνάγκη ἄρα κάμπτεσθαι τὸ προϊόν, καὶ  κάμψαν ἅμα ἐκτείνειν θάτερον, ἐκκλίνειν τε καὶ διαβεβηκέναι καὶ ἐπὶ τῆς καθέτου μένειν· ἰσοσκελὲς γὰρ γίνεται τρίγωνον τὰ κῶλα, καὶ ἡ κεφαλὴ γίνεται κατώτερον, ὅταν κάθετος ᾖ ἐφ’ ἧς βέβηκε. τὰ δ’ ἄποδα τὰ μὲν κυμαίνοντα προέρχεται (τοῦτο δὲ διττῶς συμβαίνει· τὰ μὲν γὰρ ἐπὶ  τῆς γῆς, καθάπερ οἱ ὄφεις, τὰς καμπὰς ποιεῖται, τὰ δ’ εἰς τὸ ἄνω, ὥσπερ αἱ κάμπαι), ἡ δὲ κύμανσις καμπή ἐστι· τὰ δ’ ἰλυσπάσει χρώμενα, καθάπερ τὰ καλούμενα γῆς ἔντερα καὶ βδέλλαι. ταῦτα γὰρ τῷ μὲν ἡγουμένῳ προέρχεται, τὸ δὲ λοιπὸν σῶμα πᾶν πρὸς τοῦτο συνάγουσι, καὶ τοῦ- τον τὸν τρόπον εἰς τόπον ἐκ τόπου μεταβάλλουσι. φανερὸν δ’ ὅτι εἰ μὴ αἱ δύο τῆς μιᾶς μείζους ἦσαν, οὐκ ἂν ἐδύναντο κινεῖσθαι τὰ κυμαίνοντα τῶν ζῴων. ἐκταθείσης γὰρ τῆς καμ- πῆς, εἰ ἴσην κατεῖχεν, οὐθὲν ἂν προῄεσαν νῦν δ’ ὑπερβάλλει

10

15 16b

20

25

30 709b1

10 κινεῖται] γίνεται Z   11 τοῖς] ἐν τοῖς Z   12 δύναιτ᾽ ἂν Z Ca S N V : δύναται Y P  14 κόνεως] κώνωσεως (sic) Z1, κόνωσεως (sic) Z2 | γὰρ] γάρ ἐστι Z   16 ὀρθήν U S Y N V P : ὀρθόν Z Ca  16b εἰ ἄκαμπτον ἔσται τὸ κινούμενον εἰς τὸ πρόσθεν Y Jaeger : ἢ διακάμπον ἔσται κινούμενον εἰς τὸ πρόσθεν Ca : εἰ (ex ἢ) ἄκαμπτον ἔσται τὸ κινούμενον Z : om. U S N V P Bekker  16b ἢ Z2 Ca Y P : om. U S N V   16b–17 καταπεσεῖται ἐλάττονος τῆς ὀρθῆς γινομένης U S, γρ. pro ἄκαμπτον ἔσται τὸ κινούμενον Z2 : καταπεσεῖται τῆς ὀρθῆς ἐλάττονος γινομένης N V P : καταπεσεῖται ἐκείνως τῆς ὀρθῆς γινομένης Ca : καταπεσεῖται ἐκτὸς τῆς καθέτου γινόμενον Y   18 σκέλους] τοῦ σκέλους Z2  19 δυνήσεται] οὐ δυνήσεται Z1, corr. Z2  20 τὸ Z1 U Y P Bekker : τό τε Z2 Ca S N V Jaeger   21 κάμψαν Z U Ca S N V : κάμψαντα Y P | ἅμα] ταδ᾽ ἅμα Z1, corr. Z2  21–22 ἐκκλίνειν τε καὶ διαβεβηκέναι καὶ ἐπὶ τῆς καθέτου μένειν Y : ἐκκλίνειν καὶ διαβεβηκέναι τε καὶ ἐπὶ τῆς καθέτου μένειν Z : τὸν πόδα ἐπὶ τῆς καθέτου μενούσης P : οm. U Ca S N V   22 γίνεται] om. Y   23 τὰ κῶλα] τὰ κῶλα τοῦ ζῴου P   24 ᾖ] εἴη Z   26 ποιεῖται Z U Ca S Y P : ποιεῖσθαι N V   27 ὥσπερ Z U Ca S Y P : καθάπερ N V | δὲ Z Y P : δ᾽ αὖ U Ca S N V Z3  28 ἰλυσπάσει Z2 U S N V P : δινήσει Z1 : ἰλυ Ca : εἰλύσει Y | καθάπερ τὰ καλούμενα] οm. Z1, add. Z2  29 μὲν Z U Ca S Y P : expunxit Z2 : om. N V | προέρχεται U Ca S Y N V : προέρχονται Z P   30 πρὸς] εἰς Z   32 μὴ] μήτε Z1, corr. Z2  709b2 δ᾽] θ᾽ U



9, 709a10–b2



but it is untrue.) Such a crawling movement involves a bending in the shoulders or the hips. But nothing could progress upright in this way continuously and safely, but would only move like men in the wrestling schools who convey themselves forward through the dust on their knees. The reason is that the upper portion of the body is big, so the leg must be long; consequently, there must be a bending. Since a standing position is perpendicular , if that which moves forward does not bend, it will either fall as the right angle becomes less, or else it will not advance at all. If one leg is at right angles and the other is advanced, the latter will be at once equal and greater; it will be equal to the leg at rest and also to the hypotenuse of the right-angled triangle. Therefore, that which goes forward must bend, and while bending one, extend the other leg, and incline forward at the same time and make a stride and remain above the perpendicular; for the legs form an isosceles triangle and the head goes lower when it is perpendicular to the triangle’s base. Some footless animals advance by undulations (this happens in two 709a24 ways: for some, the bending is upon the ground, such as with snakes, while for others it is up and down, such as with caterpillars) and undulation is bending. Others move by oozing, such as what are called earthworms and leeches. For these advance with one part leading the way, and then drawing the remainder of the body to them, and in this way they change from place to place. It is clear that, if the two lines they form were not greater than the one, 709a31 movement would be impossible for undulating animals. The reason is that, when the bend is extended, they would not have made any advance, if it subtended an equal line; as it actually is, the line is longer when it is



9, 709b3–10, 709b28

ἐκταθεῖσα, καὶ ἠρεμήσαντος τούτου ἐπάγει τὸ λοιπόν. ἐν ἁπάσαις δὲ ταῖς λεχθείσαις μεταβολαῖς τὸ κινούμενον ὁτὲ μὲν ἐκτεινόμενον εἰς εὐθὺ προέρχεται, ὁτὲ δὲ συγκαμπτόμενον,  τοῖς μὲν ἡγουμένοις μέρεσιν εὐθὺ γινόμενον, τοῖς δ’ ἑπομένοις συγκάμπτον. ποιεῖται δὲ καὶ τὰ ἁλλόμενα πάντα κάμψιν ἐν τῷ ὑποκειμένῳ μέρει τοῦ σώματος, καὶ τοῦτον τὸν τρόπον ἔχοντα ἅλλεται. καὶ τὰ πετόμενα δὲ καὶ τὰ νέοντα, τὰ μὲν τὰς πτέρυγας εὐθύνοντα καὶ κάμπτοντα πέταται, τὰ  δὲ τοῖς πτερυγίοις, καὶ τούτων τὰ μὲν τέτταρσι τὰ δὲ δυσίν, ὅσα προμηκέστερα τὴν μορφήν, ὥσπερ τὸ τῶν ἐγχελύων γένος· τὴν δὲ λοιπὴν κίνησιν ἀντὶ τῶν δύο πτερυγίων τῷ λοιπῷ τοῦ σώματος καμπτόμενα νεῖ, καθάπερ εἴρηται πρότερον. οἱ δὲ πλατεῖς τῶν ἰχθύων τῇ μὲν τῷ πλάτει χρῶν- ται τοῦ σώματος ἀντὶ πτερυγίων, τῇ δὲ πτερυγίοις δυσί. τὰ δὲ πάμπαν πλατέα, καθάπερ ὁ βάτος, αὐτοῖς τοῖς πτερυγίοις καὶ ταῖς ἐσχάταις τοῦ σώματος περιφερείαις εὐθύνοντα καὶ κάμπτοντα ποιεῖται τὴν νεῦσιν. [10] Ἀπορήσειε δ’ ἄν τις ἴσως πῶς κινοῦνται τέτταρσι ση- μείοις οἱ ὄρνιθες, ἢ πετόμενοι ἢ πορευόμενοι, ὡς εἰρημένου ὅτι πάντα τὰ ἔναιμα κινεῖται τέτταρσιν· οὐκ εἴρηται δέ, ἀλλ’ ὅτι οὐ πλείοσιν. οὐ μὴν ἀλλ’ οὔτ’ ἂν πέτεσθαι δύναιντο ἀφαιρεθέντων τῶν κώλων οὔτε πορεύεσθαι τῶν πτερύγων ἀφαιρεθεισῶν, ἐπεὶ οὐδ’ ἄνθρωπος βαδίζειν μὴ κινῶν τι τοὺς ὤμους. ἀλλὰ πάντα γε, καθάπερ εἴρηται, κάμψει καὶ ἐκτάσει ποιεῖται τὴν μεταβολήν· ἅπαντα γὰρ εἰς τὸ ὑποκείμενον μέχρι τινὸς οἷον εἰς ὑπεῖκον προέρχεται, ὥστ’ ἀ-

5

10

15

20

25

3 ἁπάσαις Z Y P Jaeger : πάσαις U Ca S N V Bekker   5 συγκαμπτόμενον Z P : συγκαπτόμενον U Y S Ca N V   7 συγκάμπτον Z P : συγκάπτον U S Ca N V : συγκαπτόμενον Y : συγκαμπτόν Bekker Jaeger | τὴν Jaeger cum Mich. : om. codd. Bekker   9 τὰ νέοντα] νέοντα Y   10 τὰς Z U Ca S N V : om. Y P | πέταται Z1 U Ca S Y P : πέτεται N Z3 : πέτονται V   11 τέτταρσι] τέτρασι Y   12 ἐγχελύων Z2 P Jaeger : ἐγχέλεων Z1 U Ca S Y N V Bekker   14 νεῖ] κινεῖ S   15 τῇ] πῆ V   16 πτερυγίων] τῶν πτερυγίων Z2 | τῇ] πῆ V   17 πάμπαν] πάντα Z1, corr. Z2 | ὁ βάτος Z U Ca S Y P : βάτος V : οm. N   19 ποιεῖται τὴν νεῦσιν Z Ca Y N V P : τὴν νεῦσιν ποιεῖται U S   20 τέτταρσι Z U S N V P : τέτρασι Ca Y   21 οἱ Z Ca Y P : om. U S N V | πετόμενοι ἢ πορευόμενοι Z U S N V : πετoύμενοι ἢ πορευόμενοι Ca : πορευόμενοι ἢ πετόμενοι Y P   22 τὰ ἔναιμα κινεῖται] κινεῖται τὰ ἔναιμα Z | τέτταρσι Z U S N V P : τέτρασι Ca Y   23 οὐ μὴν Z U Ca S N V : om. Y P | δύναιντο Y, ex δύναιτο V : δύναιτο Z U Ca S N P   25 βαδίζειν Z Y : βαδίζει U Ca S N V P   25–26 τι τοὺς ὤμους S Jaeger : τι τοῦ σώματος Z1 U : τοὺς ὤμους Z2 Ca Y N V P Bekker   26 γε] om. S   26–27  εἴρηται … τὸ] οm. Z1, add. Z2 mg  28 oἷον εἰς ὑπεῖκον προέρχεται] καὶ οἱονεὶ συνυπεῖκον προέρχενται (sic) Z1 (προέρχεται Z2)



9, 709b3–10, 709b28



extended, and then this part stays still and draws up the remainder. In all the aforementioned changes, that which moves advances by first extending itself straight and then by curving itself; it straightens itself with its leading part and curves itself in the parts which follow. All jumping animals as well make a bend in the lower part of the body, 709b7 and jump in this manner. So too flying and swimming animals progress, the one straightening and then bending their wings to fly, the other their fins to swim. Some of the latter have four fins and others, those with a longer shape, for instance eels, have two. These move by substituting a bending of the rest of their body for the missing pair of fins, as we have already said. Flat fish use two fins, and the flat part of their body instead of the second pair of fins. Really flat fish, like the ray, produce their swimming with the actual fins and with the outer periphery of their body, alternately bending and straightening.

[10] But someone might perhaps be puzzled as to how birds move by 709b20 means of four points, either when flying or when progressing on land, thinking that we said that all the blooded animals move by means of four points. But actually we did not say that; but rather that they do not move by more than four . However, birds would not be able to fly if their legs were taken away, nor progress on land if their wings were taken away, since nor even could a human being walk if it were not moving its shoulders a little. But all, at any rate, make their change by means of bending and 709b26 straightening, as was said. For all go forward into what is beneath them, up to a point, that is, into what yields. So that even if the bending does



10, 709b29–710a19

ναγκαῖον, εἰ μὴ καὶ κατ’ ἄλλο μόριον γίνεται ἡ κάμψις, ἀλλ’ ὅθεν γε ἡ ἀρχὴ τοῖς μὲν ὁλοπτέροις τοῦ πτεροῦ, τοῖς δ’  ὄρνισι τῆς πτέρυγος, τοῖς δ’ ἄλλοις τοῦ ἀνάλογον μορίου, καθάπερ τοῖς ἰχθύσι· τοῖς δ’, ὥσπερ οἱ ὄφεις, ἐν ταῖς καμπαῖς τοῦ σώματός ἐστιν ἡ ἀρχὴ τῆς κάμψεως. τὸ δ’ ὀρροπύ- γιόν ἐστι τοῖς πτηνοῖς πρὸς τὸ κατευθύνειν τὴν πτῆσιν, καθάπερ τὰ πηδάλια τοῖς πλοίοις. ἀναγκαῖον δὲ καὶ ταῦτα ἐν τῇ προσφύσει κάμπτειν. διόπερ τά τε ὁλόπτερα καὶ τῶν σχιζοπτέρων οἷς τὸ ὀρροπύγιον ἀφυῶς ἔχει πρὸς τὴν εἰρημέ- νην χρῆσιν, οἷον τοῖς τε ταῷς καὶ τοῖς ἀλεκτρυόσι καὶ ὅλως τοῖς μὴ πτητικοῖς, οὐκ εὐθυποροῦσι· τῶν μὲν γὰρ ὁλοπτέρων ἁπλῶς οὐθὲν ἔχει ὀρροπύγιον, ὥστε καθάπερ ἀπηδάλιον πλοῖον φέρεται, καὶ ὅπου ἂν τύχῃ ἕκαστον αὐτῶν προσπίπτει, ὁμοίως τά τε κολεόπτερα, οἷον κάνθαροι καὶ μηλολόνθαι,  καὶ τὰ ἀνέλυτρα, οἷον μέλιτται καὶ σφῆκες. καὶ τοῖς μὴ πτητικοῖς ἀχρεῖον τὸ ὀρροπύγιόν ἐστιν, οἷον τοῖς τε πορφυρίωσι καὶ ἐρωδιοῖς καὶ πᾶσι τοῖς πλωτοῖς· ἀλλ’ ἀντὶ τοῦ ὀρροπυγίου πέτανται τοὺς πόδας ἀποτείνοντα, καὶ χρῶνται ἀντ’ ὀρροπυγίου τοῖς σκέλεσι πρὸς τὸ κατευθύνειν τὴν πτῆσιν. βρα- δεῖα δ’ ἡ πτῆσις τῶν ὁλοπτέρων ἐστὶ καὶ ἀσθενὴς διὰ τὸ μὴ κατὰ λόγον ἔχειν τὴν τῶν πτερῶν φύσιν πρὸς τὸ τοῦ σώματος βάρος, ἀλλὰ τὸ μὲν πολύ, τὰ δὲ μικρὰ καὶ ἀσθενῆ. ὥσπερ ἂν οὖν εἰ ὁλκαδικὸν πλοῖον ἐπιχειροίη κώπαις

30 710a1

5

10

15

29 καὶ Z1 U Ca S N V P : om. Z2 Y | ἡ Z Y P : om. U Z2 Ca S N V   30 γε : om. Z Y P | πτεροῦ] πτιλοῦ Z1, corr. Z2  31 τῆς om. U   32 περ … ὄφεις] om. Z1, add. Z2   710a1 ὀρροπύγιον U Ca S N V Jaeger : οὐροπύγιον Z Y P Bekker   1–2 ὀρροπύγιον … πτηνοῖς] om. Z1, add. Z2  3 τοῖς] ἐπὶ τοῖς Z   4 διόπερ U Ca N Y P : διότιπερ U S : διὸ V | ὁλόπτερα] ὁλόποδα S  4–5 τῶν σχιζοπτέρων Z1 P : τὰ σχιζόπτερα Y : τοῖς σχιζοπτέροις U : τισι σχιζοπτέροις Ca S N V Z3  5 οἷς Z1 : τισὶ P : om. U Ca S Y N V Z3 | ὀρροπύγιον U Ca S N V Jaeger : οὐροπύγιον Y P Bekker | ἀφυῶς] ἀφυὲς Y   6 ταῷς U : ταῷσι Ca Y : ταοῖς Z N V P : τούτοις S   7 οὐκ] καὶ οὐκ Z2 | ὁλοπτέρων Z2 Ca N V1 Y P : κολεοπτέρων Z U S V2  8 οὐθὲν Z U S Y P : οὐδὲν Ca N V | ὀρροπύγιον U Ca S N V Jaeger : οὐροπύγιον Y P Bekker | ἀπηδάλιον πλοῖον U Z2 Ca S Y N V P : ἀπήδαλον πλοῖον Z1 Bekker Jaeger   9 καὶ Z U Y P : om. Z3 Ca S N V   10 δὲ add. Jaeger : om. libri | κολεόπτερα Z U Ca S N : κουλεόπτερα V Y P | οἷον U Ca S N V : οἷον οἱ Z Y P | καὶ Z U Ca S N V : καὶ αἱ Y P   11 μέλιτται] μέλιττα U   12 ὀρροπύγιον U Ca S N V Jaeger : οὐροπύγιον Y P Bekker   13 ἀλλ᾽] om. P   14 πέτανται Z U Ca S Y P : πέτονται N V Z3 | ἀποτείνοντα] ἀποτείναντα Y   16 ὁλοπτέρων] κολεοπτέρων Z1, κουλεοπτέρων fecit Z3  19 ὁλκαδικὸν] ὀλκαδὸν Z1, ὀλκαδικὸν fecit Z3 | ἐπιχειροίη κώπαις] κώπαις ἐπιχειροίη Y



10, 709b29–710a19



not come about in another part, nevertheless it is necessary for it to come about at least from where the origin is – the origin of the membranous wing in the case of whole-winged animals, of the feathered wing in the case of birds, and of the analogous part in the others, such as fish. And for some, like snakes, the origin of the bending is in the joints of the body. And the tail in winged animals is for the purpose of keeping course in 710a1 flight, just like the rudders in ships. And it is necessary for these too to bend at the point of attachment. This is precisely why certain animals do not move in a straight course, namely whole-winged animals and those among the split-winged animals whose tails are not naturally suited for the aforementioned use, such as peacocks and domestic fowl, and generally speaking the birds that are not flyers. For of the whole-winged animals in general, none has a tail, so that they are borne along just like a rudderless ship, and each of them collides with whatever it happens upon; in a similar way both the sheath-winged animals, like dung beetles and cockchafers, and the non-sheathed ones, like bees and wasps. And in the birds that are not flyers the tail is useless, for example, in purple gallinules and herons and all the swimming birds. But instead of the tail they fly by stretching out their feet, and they use their legs instead of their tail for the purpose of keeping course in flight. And the flight of whole-winged animals is slow and weak, due to the 710a15 fact that the nature of their wings is not proportionate to the weight of their body, but instead this is great whereas their wings are small and weak. So, just as if a cargo ship were to try to make its voyage using oars,



10, 710a20–11, 710b10

ποιεῖσθαι τὸν πλοῦν, οὕτω ταῦτα τῇ πτήσει χρῆται. καὶ ἡ  ἀσθένεια δὲ αὐτῶν τε τῶν πτερῶν καὶ ἡ τῆς ἐκφύσεως συμβάλλεταί τι πρὸς τὸ λεχθέν. τῶν δ’ ὀρνίθων τῷ μὲν ταῷ τὸ ὀρροπύγιον ὁτὲ μὲν διὰ τὸ μέγεθος ἄχρηστον, ὁτὲ δὲ διὰ τὸ ἀποβάλλειν οὐθὲν ὠφελεῖ. ὑπεναντίως δ’ ἔχουσιν οἱ ὄρνιθες τοῖς ὁλοπτέροις τὴν τῶν πτερῶν φύσιν, μάλιστα δ’ οἱ  τάχιστα αὐτῶν πετόμενοι. τοιοῦτοι δ’ οἱ γαμψώνυχες· τούτοις γὰρ ἡ ταχυτὴς τῆς πτήσεως χρήσιμος πρὸς τὸν βίον. ἀκόλουθα δ’ αὐτῶν ἔοικεν εἶναι καὶ τὰ λοιπὰ μόρια τοῦ σώματος πρὸς τὴν οἰκείαν κίνησιν, κεφαλὴ μὲν ἁπάντων μικρὰ καὶ αὐχὴν οὐ παχύς, στῆθος δ’ ἰσχυρὸν καὶ ὀξύ, ὀξὺ  μὲν πρὸς τὸ εὔτονον εἶναι, καθάπερ ἂν εἰ πλοίου πρώρα λεμβώδους, ἰσχυρὸν δὲ τῇ φύσει τῆς σαρκός, ἵν’ ἀπωθεῖν τε δύνηται τὸν προσπίπτοντα ἀέρα· καὶ τοῦτο δρᾷ ῥᾳδίως καὶ μὴ  μετὰ πόνου· τὰ δ’ ὄπισθεν κοῦφα καὶ συνήκοντα πάλιν εἰς στενόν, ἵν’ ἐπακολουθῇ τοῖς ἔμπροσθεν, μὴ σύροντα τὸν ἀέρα διὰ τὸ πλάτος. [11] Καὶ περὶ μὲν τούτων διωρίσθω τὸν τρόπον τοῦτον, τὸ δὲ  μέλλον ζῷον ὀρθὸν βαδιεῖσθαι διότι δίπουν τε ἀναγκαῖόν ἐστιν εἶναι, καὶ τὰ μὲν ἄνω μέρη τοῦ σώματος κουφότερα ἔχειν τὰ δ’ ὑφεστῶτα τούτοις βαρύτερα, δῆλον· μόνως γὰρ ἂν οὕτως ἔχον οἷόν τ’ εἴη φέρειν ἑαυτὸ ῥᾳδίως. διόπερ ἄνθρωπος μόνον ὀρθὸν τῶν ζῴων ὢν τὰ σκέλη κατὰ λόγον ἔχει πρὸς

20

25

30 710b1

5

10

20 χρῆται Z U Ca S Y P : χρῶνται N V   21 δὲ Z1 U Ca Y P : om. U S Z3 : δ᾽ ἡ N V | ἡ] om. Z1, add. Z2  23 ταῷ τὸ] om. Z1, add. Z2 | ὁτὲ μὲν Z U Ca Y P : om. U S N V | διὰ τὸ μέγεθος ἄχρηστον Z U : ἄχρηστον διὰ τὸ μέγεθος Ca S N V : διὰ μέγεθος ἄχρηστον Y P   25 ὁλοπτέροις C a Y N V P : κολεοπτέροις Z 1 U S : κουλεοπτέροις Z 2  27 χρήσιμος] ὡς χρήσιμος Z   29 οἰκεῖαν Z1 U Ca S P Bekker : ταχεῖαν Y N V Z3 : ὠκεῖαν corr. Jaeger | μὲν] μὲν οὖν P   30 ἰσχυρὸν καὶ ὀξὺ Ca Y P : οm. Z U S N P   31 εὔτονον] εὔτολμον Z | εἶναι] om. P | καθάπερ Z U Ca S Y P : καὶ καθάπερ N V P | πλοίου πρώρα] πρώρα πλοίου P  32 δὲ] τε Z | φύσει] περιφύσει Z1, corr. Z3 | τε Z1 Y Jaeger : οm. U Ca S P N V Z3 Bekker   710b1 τὸν] καὶ τὸν Z1, corr. Z 3 | τοῦτο Z 1 U C a S Y P : τοῦτον N V Z 3 | δρᾷ Z C a Y P Jaeger : om. U S N V Bekker  1–2 μὴ μετὰ πόνου Z1 Ca N V Y P : μετὰ τόνου U Z2 : μετὰ πόνου S   2 ὄπισθεν U Ca S N V: ὄπισθε Z : ὀπίσθια Y P   3–4 τὸν ἀέρα διὰ τὸ πλάτος] διὰ τὸ πλάτος τὸν ἀέρα V   6 ὀρθὸν] ὀρθῶς Z3  7 μέρη τοῦ σώματος Z U S Jaeger : τοῦ σώματος μέρη Ca Y N V P   8 ἂν οὕτως U Ca S N V : οὕτως ἂν Z Y : οὕτως P   10 μόνον ὀρθὸν τῶν ζῴων ὢν U Z2 S : μόνον ὀρθὸν ζῷον ὂν Z1 : μόνον ὀρθὸν τῶν ζῴων ὂν Y P : ὀρθὸν ὢν μόνος τῶν ζῴων Ca : μόνος ὀρθὸς τῶν ζῴων ὢν N V | κατὰ λόγον ἔχει] ἔχει κατὰ λόγον Ca



10, 710a20–11, 710b10



so in this way do these animals use flight. And also the weakness both of the wings themselves and of their outgrowth contributes something to what we mentioned. Among birds, the peacock’s tail is sometimes useless because of its size, and sometimes of no benefit because of the shedding. But birds are in the opposite condition to the whole-winged animals 710a24 with regard to the nature of their wings, and especially the swiftest flyers among them. Such are the crook-taloned birds. For swiftness of flight is useful to their way of life. And the remaining parts of their body seem to conform to their proper movement: in all of them, their head is small, their neck is not thick, and they have a strong and sharp breast – sharp so as to be forceful, as if it were the prow of a lembos-type ship, and strong by virtue of the nature of its flesh, so that it is able to push away the air hitting against it, and does this easily and not laboriously. And the rear parts are light and come together again to a taper, so that they follow after the front parts without dragging the air because of their breadth.

[11] Let it be determined in this way concerning these matters. But why 710b5 it is necessary for the animal that is to walk upright to be two-footed, and for it to have the upper parts of its body lighter, and the parts set below these heavier, is evident. For only if organized like this would it be able to carry itself easily. This is why a human being, being the only upright animal, has legs that – in proportion to the upper parts of its body – are the



11, 710b11–711a3

τὰ ἄνω τοῦ σώματος μέγιστα τῶν ὑποπόδων καὶ ἰσχυρότατα. δῆλον δὲ ποιεῖ τοῦτο καὶ τὸ συμβαῖνον τοῖς παιδίοις· οὐ γὰρ δύναται βαδίζειν ὀρθὰ διὰ τὸ πάντα νανώδη εἶναι καὶ μείζω καὶ ἰσχυρότερα ἔχειν ἢ κατὰ λόγον τὰ ἄνω μέρη τοῦ σώματος τῶν κάτωθεν. προϊούσης δὲ τῆς ἡλικίας αὔξησιν λαμβάνει τὰ  κάτω μᾶλλον, μέχρι οὗπερ ἂν λάβωσι τὸ προσῆκον μέγεθος, καὶ ποιοῦνται τότε τοῖς σώμασι τὴν βάδισιν τὴν ὀρθήν. οἱ δ’ ὄρνιθες κοῦφοι ὄντες δίποδές εἰσι διὰ τὸ ὄπισθεν αὐτοῖς τὸ βάρος εἶναι, καθάπερ ἐργάζονται τοὺς ἵππους τοὺς χαλκοῦς τοὺς τὰ πρόσθια ᾐρκότας τῶν σκελῶν. αἴτιον δὲ μάλιστα τοῦ  δίποδας ὄντας δύνασθαι ἑστάναι τὸ ἔχειν τὸ ἰσχίον ὅμοιον μηρῷ καὶ τηλικοῦτον ὥστε δοκεῖν δύο μηροὺς ἔχειν, τόν τ’ ἐν τῷ σκέλει πρὸ τῆς καμπῆς καὶ τὸν πρὸς τοῦτο τὸ μέρος ἀπὸ τῆς ἕδρας· ἔστι δ’ οὐ μηρὸς ἀλλ’ ἰσχίον. εἰ γὰρ μὴ τηλικοῦτον ἦν, οὐκ ἂν ἦν ὄρνις δίπους. ὥσπερ γὰρ τοῖς ἀνθρώ- ποις καὶ τοῖς τετράποσι ζῴοις, εὐθὺς ἂν ἦν ἀπὸ βραχέος ὄντος τοῦ ἰσχίου ὁ μηρὸς καὶ τὸ ἄλλο σκέλος· λίαν οὖν ἂν ἦν τὸ σῶμα πᾶν προπετὲς αὐτῶν. νῦν δὲ μακρὸν ὂν μέχρι ὑπὸ μέσην παρατείνει τὴν γαστέρα, ὥστ’ ἐντεῦθεν τὰ σκέλη ὑπερηρεισμένα φέρει τὸ σῶμα πᾶν. φανερὸν δ’ ἐκ τούτων  καὶ ὅτι ὀρθὸν οὐκ ἐνδέχεται τὸν ὄρνιθα εἶναι ὥσπερ τὸν ἄνθρωπον· ἡ γὰρ τῶν πτερῶν φύσις ὡς ἔχουσι τὸ σῶμα νῦν οὕτως αὐτοῖς χρήσιμός ἐστιν, ὀρθοῖς δ’ οὖσιν ἄχρηστος ἂν ἦν, ὥσπερ γράφουσι τοὺς ἔρωτας ἔχοντας πτέρυγας. ἅμα γὰρ τοῖς εἰρημένοις δῆλον ὅτι οὐδ’ ἄνθρωπον, οὐδ’ εἰ ἄλλο τι τοιοῦτόν

15

20

25

30 711a1

10–14 ἔχει … λόγον] om. N   11 τὰ ἄνω] τὸ ἄνω V | ἰσχυρότατα] ἰσχυρότερα V   13 δύναται Z U Ca S N V Jaeger : δύνανται Y P Bekker   14 ἢ κατὰ λόγον Z U Ca S N V : om. Y P   15 κάτωθεν] κάτω Z1, corr. Z2  16 μέχρι οὗπερ ἂν U Ca N V Jaeger : μέχρι οὗπερ S : μέχρι περ ἂν Z Bekker : μέχρι περ ἂν οὗ Y P   17 τότε τοῖς σώμασι Ca S V Y P : τοῖς σώμασι τότε Z U : τότε τοῖς σώματος N | τὴν ὀρθήν Z U Ca S : ὀρθὴν N V Y P   18 κοῦφοι] κυφοι Z1, corr. Z2   22 τόν τ᾽ U Z2 Ca S N V P : τό τ᾽ Z1 Y   23 καὶ τὸν U Z2 Ca V S Y P : καὶ τὸ Z1 Y | τὸ U Z2 Ca V S Y P : om. Z1 Y   27 ἂν ἦν] ἦν ἂν S   28 μακρὸν ὂν Bekker Jaeger : μακρὸς ὢν codd.   29 παρατείνει τὴν γάστερα] τὴν γάστερα παρατείνει S   30 ὑπερηρεισμένα U Ca S V P : ὑπηρεισμένα Z Y P   31 τὸν ὀρνιθα εἶναι Z Ca Y N V P : εἶναι τὸν ὀρνιθα U S   32 ἔχουσι] ἔχουσα S   32–711a 1 τὸ σῶμα … αὐτοῖς] οm. Z   1 οὕτως αὐτοῖς U Ca S V Y : οὕτως εὐθὺς αὐτοῖς N P : οὕτως αὐτοῖς εὐθὺς V   2 γράφουσι τοὺς ἔρωτας Z U Ca S Y P : τοὺς ἔρωτας γράφουσι V : τοὺς ἔρωτάς φησι N   3 ὅτι Z U Ca S Y P : ὡς N V | οὐδ᾽ ἄνθρωπον] οὐδὲ τὸν ἄνθρωπον Y | εἰ Y P Jaeger : ἂν U Z2 S N V : om. Z Ca Bekker | ἄλλο τι U Z2 Ca S N V P : τι ἄλλο Y : ἄλλο om. Z  



11, 710b11–711a3



biggest and strongest of the footed animals. And what happens in the case of little children also makes this evident. For they are unable to walk upright, because of the fact that they are all dwarfish, and have the upper parts of their body bigger and stronger in proportion to the lower parts. But as they get older the lower parts undergo more growth, up to the point at which they acquire the proper size, and then they produce with their bodies upright walking. But birds, being light, are two-footed because of the fact that their 710b17 weight is situated at the back, just as people make bronze horses that have raised up their front legs. But most of all the cause of their being able to stand, despite being two-footed, is their having an ischium like a thigh and so large that they seem to have two thighs, one in the leg before the joint, and one going toward this part from the rump. But it is not a thigh, but rather an ischium. If it were not so large, a bird would not be two-footed. For just as with humans and four-footed animals, the thigh and the rest of the leg would begin immediately from the ischium, being short. Then their whole body would be inclined too far forward. But as it is, being long, the ischium stretches up to beneath the middle of the belly, so that from here the legs, having been put under as a support, carry the whole body. And it is also clear from these things that it is not possible for the bird 710b30 to be upright in the way that the human is. For, as birds now have their body, the nature of the wings is useful to them, but it would be useless to them were they upright, just as people paint Erotes with wings. For, at the same time, it is evident from what has been said that neither a human



11, 711a4–12, 711a28

ἐστι τὴν μορφήν, δυνατὸν εἶναι πτερωτόν, οὐ μόνον ὅτι πλείοσι σημείοις κινήσεται ἢ τέτταρσιν ἔναιμον ὄν, ἀλλ’ ὅτι ἄ- χρηστος αὐτοῖς ἡ τῶν πτερύγων ἕξις κατὰ φύσιν κινουμένοις· ἡ δὲ φύσις οὐθὲν ποιεῖ παρὰ φύσιν. [12] Ὅτι μὲν οὖν εἰ μὴ κάμψις ἦν ἐν τοῖς σκέλεσιν ἢ ἐν ταῖς ὠμοπλάταις καὶ ἰσχίοις, οὐθὲν οἷόν τ’ ἦν ἂν τῶν ἐναίμων καὶ ὑποπόδων προβαίνειν, εἴρηται πρότερον, καὶ ὅτι κάμ- ψις οὐκ ἂν ἦν μηθενὸς ἠρεμοῦντος, ὅτι τε ἐναντίως οἵ τε ἄνθρωποι δίποδες ὄντες καὶ οἱ ὄρνιθες τὴν τῶν σκελῶν ποιοῦνται κάμψιν, ἔτι δὲ τὰ τετράποδα ὑπεναντίως καὶ αὑτοῖς καὶ τοῖς ἀνθρώποις. οἱ μὲν γὰρ ἄνθρωποι τοὺς μὲν βραχίονας κάμπτουσιν ἐπὶ τὰ κοῖλα, τὰ δὲ σκέλη ἐπὶ τὸ κυρτόν,  τὰ δὲ τετράποδα τὰ μὲν πρόσθια σκέλη ἐπὶ τὸ κυρτόν, τὰ δ’ ὀπίσθια ἐπὶ τὸ κοῖλον· ὁμοίως δὲ καὶ οἱ ὄρνιθες. αἴτιον δ’ ὅτι ἡ φύσις οὐθὲν δημιουργεῖ μάτην, ὥσπερ εἴρηται πρότερον, ἀλλὰ πάντα πρὸς τὸ βέλτιστον ἐκ τῶν ἐνδεχομένων. ὥστ’ ἐπεὶ πᾶσιν ὅσοις ὑπάρχει κατὰ φύσιν ἡ κατὰ τόπον  μεταβολὴ τοῖν σκελοῖν, ἑστῶτος μὲν ἑκάστου τὸ βάρος ἐν τούτῳ ἐστί, κινουμένοις δ’ εἰς τὸ πρόσθεν δεῖ τὸν πόδα τὸν ἡγούμενον τῇ θέσει κοῦφον εἶναι, συνεχοῦς δὲ τῆς πορείας γινομένης αὖθις ἐν τούτῳ τὸ βάρος ἀπολαμβάνειν, δῆλον ὡς ἀναγκαῖον ἐκ τοῦ κεκάμφθαι τὸ σκέλος αὖθις εὐθὺ γίνεσθαι  μένοντος τοῦ τε κατὰ τὸν προωσθέντα πόδα σημείου καὶ τῆς κνήμης. τοῦτο δὲ συμβαίνειν ἅμα καὶ προϊέναι τὸ ζῷον εἰς τοὔμπροσθεν μὲν ἔχοντος τὴν καμπὴν τοῦ ἡγουμένου σκέλους

5

10

15

20

25

4 τὴν μορφήν, δυνατὸν εἶναι] τὴν μορφὴν δυνατόν ἐστι Y   5 κινήσεται ἢ τέτταρσιν] ἢ τέταρσιν κινήσεται S | ἢ τέτταρσιν ἔναιμον ὂν] ἔναιμον ὂν ἢ τέτρασιν Y | τέτταρσιν P, τέταρσιν S : τέτρασιν Z U Ca N V Y | ὂν] om. Z1, add. Z2  6 ἕξις] γρ. φύσις Z2  7 οὐθὲν scripsi : οὐδὲν codd. Bekker Jaeger  8 ταῖς U Z2 Ca S N V : τοῖς Z1 Y P   9 ἂν Z Y P : om. U Ca S N V   10 κάμψις] αἱ κάμψεις Y   11 οὐκ ἂν] οὐθὲν Z1, corr. Z2 | ἦν] ἦσαν Y | μηθενὸς U Ca S N V P : μηδενὸς Z Y Bekker Jaeger | ὅτι τε U Ca S N V : ὅτι δὲ Z, ὅτι δ᾽ P : δῆλον ὅτι Y | οἵ τε Z Y P : οἱ U Ca S N V   11–12 ἄνθρωποι … καὶ οἱ ὄρνιθες Z U Ca S Y P : ὄρνιθες καὶ οἱ ἄνθρωποι N V   13 ἔτι δὲ Z Y P : ἔτι U S : ἔτι δὲ καὶ Ca : ὅτι δὲ N V   14 καὶ] om. Z1, add. Z2 | μὲν Z U Ca S Y P : om. N V   16–17 τετράποδα … τὰ δ᾽] οm. Z1, add. Z2  17 ὀπίσθια] ὀπίσθια σκέλη Z   18 δ᾽ ὅτι ἡ φύσις oὐθὲν] ἡ φύσις oὐδὲν γὰρ Z1, corr. Z2 | οὐθὲν scripsi : οὐδὲν codd. Bekker Jaeger   20 ἐπεὶ] ἐπὶ Z1, corr. Z2 | ὅσοις] οἷς P   25 ἐκ τοῦ] expunxit Z2 | εὐθὺ Z1 Y Jaeger : τε εὐθὺ U Z2 Ca S N V P Bekker  26 τε U Ca Y N V P : του Z1, corr. Z2 : om. S | προωσθέντα Z U2 Ca N V P : προσθέντα U1 S : προτεθέντα Y   27 συμβαίνειν ἅμα] om. Z1, συμβέβηκεν ἅμα Z2, συμβαίνειν ἅμα Z3 | προϊέναι] προσιέναι Z1  28 τοὔμπροσθεν U Ca N V Bekker : τὸ ἔμπροσθεν Z S Y P Jaeger



11, 711a4–12, 711a28



nor anything else of this kind of form can be winged, not only because it will then move by means of more than four points, despite being blooded, but also because the possession of wings would be useless to them in moving naturally. But nature does nothing contrary to nature. [12] It has already been established that, if there were no bending in the 711a8 legs or in the shoulders and hips, no blooded and footed animal could advance;2 that there would be no bending if nothing were at rest,3 and that human beings and birds, though both are two-footed, bend their legs in opposite directions;4 moreover, that four-footed animals bend their legs in opposite directions both with respect to themselves and to human beings.5 Indeed, human beings bend their arms toward the concave and their legs toward the convex, while four-footed animals bend their front legs toward the convex and their back legs toward the concave; birds also do the same. The cause of this is that nature crafts nothing in vain, as has been said before, but everything with a view to achieve what is best from among the available possibilities.6 Therefore, since in the case of all the animals that by nature engage in 711a20 change with respect to place by means of two legs, when each leg is standing the weight is on this one, but when the animals move forward the leading foot must be in a light position, and as progression continues the weight must fall on this foot again, it is evidently necessary for the leg, once it has been bent, to become straight again while the point of motion of the foot that has been thrust forward, and the shin, remain at rest. And it is possible for this to happen and, at the same time, for the animal to advance if the leading leg has a forward bend; but if the leading 2

5 6 3

4

IA 9, 708b24–709a7. IA 9, 708b21–24. IA 1, 704a18–20. IA 1, 704a22–b5. IA 2, 704b15–17; IA 8, 708a9–12.



12, 711a29–b22

δυνατόν, εἰς τοὔπισθεν δ’ ἀδύνατον. οὕτω μὲν γὰρ προενεχθέντος τοῦ σώματος ἡ ἔκτασις τοῦ σκέλους ἔσται, ἐκείνως δ’  ἀνενεχθέντος. ἔτι δ’ εἰς τὸ ὄπισθεν μὲν τῆς καμπῆς οὔσης διὰ δύο κινήσεων ἐγίγνετ’ ἂν ἡ τοῦ ποδὸς θέσις ὑπεναντίων τε αὑταῖς, καὶ τῆς μὲν εἰς τὸ ὄπισθεν τῆς δὲ εἰς τὸ ἔμπροσθεν· ἀναγκαῖον γὰρ ἐν τῇ συγκάμψει τοῦ σκέλους τοῦ μὲν μηροῦ τὸ ἔσχατον εἰς τοὔπισθεν προάγειν, τὴν δὲ κνήμην ἀπὸ τῆς καμπῆς εἰς τὸ ἔμπροσθεν τὸν πόδα κινεῖν εἰς τὸ ἔμπροσθεν δὲ τῆς καμπῆς οὔσης, οὔθ’ ὑπεναντίαις κινήσεσι μιᾷ τε τῇ εἰς  τὸ ἔμπροσθεν ἡ λεχθεῖσα πορεία συμβήσεται. ὁ μὲν οὖν ἄνθρωπος δίπους ὢν καὶ τὴν κατὰ τόπον μεταβολὴν κατὰ φύσιν τοῖς σκέλεσι ποιούμενος διὰ τὴν εἰρημένην αἰτίαν κάμπτει εἰς τὸ ἔμπροσθεν τὰ σκέλη, τοὺς δὲ βραχίονας ἐπὶ τὸ κοῖλον εὐλόγως· ἄχρηστοι γὰρ ἂν ἦσαν καμπτόμενοι εἰς τοὐ- ναντίον πρός τε τὴν τῶν χειρῶν χρῆσιν καὶ πρὸς τὴν τῆς τροφῆς λῆψιν. τὰ δὲ τετράποδα καὶ ζῳοτόκα τὰ μὲν ἔμπροσθεν σκέλη, ἐπειδὴ ἡγεῖταί τε τῆς πορείας αὐτῶν καὶ ἔστι ταῦτ’ ἐν τῷ μέρει τῷ ἔμπροσθεν τοῦ σώματος, ἀνάγκη κάμπτειν ἐπὶ τὴν περιφέρειαν διὰ τὴν αὐτὴν αἰτίαν ἥνπερ καὶ  οἱ ἄνθρωποι κατὰ γὰρ τοῦτο ὁμοίως ἔχουσι. διόπερ καὶ τὰ τετράποδα κάμπτουσιν εἰς τὸ πρόσθεν τὸν εἰρημένον τρόπον. καὶ γὰρ οὕτως μὲν αὐτῶν τῆς κάμψεως γινομένης ἐπὶ πολὺ δυνήσονται τοὺς πόδας μετεωρίζειν· ἐναντίως δὲ κάμπτοντες μικρὸν ἀπὸ τῆς γῆς ἂν αὐτοὺς ἐμετεώριζον διὰ τὸ τόν τε  μηρὸν ὅλον καὶ τὴν καμπήν, ἀφ’ ἧς ἡ κνήμη πέφυκεν, ὑπὸ τῇ γαστρὶ γίγνεσθαι προϊόντος αὐτοῦ. τῶν δ’ ὄπισθεν σκελῶν εἰ

30 711b1

5

10

15

20

29 μὲν] om. Y | προενεχθέντος] προσενεχθέντος Z1  31 ἔτι] ἐπεὶ S | δ᾽ Y P, δὲ Ca S N V : om. Z U   32 ὑπεναντίων τε αὑταῖς Jaeger : ὑπεναντίως τε αὐταῖς Z2 Ca S Y N V P Bekker : ὑπεναντίως δὲ αὐταῖς Z1 U   711b1 τῆς δὲ εἰς τὸ ἔμπροσθεν] om. Z1, add. Z2  2 συγκάμψει Z Ca S N V P : συγκάψει U Y   3 τοὔπισθεν Z U S Y P : τὸ ὄπισθεν Ca N V   4 τὸν πόδα κινεῖν] κινεῖν τὸν πόδα U   5 οὔθ᾽ U : οὔτε Z Ca S V Y P : οὔτε ἐν N | ὑπεναντίαις κινήσεσι U Ca S N V : ὑπεναντίως ταῖς κινήσεσι Z Y P | τε τῇ] τῇ Z2  7–8 κατὰ φύσιν] om. Z   9 εἰς τὸ] εἰς τὰ S   10 ἂν] om. S | ἦσαν] εἴησαν Z1, corr. Z2 | εἰς U Ca N V : om. Z S Y P | τοὐναντίον] τὸ ἐναντίον V   12 ζῳοτόκα Z Ca Y N V P Bekker : τὰ ζῳοτόκα U S Jaeger | μὲν Z Y P : om. U Ca S N V   15 ἐπὶ] διὰ Z1, corr. Z2 | διὰ] καὶ διὰ Z1, corr. Z2  17 κάμπτουσιν Z Ca S Y P : κάμπτει U N V | εἰς τὸ πρόσθεν τὸν εἰρημένον τρόπον] τὸν εἰρημένον πρόσθε (sic) τρόπον Z   18 αὐτῶν τῆς κάμψεως Z U C a S N V Jaeger : τῆς κάμψεως αὐτῶν Y P Bekker | ἐπὶ πολὺ] ἐπὶ τὸ πολὺ P   19 δυνήσονται] δυνήσεται Z1, corr. Z2  20 αὐτοὺς ἐμετεώριζον Z U Ca S Y P : ἐμετεώριζον αὐτοὺς N : ἑαυτοὺς ἐμετεώριζον V   21–22 ὑπὸ τῇ γαστρὶ γίγνεσθαι] γίγνεσθαι ὑπὸ τῇ γαστρί Y



12, 711a29–b22



leg has a backward bend, it is impossible. The reason is that in the former case the extension of the leg will occur as the body moves forward but in the latter case as the body moves backward. Moreover, if the bend were backward, the positioning of the foot would take place through two movements that are contrary to each other, namely one backward and the other forward. The reason is that it is necessary, in the bending-together of the leg, for the last part of the thigh to advance backward and for the shin to move the foot forward from the point of bending. But if the bend is forward, the said progression will occur by means of a single forward movement rather than by means of two contrary movements. The human being, because it is two-footed and naturally makes change 711b6 with respect to place by means of legs, bends its legs forward due to the stated cause, and its arms backward for good reason: if the arms were bent in the opposite direction, they would be useless for the employment of the hands and also for taking food. It is necessary for the live-bearing, four-footed animals to bend their 711b12 front legs toward the circumference, since these legs initiate their progression and are situated in the front part of their bodies, and on account of the same cause as human beings do. In this respect they, indeed, resemble each other. That is why the four-footed animals, too, bend their legs forward in the stated manner. And if their leg-bending happens in this way, they will be able to raise their feet from the ground, whereas if they bent their legs in the opposite direction, they would raise them only a little because the entire thigh and the joint from which the shin by nature grows would be underneath the belly as the animal advanced. On the

12, 711b23–13, 712a13



μὲν ἦν εἰς τὸ ἔμπροσθεν ἡ κάμψις, τῶν ποδῶν ὁ μετεωρισμὸς ὁμοίως ἂν αὐτοῖς εἶχε τοῖς προσθίοις (ἐπὶ βραχὺ γὰρ ἂν ἐγίγνετο καὶ τούτοις κατὰ τὴν ἄρσιν τῶν σκελῶν, τοῦ τε  μηροῦ καὶ τῆς καμπῆς ἀμφοτέρων ὑπὸ τὸν τῆς γαστρὸς τόπον ὑποπιπτόντων), εἰ δ’ εἰς τὸ ὄπισθεν, καθάπερ καὶ νῦν κάμπτουσιν, οὐθὲν ἐμπόδιον αὐτοῖς γίγνεται πρὸς τὴν πορείαν ἐν τῇ τοιαύτῃ κινήσει τῶν ποδῶν. ἔτι τοῖς γε θηλαζομένοις αὐτῶν καὶ πρὸς τὴν τοιαύτην λειτουργίαν ἀναγκαῖον ἢ βέλτιόν  γ’ οὕτω κεκάμφθαι τὰ σκέλη· οὐ γὰρ ῥᾴδιον τὴν κάμψιν ποιουμένων ἐντὸς ὑφ’ αὑτὰ ἔχειν τὰ τέκνα καὶ σκεπάζειν. [13] Ὄντων δὲ τεττάρων τρόπων τῆς κάμψεως κατὰ τοὺς  συνδέσμους (ἀνάγκη γὰρ κάμπτειν ἢ ἐπὶ τὸ κοῖλον καὶ τὰ πρόσθια καὶ τὰ ὀπίσθια, καθάπερ ἐφ’ οἷς Α, ἢ ἐπὶ τοὐναντίον ἐπὶ τὸ κυρτόν, καθάπερ ἐφ’ οἷς Β, ἢ ἀντεστραμμένως μὴ ἐπὶ τὰ αὐτά, ἀλλὰ τὰ μὲν πρόσθια ἐπὶ τὸ κυρ- τόν, τὰ δ’ ὀπίσθια ἐπὶ τὸ κοῖλον, καθάπερ ἐφ’ οἷς τὸ Γ, ἢ τοὐναντίον τούτοις τὰ μὲν κυρτὰ πρὸς ἄλληλα, τὰ δὲ κοῖλα ἐκτός, καθάπερ ἔχει ἐφ’ οἷς τὸ Δ), A

B



25

30

712a1

5



ὡς μὲν ἔχει ἐφ’ οἷς τὸ Α ἢ τὸ Β, οὐθὲν κάμπτεται οὔτε τῶν διπόδων οὔτε τῶν τετραπόδων, ὡς δὲ τὸ Γ, τὰ τετράποδα, ὡς δὲ τὸ Δ, τῶν μὲν τετραπόδων οὐθὲν πλὴν ἐλέφας, ὁ δ’ ἄνθρωπος τοὺς βραχίονας καὶ τὰ σκέλη· τοὺς μὲν γὰρ ἐπὶ τὸ κοῖλον κάμπτει, τὰ δὲ σκέλη ἐπὶ τὸ κυρτόν. ἀεὶ δ’ ἐναλλὰξ ἐναντίως ἔχει

10

22–23 εἰ … ποδῶν] om. Z1, add. Z2 | μὲν] μὴ P   23 ἔμπροσθεν] πρόσθεν U | ὁ Z1 Y P : oὐ S : ἂν U Z2 Ca N : om. V   24 ὁμοίως ἂν Z Y P : ὁμοίως U Ca S N : ἂν ὁμοίως V   25 ἐγίγνετο] ἐγίγνοντο U | καὶ Z2 Ca S Y N V P : τούτων καὶ U : om. Z | τοῦ τε U Ca Y N V P : καὶ τοῦ Z : τοῦ γε S   26 καὶ] om. Z1, add. Z2  28 αὐτοῖς γίγνεται] γίγνεται αὐτοῖς U   29 τῶν] om. Z1, add. Z2 | ἔτι] ἔτι δὲ Y   31 γ᾽ Ca Y N V P : τ᾽ Z1 : om. Z2 U S   32 ὑφ᾽ αὑτὰ Z U S N P : ὑφ᾽ αὑτῶν Ca : ὑφ᾽ αὑτοῖς Y V   712a2 συνδέσμους] συνδυασμους Z1, corr. Z3 | γὰρ] om. V   3 τὰ ὀπίσθια Z Ca S Y N P : ὀπίσθια U V | καθάπερ Z Ca Y N V P : οἷον S : om. U | Α] τὸ Α V   3–4 A … ἐφ᾽ οἷς] om. Y   4 τὸ κυρτὸν : τῷ κυρτῷ V | καθάπερ] οἷον P   5 μὴ Ca N V P Jaeger : καὶ μὴ Z U Y Bekker : om. S   6 τὸ Γ] Γ Z   8 ἐκτὸς Z Y : om. U Ca S N V P | καθάπερ … Δ] om. Y | figuram restitui : non habent codd. | μὲν] μὲν οὖν Y   9 ἢ τὸ Β U Z2 Ca S N V P : om. Z1 Y | οὔτε τῶν διπόδων οὔτε] οὐδὲ τῶν διπόδων οὐδὲ P   9–10 oὔτε τῶν τετραπόδων] om. Z1, add. Z2  10 τὰ] om. S   12–13 τοὺς μὲν … σκέλη] om. P   13 ἔχει U Ca S N V : ἔχουσι Z Y P



12, 711b23–13, 712a13



other hand, if the bending of the back legs were forward, the raising of these feet would be similar to the raising of their front feet; in fact, bending during leg-raising would also be short in the case of these animals, since both the thigh and the joint would fall under the belly region; but if the bending of the rear legs is backward, as it actually is, no obstacle to progression happens to them in such a movement of the feet. Moreover, it is necessary, or at any rate better, for the legs to be bent in this way in the case of nursing animals also in view of this function; for it would not be easy for them to have their children underneath and to protect them if their legs were bent inward. [13] There are four ways of bending at the joints. It is necessary for both 712a1 front and back limbs either to bend toward the concave, as in figure A, or the other way round, namely toward the convex, as in figure B, or inversely not both in the same direction but the front limbs toward the convex and the back limbs toward the concave, as in C, or in a manner contrary to this, each pair of limbs toward the convex in relation to the other with their concave facing outward as is the case with D. A

B

C

D

No two-footed or four-footed animal is bent as is the case with A or B, but four-footed animals are bent as in C, whereas no four-footed animal, apart from the elephant, is bent as in D; the human being, on the other hand, bends its arms and legs in this way: indeed, its arms toward the concave and its legs toward the convex.



13, 712a14–14, 712b7

τὰ κῶλα τὰς κάμψεις τοῖς ἀνθρώποις, οἷον τὸ ὀλέκρανον ἐπὶ τὸ κοῖλον, ὁ δὲ καρπὸς ἐπὶ τὸ κυρτόν, καὶ  πάλιν ὁ ὦμος ἐπὶ τὸ κυρτόν· ὡσαύτως δὲ καὶ ἐπὶ τῶν σκελῶν ὁ μηρὸς ἐπὶ τὸ κοῖλον, τὸ δὲ γόνυ ἐπὶ τὸ κυρτόν, ὁ δὲ ποὺς τοὐναντίον ἐπὶ τὸ κοῖλον. καὶ τὰ κάτω δὴ πρὸς τὰ ἄνω φανερὸν ὅτι ἐναντίως· ἡ γὰρ ἀρχὴ ὑπεναντίως, ὁ μὲν ὦμος ἐπὶ τὸ κυρτόν, ὁ δὲ μηρὸς ἐπὶ τὸ κοῖλον· διὸ καὶ  ὁ μὲν ποὺς ἐπὶ τὸ κοῖλον, ὁ δὲ καρπὸς τῆς χειρὸς ἐπὶ τὸ κυρτόν. [14] Αἱ μὲν οὖν κάμψεις τῶν σκελῶν τοῦτόν τε τὸν τρόπον ἔχουσι καὶ διὰ τὰς αἰτίας τὰς εἰρημένας, κινεῖται δὲ τὰ ὀπίσθια πρὸς τὰ ἔμπροσθεν κατὰ διάμετρον· μετὰ γὰρ τὸ  δεξιὸν τῶν ἔμπροσθεν τὸ ἀριστερὸν τῶν ὄπισθεν κινοῦσιν, εἶτα τὸ ἀριστερὸν τῶν ἔμπροσθεν, μετὰ δὲ τοῦτο τὸ δεξιὸν τῶν ὄπισθεν. αἴτιον δ’ ὅτι εἰ μὲν τὰ ἔμπροσθεν ἅμα καὶ πρῶτον, διέσπαστο ἂν ἢ καὶ προπετὴς ἂν ἐγίνετο ἡ βάδισις καὶ οἷον ἐφελκομένοις τοῖς ὄπισθεν. ἔτι δ’ οὐ πορεία ἀλλὰ ἅλ- σις τὸ τοιοῦτον· χαλεπὸν δὲ συνεχῆ ποιεῖσθαι τὴν μεταβολὴν ἁλλόμενα. σημεῖον δέ· ταχὺ γὰρ ἀπαγορεύουσι καὶ νῦν τῶν ἵππων ὅσοι τὸν τρόπον τοῦτον ποιοῦνται τὴν κίνησιν, οἷον οἱ πομπεύοντες. χωρὶς μὲν οὖν τοῖς ἔμπροσθεν καὶ ὄπισθεν διὰ ταῦτα οὐ ποιοῦνται τὴν κίνησιν· εἰ δὲ τοῖς δεξιοῖς  ἀμφοτέροις πρώτοις, ἔξω ἂν ἐγίγνοντο τῶν ἐρεισμάτων καὶ ἔπιπτον ἄν. εἰ δὴ ἀνάγκη μὲν ἢ τούτων τῶν τρόπων ὁποτερονοῦν ποιεῖσθαι τὴν κίνησιν ἢ κατὰ διάμετρον, μὴ ἐνδέχεται δ’ ἐκείνων μηδέτερον, ἀνάγκη κινεῖσθαι κατὰ διάμετρον· οὕ- τω γὰρ κινούμενα ὥσπερ εἴρηται οὐδέτερα τούτων οἷόν τε πάσχειν. καὶ διὰ τοῦτο οἱ ἵπποι καὶ ὅσα τοιαῦτα, ἵσταται

15

20

25

30

712b1

5

14 ὀλέκρανον Z U S Y V : ὠλέκρανον Ca N P   15 ἐπὶ τὸ κοῖλον, ὁ δὲ καρπὸς ἐπὶ τὸ κυρτόν Y V : ἐπὶ τὸ κυρτόν, ὁ δὲ καρπὸς ἐπὶ τὸ κοῖλον Z1 : ἐπὶ τὸ κοῖλον U Z2 S N P : κυρτόν, τὸ δ᾽ ἐντὸς κοῖλον Ca  16 ὁ Z Ca S Y V P : om. U N | ἐπὶ τὸ] om. N | καὶ ἐπὶ Z Y P : καὶ U Ca S N V  17–21 τὸ δὲ … κοῖλον] om. N   18 τὰ κάτω] κάτω Ca | δὴ] δὲ V   19 ἐναντίως] ἐναντίον Z | μὲν] δὲ Y   20–22 ὁ δὲ μηρὸς … κυρτόν] ὁ δὲ ποῦς τοὐναντίον ἐπὶ τὸ κοῖλον, καὶ τὸ μὲν ὠλέκρανον ἐπὶ τὸ κοῖλον, τὸ δὲ γόνυ ἐπὶ τὸ κυρτόν, καὶ ὁ μὲν καρπὸς τῆς χειρὸς ἐπὶ τὸ κυρτόν, ὁ δὲ ποῦς τοὐναντίον ἐπὶ τὸ κοῖλον V   20–21 διὸ καὶ ὁ μὲν ποῦς ἐπὶ τὸ κοῖλον Z Y : om. U Ca S P  27 τοῦτο] om. P  28 πρῶτον] ὄπισθεν Y   29 διέσπαστο Z U S V Jaeger : διεσπᾶτο Y P Bekker, διεσπατο Ca, διέσπατο N | ἢ U Ca S N V P : om. Z Y | ἂν Z1 U Ca S N V P : del. Z2, om. Y  30 ἅλσις] καὶ ἅλσις S   32 ἁλλόμενα] ἁλλόμενον Y   712b1 οὐ Z2 U2 Ca S N V P : om. Z1 U1 Y   2 πρώτοις U Z2 Ca S N V P : πρῶτον Z1 Y | ἐγίγνοντο τῶν] ἐγίγνετο τὸ τῶν Y   3 ἔπιπτον ἂν Z Y P : ἀνέπιπτον U S N V : ἀνεπίληπτον Ca | τούτων τῶν τρόπων Z U Ca N V P : τοῦτον τὸν τρόπον U S   5 ἐκείνων Z2 Y N P : ἐκείνως Z1 U Ca S V | κατὰ] τὰ κατὰ Z



13, 712a14–14, 712b7



The limbs responsible for bending alternately have contrary disposi- 712a13 tions without exception in human beings: the elbow bends toward the concave whereas the wrist toward the convex, and the shoulder toward the convex, too; it is similar also with the legs: the hip bends toward the concave, the knee toward the convex and the foot, by contrast, toward the concave. And it is clear that the lower limbs bend opposite to the upper limbs; the reason is that the origin of movement bends in opposite directions: the shoulder bends toward the convex whereas the hip toward the concave. This is also why the foot bends toward the concave and the wrist of the hand toward the convex. [14] Therefore, the bendings of the legs hold in 712a23 this way and on account of the causes stated, but the back legs move diagonally relative to the front legs. After the right front leg, four-footed animals move the left back leg, then the left front leg, and after it the right back leg. The cause of this is that if they moved the front legs simultaneously 712a28 and first, they would be strained, or their walking would even be a falling forward with their back legs, as it were, dragged behind. Further, this sort of movement would not be progression but jumping; but it is hard to make change continuous through leaping. Here is a piece of evidence: even now all horses that move in this manner (for example, procession horses) quickly get tired and refuse. For these reasons, then, four-footed animals do not move separately with their front and back legs. And if they moved with both right legs first, these animals would be 712b1 outside their supporting limbs, and they would fall. If, then, it is necessary that they move in one of these two ways, or else diagonally, and neither of these first two ways is possible, it is necessary for them to move diagonally. For by moving in the manner that has been said they are able to undergo neither of these outcomes .



14, 712b8–15, 713a3

προβεβηκότα κατὰ διάμετρον, καὶ οὐ τοῖς δεξιοῖς ἢ τοῖς ἀριστεροῖς ἀμφοτέροις ἅμα. τὸν αὐτὸν δὲ τρόπον καὶ ὅσα πλείους πόδας ἔχει τεττάρων ποιεῖται τὴν κίνησιν· ἀεὶ γὰρ ἐν τοῖς τέτταρσι τοῖς ἐφεξῆς τὰ ὀπίσθια πρὸς τὰ ἔμπροσθεν κινεῖται κατὰ διάμετρον. δῆλον δ’ ἐπὶ τοῖς βραδέως κινουμένοις. καὶ οἱ καρκίνοι γὰρ τὸν αὐτὸν τρόπον κινοῦνται· τῶν πολυπόδων γάρ εἰσιν· ἀεὶ γὰρ καὶ οὗτοι κατὰ διάμετρον κινοῦνται, ἐφ’ ὅπερ ἂν ποιῶνται τὴν πορείαν. ἰδίως γὰρ τοῦτο  τὸ ζῷον ποιεῖται τὴν κίνησιν μόνον γὰρ οὐ κινεῖται ἐπὶ τὸ πρόσθεν τῶν ζῴων, ἀλλ’ ἐπὶ τὸ πλάγιον. ἀλλ’ ἐπεὶ τοῖς ὄμμασι διώρισται τὸ πρόσθιον, ἡ φύσις πεποίηκεν ἀκολουθεῖν δυναμένους τοὺς ὀφθαλμοὺς τοῖς κώλοις· κινοῦνται γὰρ εἰς τὸ πλάγιον αὐτοῖς, ὥστε τρόπον τινὰ καὶ τοὺς καρκίνους κινεῖσθαι  διὰ τοῦτ’ ἐπὶ τὸ ἔμπροσθεν. [15] Οἱ δ’ ὄρνιθες τὰ σκέλη καθάπερ τὰ τετράποδα κάμπτουσι· τρόπον γάρ τινα παραπλησίως ἡ φύσις αὐτῶν ἔχει· τοῖς γὰρ ὄρνισιν αἱ πτέρυγες ἀντὶ τῶν προσθίων σκελῶν εἰσι. διὸ καὶ κεκαμμέναι τὸν αὐτόν εἰσι τρόπον ὥσπερ ἐκεί- νοις τὰ πρόσθια σκέλη, ἐπεὶ τῆς ἐν τῇ πορείᾳ κινήσεως τούτοις ἀπὸ τῶν πτερύγων ἡ κατὰ φύσιν ἀρχὴ τῆς μεταβολῆς ἐστι· πτῆσις γάρ ἐστιν ἡ τούτων οἰκεία κίνησις. διόπερ ἀφαιρεθεισῶν τούτων οὔθ’ ἑστάναι οὔτε προϊέναι δύναιτ’ ἂν οὐθεὶς ὄρνις. ἔτι δίποδος ὄντος καὶ οὐκ ὀρθοῦ, καὶ τὰ ἔμπροσθεν  μέρη τοῦ σώματος κουφότερα ἔχοντος, ἢ ἀναγκαῖον ἢ βέλτιον πρὸς τὸ ἑστάναι δύνασθαι τὸν μηρὸν οὕτως ὑποκείμενον ἔχειν ὡς νῦν ἔχει, λέγω δ’ ὅτι εἰς τὸ ὄπισθεν πεφυκότα. ἀλλὰ μὴν εἰ δεῖ τοῦτον ἔχειν τὸν τρόπον, ἀνάγκη τὴν κάμψιν ἐπὶ τὸ κοῖλον γίνεσθαι τοῦ σκέλους, καθάπερ τοῖς τετρά- ποσιν ἐπὶ τῶν ὀπισθίων, διὰ τὴν αὐτὴν αἰτίαν ἥνπερ εἴπομεν ἐπὶ τῶν τετραπόδων καὶ ζῳοτόκων. ὅλως δὲ οἵ τε ὄρνιθες

10

15

20

25

30

713a1

9 ἀμφοτέροις] ἀμφότερα Y  10 πόδας ἔχει U Ca S N V Jaeger : ἔχει Z : ἔχει πόδας Z2 Y P Bekker | ἀεὶ U Z2 Ca S N V P : δεῖ Z1 Y | γὰρ] τὴν V   11 τέτταρσι Bekker Jaeger : τέσσαρσι Z U S V P : τέτρασι Ca Y N | πρὸς] καὶ P   12 κινεῖται U Z2 Ca S N V P : κινεῖσθαι Z1 Y   13 γὰρ Z U Ca S Y N V Bekker : δὲ P Jaeger   14 καὶ U Z2 Ca S N V P : om. Z1 Y   15 ποιῶνται] ποιοῦνται Y   16 κινεῖται] κινοῦνται Y   18 ἡ] καὶ ἡ P   19–20 εἰς τὸ πλάγιον αὐτοῖς Z U Ca S N V : αὐτοὶ εἰς τὸ πλάγιον Y P   22 τὰ τετράποδα] τετράποδα Z1, add. τὰ Z3  24 γὰρ ὄρνισιν U Ca S N V P : τετράποσιν Z Y | αἱ Ca S Y N V P : αἱ γὰρ Z : om. U | αὐτῶν post πτέρυγες add. Z2 | τῶν προσθίων σκελῶν] τῶν σκελῶν τῶν προσθίων Y   25 καὶ Z1 Y P : om. U Z2 Ca S N V | γὰρ post ὥσπερ add. Z2  26 ἐπεὶ U Ca S N V : ἐπὶ Z2 P : ἔτι Z1 Y | τῇ] om. Y   27 πτερύγων Ca Y N V P : πτερυγίων Z U S   30 ἔτι] ἐπεὶ V   33 ὡς νῦν ἔχει Z Y P : om. U Ca S N V   34 δεῖ Z Y Jaeger : ἔδει U Ca S N V P Bekker   713a2 εἴπομεν] εἶπον V   3 δὲ] τε S



14, 712b8–15, 713a3



And, on account of this, horses and similar animals stand having stepped forward diagonally and not with right or left legs both simultaneously. Moreover, the animals with more than four feet also make their change 712b9 in this same way. For, in the next set of four feet, the back legs always move diagonally relative to the front legs. This is clear in animals that move slowly: crabs, too, move in the same manner, for they are among the many-footed animals. They also always move according to the diagonal in whichever direction they progress. Indeed, this animal makes its motion in a unique way; it is the only 712b15 animal that does not move forward but obliquely. But since forward is distinguished in relation to the eyes, nature has made the crab’s eyes able to follow its limbs. For they move for its own benefit obliquely, and so, because of this, crabs also move forward in a way. [15] Birds bend their legs just as four-footed animals do . For in a way the nature of their legs is nearly the same; in birds the wings are there instead of front legs, which is also why they are bent in the same way, since, regarding the motion involved in progression, the natural principle of the change is from the wings. The reason is that their proper motion is flight, which is also why if these wings were taken away, no bird would be able to stand or move forward. Further, since the bird is two-footed and not upright, and the front 712b30 parts of its body are lighter, it is either necessary or better for its being able to stand that the thigh lie under as actually is the case. (I mean growing toward the back.) But if it must have this direction, it is necessary that the bending of the leg be toward the concave as in the back legs of four-footed animals, and on account of the same cause we stated in the case of four-footed egg-laying animals.



15, 713a4–25

καὶ τὰ ὁλόπτερα τῶν πετομένων καὶ τὰ ἐν τῷ ὑγρῷ νευστικά, ὅσα αὐτῶν δι’ ὀργάνων τὴν ἐπὶ τοῦ ὑγροῦ ποιεῖται πο- ρείαν, οὐ χαλεπὸν ἰδεῖν ὅτι βέλτιον ἐκ πλαγίου τὴν τῶν εἰρημένων μερῶν πρόσφυσιν ἔχειν, καθάπερ καὶ φαίνεται νῦν ὑπάρχειν αὐτοῖς ἐπί τε τῶν ὀρνίθων καὶ τῶν ὁλοπτέρων. ταὐτὸ δὲ τοῦτο καὶ ἐπὶ τῶν ἰχθύων· τοῖς μὲν γὰρ ὄρνισιν αἱ πτέρυγες, τοῖς δ’ ἐνύδροις τὰ πτερύγια, τὰ δὲ πτίλα τοῖς  ὁλοπτέροις ἐκ τοῦ πλαγίου προσπέφυκεν. οὕτω γὰρ ἂν τάχιστα καὶ ἰσχυρότατα διαστέλλοντα τὰ μὲν τὸν ἀέρα τὰ δὲ τὸ ὑγρὸν ποιοῖτο τὴν κίνησιν· εἰς γὰρ τὸ ἔμπροσθεν καὶ τὰ ὄπισθεν μόρια τοῦ σώματος ἐπακολουθοίη ὑπείκοντι φερόμενα τὰ μὲν ἐν τῷ ὑγρῷ τὰ δ’ ἐν τῷ ἀέρι. τὰ δὲ  τρωγλοδυτικὰ τῶν τετραπόδων καὶ ᾠοτόκων, οἷον οἵ τε κροκόδειλοι καὶ σαῦροι καὶ ἀσκαλαβῶται καὶ ἑμύδες τε καὶ χελῶναι, πάντα ἐκ τοῦ πλαγίου προσπεφυκότα τὰ σκέλη ἔχει καὶ ἐπὶ τῇ γῇ κατατεταμένα, καὶ κάμπτει εἰς τὸ πλάγιον, διὰ τὸ οὕτω χρήσιμα εἶναι πρὸς τὴν τῆς ὑποδύ- σεως ῥᾳστώνην καὶ πρὸς τὴν ἐπὶ τοῖς ᾠοῖς ἐφεδρείαν καὶ φυλακήν. ἔξω δ’ ὄντων αὐτῶν, ἀναγκαῖον τοὺς μηροὺς προσστέλλοντα καὶ ὑποτιθέμενα ὑφ’ αὑτὰ τὸν μετεωρισμὸν τοῦ ὅλου σώματος ποιεῖσθαι. τούτου δὲ γινομένου κάμπτειν αὐτὰ οὐχ οἷόν τε ἄλλως ἢ ἔξω.

5

10

15

20

25

4 ὁλόπτερα U Ca S N V P : κολεόπτερα Z1 Y, κουλεόπτερα Z2 | νευστικά U Z2 Ca Y N V P : ἰθυντικά Z1 : καὶ S   7 φαίνεται νῦν] νῦν δύναται Z1, νῦν φαίνεται Z2  8 ὁλοπτέρων] κολεοπτέρων Z 1, κουλεοπτέρων Z 2  9 τῶν ἰχθύων] τοῖς ἰχθύσι V   10–11 τὰ δὲ … προσπέφυκεν del. Jaeger | πτίλα τοῖς ὁλοπτέροις] κολεα τοῖς κολεοπτέροις Z1, πτίλα τοῖς κουλεοπτέροις Z 2   12 διαστέλλοντα] Z e corr. , διαστέλλοντα Jaeger  13 ποιοῖτο] δι᾽ ὧν ποιοῖντο Z | εἰς Z U Ca S N P V2 : εἰ Y V1 | γὰρ Ca Y N V P : om. Z U S | τὸ] τὰ Y | τὰ Y Jaeger : τὸ U Ca S N P Bekker : om. Z N   14 τὰ ante μόρια add. Z2 | ἐπακολουθοίη Bekker Jaeger : ἐπακολουθείη U Ca S N V P : ἐπακολουθοῖεν Z : ἐπακολουθεῖεν Y | ἂν add. Bekker : om. codd.   15 μὲν] δ᾽ Y   16 τρωγλοδυτικὰ Z Y P Bekker : τρωγλόδυτα U Ca S N V Jaeger | καὶ] ἢ Z  17 σαῦροι Z U S : σαῦραι Ca N V : αἱ σαῦραι Y P | καὶ ἑμύδες Z1 P : καὶ αἱ μύδες Z2 Ca Y : αἱ μύδες U S N V   18 χελῶναι] αἱ χελῶναι Z | τοῦ] τε τοῦ fecit Z2  19 καὶ (pr.) U Z2 Ca S N P : om. Z1 Y V | κατατεταμένα Ca S N V P : κατατεταγμένα Z U S   20 τὸ οὕτω χρήσιμα εἶναι πρὸς Z Ca P : τὸ χρήσιμα εἶναι πρὸς Y : om. U S N V | ὑποδύσεως] ὑποδέσεως S  21 ἐπὶ] ἐν Z | ἐφεδρείαν Z Ca N V1 P : ἐφεδρίαν Y V2 : ὑφεδρίαν U S   22 ἀναγκαῖον U Z2 C a S N V P : τοῦ σώματος ἀναγκαῖον Z 1 Y   23 προσστέλλοντα Jaeger cum Mich. : προστέλλοντα Z U Ca S V P Bekker : προστίλλοντα Y : προσέλοντα N | ὑφ᾽ αὑτὰ U Ca N V P, ὑφ᾽ ἑαυτὰ S : ὑπ᾽ αὐτὰ Z Y   24 ὅλου Z Y P : οm. U Ca S N V | γινομένου] κινουμένου Y | ἄλλως post αὐτὰ add. Z2



15, 713a4–25



Generally, in birds, whole-winged animals, and animals capable of 713a3 swimming in water (the ones that progress through water by means of instrumental parts), it is not difficult to see that it is better for the attachment of the aforementioned parts to be oblique, just as it actually appears to be the case in both birds and whole-winged animals. The same also applies to fish. The wings in birds, the fins in aquatic animals, and the wings in whole-winged animals are attached obliquely. In this way they would move most quickly and forcefully by severing the air and the water; especially the back parts of the body would follow forward as they are carried along in a yielding medium, the ones in water and the others in air. The hole-dwellers among the egg-laying, four-footed animals (for 713a16 example, crocodiles, lizards, spotted lizards, freshwater tortoises, and sea turtles) all have their legs attached obliquely and stretched out on the ground, each leg-bending obliquely. For this reason their legs are therefore useful for ease of retiring underground and for sitting upon and guarding eggs. Also, because project outward they must necessarily draw their thighs close and put them underneath themselves to make their entire body rise. Because of this they cannot bend their legs otherwise than outward.



16, 713a26–17, 713b18

[16] Τὰ δ’ ἄναιμα τῶν ὑποπόδων ὅτι μὲν πολύποδά ἐστι καὶ οὐθὲν αὐτῶν τετράπουν, πρότερον ἡμῖν εἴρηται· διότι δ’ αὐτῶν ἀναγκαῖον ἦν τὰ σκέλη πλὴν τῶν ἐσχάτων ἔκ τε τοῦ πλαγίου προσπεφυκέναι καὶ εἰς τὸ ἄνω τὰς καμπὰς ἔχειν, καὶ αὐτὰ ὑπόβλαισα εἶναι εἰς τὸ ὄπισθεν, φανερόν·  ἁπάντων γὰρ τῶν τοιούτων ἀναγκαῖόν ἐστι τὰ μέσα τῶν σκελῶν καὶ ἡγούμενα εἶναι καὶ ἑπόμενα. εἰ οὖν ὑπ’ αὐτοῖς ἦν, ἔδει αὐτὰ καὶ εἰς τὸ ἔμπροσθεν καὶ εἰς τὸ ὄπισθεν τὴν καμ- πὴν ἔχειν, διὰ μὲν τὸ ἡγεῖσθαι εἰς τὸ ἔμπροσθεν, διὰ δὲ τὸ ἀκολουθεῖν εἰς τὸ ὄπισθεν. ἐπεὶ δ’ ἀμφότερα συμβαίνειν ἀναγκαῖον αὐτοῖς, διὰ τοῦτο βεβλαίσωταί τε καὶ εἰς τὸ πλάγιον ἔχει τὰς καμπάς, πλὴν τῶν ἐσχάτων· ταῦτα δ’  ὥσπερ πέφυκε μᾶλλον, τὰ μὲν ὡς ἑπόμενα τὰ δ’ ὡς ἡγούμενα. ἔτι δὲ κέκαμπται τὸν τρόπον τοῦτον καὶ διὰ τὸ πλῆθος τῶν σκελῶν· ἧττον γὰρ ἂν οὕτως ἐν τῇ πορείᾳ ἐμπόδιά τε αὐτὰ αὑτοῖς εἴη καὶ προσκόπτοι. ἡ δὲ βλαισότης αὐτοῖς ἐστι διὰ τὸ τρωγλοδυτικὰ εἶναι πάντα ἢ τὰ πλεῖστα· οὐ  γὰρ οἷόν τε ὑψηλὰ εἶναι τὰ ζῶντα τὸν τρόπον τοῦτον. [17] Οἱ δὲ καρκίνοι τῶν πολυπόδων περιττότατα πεφύκασιν· οὔτε γὰρ εἰς τὸ πρόσθεν ποιοῦνται τὴν πορείαν πλὴν ὥσπερ εἴρηται πρότερον, πολλούς τε τοὺς ἡγουμένους ἔχουσι μόνοι τῶν ζῴων. τούτου δ’ αἴτιον ἡ σκληρότης τῶν ποδῶν, καὶ ὅτι οὐ χρῶνται  νεύσεως χάριν αὐτοῖς ἀλλὰ πορείας· πεζεύοντα γὰρ διατελοῦσι. πάντων μὲν οὖν τῶν πολυπόδων εἰς τὸ πλάγιον αἱ καμπαί, ὥσπερ καὶ τῶν τετραπόδων ὅσα τρωγλοδυτικά (τοιαῦτα δ’

30 713b1

5

10

15

27 τετράπουν Ca S Y N V P : τετράποδον Z U   28 ἦν U Ca S N V P : ἐστι Z Y | τὰ σκέλη Z Y P : om. U Ca S N V | τε] om. Z1, add. Z2  29 καὶ] καὶ ἄλλοτε Y | ἄνω Z V : ἄνω τε U Ca S Y : πλάγιον P : πλάγιον τε N   30 εἰς τὸ ὄπισθεν] om. V   31 ἐστι] ἐπὶ S  32 ὑπ᾽ Z1 : τοῦτ᾽ U Z2 Ca S Y N V P   713b1 αὐτὰ καὶ εἰς τὸ ἔμπροσθεν U Z2 Ca Y N V P : καὶ εἰς τὸ ἔμπροσθεν αὐτὰ S : καὶ εἰς τὸ ἔμπροσθεν Z   2 ἔμπροσθεν] πρόσθεν Y   2–3 δὲ τὸ ἀκολουθεῖν Z U Ca Y N V : δὲ τὸ ἀκόλουθον S : τὸ ἀκολουθεῖν δὲ P   4 τοῦτο] τὸ Ca | βεβλαίσωται Z U Y : βεβλαισῶσθαι Ca S V P : βεβαιῶσθαι N | τε Z Ca Y P : om. U S N V   5 ἔχει Z U Y P : ἔχειν Ca S N V   6 ὥσπερ] ὡς Y   7 καὶ] om. Z1, add. Z2  8 ἂν οὕτως Z U S Y P : οὕτως ἂν Ca N V   9 προσκόπτοι Z1 U Ca S Y P : προσκόπτει Z2 V : προσκόποι N | δὲ Z Jaeger : τε U Ca S Y N V P Bekker  9–10 αὐτοῖς ἐστι U Ca S N V P : αὐτῶν ἐστι Z Bekker : ἐστὶν αὐτοῖς καὶ Y   11 εἶναι τὰ ζῶντα Z Y P : εἶναι U Ca S N : ὄντα V   12 περιττότατα Z U Ca S P : περιττότατοι Y V : περιττώματα N   13 ὥσπερ] ὡς Z   14 πόδας suppl. Jaeger : om. codd.   15 ἡ] om. Y | οὐ χρῶνται Z Jaeger : χρῶνται οὐ U Ca S Y N V P Bekker   17 τὸ πλάγιον Z U Ca S N P: πλάγιον Y : τὰ πλάγια V   18 τρωγλοδυτικά Z Y P Bekker : τρωγλόδυτα U Ca S N V Jaeger



16, 713a26–17, 713b18



[16] We have already established that the footed animals that are blood- 713a26 less are many-footed, and that none of them is four-footed.7 And the reason why their legs, except the extreme pairs, are necessarily attached obliquely, bend upward, and are themselves somewhat bowed backward is clear. For in all such animals the intermediate legs must both lead and follow. If, therefore, the legs were underneath them, they would have to bend the legs both forward and backward; forward in order to lead, and backward in order to follow. But since have to do both, for this reason their legs are bowed and bend obliquely, except for the extreme pairs; these are more in accordance with nature, since the first pair leads and the last pair follows. Moreover, their legs are bent in this particular way also because of their number; for they would thus be less likely to be an obstacle to one another in movement and collide with one another. And the reason why these animals are bow-legged is that either all or most of them are hole-dwellers; for animals that live in this way cannot be tall.

[17] Crabs are the most oddly constituted in nature among the many- 713b11 footed animals; the reason is that they do not progress forward except in the sense already mentioned,8 and they alone among animals have many leading legs. The cause of is the hardness of their feet and the fact that they use them not for swimming but for progressing on land; for their mode of living is to proceed along the ground. All the many-footed animals, therefore, have their bends obliquely like the four-footed animals that are hole-dwellers 7

IA 7, 707a20–23. IA 14, 712b13–21.

8



17, 713b19–714a6

ἐστὶν οἷον σαῦραι καὶ κροκόδειλοι καὶ τὰ πολλὰ τῶν ᾠοτοκούντων αἴτιον δ’ ὅτι τρωγλοδυτεῖ, τὰ μὲν τοῖς τόκοις,  τὰ δὲ καὶ τῷ βίῳ παντί). ἀλλὰ τῶν μὲν ἄλλων βλαισοῦνται τὰ κῶλα διὰ τὸ μαλακὰ εἶναι, τῶν δὲ καράβων ὄντων σκληροδέρμων οἱ πόδες εἰσὶν ἐπὶ τῷ νεῖν καὶ οὐ τοῦ βαδίζειν χάριν· τῶν δὲ καρκίνων ἡ κάμψις εἰς τὸ πλάγιον, καὶ οὐ βεβλαίσωται, ὥσπερ τοῖς τε ᾠοτόκοις τῶν τετρα- πόδων καὶ τοῖς ἀναίμοις καὶ πολύποσι, διὰ τὸ σκληρόδερμα εἶναι τὰ κῶλα καὶ ὀστρακώδη ὄντι οὐ νευστικῷ καὶ τρωγλοδύτῃ· πρὸς τῇ γῇ γὰρ ὁ βίος. καὶ στρογγύλος δὲ τὴν μορφήν, καὶ οὐκ ἔχων ὀρροπύγιον ὥσπερ ὁ κάραβος· πρὸς τὴν νεῦσιν γὰρ τοῖς καράβοις χρήσιμον, ὁ δ’ οὐ νευ- στικός. καὶ ὅμοιον δὴ τῷ ὄπισθεν τὸ πλάγιον ἔχει μόνος διὰ τὸ πολλοὺς ἔχειν τοὺς ἡγουμένους πόδας· τούτου δ’ αἴτιον ὅτι οὐ κάμπτει εἰς τὸ πρόσθεν οὐδὲ βεβλαίσωται. τοῦ δὲ μὴ  βεβλαισῶσθαι τὸ αἴτιον πρότερον εἴρηται, ἡ σκληρότης καὶ τὸ ὀστρακῶδες τοῦ δέρματος. ἀνάγκη δὴ διὰ ταῦτα πᾶσί τε προηγεῖσθαι καὶ εἰς τὸ πλάγιον, εἰς μὲν τὸ πλάγιον ὅτι εἰς τὸ πλάγιον ἡ κάμψις, πᾶσι δ’ ὅτι ἐνεπόδιζον ἂν  οἱ ἠρεμοῦντες πόδες τοῖς κινουμένοις.

20

25

30 714a1

5

19 ἐστὶν Z U Ca S N V : εἰσὶν Y P | σαῦραι] σαῦροι Z | τὰ Ca S Y P N V : οm. Z U | πολλὰ Z U Ca S V P : λοιπὰ Y N | ᾠοτοκούντων Ca S Y P N V : ζωιοτοκούντων Z U   20 τρωγλοδυτεῖ Z1 : τρωγλόδυτα U Z 2 C a S N V : τρωγλοδυτικά Y P   21 βλαισοῦνται Z Y P N V Jaeger : βλαισοῦται U Ca S Bekker   23 τῷ Z1 : τὸ U Z2 Ca S Y N V P   25 οὐ βεβλαίσωται Bekker : ὅτι βεβλαίσωται U Z2 Ca S N V P : οὐκ ἐκβλαισοῦται Z1 Y | τε Z Y P : om. U Ca S N V   26 ἀναίμοις] ἐναίμοις Y   27 ὄντι Z Ca Y N V P : ὅτι U S   28 στρογγύλος δὲ U Ca S N V : ὅτι στρογγύλος Z Y P   29 ὀρροπύγιον Z U C a S N V P Jaeger : οὐροπύγιον Y Bekker   30 χρήσιμον U Ca Y N V P : χρήσιμος Z : χρήσιμα S   31 δὴ Z2 Ca Y N V P Jaeger : δὲ Z1 U S Bekker | τῷ ὄπισθεν τὸ πλάγιον ἔχει] τὸ ὄπισθεν καὶ τὸ πλάγιον ἔχοι Z1, corr. Z2 | μόνος] μέρος U  32 τοὺς] om. Y | ἡγουμένους U Ca V : ἡγεμόνας Z S Y N P Bekker Jaeger   714a1 oὐ κάμπτει εἰς τὸ πρόσθεν οὐδὲ βεβλαίσωται] οἱ εἰς τὸ πρόσθεν οὐ βεβλέσωνται Z1, corr. Z2 | μὴ Z Y : om. U Ca S N V P   2 πρότερον εἴρηται Z U S : εἴρηται πρότερον Ca Y N V P   3 δὴ Z2 Ca S Y N P : om. Z1 U : δὲ V   4 τε προηγεῖσθαι] προηγεῖσθαί τε P | εἰς τὸ πλάγιον U Z2 Ca S N P : εἰς τὰ πλάγια V : εἰς πλάγιον Z1 : πλάγιον Y   5 κάμψις U Ca S N V : κάμψις πλεῖον Za.c., πλείων Zp.c. : κάμψις πλείοσιν Y : κάμψις ἐστί P | πᾶσι] αἴτιον Y   6 πόδες] οm. V   6 ψηττοειδεῖς Z Y P : ψιττοειδεῖς U Ca S N V



17, 713b19–714a6



(­lizards, for instance, and crocodiles, and most four-footed animals that lay eggs are of this nature; the cause is that they live in holes, some only during the breeding season, others throughout their whole life). But the limbs of the other are bowed because they are soft-skinned. By contrast, the crayfish are hard-skinned and their legs are used for swimming and not for walking. The bending of the crab’s limbs is oblique and they are not bowed, as is the case with the four-footed animals that lay eggs and the bloodless animals, that is, the many-footed. The reason is that the crab, which is not a swimmer and lives in holes (for it lives on the ground), has hard-skinned and hard-shelled limbs. In addition, it is round in shape and does not have a tail like the crayfish; the reason is that a tail is useful to the crayfish for swimming, whereas the crab is not a swimmer. And, of course, the crab is the only animal in which the side is like a hind part, because it has many leading feet; the cause of this is that its limbs do not bend forward and are not bowed. The reason why they are not bowed was stated before, namely the crab’s skin is hard and like that of a hardshelled animal.9 For these reasons, it must lead off with all its legs and obliquely; obliquely because the bending is oblique, and with all its legs because otherwise those at rest would be an obstacle to those moving. 9

IA 17, 713b14–16.



18, 714a6–b7

[18] Οἱ δὲ ψηττοειδεῖς τῶν ἰχθύων, ὥσπερ οἱ ἑτερόφθαλμοι βαδίζουσιν, οὕτω νέουσι· διέστραπται γὰρ αὐτῶν ἡ φύσις. οἱ δὲ στεγανόποδες τῶν ὀρνίθων νέουσι τοῖς ποσί, καὶ διὰ μὲν τὸ τὸν ἀέρα δέχεσθαι καὶ ἀναπνεῖν δίποδές εἰσι, διὰ δὲ τὸ ἐν ὑγρῷ τὸν βίον  ἔχειν στεγανόποδες· ἀντὶ πτερυγίων γὰρ χρήσιμοι οἱ πόδες αὐτοῖς τοιοῦτοι ὄντες. ἔχουσι δὲ τὰ σκέλη οὐχ ὥσπερ οἱ ἄλλοι κατὰ μέσον, ἀλλ’ ὄπισθεν μᾶλλον· βραχυσκελῶν γὰρ αὐτῶν ὄντων ὄπισθεν ὄντα πρὸς τὴν νεῦσιν χρήσιμα. βραχυσκελεῖς δ’ εἰσὶν οἱ τοιοῦτοι διὰ τὸ ἀπὸ τοῦ μήκους τῶν σκε- λῶν ἀφελοῦσαν τὴν φύσιν προσθεῖναι τοῖς ποσί, καὶ ἀντὶ τοῦ μήκους πάχος ἀποδοῦναι τοῖς σκέλεσι καὶ πλάτος τοῖς ποσί· χρήσιμοι γὰρ πλατεῖς ὄντες μᾶλλον ἢ μακροὶ πρὸς τὸ ἀποβιάζεσθαι τὸ ὑγρόν, ὅταν νέωσιν. εὐλόγως δὲ καὶ τὰ μὲν πτηνὰ πόδας ἔχει, οἱ δ’  ἰχθύες ἄποδες· τοῖς μὲν γὰρ ὁ βίος ἐν τῷ ξηρῷ, μετέωρον δ’ ἀεὶ μένειν ἀδύνατον, ὥστ’ ἀνάγκη πόδας ἔχειν· τοῖς δ’ ἰχθύσιν ἐν τῷ ὑγρῷ ὁ βίος, καὶ τὸ ὕδωρ δέχονται, οὐ τὸν ἀέρα. τὰ μὲν οὖν πτερύγια χρήσιμα πρὸς τὸ νεῖν, οἱ δὲ  πόδες ἄχρηστοι. εἰ δ’ ἄμφω εἶχον, ἄναιμοι ἂν ἦσαν. ὁμοίως δ’ ἔχουσιν οἱ ὄρνιθες τρόπον τινὰ τοῖς ἰχθύσι· τοῖς μὲν γὰρ ὄρνισιν ἄνω αἱ πτέρυγές εἰσι, τοῖς δὲ πτερύγια δύο ἐν τῷ πρανεῖ· καὶ τοῖς μὲν ἐν τοῖς ὑπτίοις οἱ πόδες,  τοῖς δὲ ἔν τε τοῖς ὑπτίοις καὶ ἐγγὺς τῶν πρανῶν πτερύγια τοῖς πλείστοις καὶ οἱ μὲν ὀρροπύγιον ἔχουσιν, οἱ δ’ οὐραῖον.

10

15

20

714b1

5

9 καὶ Z Y P : om. U Ca S N V | μὲν Z Y P : μέντοι U Ca S N V   10–11 τὸν βίον ἔχειν Z2 Y : om. Z : τὸν βίον U Ca S N V : τὸν βίον εἶναι P   11 πτερυγίων Z U Ca S N V : πτερύγων Y P   12 αὐτοῖς] αὐτῶν Y   14 νεῦσιν Ca Y N V P : σύννευσιν Z U S   16 τοῖς ποσί U Z2 S Jaeger : τοὺς πόδας Z1 Y : εἰς τοὺς πόδας Ca N V P Bekker   18 πλατεῖς Z1 Jaeger : παχεῖς U Z2 Ca S Y V P Bekker : ταχεῖς N : “fortasse παχεῖς καὶ πλατεῖς legendum est” Jaeger   20 τὰ … ἔχει] τὸ τὰ … ἔχειν S   21 ὁ] om. V   714b1 χρήσιμα] καὶ χρήσιμα P   3 ὁμοίως Z Ca Y N V P : ὅμως U S   4 γὰρ] om. S | εἰσι U Z2 Ca S N V P : δύο Z1 Y   5 ἐν τοῖς ὑπτίοις οἱ πόδες S N V P : οἱ πόδες ἐν τοῖς ὑπτίοις U : οἱ πόδες Z Y : ἐν τοῖς ὑπτίοις, omisso οἱ πόδες … πλείστοις (7) Ca  6 τοῖς δὲ ἔν τε … πτερύγια] ἔν τε τοῖς ὑπτίοις καὶ ἐγγύς, τοῖς δὲ τὰ πτερύγια ἐγγὺς τῶν πρανῶν Z | πτερύγια] πτερυγίων Y   7 ὀρροπύγιον Z U Ca S N V P Jaeger : οὐροπύγιον Y Bekker | ἔχουσιν] om. Z1, add. Z2  



18, 714a6–b7



[18] Flatfish swim as one-eyed humans walk, for their nature is warped. 714a6 Web-footed birds swim with their feet, and because they take in air and breathe, they are two-footed; but because they live in water, they are webfooted, for their feet, being of this kind, are of service to them in place of fins. They do not have their legs, as the other birds do, in the center , but placed rather toward the back; for since they are shortlegged, their legs, being set back, are useful for swimming. Birds of this kind are short-legged because nature has taken away from the length of their legs and added to their feet; for, being broad, they are more useful than if they were long, in order to force away the water when they are swimming. It is also for a good reason that winged animals have feet, but fish have 714a20 none. The former live on dry land and cannot always remain up in the air, and so necessarily have feet. Fish, on the contrary, live in water and take in water and not air. Their fins, then, are useful for swimming, whereas feet would be useless. If they had both, they would be bloodless. Nonetheless, birds in a way resemble fish. For birds have their wings in the upper part of their bodies and fish have two fins in their fore-part; birds have feet on their under-part and most fish have fins in their underpart and near their front fins; also, birds have a tail, fish a tailfin.



19, 714b8–23

[19] Περὶ δὲ τῶν ὀστρακοδέρμων ἀπορήσειεν ἄν τις τίς ἡ κίνησις καί, εἰ μὴ ἔχουσι δεξιὸν καὶ ἀριστερόν, πόθεν κινοῦνται· φαίνονται δὲ κινούμενα. ἢ ὥσπερ ἀνάπηρον δεῖ τιθέναι πᾶν τὸ τοιοῦτον γένος καὶ κινεῖσθαι ὁμοίως οἷον εἴ τις ἀποκόψειε τῶν ὑποπόδων τὰ σκέλη, ὥσπερ ἡ φώκη καὶ ἡ νυκτερίς καὶ γὰρ ταῦτα τετράποδα, κακῶς δ’ ἐστί. τὰ δ’ ὀστρακόδερμα κινεῖται μέν, κινεῖται δὲ παρὰ φύσιν· οὐ γάρ ἐστι κινητικά, ἀλλ’ ὡς μὲν μόνιμα καὶ προσπεφυκότα  κινητικά, ὡς δὲ πορευτικὰ μόνιμα. ἔχουσι δὲ φαύλως καὶ οἱ καρκίνοι τὰ δεξιά, ἐπεὶ ἔχουσί γε. δηλοῖ δ’ ἡ χηλή· μείζων γὰρ καὶ ἰσχυροτέρα ἡ δεξιά, ὡς βουλομένων διωρίσθαι τῶν ἀριστερῶν καὶ τῶν δεξιῶν.  Τὰ μὲν οὖν περὶ τῶν μορίων, τῶν τ’ ἄλλων αὐτῶν καὶ τῶν  περὶ τὴν πορείαν τῶν ζῴων καὶ περὶ πᾶσαν τὴν κατὰ τόπον μεταβολήν, τοῦτον ἔχει τὸν τρόπον· τούτων δὲ διωρισμένων ἐχόμενόν ἐστι θεωρῆσαι περὶ ψυχῆς.

10

15

20

10 κινούμενα Z2 Ca S Y N V P : κινούμενοι Z1 U   11 oἷον] ὡς Z | ἀποκόψειε] ἀποκόψει S    13 δ᾽ (alt.)] om. Z1, add. Z2  14 κινεῖται δὲ παρὰ φύσιν] παρὰ φύσιν δέ Y   15 μὲν Z Y P : om. U Ca S N V | προσπεφυκότα Y P : πεφυκότα Z U Ca S N V   17 ἔχουσι] ἔχει V | γε Z Y P : om. U Ca S N V   19 τῶν ἀριστερῶν καὶ τῶν δεξιῶν Z U Ca S N V : τῶν δεξιῶν καὶ τῶν ἀριστερῶν Y P   20 αὐτῶν U Ca S N V P Jaeger : οm. Z Y Bekker | καὶ τῶν] om. U   23 ψυχῆς] ζωῆς Zmg



19, 714b8–23



[19] A puzzle may be raised as to what the motion of hard-shelled ani- 714b8 mals is, and whence they move if they have no right and left; for they appear to be moving. The answer is that we must treat all this kind as if it were maimed, that is, as moving as a footed animal would do if one were to cut its legs off, like the seal and the bat, which are four-footed but are malformed. Now, the hard-shelled animals do move, but they move in a way contrary to nature. They are not really animals able to move, but if you regard them as stationary and attached by growth, they are able to move; and if you regard them as able to engage in progression, they are stationary. Crabs, too, have their right side defective, but they do have it. It can be seen in the claw; for the right claw is bigger and stronger, as though the right and the left wanted to be different. So much for our discussion of the parts – both the others and those 714b20 that have to do with the progression of animals and with all change with respect to place. Now that these points have been settled, our next task is to consider soul.

part iii

Interpretative Essays

chapter 3

De incessu animalium 1–3 The Theoretical Framework and the Beginning of the Actual Investigation Andrea Falcon D E I N C E S S U A N I M A L I U M 1– 2

The Aims of the IA and its Place in Aristotle’s Natural Science Our text opens with a general characterization of the aims of the IA: As for the parts that are useful to animals for motion with respect to place, we must investigate why each part is such as it is and for what end it belongs to them, and also the differences in the parts of one and the same animal and compared to the parts of animals different in kind. (IA 1, 704a4–9)

Let us start by stating the obvious: in Aristotle’s explanatory project, we do not have a single discussion of all the parts of animals. Rather, in addition to the official treatment of the parts of animals offered in PA II–IV, we have separate treatments of the locomotive parts and of the generative parts. While the former are studied in the IA, the latter are discussed in the course of the first book of the GA (GA I 3–16).1 Aristotle’s modus operandi calls for an explanation: Why do we need to deal separately with the parts that are useful for locomotion and with those that contribute to reproduction? This question becomes even more pressing as soon as we realize that some of the explanatory strategies adopted in the IA and in GA I 3–16 are the same as those adopted in PA II–IV. For instance, both in his account of the generative parts and in his study of the locomotive parts, Aristotle employs final causality. Let us focus on the IA. The opening statement reported above makes it clear that the main aim of the treatise is explanatory. More directly, Aristotle promises an answer to the question “Why?” 1

At the outset of the GA (GA I 1, 715a1–18) Aristotle presents the investigation he is about to launch as a combination of two studies: a study of the generative parts and a study of generation. The study of the generative parts is a leftover from the PA. It was postponed because Aristotle made a determination that a study of generative parts was going to be most successful in the context of a study of generation.





Part III Interpretative Essays

Note, however, that the emphasis is from the start on final causality: why each part is such as it is and (καί) for what end animals have it. If the καί is epexegetical, Aristotle promises an explanation of the presence in an animal of a certain locomotive part by asking for what end the animal has that part. If it is not, his promise is to explain the presence of that part by asking also for what end the animal has that part. Admittedly, we cannot decide between these two readings on merely grammatical grounds. Rather, we have to look at the actual explanations offered in the rest of the treatise. As soon as we do that, however, we realize that Aristotle’s focus is not just on final causality. But the same observation can be made in connection with the explanations advanced in PA II–IV. There, too, the search for the causes is not just a search for final/formal causes but also for material causes. While the points of contact of the IA and GA I 3–16 with PA II–IV are important and obvious, they do not mandate a common treatment of all the parts of animals in a single work – namely, a PA augmented by a study of the parts of animals useful to locomotion and reproduction. I leave aside the reasons that may have motivated Aristotle to integrate the study of the generative parts into his study of generation,2 and concentrate on those that may have led him to opt for a separate treatment of the locomotive parts. I begin by noticing a mildly surprising fact. There is no evidence, outside the IA, that the treatment of the parts useful to animals for locomotion is considered a contribution to the study of parts. Rather, there is some evidence that the IA is regarded as a study of animal motion. More directly, there are four explicit references to the IA in the extant corpus of Aristotelian writings. Three of these references are found in PA. They are all references to another causal investigation; moreover, they describe the IA as a study of progression (πορεία) or motion (κίνησις).3 Either way, the reference is to a work on locomotion rather than to a work on the bodily parts involved in locomotion. Interestingly enough, the attempt to cast the content of the IA as an investigation of animal locomotion is not unique to PA. The fourth reference is found in the DC. There too Aristotle refers to the IA as to a study of the different modes of animal locomotion.4 2

Recent attempts to deal with the question of why Aristotle has opted for separate treatments of the generative parts can be found in Gotthelf–Falcon 2017: 15–34 and Lefebvre 2017: 35–55. 3 PA IV 11, 690b15 and IV 11, 692a17: πορεία; PA IV 13, 696a11–12: πορεία and κίνησις. The reference to a study of πορεία and κίνησις looks like a conscious attempt to present the content of the IA as a study of a special type of motion, namely animal progression. I will return to the difference between progression (πορεία) and motion (κίνησις) in due course. (See below, my commentary on IA 3.) 4 DC II 2, 284b13: motions.



3 De incessu animalium 1–3



There is a fifth, indirect reference to the IA at the outset of the MA: In regard of the movement of animals, namely what features belong to each particular kind of them, and what the differences and what the causes for each of their particular features are, we have investigated exhaustively elsewhere. Now the common cause of moving with any motion whatsoever is to be investigated in general. The reason is that some animals move by flying, some by swimming, some by walking, and some in other ways. (MA 1, 698a1–7)

Aristotle promises an account that will explain animal motion by pointing to “the common cause.” This account will explain the movement of animals “in general” by giving the common cause of their moving with “any movement whatsoever.” This is exactly what Aristotle does in the rest of the MA. However, he makes it clear that this account is to be integrated with a detailed account of the motion belonging to each kind of animal. I speak of a “detailed account” because Aristotle seems to distinguish between two approaches to animal motion: while the first focuses on what is common or general, the second is concerned with the differences in animal locomotion. Dealing with the differences in animal locomotion requires engaging in a study of the ways in which animal locomotion is realized in the various kinds of animals. The latter is traditionally taken to be the approach adopted in the IA, which has a great deal to say on the bodily parts involved in bodily displacement. If so, the IA and the MA would work together to provide an integrated account of animal motion.5 A full discussion of the opening lines of the MA goes emphatically beyond the scope of this chapter.6 What matters here is that a tension can be felt between the way in which Aristotle describes the IA at the outset of the MA (as well as in the PA and in the DC) and the emphasis that he places on parts in the opening statement of the IA. We can try to reduce this tension by saying that in the IA we are interested in the parts of animals insofar as they are useful for locomotion, and in this sense our study of those parts is also a study of animal locomotion. Still, a tension remains between two ways of describing the explanatory project attempted in the IA. While the first considers this work a contribution to 5

It is in fact not unusual to find these two works placed together in recent translations: see, for instance, Preus 1981; Kollesch 1985; and Morel 2013. Of course, this passage from the MA provides the theoretical motivation for this editorial practice. We have already seen that this passage may have influenced Bekker in his decision to adopt the sequence PA, IA, and MA. See my essay on the reception of the IA (ch. 2). 6 I refer the reader to Falcon 2017a: 215–235.



Part III Interpretative Essays

the study of the parts of animals, the second regards it as a study of animal locomotion. I note, in passing, that the very end of the treatise reinforces the idea that the PA and the IA work together in the explanation of the parts of animals: So much for our discussion of the parts – both the others and those that have to do with the progression of animals and with all change with respect to place. (IA 19, 714b20–22)

Also in light of this second passage, it is not surprising that the IA is often considered an appendix of the PA.7 Quite the opposite: the reader who is engaged in a close reading of the IA finds it natural to assimilate this work to the PA.8 Yet we should not be blind to the fact that the IA is kept separate from the PA. This fact cannot be explained away as a mere accident of the manuscript transmission, or as the result of a later editorial decision. On the contrary, our work begins with a clear set of questions – namely, the agenda of the IA – to be answered on the basis of a theoretical framework explicitly introduced for this purpose. Put differently, even when Aristotle thinks of the IA as a study of parts rather than a study of animal locomotion, he has reasons to think that this study is not a mere continuation of the one offered in PA II–IV.9 A final note on the opening statement of the IA is in order. Aristotle makes it clear that he is not only interested in why animals have certain parts but also why they have them in the particular way they do. This is how we should understand the second part of the passage quoted above. There, we are told that our task is to understand why the different parts involved in locomotion differ within the same animal (e.g., front and hind legs in four-footed live-bearing animals) and across different kinds 7

See, for instance, Preus 1981: “Obviously [the IA] was written in conjunction with the PA. This is shown not only by the first words of this treatise, but also by the similar way in which the material is treated” (147). Even a quick glance at the different and at times competing arrangements adopted in the early printed editions of Aristotle suggests that all the editors of Aristotle have tried to save the link that Aristotle establishes between the IA and the PA. I refer the reader to the data I have collected in dealing with the reception of the IA (ch. 2). 8 The most explicit attempt to defend this approach is found in Leonico Tomeo 1523. For more information on his translation, which is accompanied by a running commentary, I refer the reader to chapter 2. Here suffice it to say that Niccolò Leonico Tomeo felt compelled to discuss the place of our work in the Aristotelian project because he offered his translation of the IA as part of a translation of the so-called Parva naturalia and the MA. 9 At this point the reader should be able to fully appreciate why the position of the IA in Aristotle’s natural philosophy is not easy to determine. Indirect evidence of this difficulty is the unstable position of the IA in the Aristotelian corpus. In chapter 2, I offer a brief survey of this question in connection with the position of the IA in the early printed editions of Aristotle.



3 De incessu animalium 1–3



of animals. The implications of this approach will become fully apparent as soon as we turn to the agenda of the IA.

The Initial Agenda of the IA The general characterization of the aims of the IA is followed by a list of the matters to be investigated. We can take this list to be the initial agenda of the IA. Clearly, the task that Aristotle envisions for himself in the opening statement is complete only if we have an answer to the following eleven questions: [Q1] What are the fewest points at which animals move? [Q2] Why do blooded animals move by means of four points while bloodless animals move by means of more than four points? [Q3] Why, in general, are some animals footless, some two-footed, some four-footed, and some many-footed? [Q4] Why do all the animals that have feet have an even number of feet? [Q5] Why, in general, are the points by means of which animals move even in number? Furthermore (ἔτι), [Q6] Why are human beings and birds two-footed whereas fishes are footless? [Q7] Why do human beings and birds, being both two-footed, bend their legs in opposite directions (human beings convexly while birds concavely)? [Q8] Why do human beings bend their legs and arms in opposite directions? [Q9] Why do the four-footed animals that are live-bearing bend their legs in the opposite way to human beings and also in opposite ways with respect to themselves (front legs convexly while hind legs concavely)? [Q10] Why do the four-footed that are egg-laying bend their legs in a way unique to them, namely laterally away from their body? [Q11] Why do four-footed animals move their legs diagonally? A couple of comments are in order. First, Aristotle does not introduce [Q8], [Q9], and [Q10] as questions but rather as statements of the way things are, but it is clear that he conceives of his task as providing an explanation of those statements. Hence, I have felt entitled to rephrase



Part III Interpretative Essays

those statements as questions to be answered in the course of the actual investigation conducted in the IA. Second, I divided the eleven questions into two groups because there is some evidence that we are presented with a carefully organized set of questions. To begin with, the agenda is divided into two blocks of questions that are clearly demarcated. I take ἔτι at 704a16 to be evidence of this division. Moreover, there is some evidence that the agenda is organized from the more to the less general questions. This is certainly the case in the first set of questions. We begin with questions concerned with the number of points and continue with questions that have to do with the number of feet. The questions dealing with the number of points – [Q1], [Q2], and [Q5] – are more general, and indeed more abstract, than the questions that regard the number of feet – [Q3] and [Q4]. The second set of questions deals with the presence (or absence) of locomotive parts in different kinds of animals. In addition to being concerned with specific kinds of animals, these questions are also about bending. Last but not least, the reader should recall the general characterization of the aims of the IA offered at the outset of the investigation. There, we are told that we will look not only at the differences in the different kinds of animals but also at the differences in the same kind of animal. An example may help illustrate this point. Consider the question about the bending of legs and arms in the human being – [Q8] – or the questions about the bending in four-footed animals, livebearing or egg-laying animals – [Q9] and [Q10].10 Answering these questions will contribute to our understanding of how the different parts involved in locomotion differ within the same animal. These considerations do not exhaust the topic of how the agenda was generated but they should go some way toward illuminating the criteria adopted for its organization. It remains to be seen whether there are deeper reasons for dividing our agenda into two parts. For example, geometrical language and geometrical arguments are conspicuous in Aristotle’s treatment of bending. IA 9 is a very clear example of the application of mathematical science.11 One may wonder whether the conscious application of geometry to the study of animal locomotion has an impact on how Aristotle organizes his agenda. Be this as it may, we are not 10

The live-bearing four-footed animals and the egg-laying four-footed animals are two large kinds (µέγιστα γένη) of animals whose existence and significance for Aristotle’s explanatory project is established in HA I 6, 490b19–23. Full discussion in Gotthelf 2012c: 293–306 (especially 299– 302). 11 In his essay on IA 9, Chris Frey speaks of “mathematical kinesiology” in connection with the main argument offered in this chapter. Helpful remarks on this topic can be found also in Angioni 2018: 53–71.



3 De incessu animalium 1–3



presented with a random list of questions but rather with a reasoned agenda that is also theoretically organized in a certain way. What we see when we see an animal moving from one place to another is movement that is organized by kind and described by adopting, inter alia, a geometrical language. Admittedly, the geometrical language that Aristotle adopts in the second part of his agenda is far from being crystal clear. Aristotle does not simply contrast bending toward the convex (ἐπὶ τὸ κυρτόν) and bending toward the concave (ἐπὶ τὸ κοῖλον). He also speaks of bending toward the circumference (ἐπὶ τὴν περιφέρειαν), and bending toward the convex part of the circumference (ἐπὶ τὸ κυρτὸν τῆς περιφερείας). This complicates things. How are we expected to understand these expressions? In all probability, Aristotle refers to the circumference that we can draw around the body of an animal if we take its heart (or the analogue to the heart) as its center. Consider the case of the four-footed animals that bend their front and hind legs in opposite directions – namely, their front legs toward the convex part of the circumference and their hind legs toward the concave part. In this case, we are talking of a circumference with an intrinsic orientation that mirrors the orientation of the living body. We can adopt this idea to make sense of what Aristotle says with reference to the bending of the human knees and arms. Again, we have to imagine that we can draw a circumference around the body of the human being. Moreover, since Aristotle is contrasting bending toward the circumference and bending toward the concave, we can understand “toward the circumference” as an abbreviation of “toward the convex part of the circumference.” Figure 3.1 illustrates what Aristotle may have in mind.12

Figure 3.1  Bending toward the circumference 12

Thanks to Christopher Frey for drawing this diagram.



Part III Interpretative Essays

Aristotle ends his agenda by returning to the overriding explanatory aim of the investigation he is about to launch: The causes of all these facts, and of other facts similar to these, are to be studied: for that things are in this way is evident from natural research; we must now investigate why. (IA 1, 704b8–11)

At least two observations are in order. The first pertains to the addition of the words “and of other facts similar to these.” Evidently, the agenda does not contain all the questions relevant to a complete study of animal locomotion. It is telling that in the course of his investigation Aristotle poses other questions that naturally arise from dealing with cases of locomotion that challenge his basic account of how animals move from place to place. This observation does not entail that the eleven questions Aristotle offers at the outset of his investigation are merely a way of starting off his discussion, or that they are a set of questions that occurred to him before engaging in the actual investigation. In the introduction to the volume, I have argued that while the eleven questions that are found in his agenda are not a complete and definitive list of questions, they provide a convenient platform from which to engage in an intelligent investigation of how animals move from place to place.13 Hence, my decision to refer to it as the initial agenda of the IA. The second observation has to do with the expression “natural research” (φυσικὴ ἱστορία). This expression need not be either a reference to a lost HA-like investigation or an alternative title for the extant HA. It can be taken to be a reference to the activity of collecting and establishing the relevant facts (ἱστορία).14 Here by “facts” I mean what we see when we see an animal moving. But what is the intended force of the addition of the qualification “natural” (φυσική)? To answer this question we only have to keep reading a few words into the next chapter.15 The expression “natural research” (φυσικὴ ἱστορία) is meant to work in tandem with “natural investigation” (φυσικὴ μέθοδος), which is found a couple of lines below (IA 2, 704b13). By adopting this language, and speaking of natural investigation and natural research, Aristotle wants to 13

See “Introduction: Explanatory Strategies in the De incessu animalium” (ch. 1). Aristotle could refer to another fact-establishing investigation (what I have described as an HA-like investigation), but I think that ἱστορία (in the singular) points to the activity of establishing the relevant facts rather than to the result of that activity – namely, the facts that are collected and organized for their subsequent explanation. The most obvious way to refer to the latter would be by means of ἱστορίαι (in the plural). 15 The reader should bear in mind that chapter division is a convention that does not go back to Aristotle. 14



3 De incessu animalium 1–3



convey the message that his study of the locomotive parts is part and parcel of his general study of nature. As a result, it is going to be a robustly philosophical study. Here, by “robustly philosophical,” I mean a study that employs the explanatory principles adopted for the study of nature as a whole. This is also the reason why it is important to look at the IA as a good example of the methodology that Aristotle applies to his study of natural phenomena. The relevance of this last point becomes apparent as soon as we turn to the second chapter of the IA.

The Explanatory Principles Adopted in the IA What makes the IA an especially interesting case study for anyone who is seriously interested in the explanatory commitments that shape and control Aristotle’s investigation of nature is the degree of explicitness by which Aristotle first introduces and then uses a theoretical framework in the attempt to give his causal explanations of the relevant facts. IA 2 is a case in point. Aristotle’s primary goal in IA 2 is to provide the reader with explanatory resources for the investigation that he is about to launch. In this sense, the chapter is still a preliminary text. Having stated the overall aims of the work, and having offered a set of questions that gives both clear boundaries and a strong structure to the investigation to come, Aristotle proceeds to introduce the explanatory principles that he will employ in answering those questions. The first thing to note is that the language is carefully crafted to convey the message that the explanatory principles Aristotle is about to introduce have a significance that goes beyond the study of the locomotive parts of animals. By applying these principles, Aristotle inscribes the IA into a larger explanatory project – his natural philosophy: We begin our investigation by positing those that we are used to employing often in natural investigation, assuming that things occur in the same manner in all the works of nature. (IA 2, 704b12–14)

What Aristotle tells us in this opening statement is fully compatible with the fact that two of the three explanatory principles he is about to introduce find their clearest and most paradigmatic application in the study of animals. Moreover, it is very likely that they were first discovered and formulated in the study of animals and then extended to other areas of natural philosophy on the assumption that “things occur in the same manner in all the works of nature.” I would like to elaborate on these points.



Part III Interpretative Essays

Let us begin our discussion of the three explanatory principles by concentrating on the claim that nature does nothing in vain.16 This is a teleological principle that Aristotle invokes several times in the course of his natural investigation.17 However, this principle has its fullest and most precise formulation in the IA, so it is not surprising that this formulation and its subsequent application in the course of the study of the locomotive parts of animals has received considerable attention lately.18 Let’s focus on how the principle is formulated: nature does nothing in vain but always what is best from among the possibilities for the substance of each kind of animal, which is why if it is best in a certain way, it is also in this way according to nature. (IA 2, 704b15–18)

The reader will find an extensive discussion of this principle and its application in the IA in the essays by Stasinos Stavrianeas, Spyridon Rangos, and Sarah Ruth Jansen (chs. 6, 9, and 10). Here I am content to formulate the two main questions that quite naturally arise from how the principle is formulated. First of all, it is far from clear whether Aristotle is offering a single unified principle or rather two separate principles. I am tempted to say that the shorter formulation of the principle (“nature does nothing in vain”) is here articulated, and indeed illuminated, in terms of what is best or optimal for the substance of each kind of animal. 19 However, the alternative reading cannot be ruled out at this early stage of the investigation. In fact, the reader will find out that Sarah Jansen takes the view that Aristotle is operating with two teleological principles. While the first is a negative principle (“nature does nothing in vain”), the second is a positive principle (“nature realizes what is best from the available possibilities for each kind of animal”). By contrast, Spyridon Rangos holds that we have one and only one principle. The second interpretative question has to do with the claim that nature realizes what is best or optimal with respect to a certain range of possibilities (ἐνδεχόµενα). How are we expected to understand these possibilities? 16

The nature in question is not a cosmic nature but rather the internal source of change and rest introduced in Phys. II 1, 192b20–23. 17 For instance: DC I 4, 271a33; DC II 11, 291b13–14; DA III 12, 434a31; PA II 13, 658a8–9; PA III 1, 661a23–24; GA II 5, 741b4–5; GA II 6, 744a36–37. It is worth noting, in passing, that this explanatory principle is invoked also by Theophrastus in his study of plants (CP I 1.1). 18 The most recent discussions of this teleological principle are Lennox 2001b: 205–223; Leunissen 2010: Chapter IV (especially 119–121 and 124–135); Henry 2013: 225–263; Morel 2016: 9–30. 19 See Henry 2013: 230.



3 De incessu animalium 1–3



At the very least, we can say that Aristotle has in mind natural rather than logical possibilities because not all that is logically possible is possible in nature. It does not take long to realize that this is a first approximation at best because in the passage quoted above Aristotle fixes the range of what is possible with reference to the substance (οὐσία) of each kind of animal. By “substance” Aristotle means the essential nature of the animal – what it is for an animal to be that kind of animal that it is. I illustrate this point with the help of an example. In the course of the IA, Aristotle seems to assume that the elongated shape of snakes is a basic fact that pertains to their essential nature along with other features such as being blooded. In other words, at least for Aristotle, to be a snake is to have an elongated body and to be a blooded animal. By his lights, these facts about the nature of snakes are basic and as such pose constraints on the sort of locomotive tools that nature can design to help them move around in the most efficient way.20 Further light on how the range of what is possible is to be determined can be shed only by looking at how Aristotle applies this general principle in the course of the study of the locomotive parts of animals. What matters here is to realize that the explanatory principle that nature does nothing in vain, which Aristotle may have discovered and formulated in the course of the study of the parts of animals, is invoked to generate explanations in other areas of natural philosophy. For instance, Aristotle adopts it in the attempt to offer explanations in the context of his celestial physics.21 Obviously, what licenses this explanatory strategy is the assumption – explicitly made in the IA – that “things occur in the same manner in all the works of nature.” The second explanatory principle that Aristotle introduces for the study of the locomotive parts of animals is a fairly abstract but powerful way of thinking about living bodies. To my knowledge, Aristotle is the first to think of a living body as having “functional directions.” My translation is an imperfect attempt to render a creative use on the part of the Aristotle of the word διάστασις. The original meaning of this word is “dimension.” For example, a body is a magnitude that extends into three dimensions (διαστάσεις).22 However, Aristotle does not use this word to 20

I refer the reader to the essay by Stasinos Stavrianeas (ch. 6) for more on this point. Leunissen 2009: 215–237 (cf. Leunissen 2010: 168–174). 22 DC I 1, 268b6–7. Cf. Top. VI 5, 142b24–25. Aristotle can also employ the term διαστήµατα to refer to the three dimensions as in Phys. IV 1, 209a4–5. But it is clear that, at least in this case, Aristotle is concerned with the idea that a dimension is always an interval between two extremes. This explains his choice of the word διάστηµα. 21



Part III Interpretative Essays

refer to bodies in general. His primary focus is on living bodies. A living body has to be organized in a certain way in order to support life. More directly, a living body can display up to six διαστάσεις, divided into the following pairs: up/down, front/back, and left/right. As will become apparent in due course, a living body that is able to engage in locomotion requires all these three pairs. In other words, an upper and a lower part, a front and a back, and finally a right and a left side must be present to the living bodies that are endowed with the power to displace themselves. In this sense, these living bodies display the maximum degree of complexity. A living body is minimally articulated into an upper and a lower part. The entry of nourishment is in the upper part. I have provided an outline of the Aristotelian doctrine, which is here introduced without further elaboration.23 I conclude this brief discussion by pointing out that Aristotle adopts (and indeed adapts) this doctrine in dealing with the first heaven (also known as the heaven of the fixed stars) in DC II 2.24 He makes it clear that the doctrine of the functional direction is native, and indeed proper, to the study of animals. 25 While designed for the study of perishable life – namely, life as encountered here on earth – these functional directions can be employed to generate explanations in the course of celestial physics. Aristotle can do so because he is relying, once again, on the assumption that “things occur in the same manner in all the works of nature.” The third principle that Aristotle introduces in this chapter is a principle of motion: pushing and pulling are the sources of all per se motion in place. Aristotle has established this principle in the context of his most general treatment of motion. The proof that all motion in place is per se or per accidens, and all per se motion in place can be reduced to pushing and pulling is given in Phys. VII 2, where Aristotle takes upon himself the task of showing that in fact all motion in place can be reduced to a case of pushing and pulling.26 If this is correct, we can immediately make the following, rather obvious, observation: while the first two explanatory principles are native to Aristotle’s study of animals, this third principle has its origin in Aristotle’s most general discussion of motion – what the subsequent tradition has considered Aristotle’s On

23

Aristotle discusses this doctrine in IA 4. I refer the reader to the interpretative essay in this volume by Panos Dimas (ch. 4). 24 His implicit assumption is that this heaven is a living body. 25 Full discussion in Lennox 2009: 187–214. 26 Phys. VII 2, 243a11–244a6.



3 De incessu animalium 1–3



Motion.27 This is also a relatively unsurprising development. To the extent that Aristotle is about to launch an investigation of animal locomotion, and not merely an investigation of locomotive parts, Aristotle reminds us (and himself ) that a general principle of motion is relevant to the project. DE INCESSU ANIMALIUM 3

The Beginning of the Actual Investigation IA 1 and 2 are best understood as a prolegomenon in which Aristotle provides the reader with his working agenda, as well as the explanatory principles, that he plans to use in his attempt to explain the relevant facts of animal locomotion. By contrast, IA 3 marks the beginning of Aristotle’s actual investigation. The train of thought in this stretch of text goes something like this: Aristotle first establishes that there has to be some underlying thing that functions as support for the animal that moves from one place to another; then he shows that there has to be some complexity in the animal itself. This second claim is illustrated by looking at the mechanics of jumping. I will organize my comments around the two key claims made in this chapter. 1. There has to be some underlying thing that functions as a support for any kind of change of place. Animals that move do so either by means of their whole body all at once as in the case of the animals that jump or part by part as in the case of the animals that are engaged in progression (πορεία). The first thing to note about this division is the rather surprising result that the case of animal motion I rendered (following William of Moerbeke) as “progression” is restricted to the case of motion by means of locomotive parts. Of course, this is not the only, or even the most obvious, meaning of πορεία.28 Still, this is the mode of bodily displacement that Aristotle would like to study in the rest of his investigation. In this sense, there can be progression on land or in water because in both cases animals make use of locomotive 27

We need not worry here about the different ways in which the tradition has considered the project of a general treatment of motion – namely, whether the Περὶ φύσεως begins with Phys. V and extends to Phys. VIII and how exactly (if at all) Phys. VII fits within this project. 28 Plato, for instance, speaks of the motion of the planets as a case of πορεία (Tim. 39 B 4 and 39 D 8). Of course, his choice has to do with the complex and apparently erratic motion of the planets.



Part III Interpretative Essays

parts. While the former kind of progression may require feet, the latter does not.29 In connection with this last point, we should keep in mind that our investigation has been presented as a study of the parts that are useful to animals for motion. At this point, we are able to see the link that Aristotle would like to establish between locomotive parts and πορεία. We can also appreciate how precise his language is at the end of his investigation, where we are told that the investigation of the other parts and the parts involved in the progression (πορεία) of animals and all change of place (µεταβολὴ κατὰ τόπον) has come to its natural conclusion.30 We can see why, in particular, Aristotle is careful to distinguish between change of place and progression. Progression is a specific case of motion with respect to place. Last but not least, we can also appreciate why our treatise as a whole is transmitted in the manuscript tradition as a work on πορεία. While there is no reason to think that this title goes back to Aristotle,31 it points to a key aspect of this investigation: Aristotle is concerned, first and foremost, with progression by means of locomotive parts. Let us return to the first claim made in our chapter. This claim has a significance that goes beyond progression (πορεία). For any kind of motion in place, there has to be an underlying thing that functions as support. Aristotle illustrates his claim by pointing to cases where the underlying thing fails to function as an adequate point of support. The discussion of these cases, as well as the general point that there has to be a point against which the animal can support itself, is made also at the outset of the discussion of animal motion offered in the MA. I have in mind the extensive discussion offered in MA 2, where Aristotle insists that there must be something outside the animal that provides support. His examples of progression on sand or on mud illustrate what happens when the support fails to perform its function. There is no doubt that the contents of IA 3 and MA 2 overlap.32 This is not surprising if we consider that both investigations are concerned with animal motion and both start from a discussion of what makes motion in place possible. For πορεία as progression on land, see IA 8, 708a25 and 708b6; IA 12, 711a23; IA 17, 713b13 and 16. In all these cases, the Greek term is used to designate progression on land by feet. This is the paradigmatic case of πορεία. If my reading is correct, however, this term must also apply to progression on land without feet (after all, snakes are footless animals). For πορεία as progression in water, see IA 15, 713a5–6. 30 IA 19, 714b20–22. 31 We have already seen that Aristotle can refer to the IA as to a work on progression (πορεία) or on motion (κίνησις). 32 The language adopted in MA 2 and IA 3 also overlaps. In both chapters, for instance, Aristotle uses the verb ἀποστηρίζεσθαι (MA 2, 699a 5, 7–8, and 9; cf. IA 3, 705a8). 29



3 De incessu animalium 1–3



2. The animal that moves displays a minimal level of complexity: one part is acted upon by the other by being compressed, the other acts on it by pressing. Hence, no motion from place to place is possible without this minimal level of complexity. The mechanics of motion is explained by focusing on jumping. We may wonder why the focus is on this particular case. We should keep in mind, first of all, that we are still looking for a general conclusion regarding what makes all motion in place possible. Secondly, jumping is a more difficult case than progression “part by part.” It may even be regarded as a potential counterexample to the overall conclusion reached in this chapter, namely that motion in place requires a minimal level of complexity. In jumping the whole body appears to move all at once. Aristotle shows that even in jumping there is some complexity: the animal that moves by jumping produces the jump by supporting itself on its own upper part and on what is below its feet. Aristotle elaborates on his analysis of jumping by looking at more familiar cases such as jumping with weights or running.33 The outcome of this analysis is captured by the following key statement: what moves produces motion with respect to place by making use of at least two parts as its tools: (a) the first, as it were, compressing, and (b) the compressed. We are back to the language of parts, which are regarded as tools or instruments to produce motion. By speaking of parts as instruments or tools, I am relying on what is nowadays a standard reading of the Greek ὀργανικόν.34 What Aristotle has primarily in mind are parts such as wings or feet.35 His language is meant to convey the message that those parts are to be understood in light of the capacity for motion with respect to place in the form of progression rather than vice versa. In other words, it is because the animal has a certain capacity for motion that it is equipped with certain locomotive parts, which are like tools or instruments for the exercise of that capacity.36 Aristotle ends his analysis of motion with a corollary: nothing that is partless can move itself because it does not display the complexity of what is to be acted on and what is to act on it. Note that the notion of 33

For a discussion of ancient long-jumping (always with weights), I refer the reader to Gardiner 20023: 144–153. 34 For helpful comments on what “organic” means (in connection with “organic body” as featured in the most common account of the soul, DA II 1, 412b4–6) see Menn 2002, especially 107–117. 35 See IA 4, 705b22–23. 36 Aristotle’s explanatory approach is stated very clearly at the outset of the discussion of generative parts: it is because the male and the female have different capacities for generation that they are equipped with different bodily parts (including organs or instruments for mating). Cf. GA I 2, 716a17–35.



Part III Interpretative Essays

part adopted in this corollary is fairly abstract: what is active and what is passive are regarded as parts involved in motion. By recasting the parts involved in motion in terms of an agent and a patient, Aristotle makes contact with a key theorem of his natural philosophy: all change requires an active and a passive component. When both components are present, and there is no impediment, change is brought about. The official presentation of this theorem is GC I 7. There, Aristotle is concerned with providing the conceptual resources for the study of the sublunary world. However, he is eager to make reference to this principle in order to show that what happens locally is an instance of a more general truth about nature. Our passage is an instance of this more general phenomenon. What is at stake is the integration of our investigation into a larger explanatory project. We have seen that this is a major concern that Aristotle signals before introducing the theoretical framework in IA 2.

chapter 4

De incessu animalium 4 Aristotle’s Conception of Dimension Panos Dimas Introduction IA 4 is dedicated to an account of the concept referred to by the Greek term διάστασις, usually, and correctly, rendered in English as “dimension.” Though the concept is explicated in some detail in the IA, this work is not the only place where Aristotle engages with it. Dimension is also discussed in the DC. When compared to what we read in the IA, the picture painted in the DC contains elements that might give the impression that Aristotle is of two minds about the concept of dimension. It is imperative, therefore, that before dealing with the account offered in the IA we analyze statements regarding dimension made in the DC. We need to ask if there is a discrepancy between the views expressed in these works; and if not, we need to define the nature of their relationship. I am concerned with these questions mostly in the first two sections. In the subsequent sections I turn to the IA to argue that Aristotle conceives of dimension as possessing a solid ontological footing, one grounded on things that are capable of independent existence and have the principle of their motion in themselves – namely, natural substances. Indeed, the DC makes it plain that the principles governing the movement of beings able to move themselves, and in particular the directionality of such movement, is the place to look in order to achieve a proper account of dimension. It is for this reason, as I see it, that the project of actually offering such an account is executed in detail in the IA. Now, the IA presupposes but does not make explicit why a proper account of dimension is indelibly linked to, and dependent on, facts about ensouled beings. The DC rectifies this omission by explaining why the IA is the proper place for a detailed exposition of the concept of dimension. By so doing it provides a framework for interpreting the account of the IA, or so I argue. In the remainder of this chapter, I look at Aristotle’s actual use of the principles of motion of self-moving substances and its directionality in accounting for dimension. I am grateful to all the participants in the De incessu workshop for their useful comments and observations. In particular, the interventions by Andrea Falcon in earlier drafts of the present contribution have been invaluable.





Part III Interpretative Essays

Body, Magnitude, and Dimension The discussion in the DC explicitly points to the IA as the place most fitting for a proper treatment of the notion of dimension.1 However, reflection on some uses of the term “dimension” in the DC promises to throw light on exactly why it is the IA that most fittingly furnishes the account of the thing to which the term refers. Interestingly, the DC speaks of the dimensions as being three in number, whereas we shall see that the IA says that they are six.2 At first sight, these statements may seem to constitute evidence for the existence of two diverging views on the topic of dimension in the Aristotelian corpus. What reinforces the impression that the two accounts are importantly distinct is that the DC appears occasionally to be using the term “dimension” with reference to spatiality quite generally.3 The reason is that, in the DC, all three dimensions are spoken of as properties of magnitude. If we take into account that for Aristotle the hallmark property of magnitude is extension, it might seem that his conception of dimension is not that different from the way moderns tend to think of this notion. As we are going to see, however, such a conception is alien to the one underlying the statements in the DC. In fact, it is hardly likely that in the DC Aristotle would refer his reader for a fuller explication of the notion of dimension to a work in which he presents a view of that notion that is significantly different from the one he holds while he writes the DC. Quite the contrary: Aristotle is careful to make the point that a study of dimension is already in place when he writes the DC, and he even refers his reader to the work where a treatment of the concept of dimension is to be found.4 As a result, we have to understand how the statements from the two works on the number of dimensions are only seemingly divergent. The first thing we need to do, therefore, is to get a view as to whether Aristotle in the DC understands the term “dimension” in an ontologically austere fashion, i.e., as merely referring to properties of extension and nothing else, or whether his way of speaking is indicative of an 1

If right and left are present in something, we should look for the principles that make them present (DC II 2, 284b12–18), and the principles should first be sought in animals rather than in the heavens. Moreover, left and right are, in a sense, principles (DC II 2, 284b18–24). For a view on the connection between DC and IA with respect to the notion of dimension see also Lennox 2009: 187–214. 2 DC II 2, 284b20–24. 3 Which the modern reader might think is close to the Cartesian sense of the term. For a discussion of the Aristotelian and Cartesian conceptions of body see Falcon 2016: 423–436 (especially 433– 435). 4 DC II 2, 284b13–14.



4 De incessu animalium 4



ontologically richer conception. A slightly closer look at the DC is sufficient to make the reader suspect that the way Aristotle conceives of magnitude has ontological roots of a kind that is alien to a conception of dimension as merely a property of extension. Tellingly, he makes the point that a detailed account of dimension is properly the task of a study of the motion of animals. This would be a strange thing to say if Aristotle conceived of dimension merely as a property of extended magnitude. He furthermore states that the three dimensions have names – “length,” “breadth,” and “depth.”5 Giving each dimension a name that properly belongs to it suggests, if it does not imply, that the three dimensions are to be distinguished in more ways than just numerically, even though it is true of all three of them that they are dimensions. Each one, we are led to believe, is distinct from any one of the other two in such a way that, for instance, using the name “length” to point to the dimension named breadth would be a mistake of misidentification. Assuming, as we reasonably should, that each name is appropriate for the dimension it names and only that, we should also assume that it refers to things regarding its referent that are not to be found in any of the other two dimensions that have different names. It follows then that each name, in addition to identifying a property of extension, identifies distinguishing marks of its referent over and above what makes its referent numerically distinct from the other two dimensions. This result goes hand in hand with the fact that nothing has independent existence merely as magnitude or space for Aristotle. To be sure, the DC often, if not mainly, speaks of dimensions by reference to magnitude. But notice that the very statement with which this work opens makes magnitude, which is the subject matter of geometry, inextricably part of the larger scientific domain known as the science concerning nature (ἡ περὶ φύσεως ἐπιστήμη).6 Actually, to characterize this science, Aristotle goes on to mention three elements: (1) bodies (σώματα) and magnitudes (μεγέθη), (2) their changing properties (πάθη) and motions (κινήσεις), and (3) the principles of these things as being parts of substance (οὐσία). There is movement from place to place because a body is what is so displaced, and there are other kinds of change because a body receives and loses properties. Body is also what yields to the principles that make it a constituent of that substance. The order of presentation of these elements is ascending in terms of ontological dependence; there are bodies in 5

DC II 2, 284b25–26. DC I, 1, 268a1–7.

6



Part III Interpretative Essays

which properties reside, and there is motion of bodies because there are substances. We should not underestimate the significance of mentioning magnitude together with body in (1). As a matter of fact, only the term “body” names something that actually exists in nature. “Magnitude” does not refer to anything over and above body. Magnitudes are mentioned together with, and not independently of, bodies because they are nothing other than bodies considered apart from sensible attributes and motion.7 Magnitudes are mentioned at the outset of the DC, in a remark about the subject matter of the science of nature, because a magnitude is nothing other than a body the way mathematicians study it. More precisely, the object of study of the mathematicians is body in an idealized form. Idealizing body in this way makes possible, among other things, the study of dimension in the abstract by isolating magnitudes of fewer than three dimensions. For instance, mathematicians can speak of line itself by itself, and they identify it as something that has only one dimension; they can show that two lines combine to form two-dimensional planes and three lines combine to form three-dimensional solids. In so doing, mathematicians may seem to treat lines, planes, and geometrical solids as separate (χωριστά), that is, as if they had independent existence from body. But they are not separate in this sense, and it is a mistake if they are thought of as being separate, according to Aristotle; they exist only in body, and by studying them in isolation through abstraction in thought we learn things about body; so in nature, lines are the limits of surfaces and surfaces are the limits of bodies, and they only occur as such. Indeed, it is furthering our understanding of body that makes treating them in the abstract legitimate, according to Aristotle. Dimension understood as a property of magnitude that is the object of study of mathematics allows no room for directionality. Nor are the three dimensions composing magnitude thus conceived distinct in any other way than numerically. One may think of them as merely lines placed according to a rule, and what line is placed where makes no difference. This is not what dimension is as Aristotle conceives of it. For if, as we have seen, there is no magnitude without body, it seems plausible to think that dimension pertains to magnitude because it pertains to body. Being a constituent of substance, however, body is ontologically posterior 7

The term “solid” (στερεόν) applies specifically to mathematical three-dimensional objects, but “body” can be used to refer to both the sensible bodies that comprise the sensible world around us and to the three-dimensional solids studied by the mathematicians. Body of the latter sort is also referred to as intelligible body.



4 De incessu animalium 4



to substance, which is prior, and that implies that the body of a particular substance is the body it is because of the substance of which it is a constituent. What these considerations suggest is that a proper study of dimension must look for it as it manifests itself in body as a constituent of substance.

Up and Down IA 4 begins as follows: The dimensions by which animals are naturally determined are six in number: the up and the down, the front and the back, and also the right and the left. Furthermore, all living beings have the up and the down, for the up and the down are found not only in animals but also in plants. Now, this distinction is one of function, and not merely of position relative to the earth and the heavens. (IA 4, 705a26–31)

We need to pay attention to three claims made in this passage. The first is the straightforward claim that there are six dimensions: the up and the down, the front and the back, and the right and the left. They are spoken of as constituting three pairs. This suggests that each one of the three dimensions of the DC is mapped onto a pair in the IA. In this section, I concentrate on the first pair, the up and the down, hoping to get a better sense of how one of the three dimensions mentioned in the DC maps onto one of the three pairs offered in the IA. The second claim is that there is an up and a down that somehow belong to all living things. It will be the task of later sections to clarify the sense of belonging. The third claim is that we are to distinguish between two ways in which up and down are to be spoken of. On the one hand, we find an up and a down that belong to living things and they are said to be present according to function. On the other, we have an up and a down with reference to position relative to the earth and the heavens. It is the latter that will occupy us in this section, that is, the positional account of up and down, as I am going to call it. In the DC, Aristotle observes that we may, and normally do, identify a side of something, for instance a stone, as being the up and the opposite side of that as the down. He then points out that if we should turn the stone upside down, we could still be identifying one of its two sides as being the up and the other as being the down of the stone. However, the sides we would now be thus identifying would not be the same as those we identified before turning the stone upside down.8 Thus, talking about 8

DC II 2, 285a1–11.



Part III Interpretative Essays

the sides of the stone before turning it, we appear to be using the terms “up” and “down” as determinations of the stone itself. Indeed, these terms pick out two of its sides, but turning the stone makes it clear that the sides the terms identify can vary and so these terms cannot be getting their sense from features that belong to the stone itself. It appears that the sides of the stone are distinguished by virtue of the fact that they are facing different regions, and “up” and “down” succeed in identifying sides of the stone by bringing out this fact. But what exactly is the fact that they bring out? Addressing the issues that relate to this question in DC I, 2 Aristotle offers some interesting remarks on the so-called simple bodies. The simple bodies, he says, are “those that have the principle of motion in accordance with nature.”9 Depending on how they move they can be said to be heavy or light. To be heavy is to have a nature such as to be moving toward the center; correspondingly, to be light is to have a nature such as to be moving away from the center.10 The center is the earth, and the previous chapter of the DC has given names to what is here spoken of as “toward the center” and “away from the center”: “I call ‘up’ (ἄνω) the away from the center, ‘down’ (κάτω) the one toward the center.”11 “Up” and “down” designate directions, and so we may render ἄνω as “upward” and κάτω as “downward.” Due to their being light, then, fire and air move upward (with fire being lighter than air), because that is what it is to be light, whereas earth and water move downward because they are heavy (with earth being heavier than water), and that is what it is to be heavy. But it is not the motion of the simple bodies that furnishes the terms “upward” and “downward” with sense; rather, ἄνω and κάτω refer to objectively specifiable places in the universe, and it is because they move toward these places that the simple bodies are said to be moving ἄνω or κάτω. “Down” refers to the earthly sphere and “up” to the outermost firmament of the superlunary region that encircles the universe. The direction from the earth to the firmament is upward, whereas downward is the direction from the firmament to the earth. Any point anywhere on the earthly sphere is down and the direction toward it is downward, whereas any point on the outermost firmament is up and the direction toward it (and away from the down) is upward. According to 9

DC I 2, 268b28. The simple bodies are the so-called elements: earth, water, air, and fire. In nature, they mostly occur as variously mixed with one another. The kind of motion displayed by the mixtures is directly proportional to the ratio of their mixtures. 10 DC I 3, 269b23–24. 11 DC I 2, 268b21–22.



4 De incessu animalium 4



Aristotle, the sense of “up” and “down” is objectively grounded on positionally stable regions of the universe that determine objectively what it is to be up and what it is to be down. Aristotle says that ἄνω and κάτω “constitute a difference, indeed a contrary with respect to place.”12 In addition to designating direction, then, ἄνω and κάτω designate the places to which the upward and downward directions respectively point. Therefore, the terms can refer both to a place and the direction toward that place, and so “up” and “down” can have the sense of both a place and a direction. In the DC, “up” and “down” refer to regions of the universe as well as to the directions of motion toward these regions. The IA names both the up and the down dimensions, and the DC gives us a hint of how the statements from these two works are related. It also offers an indication of how to approach the text of the IA. We have seen that, at the outset of the IA, Aristotle groups up and down, right and left, and front and back in pairs. Now, early on in DC I, we read that body, as we encounter it in nature, is perfect and that to be perfect in the sense intended is to have the three dimensions – namely, length, breadth, and depth.13 In DC II, Aristotle calls the three pairs mentioned above dimensions of the perfect body. He goes on to say that each pair comprises the principles of the dimensions considered as being three in number, namely, length, breadth, and depth.14 As set of principles, each pair is constitutive of the dimension of which it is the set of principles. So, to speak of the dimension concerning us presently, length is the intelligible line connecting up and down, and up and down are constitutive of the dimension of length in this sense and of nothing else.15 Anywhere on the earthly sphere, length so understood is the same. We may say that it is the intelligible perpendicular line on a tangential at any point of the earthly sphere that extends all the way to the outer firmament. On this construal, “down” refers to the point at which the perpendicular line meets the tangential and “up” to the point where it meets the outer firmament. But “up” and “down” may also refer to the upward or downward direction along length. Consequently, “length” does not name just any length in any direction, but names the line linking those two regions specified 12

DC I 4, 271a5–6. DC I 1, 268a23–24. DC II 2, 284b21–26. 15 The connection of “length” with “up and down” may seem counterintuitive: the term used by Aristotle is μῆκος, which can also be translated “height.” There is no English word that exactly corresponds. 13

14



Part III Interpretative Essays

by up and down, and in the way we explained previously. We are going to see in later sections that we have similar accounts for the other dimensions. As importantly, speaking about the dimensions sometimes as three and sometimes as six is not a sign of being of two minds about their number, but merely a consequence of the fact that directional orientation is a constitutive part of the notion of dimension, as Aristotle conceives of it. Though there is only one line that is length, there are two different directions in which it can be traversed. Insofar as dimension is indicative of direction, it is equally permissible to speak of length as well as of up and down as dimensions.

Two Accounts of Up and Down So far we have seen that Aristotle uses up and down in terms of position by reference to (1) objectively specifiable positions in a finite universe, or (2) directional orientations, where up is looking toward the heavens and down toward the earth.16 We may call this “the positional account” of the up and the down, and the up and the down thus accounted for the positional up and down – referring to the positions themselves or to directions toward them. If one now expects objectively specifiable points of reference similar to those constituting length to constitute breadth and depth, one will be disappointed. Unlike up and down, there do not seem to be points in the Aristotelian universe to which front and back, and right and left, refer. Still, we can speak of the right and the left, the front and the back, of a stone in just the way we spoke previously of its up and down. As importantly, we do in fact seem to have a firm sense of what front, back, right, and left are. At the same time, however, when attributing these orientations to lifeless things, we do so by making ourselves our point of departure. Furthermore, we often use ourselves as points of reference when speaking of up and down.17 And we speak of up and down when we use ourselves in ways that do not agree with the positional up and down. For instance, we say that to undergo what is known as inversion therapy, human beings must hang upside down. Making sense of the phrase “upside down,” as we in fact do in this instance, presupposes quite precisely that we assume there is an up and a down that properly belongs to the person undergoing the therapy and that in this particular instance the 16

See DC I 2, 268b21–22 (“I call ‘up’ (ἄνω) the away from the center, ‘down’ (κάτω) the one toward the center”), where “center” (μέσον) refers to the position occupied by the earth. 17 Which is what Aristotle in fact does at DC II 2, 285a11–12.



4 De incessu animalium 4



person’s up is not aligned – or rather is at odds – with the up and down in relation to the earth and the heavens. It seems then that this person, and in just the same way all of us, have an up that belongs to us, and that may on occasion be pointing in the direction of the positional down, and we have similarly a down that may be pointing in the direction of the positional up. So, the dimension of length affords an alternative account. As the case of inversion therapy suggests, references for the terms “up” and “down” can be provided independently of the positional specifications that we saw previously that these terms afford. However, the way in which length is determined remains structurally the same. It is the line whose limits are the up and the down. Finally, we should note that though the referents of “up” and “down” that constitute length may vary, the way in which they are specified is neither arbitrary nor relational. On the contrary, it is grounded in objective facts of nature. We know the way in which this is so in the case of the positional account of length: it is the intelligible line having as limiting points the up and down considered as places (τόποι). When we later look at the alternative account of length that the above paragraphs have given us grounds to expect, it will be confirmed beyond doubt that the same is valid with respect to the two other references of “up” and “down.” As importantly, while ascertaining the set of objective facts on which the alternative account of length is grounded, it is reasonable to entertain the expectation that these are facts of a type that may facilitate the provision of directional orientations that also ground breadth and depth.

Principle(s) of Dimensions The above implies that we can speak of length in two distinct ways, even though the line that is the length of a human being in an upright position is not significantly different than the line that is the positional length. Mere reference to the dimension “length” is blind to the fact that there are different ways in which the terms “up” and “down” single out their referent. This is one part of the motivation for Aristotle’s original remark that up and down are principles of length – and even more so for the pairs left and right, and front and back, of the dimensions of breadth and depth that, unlike length, have no positional account. To have a line we need two limiting points. If the positions of the limiting points are specified, so is the line that extends between them and its position. The limiting points define the line extending between them and are as such



Part III Interpretative Essays

constitutive of, and ontologically prior to, it. It is in this respect that both the up and the down are the principles of length. But also, and here is the other part of Aristotle’s motivation, insofar as directionality is a feature of dimension on his account, the two limits defining a specific dimension also constitute the reference points necessary for specifying either of the two possible directions along the dimension in question. Immediately after calling the up and the down principles, Aristotle goes on to say that the up is the principle – that is, the ἀρχή – evidently implying that length has only one principle. Aristotle implies the same about breadth and depth.18 He obviously does not abandon the thought that up and down are the two limiting points by notionally connecting which we obtain length. Aristotle may well, however, have been thinking as follows: If a condition obtains such that knowing the position of one limiting point is sufficient for inferring the position of the second limiting point to which the line starting from the first extends, then only one of the two points is the principle of the line in question strictly speaking, namely that which is the basis for the inference. Indeed, Aristotle holds contraries to be such that if one is demarcated so is the other, and the limiting points of a straight line are contraries.19 If so, knowing that one contrary is the up, it is easy to infer that the other is the down. Correspondingly, knowing that one of the contraries is the down we may infer that the other is the up. If it were to be the case that one of the two enjoys priority over the other in some sense, ontological economy would push for declaring this the principle (ἀρχή). Does either of these two contraries enjoy such priority? A principle is fundamental with respect to the being of which it is the principle. It is that without which the thing in question would not be what it is and, therefore, would not be. With respect to the being of anything that is, then, it is reasonable that that to which this thing owes its being is honorable and noble. A principle, Aristotle says, is honorable or noble (τίμιον).20 Now, in terms of nobility, of the two places to which “up” 18

As we are about to see, Aristotle goes on to say, explicitly, that the right is the principle of breadth and that the front is the principle of depth. So, each one of the three dimensions has one principle (DC II 2, 284b24–25). 19 Cf. DC I 6, 273a9–10: “if one of the contraries is defined, the other will be defined as well.” Since dimension is conceived as a line, either one of its two ends will be the contrary of the other. 20 IA 5, 706b12–13. Jim Lennox (in Lennox 2009: 191) correctly observes that the passage from DC II 2, 285a11–12 does not have the definite article before principle (ἀρχή). But I am unsure whether we are to take this, as he suggests, as a sign that “up,” “front,” and “right” are not meant to name the principles of length, depth, and breadth but merely one of the two principles of each. First, the absence of the definite article in Greek does not imply that it is not assumed. Second, and more importantly, even though the principles of each dimension are rightly said to be two, it is also the



4 De incessu animalium 4



and “down” respectively refer, the former seems superior – if for no other reason, because it shares in the nobility of the place to which it refers. The up is fixed at the outer firmament, which is subject to the most perfect of those motions that a perceptible entity performs, namely circular motion. The center, on the other hand, i.e., the place of the down, is immobile and for this reason base.21 These are sufficient reasons for declaring one of the two defining limiting points of the line that the dimension length is in its positional specification as the principle, and to declare the up to be the one in question. As we proceed we will identify principles for the functional account of length, as well as the dimensions of breadth and depth.

Dimensions and Movement At the beginning of DC II 2, Aristotle criticizes the Pythagoreans for claiming that the universe – “the body of the whole” as he calls it – has a right and a left only.22 They are correct in thinking that it has a right and a left, but they are wrong in claiming that it has only that distinction. More precisely, they fail to recognize that the universe, being the kind of thing that it is, is also subject to other principles, and that these other principles are of the same kind as the left and the right. The other principles in question are the up and the down. On Aristotle’s account, if something has the distinction right and left, it also has the distinction up and down. Front and back are not mentioned but, as we will see in due time, just as up and down are implied by right and left, so are right and left – and up and down – implied by front and back. For further elaboration on this topic, Aristotle refers his reader to the work on the movement of animals because, according to him, an account of dimensions is the task of biology. However cursorily, in the DC Aristotle states the main results reached in his study of the movement of animals, namely that in some animals we find all three dimensions, in other animals some of them, and in plants we find only the up and the down. I will return to the distribution of the dimensions in due course. Here, I only note that Aristotle’s warrant for the claim that having a left case that one item of each pair is more of a principle, and should for that reason be considered the principle. For, immediately after making the point in IA 5 that the principle is noble, Aristotle goes on to say that the up is nobler than the down, the front is nobler than the back, and the right is nobler than the left (706b13–14). 21 One sufficient reason is that the circular motion to which heaven is subject is eternal, whereas the center is the place at which the bodies that naturally seek it come to rest. See DC II 2, 286a15–18; II 4, 288a22–26. 22 DC II 2, 284b6–10.

Part III Interpretative Essays



and a right implies that the universe has also an up and a down is that the only account that applies to the universe as regards dimension is the one offered in the work on the movement of animals. The implication is that, for Aristotle, the universe has an up and a down and a left and a right because it is alive. So even though on the positional account up and down refer to places (τόποι), these places are dimensions of the universe and the universe has them as such because it is alive. In the DC, Aristotle states something that he does not explicitly say in the IA: And that is why it is not in every body (ἐν ἅπαντι σώματι) that we ought to search for the up and the down, the left and the right, and the front and the back, but only in those that have the principle of motion in themselves because they are ensouled. For nowhere in the inanimate things do we see the whence of the starting point of motion. (DC II 2, 284b30–34)

This is a sweeping claim to the effect that dimensions as conceived of by Aristotle are found only in living things. This claim is not made in the IA because there is no need for it. The IA is a work on the progression of animals, so de facto the context of the investigation is confined to ensouled beings. However, precisely because the IA does not say whether inanimate things possess dimensions, it is necessary that it be made clear that, on Aristotle’s account, they do not. Furthermore, Aristotle is careful to specify that dimensions are found in a body that moves on account of its being ensouled. He specifies that self-motion is distinguished from other kinds of motion in which a body may be involved. So, inanimate things can be in motion because other things move them. But if the universe were to contain only things that need other things to move them, there would be no locomotion in it. This is the reason why simple bodies do not possess dimensions, even though they move on their own account. True, the four simple bodies in the sublunary region move upward or downward – and the heavenly bodies in a circle. More directly, two of the simple bodies move upward and two downward, and this is due to their seeking their natural place of rest as light and heavy things respectively. Given that in the sublunary region we find compounds of simple bodies, their motion is the natural motion of the simple body that is the predominant one in the compound.23 But the motion of compounds too is either upward or downward, for they are compounds of simple bodies and the motions they can 23

DC I 2, 268b28–269a2.



4 De incessu animalium 4



engage in are either of these two. Any other kind of motion in which they may be involved is caused by things other than them. Consequently, if the self-movers were ever to come to a rest, the simple bodies would move toward their respective natural places, eventually to come to rest there, never to move again, according to Aristotle. For, when he says that we do not see the starting point of motion of the inanimate things, he means that once they come to rest in their own natural place, they cannot set themselves in motion by themselves. It should not be overlooked how strong the claim Aristotle has made is. An inanimate body – or, for that matter, a magnitude – considered by itself does not have dimensions. Clearly, Aristotle’s conception of dimension is radically different from ours. And it is supported with the argument that has already proven useful to us: we speak of the right or left of inanimate things – as we also do, we may add, of their front and back, which too are dimensions – but do so because they correspond with, or lie opposite to, our own right and left. Our attributions of dimensions to these inanimate things are derivative. In other words, just being a body and having magnitude does not make anything an owner of dimensions as conceived by Aristotle. Dimensions can be attributed to inanimate objects derivatively. Properly they are attributed only to those entities on account of which the derivative attributions are made. The latter entities are ensouled beings, for they have the principle of motion in themselves. This confirms my previous claim that the facts grounding the account of dimensions concern natural substances. It is convenient, and useful with respect to the ordo cognoscendi, to speak of one-dimensional lines, or two-dimensional planes as do the mathematicians. With respect to the ordo essendi, however, ensouled beings are the place to look for dimensions. There would be no dimensions in the universe, according to Aristotle, if there were no ensouled body in it. Nor would the universe itself have dimensions if it were not itself ensouled. If one wonders what these entities have that makes them proper owners of dimensions, Aristotle drops a hint when he says that the up and the down, the left and the right, the front and the back are starting-points of the motions of the entities that have these motions.24 By “these motions,” Aristotle means all kinds of locomotion of which ensouled beings are capable.25 These include the simple motions that comprise the rectilinear 24

DC II 2, 285a27–29. DC I 2, 268b17–19.

25



Part III Interpretative Essays

motion of the simple bodies and the circular motion of the heavenly ones.26 But the kind of motion Aristotle has in mind is that which ensouled beings alone are capable of. For the beings having this motion he speaks of at this juncture are those that have the principle of motion in themselves on account of being ensouled and he includes heaven among them.27 The entities to which he refers as having these motions, then, are those that can engage in all the various types of motion by themselves and not by being forced to do so by something else. I will carry on referring to them also as self-movers, by which I mean ensouled beings, that is, the natural substances.28 They are the ultimate causes of motion from one place to another in the sublunary region. Now, body is one constituent of substance and, we have seen, does not have the power to move itself. This provides an inferential path to the well-known Aristotelian thesis that the movement in which substances engage by themselves is due to their soul. Since they move and their motion is not due to their body, it must be due to the fact that they live, and that they do is owed to their soul. Combined with the explicit statement that dimension is the starting-point of the motions of the entities that have them, this makes the soul the locus of the facts that ground the principles of dimensions as conceived of by Aristotle, and to which our analysis must now turn.

Functional Up and Down IA 4 confirms early on that length affords a specification other than the positional. It does not explain how this is so; it simply states that it is so: Furthermore, all living beings have the up and the down, for the up and the down are found not only in animals but also in plants. Now, this distinction is one of function, and not merely of position relative to the earth and the heavens. For that from which food and growth is distributed is the up, while that to which it is distributed and in which it ends is the down. (IA 4, 705a29–b1)

We are told that there are two accounts of the up and the down. In addition, we are offered an intriguing statement to the effect that the nonpositional (οὐ θέσει) up and down is possessed not only by animals but 26

DC I 2, 268b29–269a2. DC II 2, 285a30: “the heaven is alive and has the principle of motion.” 28 On this view, which must be left unargued here, it follows that the simple bodies are not ­substances. 27



4 De incessu animalium 4



also by plants. The implication is that there are dimensions the plants do not have. Moreover, we are also told that the alternative, non-positional account is one according to function (ἔργῳ). We may follow Aristotle in calling this alternative account of length functional. The fundamental difference between ensouled beings and inanimate things is life: while the former beings are alive, the latter are not. The mention of function directs us to the part of the living thing that is responsible for the basic functions that sustain it in the kinds of life activities its nature permits. Life manifests itself in many ways, according to what Aristotle tells us in the DA. The central ones are thought, sense-perception, movement and rest with respect to place, in addition to movement involved in getting nutriment, growth, or decay. We say that something is alive if any one of these is present in it. So, for instance, plants are alive, even though only one of the functions named above, namely the one involved in getting nutriment, growing, or declining is present in them. Animals, on the other hand, have sense-perception, and some of them, we may add, have intellect and cognition.29 Now, the thinking function that the faculty of intellectual cognition supports is the most noble. That this is so, as well as the reason why, is well attested in the PA.30 The human animal that has this function is the only one whose nature (φύσις) and substance (οὐσία) is divine (θεία), the reason being that to think and to be intelligent is most divine (θειότατον). Aristotle sums up his introductory investigation into the kind of thing that the soul is by declaring it the origin or source (ἀρχή) of the faculties of nutrition, sense-perception, thought, and movement. The presence of any of these faculties and the work they support, he says, is a sure signal that we are in the presence of a soul.31 After having determined that it is in the soul that the dimensions are ultimately grounded, and having characterized their account as functional, we have to turn to the functions that define the soul. Aristotle points out that the soul-faculty on which the functional account of the up and the down is grounded is the nutritive. It is the faculty whose work is to keep the ensouled being alive by turning the nutriment into bodily tissue. Hence, the implication in the DC that the plants have this dimension but not the others. The reason is that their soul is only capable of the nutritive function. On this account, then, not all ensouled beings have all 29

DA II 2, 413a21–b2. PA IV 10, 686a27–30. 31 DA II 2, 413b11–13. The verb used here is ὥρισται, but I take it to mean not that the functions in question define the soul, but merely that they mark the presence of one. 30



Part III Interpretative Essays

dimensions. Though the nutritive function ensures that all ensouled beings have the dimension of length, the presence of the other dimensions depends on whether the ensouled being engages in locomotion in order to gain access to its nutriment. Plants get the nutriment they need directly from the soil. As a result, they do not engage in locomotion.32 Lack of locomotion and possession of only one soul-function make it the case that trees for instance have only an up and a down, even though they have a body that stretches in more directions than up and down. This is further confirmation of the claim that dimension is not a property of magnitude, though it can be attributed to magnitude derivatively. The specifications for determining the two directional orientations constituting length in its functional account seem clear and intuitively plausible. They are: (1) “that from which food and growth is distributed” (IA 4, 705a32–33), and (2) “that to which it is distributed and in which it ends” (IA 4, 705a33–b1). The former is the up, whereas the latter is the down. It seems plausible to hold, as does Aristotle, that taking in nutriment is prior compared to its distribution to the body and to growing, as the latter depends on the former. The former enjoys functional primacy. Moreover, with the up defined in this way, it is easy to determine the other direction toward which the dimension of length stretches: it is the direction to which the nutriment that is taken in is distributed. So, in addition to its functional primacy, the up is also epistemologically potent: knowing it makes it possible to know its opposite. Therefore, the up is the principle of length. According to Aristotle, all ensouled beings grow in the direction to which the nutriment they take in is distributed to their body. The direction in which they grow too, then, is opposite to the one in which they take in nutriment. This is obviously so in the case of plants.33 They grow in the direction of the heavens, which is their down, and take their nutriment from earth. This may not be as obvious in the case of some animals. For it might seem that the direction in which two-footed and four-footed animals grow is the same as that from which they take in nutriment. But this need not be a problem for Aristotle’s position. For Aristotle could, if pressed, point to the fact that in the case of two-footed and four-footed 32

By “locomotion” I will henceforth be referring to motion from place to place other than growth. In DA II 2 we read that plants grow both up and down, where “up” and “down” are used in their positional sense. This need not be at odds with Aristotle’s position here. Though the plants’ roots do grow down, up is the direction in which plants grow most and they grow upward because doing so is their telos. The reason their roots are extending downward, on the other hand, is to facilitate the plant’s intake of nutriment, which serves the plants’ telos.

33



4 De incessu animalium 4



animals it is the part of the body other than the head that really grows – indeed the head grows very little compared with the rest of the body. It does so happen, however, that these ensouled beings support themselves on the surface of the earth and this is the reason that when they grow they appear to be stretching in the direction from which they take in nutriment. Considered from the point of view of the positional account, the down is where the plants take their food from, whereas the up is the direction in which they distribute it and grow. Aristotle does not quite explain why he calls the part from which they take in nutriment the up. Why not follow the positional account and call it down, and call the direction in which they grow up? Had he done that, the principle of length in plants would be the down. We may think of several reasons why such an approach would not appeal to Aristotle, and in particular why he could not accept that the principle of length in plants is the down. The up is the principle in the positional account, and it makes sense that the principle of both accounts of length, the positional and the functional, should be the same. Moreover, to be credible, the functional account of length must be grounded in a principle that makes a valid claim for all ensouled beings. Taking in nutriment does precisely that, for Aristotle, and it is reasonable that the principle for the functional length should carry the same name for all cases. Finally, making the positional principle of length also the functional principle of that dimension for plants would undermine the internal coherence of his account. Nothing in the universe can serve as specification for the orientations constituting the dimensions breadth and depth. Letting the positional account determine what is up in the case of plants, when it is unable to provide orientation points for determining the dimensions breadth and depth, would be ad hoc. It is equally important to note that the way in which the functional account conflicts with the positional account provides a neat way of confirming Aristotle’s ranking of the ensouled beings inhabiting the sublunary region in terms of nobility. As already noted, what is principle is also noble, more so than what is not, and in the case of the positional account the up is the principle because the heavens are nobler than the center. Similarly, in the functional account, taking in nutriment is nobler than having it distributed in the body and growing because the latter depends on the former. It becomes clear later on in the IA that the placement of the functional up in different groups of animals relative to the positional up reflects their relative nobility in terms of psychic complexity. So, in IA 5, we are



Part III Interpretative Essays

told that “the two-footed animals have their upper part lined up with the upper part of the universe.”34 That means that these animals are erect, and the human being is the animal that is most properly erect. The reason for this is that the human being is most properly two-footed according to nature.35 In PA IV 10 Aristotle states that the human being is the only animal that stands erect because it is the only one that is capable of intelligence and thought.36 A few lines further down, Aristotle links a being’s capacity for thought to its having the head in the uppermost part of their body, understood positionally. Since human beings are erect, that is where they have their head. Moreover, the functional up for animals – namely the place where they take in their nourishment – is located in their head.37 Therefore humans take in nutriment from the uppermost part of the body, understood positionally. So the degree to which the up of a species inhabiting the sublunary region is aligned with that of the universe tells us something about its nobility. The animal species whose functional up is best aligned with the positional up is indeed the noblest, as its soul is the one with the greatest functional complexity. On the other hand, the species whose up is mostly at odds with the positional up is the least noble.

Functional Front and Back Aristotle says that plants only live (IA 4, 705b8). This is his way of saying that their soul has only the nutritive function. Their living is confined to taking in nutriment and turning it into bodily tissue so as to be growing. Given that their soul possesses only the nutritive function they only have the dimension defined by the opposites up and down. It is, however, possible to attribute to them front, back, right, and left, but only in the way we attribute these dimensions to inanimate objects, derivatively, relative 34

IA 5, 706b4–5. IA 5, 706b10–11. Aristotle refers also to birds as being two-footed animals but for reasons he explains later on in IA he does not consider them to be two-footed, properly speaking. In IA 12, Aristotle points out that birds bend their legs in opposite direction to that of humans (711a11–13). In IA 15, he says that birds bend them in the way the four-footed animals bend theirs (712b22). He goes on to say, quite explicitly, that the nature of the legs of the birds is very closely similar to that of the four-footed animals and that their wings are there instead of front legs (712b23–24). A twofooted animal properly speaking stands erect (ὀρθόν). The term ὀρθόν in geometry means standing at right angles to the plane. (Cf. Euclides, 11, Def. 3: ὀρθή ἐστιν ὅταν […]). Two-footed animals are erect (ὀρθά) in that they stand at right angles to the earth, and only humans do that according to PA IV 10. 36 PA IV 10, 686a27–30. 37 PA IV 10, 686a13–14. 35



4 De incessu animalium 4



to something other than themselves and most often relative to us. The plants’ simplicity in terms of psychic function is doubly signaled in the account of dimension: (1) their up is the positional down, and (2) they have no right and left, front or back, that properly belong to them. The reason why plants do not have their own front and back, Aristotle says, is that they lack sense-perception. Animals, by which Aristotle means ensouled beings whose soul possesses the perceptive function in addition to the nutritive one, also have the dimension of depth, defined by the opposites front and back. Front and back are grounded in senseperception, and since to be an animal is to have sense-perception, all animals have a front and a back. Sense-perception is also that which determines what is their front and their back.38 As is the case with up and down, front and back have two senses when designating dimensions. Each can be used to refer to a place or to indicate a direction. Hence, Aristotle says that the animal’s front is where its sensory apparatus is located and is also the whence (ὅθεν) from which the creature derives its sensory perception. The back is the part of the body that is, and points to, the direction opposite to the front. There are several things to notice in this account. First, speaking quite generally about the placement of the sense-organs and the whence of sense impressions, as Aristotle does, leaves it wide open whether we are supposed to include touch as a determinant of the front and the back of the animal. Touch is the sense that has flesh as its organ. Humans have the sense of touch wherever they have flesh. It would be difficult, not to say impossible, to determine the front and the back in the way Aristotle instructs us to on the basis of touch alone. Presumably, the sense Aristotle has in mind when he speaks of the point of entry of sense-impressions is sight. And he might be focusing on sight because he considers it the chief sense in important respects. We know that he does so for humans elsewhere.39 He may be treating them as the paradigm case which gives him license to do so in the case of four-footed and lesser two-footed animals as well. On such an interpretation, the front and the back are easily and plausibly determinable for all two-footed and four-footed animals. Aristotle points out that there are animals capable of sense-perception that are not able to engage in locomotion. What he has in mind are most likely sessile animals, concerning which he says in HA I 1 that they are 38

IA 4, 705b11. So, in the opening lines of the Metaphysics Aristotle says that, of all sensory input, the one entering through the eyes is the most liked by humans, who prefer seeing most of all because it enables them to know many things and discern many distinctions (Meta. I 1, 980a23–27).

39



Part III Interpretative Essays

stationary (μόνιμα) and are found not on land but only in water. They live by growing on other things. The example Aristotle gives of a stationary animal with (some) sense-perception is the sponge.40 Given Aristotle’s account of dimension, the sense-perception that sponges possess commits him to attribute depth to them. However, his way of determining the front and the back seems hardly to apply in their case. Whatever sense these animals may be said to have, they have it everywhere they have their body, and it is most likely the only sense they have. It is, however, reasonable to suspect that Aristotle’s interest is not in sessile animals but rather in accounting for the constitution of the dimension of depth as found in higher forms of life. If so, it seems reasonable to proceed by focusing on the relevant psychic function as this is manifested in these forms of life, and in particular on the paradigmatic case, which is the life of the human being. On this basis, he can claim, quite plausibly, that the principle of this dimension is the front and the back is its opposite. It would then be hardly contestable that the front, as Aristotle defines it, is primordial compared to the back and hence the principle of depth. The fact that this account does not seem perfectly to suit sessile animals is probably more a sign that these animals represent for Aristotle instances of a lowly, and as such less orderly, form of life than that the account is problematic.

Functional Right and Left Further evidence that the animals capable of locomotion represent higher forms of life is that the dimensions that Aristotle calls right and left can be attributed to these animals as their functional parts (IA 4, 705b13–18). These animals alone have all three dimensions. Since the determinations left and right make their appearance in animals capable of locomotion, we should expect locomotion to be that on which these determinations – and therefore also breadth, which they define – depend.41 Indeed, Aristotle goes on to claim that the right is the part of the body from which any locomotive change undertaken by the animal begins (IA 4, 705b18–20). By this, he means not that the right is the side that changes location first, but rather that it is the one that propels the animal for relocation to ensue. Being the locus of the beginning of locomotion, the right is the principle of breadth. The left is the side 40

HA I 1, 487b9–10. As Aristotle explicitly states in DC II 2, 284b32–33.

41



4 De incessu animalium 4



opposite to the right and the one that the animal engaging in locomotion relocates. From a standstill position, and propelled as it is from the right, for locomotion to commence the left side is the first to change place. Hence, the motion originates in the side that stays put, for that is the one on which the animal supports itself to produce the impetus that pushes the left side. It is most likely humans that Aristotle has in mind while offering his account, even though the account is meant to be general and valid for all animals capable of engaging in locomotion. Right and left are the end-points of the intelligible line that is at right angles with the intelligible line whose end-points are front and back. In fact, once one can imagine the three planes constructed by length, depth, and breadth standing at right angles to each other, one has also the geometrical notion of magnitude’s three dimensions, founded in the sensible world as Aristotle conceives of it. All animals that are capable of engaging in locomotion have right and left, but in those having instrumental parts such as legs or wings for moving from place to place – and in particular progressing, which is the kind of locomotion Aristotle is interested in here – the distinction between the right and the left is more pronounced so that right and left are more readily identifiable (IA 4, 705b21–25). The animals in which this distinction is less pronounced are those without feet, such as snakes (IA 4, 705b25–29). However, insofar as these footless animals engage in locomotion of which they are the source, they too have a functionally determined right and left. In the case of these animals, too, it is the right side of the body that propels the left to relocate in order for locomotion to commence. Still, Aristotle’s sweeping claim that right is the same in all animals capable of initiating their own locomotion comes at a cost. It is far more general than the empirical evidence can support. For Aristotle’s claim, it ought to be clear, is not that right is the side from which locomotion originates, whichever that side is. His claim is straightforwardly, and clearly, the general one that right is the same in all animals and that their locomotion originates from that side (IA 4, 706a10–12). It is admittedly hard to test empirically whether this account holds for footless animals. It is as hard to have a firm opinion as to how much empirical evidence Aristotle may have used to support his claim.42 One may suspect that he considered the validity of the account with respect to 42

Nothing, however, precludes the possibility that students or collaborators of Aristotle amassed empirical evidence on his behalf. A legitimate question that may be raised in this case is how firmly such evidence corroborated his conclusions.



Part III Interpretative Essays

footed animals to be a reason for extending this account to footless animals as well. On this basis, he may have come to the view that the account is applicable to the latter without subjecting it to rigorous testing. He could support such a move with the consideration that if the account is correct for footed animals – where the two directional orientations defining breadth are clearly distinguishable due to their having limbs – it is as good as certain that it is also correct for footless animals.43 A more serious problem for this account is that there are two-footed animals that are left-sided. We know that there are left-sided people, and based on Aristotle’s analysis of how the animal sets itself in motion, these would actually be putting the right foot forward, supporting themselves on their left foot. Still, Aristotle holds that all animals have their right on the same side. To be sure, he does also say that people might occasionally do the opposite than they would on his account be expected to do, namely they do on occasion step off on the left foot to move the right forward from a standstill position, but if so this is done by chance or luck, as he says. 44 However, for Aristotle, doing something by luck implies that people do not do it regularly, but only on the odd occasion and for no reason. Left-sided people, by contrast, do things the way they do them regularly and could therefore be said to present a counterexample to Aristotle’s account. Aristotle is bound to have been aware that readers might feel uneasy with the generality of his claim. This is certainly a reason why he meticulously seeks to support it with all kinds of evidence, such as how men usually jump, or how they defend themselves in battle. In support of his claim, he also observes that when men carry burdens, they do so on their left side so as to set the right, which is the origin of motion, free. He generalizes the claim to point out that the burden of the body must not rest on the side where movement is initiated, and finds evidence for this claim among footless animals. Now, he also says that stromboid testaceans have their shells on the right (IA 4, 706a13–16). Based on his analysis one would expect that these animals would constitute a counterexample. However, these animals are said to be making the f­orward move (προέρχονται) on the opposite side (καταντικρύ) of the spiral, so that they are setting forward the left side when setting themselves in motion. Given his account of locomotion, this implies that they propel themselves 43

As the following chapters also testify, it is indeed a usual procedure of analysis in the IA to start with the higher animals as being the paradigm case and use this as a basis for claims concerning lesser animals. For more on this explanatory strategy I refer the reader to the introduction (ch. 1). 44 IA 4, 706a6–8.



4 De incessu animalium 4



using the right side. If anything, stromboid testaceans constitute a strong confirmation of Aristotle’s account of motion, for they make their forward move on the left side – and therefore must propel themselves by the right, even though they have the load on that same side.45 They, too, have a right and a left, and the same right and left as do all animals. What Aristotle may consider sufficient basis for making such a sweeping claim is that the absolute vast majority of people are right-sided, and it is indeed remarkable that so many people are, no matter their background or ethnicity. I suggest that, for him, this is evidence that firmly backs his claim that the right side is the whence of locomotion in animals that are capable of it. He could even add that given how widespread right-sidedness is among people, the few individuals that are left-sided came to be so by accident or luck, and they should not therefore be considered a counterexample to his claim. 45

IA 4, 706a14–16. Exactly the same point is made in HA IV 4, 528b8–10.

chapter 5

De incessu animalium 5–6 The Architecture of Locomotive Bodies Klaus Corcilius Introduction Like IA 2, IA 3 and 4 are still preparatory for the main task of the treatise, which is to provide scientifically adequate answers to the set of questions outlined in IA 1. As Aristotle makes clear at the outset of his treatise, they are questions that concern the workings of the parts of the body useful for animal locomotion (IA 1, 704a4–9). We will see that IA 5–6 prepare the ground for the functional account of the locomotive parts by providing an account of the basic bodily design or structure of locomotive animals insofar as they are locomotive. Roughly, this structural design consists in an articulation of parts of the body, or points of motion, according to four directions – namely, left/right and up/down – and a common origin of these locomotive parts located at the center of the animal’s body.1 In what follows I will refer to this structure as the architecture of locomotive bodies. The reason why I chose this term is that Aristotle thinks of the structural design of locomotive bodies in teleological terms. Nature designs things according to what is best and this general purposefulness of natural things applies also to the case of the natural design of locomotive bodies. Hence, the structural design of locomotive bodies is not just a structure that happens to exist but is a structure that exists for the sake of some goal. It is, in other words, an architecture.2 Once Aristotle has given his account of the architecture of locomotive bodies, he can proceed to give his functional account of the workings of the animal parts useful for the different kinds of animal progression in IA I would like to thank all the participants of the conference held in Patras for their comments and suggestions. I would also like to thank Andrea Falcon and Stasinos Stavrianeas for their patience and work as editors, and Justin Vlasits and Yue Lu for copy-editing my essay. I owe special thanks to Pavel Gregoric for his written comments on a draft of this essay. All translations are mine unless otherwise noted. 1 Aristotle discards the other two directions of the dimensional framework along the axis front/back as not in the same way relevant for animal locomotion on the ground that there is no natural backward motion (for more on this see below). 2 See PA I 1, 639b14ff. and 3, 644a7–11. I would like to thank Pavel Gregoric for reminding me of this important point.





Part III Interpretative Essays

7–19. Furthermore, as we will see, IA 5 also prepares the way for an answer to the third question on the initial agenda – why, in general, are some animals footless, some two-footed, some four-footed, and some many-footed? – by means of an explanation of the different locomotive equipment different kinds of animals possess in terms of the natural capacities they are endowed with. It is characteristic of Aristotle that in IA 5–6 he seeks to establish the basis for the following functional account of animal parts in a principled and proof-oriented way: in spite of the sometimes less than clear language he uses, Aristotle offers proofs for both the bodily articulation of locomotive animals into four parts and for their centralized functional unity in the locomotive body. Indeed, in IA 5, he offers grounds even for why certain animals are bipedal and others are not. A further remarkable characteristic seen in the IA, which is particularly salient in IA 5–6, is its unusually high level of abstraction. Aristotle is aware that locomotion is a common feature shared by many animals across biological genera such as fish, land-dwellers, and birds. But he is also aware that different animals differ greatly with regard to the particular ways in which each of them brings their locomotion about. Yet due to his general methodological commitment to “commensurately universal” explanations – namely, explanations that are as general as possible and as specific as necessary to capture the maximal extension of any given phenomenon3 – Aristotle prefers, even in this extreme case of variability, to give a common scientific treatment for all the different kinds of locomotive parts in the animal realm in one single investigation. So, on the one hand, the IA’s treatment will have to be highly abstract – it will investigate the causes of all the locomotive parts animals possess across genera – while, on the other hand, it will have to go into a considerable level of specific detail, as it will have to offer explanations for each of the different ways in which the different kinds of locomotive parts work: the workings of wings, fins, legs, and so on will each require a separate functional explanation. From a methodological point of view, then, the IA is special. Its transgeneric causal investigation of locomotive parts operates on an extremely high level of zoological abstraction, yet at the This “commensurately universal” (πρῶτον καθόλου) mode of procedure not only serves the goal of methodological economy (minimizing explanatory work and avoiding repetitions, PA I 1, 639a15– b5, and I 4, 644a25–b15; Phys. I 7, 189b31–32; DA I 1, 402b8–10), but also, and perhaps more importantly, establishes the proper hierarchical sequence of explanations: first things – that is, more general things – first. For the general methodological idea of commensurately universal explanatory proofs, see APo I 4, 73b25–74a3; a32–b3; cf. Barnes 1993 ad loc.; McKirahan 1992: 171–176, and Kullmann 2007 ad PA I 1, 639a15–b5.

3



5 De incessu animalium 5–6



same time the functional accounts it offers for the particular forms of animal progression are specific to the different kinds of locomotion these animals engage in.4 This remarkable combination of specificity and scientific abstraction required by the common treatment of animal locomotion is the topic of a methodological discussion in the opening section of PA: Yet there are probably other attributes that turn out to have the same designation, but to differ by a difference in form, for instance the locomotion of animals; it is apparent that locomotion is not one in form, because flying, swimming, walking, and crawling differ. Accordingly, the following question about how one is to carry out an examination should not be overlooked – I mean the question of whether one should study things in common according to kind first, and then later their distinctive characteristics, or whether one should study them one by one straight away. (PA I 1, 639a25–b5, trans. J. G. Lennox, slightly modified)

Aristotle observes that there is a common designation (κατηγορία) for all animal locomotion, namely “locomotion” (“progression,” πορεία, animal self-initiated displacement by means of bodily parts specifically designed for that purpose), and that in spite of this common designation the ways in which animals locomote differ from each other in form (or species: διαφέρειν δὲ τῇ κατ’ εἶδος διαφορᾷ). In view of this fact, he raises the methodological question whether one ought to explain the generic characteristics (κατὰ γένος) shared by all animals that fall under the same designation first and only then set about explaining their specific features in turn, or one ought to forgo the generic explanation and explain the corresponding features on a case-by-case basis instead, for each animal species separately. In the PA Aristotle does not explicitly answer that question. However, the very fact that there is a treatise On the Progression of Animals (De incessu animalium) that does offer a comprehensive causal investigation of the locomotive parts of animals makes it plain that he favors the first of the two alternatives. This becomes even more evident in the beginning of his treatise On the Movement of Animals (De motu animalium), where Aristotle very briefly summarizes the results of the investigation carried out in the IA: 4

In this respect, the IA operates on a level of abstraction even higher than that of the MA. The MA offers a likewise transgeneric causal explanation (of animal self-motion); however, Aristotle emphasizes that there is a common cause (κοινὴ αἰτία) of animal self-motion that is the same for flyers, swimmer, walkers, and the like (MA 1, 698a4–7). Note that there is no such claim of a common cause for the workings of the animal parts useful for locomotion in the IA. What makes the IA special from a methodological perspective is that it offers a common investigation of the workings of locomotive parts of animals even in the absence of such a unitary common cause.



Part III Interpretative Essays In regard of the movement of animals, namely what features (ὅσα) belong to each particular kind of them, and what the differences and what the causes for each of their particular features are, we have investigated exhaustively (ἁπάντων) elsewhere. (MA 1, 698a1–4)

The passage refers to a common treatment of the locomotive features (i.e., parts) of animals that at the same time gives an exhaustive account of the particular differences and causes that these parts have. As all commentators agree, this is a reference to the IA. The IA gives a common treatment of the differences and causes of the particular bodily parts that are instrumental for the various ways in which animals engage in locomotion. So we may say that the treatise investigates the functioning of locomotive parts insofar as they serve the purpose of locomotion across genera – namely, for crawlers, walkers, swimmers, and flyers alike – and that it does so, among other things, by explaining each of the various different ways in which the different kinds of locomotive animals do this.5 What is the role of IA 5–6 within this undertaking? We will see that within the IA the task of these chapters is to provide a “commensurately universal” basis and common framework for the different functional explanations of the animal parts in different locomotive genera. The chapters thus pertain more to the highly abstract and general side of the IA. If I am right, the basis and common framework provided by IA 5–6 consists in (i) the introduction of the doctrine that there is a fourfold bodily articulation of locomotive parts or points of motion in all locomotive animals; (ii) a classification of the most basic differences among types of locomotive animals on the basis of their possession of a clear bodily articulation of these parts in them (both offered in IA 5); and (iii) a proof of the thesis that there is a common origin of these locomotive parts or points of motion located at their center (IA 6). However, I should restate here that Aristotle does not always bother to spell out the premises of the arguments he provides. This is a feature typical of the causal investigations of his biological corpus. But it is worth emphasizing it here as in what follows I will endeavor to make explicit how I understand the reasoning that guided him in writing these chapters.

5

It is not easy to identify common principles for the different explanations Aristotle invokes in his explanations of the various ways in which animals use their locomotive parts in the IA. Michael of Ephesus at one point speaks of the “geometrical necessity” with which Aristotle proves his point (γεωµετρικαῖς ἀναγκαῖς, In IA 153.11). Perhaps, the fact that the principles he invokes in his explanations are, from a zoological perspective at least, transgeneric (that is, geometrical and mechanical), this is all that there is to it. The question deserves more attention than I can give to it here.



5 De incessu animalium 5–6



The Context After the brief introduction offered in IA 1, the second chapter lays down the terms for the execution of the task. It does so by introducing the general teleological framework with the famous formulation of the explanatory principle that nature does nothing in vain but “always does the best from among the possibilities for the substance of each kind of animal” (704b15–17; this principle will do important work in IA 6), and by announcing the conceptual tools and premises that are necessary for offering an adequate explanation of the phenomena of animal locomotion. They are the “dimensional” framework – the six directions: up and down, front and back, left and right – plus a reminder of the general thesis that all self-motion is based on mechanical exertion of force, that is, on some kind of pushing and pulling (ὦσις καὶ ἕλξις).6 IA 3, then, expands on the mechanical presuppositions of the theory to follow. The first and foremost presupposition is the supporting-point principle. According to this principle, every self-motion requires an external platform sufficiently stable to offer resistance so that the animal can press against it to displace itself. Aristotle argues for that principle at length in MA 2–4. In IA, he is eager to point out how the supporting-point principle is at work even inside the animal, as for instance when it jumps or runs: all self-motion requires that one part of the animal presses against another as its point of support. This requirement, in turn, presupposes an internal differentiation in the animal so that one “organic” part of it can serve as a supporting point for another (cf. IA 10, 709b26–28). Aristotle’s point is that all self-motion requires such internal differentiation and functional complexity within the animal (ἐν αὑτῷ διάληψιν, IA 3, 705a25).7 IA 4, finally, applies the general dimensional framework of IA 3 to the parts of the animal insofar as they are relevant to self-motion. Aristotle opens the chapter by saying that animals are by nature differentiated (ὁρίζεσθαι)8 according to the six directions that also structure the dimensional character of the universe (or cosmos) as a whole – up, down, front, back, left, right – and then goes on to classify animals according to whether they possess bodily articulations that correspond to these cosmic See Phys. VII 2, 243a11–244a6 and MA 10, 703a19–20. Pushing and pulling (ὦσις καὶ ἕλξις) are the basic movements on which all self-motion depends. Cf. DA III 10, 433b25f. See Gregoric–Kuhar 2014. 7 See Phys. VII 2, 243a11–244a6, which makes the same point on a more general level (κινοῦν καὶ κινούμενον) 8 Well rendered by Jutta Kollesch. See Kollesch 1985: “äußere Begrenzung.” Cf. Lennox 2009: “bounded.” 6

Part III Interpretative Essays



directions or not. In doing so, he emphasizes that we have to identify these directions relatively for each animal to its own function – ἔργον – and not absolutely to the cosmic directions. It turns out that not all animals possess all these directional articulations;9 rather, the resulting picture is a complex hierarchy, which can be captured with the help of Table 5.1. Table 5.1  Possession of bodily articulations according to directions in living things All living things (animals and plants)

All animals

Locomotive animals

Up/Down

X

X

X

Front/Back

–

X

X

Left/Right

–

–

X

While all living things – plants and animals alike – possess an internal bodily articulation according to the up and the down, which is due to the functional differentiation necessitated by the intake of nourishment from “above,” animals alone possess an internal bodily differentiation between front and back (the front being the part where the senses are located). Finally, the locomotive animals alone possess an internal bodily articulation into a left and a right side, since, as Aristotle will argue in the immediate sequel, self-motion always takes place from right to left. There is a great deal that is interesting in IA 4. Aristotle identifies an absolute dimensional framework for the cosmos as a whole (cf. DC II 2, 284b10ff.), and then uses this framework as a standard for an axiological ordering of animated bodies. This axiological ordering of kinds of animal bodies is similar to the familiar idea of the scala naturae: the higher the degree of internal differentiation – or, perhaps better, the higher the degree of articulation of bodily parts – an animal possesses, the more sophisticated, and thus “more natural,” this animal will be. The axiological aspect comes out very clearly in the second part of IA 4. There, 9

It is noteworthy that DC II 2 introduces the discussion of the six cosmic directions with an explicit reference to the discussion offered in the IA. That, however, need not be taken as problematical. While DC II 2 argues for the fact that there are directions in the cosmos, the IA seems to be invested in a causal inquiry, the goal of which is to explain why certain animals possess certain directional differentiations in their bodies and some do not. See below. For a discussion of DC II 2 and its relation to the IA, see Lennox 2009: 187–214.



5 De incessu animalium 5–6



Aristotle claims that there is a place of origin, or a principle (ἀρχή), within each of the opposite pairs of directions of the dimensional framework. They are the up, the front, and the right. This leads him to elaborate at great length on the further claim that the right side, and not the left side, is the origin and the principle of animal locomotion (705b18– 706a24). With these distinctions in place, we can now turn to IA 5. DE INCESSU ANIMALIUM 5

IA 5 further elaborates on the classification of animals according to the dimensional framework. Aristotle begins the chapter with a further refinement of the distinction outlined in IA 4 between types of internal bodily articulation according to directions in animals and juxtaposes it with the framework of the three cosmic regions or places (up, middle, and down). These regions (τόποι, 706b3) are not to be regarded as physically separate “spheres” but as the result of an abstract division of the cosmos according to the axis “up/down.” Aristotle also states reasons for the data that emerge from that juxtaposition, i.e., for the match or mismatch of bodily articulation of animals with the regions of the cosmos. His explanation will be in terms of the natural capacities of animals.

A Classification of Locomotive Animals [IA 5, 706a26–b16] In the first section (706a26–b2), Aristotle asks whether locomotive animals exhibit their internal differentiations also in distinct parts of the body. His main concern is not answering the question whether the animals under investigation possess bodily differentiations according to the cosmic directions. The fact that all locomotive animals have functionally determined bodily differentiations articulated by means of the oppositions up/down, front/back, and left/right is the main outcome of the discussion offered in IA 4. Aristotle can safely presuppose that conclusion. Now, he turns his attention to the question whether locomotive animals exhibit those bodily articulations also in parts of the body that are set apart from one other (διώρισται). Here is how Aristotle introduces his new focus: All the animals that, like human beings and birds, have the upper part set apart from the front part are two-footed (two of the four points are wings in one group, and hands and arms in the other group), whereas all the animals that have their upper and front parts in the same place are four-footed, many-footed, and footless. I use the term “foot” for



Part III Interpretative Essays a part connected with a point on the ground capable of producing motion; for the feet (πόδες) also appear to have got their name from “ground” (πέδον). Some animals, too, have their front and back parts in the same place, for example soft-bodied animals and stromboid hard-shelled animals. These animals have been discussed earlier elsewhere. (IA 5, 706a26–b2)

This time, then, the outcome is not a classification of animated beings. Rather, it is a classification of locomotive animals. In other words, it is only the last slot of the previous classification of animated things that is the focus of the discussion offered in IA 5. Aristotle here establishes a linkage between bodily differentiation in animals – namely, whether or not they possess their bodily articulations according to the dimensional framework in the form of limbs that are clearly set apart from each other – and their particular way(s) of engaging in locomotion. The juxtaposition of the criteria of possession (or absence) of separate front and back parts and possession (or absence) of separate upper and lower parts in animal bodies yields the following threefold classification of animals insofar as they are locomotive: (i) two-footed animals; (ii) four-footed, manyfooted, and footless animals; and (iii) soft-bodied animals and stromboid hard-shelled animals (that is, hard-shelled animals with a spiral-like shell). It is not entirely clear how exactly the animals grouped under (iii) present us with a distinct mode of locomotion. But we do know that Aristotle considers some hard-shelled animals such as the mollusks to be mutilated (ἀνάπηρον) self-movers: they move in the way in which “a footed animal would do if one were to cut its legs off” (IA 19, 714b11–12). This may be taken to mean that these animals, though self-movers, move in somehow unnatural ways.10 The result is a threefold classification of locomotive animals, as in Table 5.2. There are, then, two manners of natural self-motion with respect to place. There is the progression of animals whose bodies have front and back as well as up and down located in separate places, as well as the progression of animals whose bodies have the front and back and the up and down located in the same places. These natural manners of self-progression are instantiated in two large groups of animals: (1) the two-footed 10

What Aristotle has in mind becomes clear in IA 19: maimed animals such as the mollusks are “dualizers”; they appear to be self-moving from the perspective of sessile animals, but they appear to be sessile from the perspective of self-moving creatures. Aristotle makes this point as follows: “they are not really animals able to move, but if you regard them as stationary and attached by growth, they are able to move; and if you regard them as able to engage in progression, they are stationary” (IA 19, 714b14–16). For more on this point, I refer the reader to the discussion offered by Pantelis Golitsis in this volume (ch. 11).



5 De incessu animalium 5–6



animals, and (2) all other properly speaking locomotive animals – namely, the four-footed, many-footed, and footless animals. The hardshelled animals do not seem to have any natural articulation of distinct points, yet they are observed to move with respect to place, which is why Aristotle concludes that their mode of progression is unnatural (IA 19, 714b11–14). Table 5.2  Possession of separate bodily articulations in locomotive animals

Front / back located in separate places in the body Up / down located in separate places from front / back in the body

1. Soft-bodied animals and stromboid hardshelled animals

2. Four-footed, 3. Two-footed many-footed, and animals (birds and footless animals humans)

–

X

X

–

–

X

The purpose of this whole exercise, it seems, is to generate a criteriabased distinction between different manners of self-motion that are progressions with respect to place: the locomotion of two-footed animals, which seems to involve two points of motion, and the locomotion of the other self-moving animals, which seems to involve more than two points of motion (706a27–28). With this distinction, Aristotle also prepares the way for his answer to the first question taken from the catalogue of questions introduced in the first chapter: What are the fewest points at which animals move? A first, provisional answer to this question is given in IA 7. And it seems that his answer there is to some extent based on what he says here in IA 5:11 locomotive animals move on the basis of four points of motion because all locomotive animals possess bodily articulations according to the four relevant directions of animal locomotion (in different degrees of separation of corresponding bodily parts, as we have just seen). This suggests that there is an important connection between the fourfold directional framework and the corresponding bodily parts on the one hand and the four points of motion Aristotle will speak about in IA 7 and IA 10 on the other. This, in turn, raises the question of how the points of motion and the directional parts relate to each other. A possible answer 11

But Aristotle will return to this question in IA 10. His final answer to this question will be given only then. See the interpretative essay by Timothy Clarke (ch. 8).



Part III Interpretative Essays

lies in the normative and therefore also heuristic priority that Aristotle attaches to blooded animals with respect to other animals.12 As will become very clear in IA 7 (707a19–23), Aristotle regards blooded animals as a model for the analysis of the workings of the locomotive parts also in other animal genera. Thus, he explains the large number of locomotive limbs that certain animals like the centipede and other similar insects exhibit by claiming that these and such animals behave as if they were composed of many four-point locomotive systems. Hence, he regards the four points of motion as the necessary and sufficient number of locomotive points for all animals (707b2–5). And, it seems, it is because the number of locomotive parts of blooded animals matches this “right” number of locomotive points in the most articulated way that Aristotle attaches a certain normative quality to them with respect to other animals (707b5–7).13 In the course of his classification of locomotive animals into the above groups, Aristotle clarifies his understanding of “foot.” “I use the term ‘foot’,” he says, “for a part connected with a point on the ground capable of producing motion; for the feet also appear to have got their name from ‘ground’.” Here, the term “point” (σημεῖον) need not, and should not, be taken in the geometrical sense of point of contact (if the latter were the intended sense, the term στιγμή would be a more natural choice; cf. Bonitz, Index Aristotelicus s.v.). Rather, Aristotle is saying that every bodily part that connects the animal with the ground at a certain spot counts as a foot. Feet are bodily limbs that make physical contact with the ground and are naturally employed for the purpose of locomotion.14 12

G. E. R. Lloyd offers interesting discussions of most of the passages in IA in which Aristotle uses human beings as a model for other animals and explains them in terms of their heuristic function (Lloyd 1983: 36–43). On further grounds that Aristotle might have had for regarding human beings as exemplary see the next footnote. 13 He says that those animals that move on the basis of two or four points of are “constituted [and therefore unified: see the entry on συνίστημι in LSJ] according to nature in the highest degree” (707b6–7). Aristotle here even offers grounds for why blooded animals possess two or four points of motion; it is that they are creatures that are natural unities in the highest degree. Presumably, this means that their bodily articulations, though complex, match exactly the number of points necessary and sufficient for accomplishing their functional task. Everything is, so to speak, in its proper place, doing its proper work, and nothing should be added or subtracted (cf. MA 10, 703a24–b3). The thought then would be that other animals fail to meet this norm and have bodily articulations that do not match the “correct” number of points because they are to a lesser degree natural unities (which is also why some of them can be cut into halves each of which continues to be functional at least for a certain time). For a different discussion of the (rationalistic) explanatory principles at work in IA 6, cf. Henry 2013: 246–248. 14 On the special role that Aristotle assigns to humans and blooded animals in his biological investigation, see previous footnote.



5 De incessu animalium 5–6



In the next step, Aristotle lines up the functionally determined internal articulations of parts of living bodies with the three cosmic regions: Since there are three regions – an upper, an intermediate, and a lower region – the two-footed animals have their upper part lined up with the upper part of the universe; the many-footed and the footless animals lined up with the intermediate region ; plants lined up with the lower . (IA 5, 706b3–6)

While two-footed animals have their upper part – their heads – located at a place that is in line with (“in relation to,” πρός) the upper region of the cosmos, the animals grouped in the second class of natural self-movers (four-footed, many-footed, and footless animals) do not have their upper part lined up with the upper region of the cosmos but rather with its intermediate region. Plants, finally, have their functional upper part lined up with the inferior region of the cosmos. Aristotle’s explanation for this distribution of functional animal parts onto the cosmic order is in terms of the natural capacities that the different animal kinds possess – namely in terms of their natures: The reason is that plants have no power of locomotion, and their upper part is determined relative to nutriment, and their nutriment is taken from the earth. Four-footed, many-footed, and footless animals lined up with the intermediate because they are not erect. Two-footed animals, however, have theirs lined up with the upper region of the universe because they are erect, and most of all the human being; the reason is that the human being is the most natural two-footed animal. (IA 5, 706b6–10)

Aristotle argues that it is because the animals mentioned in this passage have such and such a nature that they have their determinate position relative to the absolute regions of the cosmos. As plants have a nature that does not involve the power of locomotion, they have none of the bodily articulations exhibited by natural locomotive self-movers (front/back and left/right). Their upper part lines up with the lower region of the cosmos because the upper part of any living body ought to be determined functionally with reference to the intake of food, which in plants happens through their roots (IA 4, 705a31–b8). In four-footed, many-footed, and footless animals, it is again the negative fact that they are not erect that is supposed to account for their failure to have their functional upper parts lining up with the cosmic up (the periphery). Many-footed and footless animals’ upper direction – the direction toward which their heads point



Part III Interpretative Essays

– is along the horizontal axis of the cosmos. Put differently, the reason for the fact that these locomotive bodies are oriented in the way they are lies in the natures these animals possess: they are just not the kinds of animals whose upper parts line up with the cosmic upper region. In twofooted animals, by contrast, it is the fact that they are erect that is invoked as explanation for why their upper parts are in line with the cosmic upper region. Why do human beings have an upright posture? Aristotle’s answer is clear: For it [sc. the human being] alone of the animals is upright, on account of the fact that its nature and substantial being are divine; and it is a function of that which is most divine to reason and to think. But this is not easy when much of the body is pressing down from above, since the weight makes thought and the common sense sluggish. For this reason, when their weight and bodily character becomes excessive, it is necessary that their bodies incline toward the earth, so that for stability nature placed forefeet beneath the four-footed animals, instead of arms and hands. For it is necessary that all those able to walk should have two hind limbs, and such animals become four-footed because their soul is unable to bear the weight. (PA IV 10, 686a27–b2, trans. J. G. Lennox)

This is an explanation in terms of the substantial being of humans. Very roughly, Aristotle argues that an upright posture serves the natural purpose of humankind to exercise rationality. Thus, humans’ upright posture is hypothetically necessary for the realization of the human nature.15 Aristotle here also takes a transgeneric perspective on humans and animals and compares the human posture with the posture of four-footed animals, arguing that – from the human or divine standpoint – fourfooted animals have an excessive bodily nature, which makes their forefeet necessary for their stability. Of course, this is not to say that four-footed animals are misconstructions of nature. Aristotle does not think that there is a mismatch between the substantial natures of fourfooted animals and their bodily nature. It is fitting for these animals to lack an upright posture because they also lack the substantial nature that would make an upright posture appropriate or even useful. Yet, from a transgeneric and comparative perspective it appears that their bodily nature is “excessive.” So both in the IA and the PA, Aristotle takes a comparative perspective on different animal natures and establishes a transgeneric hierarchy of 15

There is no conflict with the explanation of human upright posture in terms of heat given earlier in the PA (II 7, 653a9–32), as the latter is an explanation in terms of efficient cause, whereas the account given in the main text is an account in terms of final cause and hypothetical ­necessity.



5 De incessu animalium 5–6



living things analogous to the famous scala naturae (HA VIII 1, 588b4ff.; PA II 2, 648a2ff.; PA IV 5, 681a12ff.; GA II 1, 732a18ff.) and other passages in his works in which he ventures comparisons between different kinds of living things and their activities as, for instance, in DA II 4 (415a23–b7) and DC II 12 (292a20–293a14). It is a characteristic of these comparative passages that they apply either divine or human nature as a standard of comparison between different natures. This makes them both axiological and also metaphysical in character. The basic idea in these passages is that the closer a kind approximates the perfect divine standard (eternity in DA II 4, being or attaining goals in DC II 12), the better its nature and its way of life is. The same basic idea seems to be at work in IA 5: the better the nature of a living thing is, the more its internal bodily articulation (which implies physiological complexity) is in line with the cosmic order of things. Thus, the human being is, as Aristotle puts it, the most natural of animals because the degree of internal functional articulation of its bodily parts is the highest and at the same time most in line with the cosmic dimensions. But why is Aristotle adopting this comparative and transgeneric perspective in IA 5? It would appear that he does so because of the comparatively deep and metaphysical character of the third question on the initial agenda. In Aristotle’s view, the question “Why, in general, are some animals footless, some two-footed, some four-footed, and some manyfooted?” allows for no empirical answer. For him, the existence of different kinds of animals that exhibit these different kinds of bodily equipment with parts that are useful for locomotion is a metaphysical given. Hence, I suggest, the metaphysical tone in his answer to the third question: animals possess the number of feet that they happen to possess because of the natures they happen to have. The chapter ends with a remark about the dimensional origins (ἀρχαί) mentioned in the previous section: And it makes good sense for the origins to operate from these parts; for the origin is honorable, and the upper part is more honorable than the lower part, the front part more than the back part, and the right side more than the left side. But it is also correct to state the reverse about them and say that these parts are more honorable than their opposites because the origins are present in them. (IA 5, 706b11–16)

What Aristotle has in mind here is the origin of locomotion on the right side of the body, the head as the place of intake of nourishment, and the face at the front as the place whence sense-perception originates (see DC



Part III Interpretative Essays

II 2, 284b13ff.; apparently Aristotle has in mind primarily the distal senses of sight and hearing). Roughly, the line of reasoning here seems to be as follows: P1 In the dimensional oppositions front/back, up/down, and left/right, the front, the up, and the right are the origins of their counterparts. P2 Origins are honorable. P3 In animals, the dimensional origins operate from (and are located in) honorable parts of their body – namely in the front part, the upper part, and the right part of their body. P4 The front part of their body is more honorable than their back part, the upper part of the body is more honorable than the lower part, and the right side is more honorable than the left side. C1 Therefore: it makes good sense (it is εὐλόγως) that the (honorable) dimensional origins – the front, the up, and the right – operate from (and are located in) the more honorable parts of animal bodies, namely the face, the head, and the right side of their body. The argument aims to show that it is fitting and appropriate that the dimensional origins are located in the more honorable parts of the animal’s body. The basic claim seems to be this: it is a fact of nature that these bodily parts (the head, the face, and the right side of the body) are more honorable than their counterparts, and that therefore, independently from the fact that they are the location of the dimensional origins, it makes good sense (it is εὐλόγως) for the likewise honorable origins to be located in them. But then, surprisingly, Aristotle adds that we can also state the reverse about these parts (τὸ ἀνάπαλιν λέγειν περὶ αὐτῶν), namely that it is equally well justified to say that the head, the face, and the right side are more honorable than their counterparts on account of the fact that the dimensional origins are located in them. This statement, no doubt made by Aristotle in order to tie together as much as possible anthropological with cosmological honorability, is nevertheless puzzling. For if the claim that the honorability of the dimensional origins of the cosmos has priority over the corresponding bodily parts is supposed to be the reversal of the previous statement, then what we should expect him to have said in the previous argument is that the bodily parts in which the dimensional origins are located have priority over the latter. However, if the above reconstruction is roughly correct, there is no such statement of a priority of the more honorable bodily parts over the dimensional origins in the previous passage.



5 De incessu animalium 5–6



We do find a statement that suggests something to that effect outside of the IA, however, namely in Aristotle’s discussion of the dimensional origins of the cosmos in DC II 2: These principles [sc. left and right] have been analyzed in the discussion of the movement of animals, for the reason that they are proper to animal nature. For in some animals we find all such distinctions of parts as this of right and left clearly present, and in others some; but in plants we find only the up and the down. Now if we are to apply to the heaven such a distinction of parts, we must expect, as we have said, to find in it also that distinction which in animals is first of them all (εἰ δὲ δεῖ καὶ τῷ οὐρανῷ προσάπτειν τι τῶν τοιούτων, καὶ τὸ πρῶτον, καθάπερ εἴπομεν, ἐν τοῖς ζῴοις ὑπάρχον εὔλογον ὑπάρχειν ἐν αὐτῷ). The distinctions are three, namely, up and down, front and its opposite, right and left – all these three oppositions we expect to find in the perfect body – and each may be called a principle. (DC II 2, 284b13–20; trans. J. L. Stocks; italics are mine)

Here, Aristotle clearly implies at least an epistemic priority of the animal parts over the cosmic directions. And maybe he just takes this for granted here in the IA too.16 Maybe. In any case, IA 5 establishes that all locomotive animals exhibit a bodily architecture with internal articulations according to the directions left/right, up/down, front/back, and classifies them according to whether they possess these internal articulations in separate parts of their bodies. The result is a threefold classification of locomotive animals (insofar as they engage in locomotion): (i) twofooted animals; (ii) four-footed, many-footed, and footless animals; (iii) soft-bodied animals and stromboid hard-shelled animals. DE INCESSU ANIMALIUM 6

IA 6 is concerned with the integrated bodily architecture of locomotive animals as described so far. It is a difficult chapter, not so much for its doctrine as for the long-winded and somewhat contorted character of Aristotle’s Greek prose, which makes it hard to see what his line of reasoning is. Aristotle puts the argument in one of his notorious conditional clauses (“since … then,” ἐπεί) without stating explicitly the grounds on which the apodosis is supposed to follow. The resulting apparent vagueness of his statements has led to different translations by scholars. I shall start by presenting my interpretation and discuss their translations later. I could not find any attempt in the literature at reconstructing Aristotle’s argument as a whole. Kollesch’s remarks ad loc. come closest to an interpretation. 16

Pavel Gregoric suggests to me that Aristotle’s deep commitment to an anthropocentric framework alone would not have made him feel the need to make this statement explicit.



Part III Interpretative Essays

The Argument for the Locomotive Body as a Centralized System [IA 6, 706b18–707a16] Apart from the first sentence, which states as a result of the previous discussion that the origin of animal locomotion is situated on the right side (706b17–18), the chapter is entirely devoted to establishing a single theorem. This is the thesis that there is one common origin of motion in the animal, located in a central place at equal distance from the parts necessary for locomotion. Aristotle does not say what this common origin is. The two plausible candidates are the heart (the option favored by Morel 2013: 125) and the soul (situated in the heart, favored by Michael of Ephesus, In IA 145.2–9, 26ff.; Kollesch 1985: 114). However, the question here is not so much whether Aristotle thinks that the moving principle of the animal is the soul or the heart or both, for we know from the MA itself and also from other of his writings that he was firmly committed to cardiocentrism. Cardiocentrism is the doctrine according to which the operations of the perceptual soul take place in the heart at the center of the animal body.17 So there is no question that he thinks that the origin of animal locomotion in the sense of the moving cause is the perceptual soul and that this soul is located in the heart (or, in animals that do not possess a heart, in the analogue of the heart). The question is rather whether Aristotle wishes to talk about the soul here in the IA or not. Now, in this regard, there is good motive for Aristotle not to discuss the operations of the soul in the IA. For the explanation of the operations and activities in which the soul engages in conjunction with the body is the subject matter of a separate group of writings which Aristotle calls works “common to body and soul,” and which tradition came to call Parva naturalia. In these works, Aristotle states the causes of processes and activities such as perception, memory, remembrance, sleep, dreaming, and so on.18 Apart from the fact that they explicitly continue the investigation into the soul in the DA and that they explicitly presuppose what has been said about the soul in the DA (Sens. 1, 436a1–11), what unifies this group of treatises are two main features. These are, first, that they 17

MA 8–9, especially 703a1–3; DA I 4, 408b5–18; for a discussion of Aristotle’s cardiocentrism and its relation to his hylomorphism, see Corcilius–Gregoric 2013: 52–97 and Corcilius–Primavesi 2018. 18 Not all of these processes and activities are mental. The last few of the authentic treatises pertaining to the Parva naturalia explain the causes of biological processes that have no immediate mental side to them (De longitudine et brevitate vitae; De iuventute et senectute; De vita et morte; De respiratione, leaving out the spurious De spiritu).



5 De incessu animalium 5–6



typically investigate the moving causes of these activities and processes, which is to say that they explain how these processes come to pass and what their necessary and sufficient conditions are; and, second, the absence of explanations by way of final causes in them. This makes the Parva naturalia significantly different from the IA. The IA, as the reader will recall, aims at giving functional explanations for the parts of animals that are useful for animal locomotion. It states why, given the different ways in which animals locomote, these parts serve their purposes well. Hence, very much unlike the Parva naturalia, which is not concerned with the explanation of why animals do possess functional parts, the IA offers teleological explanations for the fact that locomotive animals have these parts. There is no explanation of processes and activities in the IA, let alone of mental processes. All of this strongly suggests that the IA does not pertain to the group of writings “common to body and soul.” We should, therefore, be reluctant to regard the location of the soul in the center of the animal locomotive body as a point that the treatise seeks to establish. As I argued in the introduction, it is much more likely that Aristotle is pursuing his functional investigation of the locomotive parts of animals insofar as they are locomotive. If we let ourselves be guided by this hypothesis Aristotle is probably talking abstractly of a bodily center here that unifies the four locomotive parts without committing to whether this center is the heart or something else.19 And regarding the soul it seems that there is simply no room for it in the explanatory project of the IA. Aristotle presents his argument in three steps. In the first step (706b18– 27), he prepares the way for the theorem by arguing in the abstract for any continuous whole capable of moving itself – namely, that it is necessary for any such continuous whole that there exists a common origin of motion located at the juncture of its parts (these being imaginary parts according to the six directions of the dimensional framework introduced in IA 4). In the second step (706b28–707a5), Aristotle applies the argument to the case at hand, arguing that of the six directions of the dimensional framework (left/right, up/down, front/back) only four are relevant for animal self-locomotion – namely, left/right and up/down. In the third and final step (707a6–16), he argues that in self-moving animals the common origin of animal locomotion is located at an equal distance from each of the parts on the left/right and up/down. 19

It is to my mind not likely that it is the trunk as a whole since the first premise of his argument almost states this much (see P1 below).



Part III Interpretative Essays The First Step Since for every continuous whole of which one part is moved while another remains at rest, in order for it to be able to move as a whole while one of its parts stands still, it is necessary, insofar as both parts are moved with opposite motions, that there exists something common, according to which these parts are continuous with each other, and that the origin of the motion of each of the two parts is located there, and likewise of the absence of motion; it is evident that, for each motion of the opposite parts, proper to the aforesaid opposite pairs, there exists a common origin at the natural juncture of the parts just mentioned. I mean the parts on the right and the left side, the upper and lower parts, and the front and back parts. (IA 6, 706b18–27)

The following, paraphrastic reconstruction seems the most plausible to me. P1 For every continuous whole of which one part is moved and another is at rest, it is necessary that there they share something common to both of them that makes them, even though they move in opposite directions, a continuous whole. P2 For every continuous whole to be able to move itself as a whole, it is necessary that the origin of the motion and rest of each of the parts (the moving part and the part at rest) be located in that same place common to both. C1 Therefore, for all opposite movements and states of rest of the parts of continuous wholes capable of moving themselves as wholes (these can be parts on the right/left, up/down, front/back), it is necessary that there exist a common origin located in a place common to all of them (this is the place at the “natural juncture” of the parts). The first premise (P1) is straightforward. It seems true to say that every continuous whole of which one part is in motion and the other at rest – that is, of which different parts engage in opposed motions – must share a common element. For, otherwise, it would not be explicable how the different parts are supposed to cohere as a physical unity in the first place. The limbs of a runner whose arms swing in opposite directions must, of course, be attached to a body part that is common to both. The second premise (P2), by contrast, seems in need of further elaboration. Why is it that the origin of the motion of different limbs of an animal is necessarily located in that part of the animal body that is common to all of them? Perhaps Aristotle intends P2 to be a conclusion but, in that case, he has failed to make explicit the grounds on which he would



5 De incessu animalium 5–6



be entitled to make the inference. A plausible candidate for such a suppressed premise which would explain P2 would be the combination of two theses Aristotle has argued for at some length in MA 1 and MA 8–9. The first is what I have above called the internal supporting-point principle, according to which every motion both of any part of the animal and of the animal as a whole requires an internal resting point from which to support itself in its motion (MA 1, 689a14–b7). The second is Aristotle’s likewise long and somewhat tortuous proof that there must exist an unmoved internal supporting point for the motion of the animal as a whole, which is located in the center and distinct from any of the moved parts of the body (MA 8, 702a22–29, and 703a3). Taken together, these two theses may warrant an inference of something like P2. However, Aristotle gives us no hint to that effect, so it is better to treat P2 simply as a premise of the argument that he does not argue for. Premises P1 and P2, as given here, do yield the conclusion C1. In effect, then, this first part of the argument establishes at the very general level of any self-moving continuous whole that, for such a thing to move itself, there must exist a common origin for the motions and states of rest of its directional parts at a place common to all of them (right/left, up/down, front/back). In the next step, Aristotle will apply this line of reasoning to the empirical case at hand, namely to animal self-movers. The Second Step Now, the distinction of front and back is not of a kind that concerns that which moves itself, because there is no animal for which backward motion is natural, nor has the moving animal any articulation according to which it can make change in each of these directions; but there exists according to the right and the left, the up and the down. This is the reason why all animals that advance using instrumental parts have them distinguished not by the differentiation of front and back, but rather by that of the remaining two pairs. The distinction of right and left is prior. The reason is that this differentiation must appear as soon as you have a division into two, while the other differentiation appears as soon as there is a division into four. (IA 6, 706b28–707a5)

A possible reconstruction of the second step of the argument goes as follows: P3 Of the three axes of the directions of the dimensional framework, the axis front/back is not relevant to the differentiation of types of



Part III Interpretative Essays

animal self-motion, as there is no natural backward motion and also no bodily articulation according to the front and the back.20 P4 There is natural animal self-motion according to the left and the right, the up and the down. [P5 Nature always does the best from the possibilities for the substance of each kind of animal.] C2 Therefore (διό), animals possess bodily parts that are instrumental for their self-motion distinguished according to the left and the right, the up and the down. Such as it stands here, the argument seems more or less straightforward as well. I should note, however, that previous translators have rendered P4 in a different way. Farquharson, Kollesch, Morel, and Forster all make P4 a claim not about the relevance of the four directions for natural selfmotion, but about the bodily articulation of animal bodies according to the four directions. This reading, however, though certainly grammatically possible, spoils the argument. Meanwhile, adding P5, as I have done, is trivial. P5 is an application of the teleological principle, introduced in IA 2, according to which “nature does nothing in vain but always what is best from among the possibilities for the substance of each kind of animal.”21 What this means, in this context, is that the design of the animal’s bodily parts is functionally determined by their natures, that is, by how they live, what they do, and how they move with respect to place, and this – given the possibilities offered by nature – in an optimal way. Given this, and given that animal natures are such as to naturally self-move along the two directional axes left/right and up/down and hence from and into these directions, it follows that nature designed their bodies in such a way as to equip them with the parts that are necessary for their motions. On this understanding, Aristotle offers us an ­explanation for the fact that self-moving animals do have articulated instrumental parts for locomotion according to the four directions (left/ right, up/down) by way of hypothetical necessity. Roughly, he argues as follows: there is natural motion of animal self-movers along two axes, and 20

Of course, Aristotle is not saying here that animal self-motion into the forward direction is not natural. He only denies that the forward/backward axis is a relevant directional distinction to be applied to the different kinds of progression because there is no natural motion into the backward direction (706b28–30). So, even if there is natural motion into the forward direction, there is no motion from and into the forward direction as there is no natural motion into the backward direction. 21 Cf. Phys. II 7, 198b3–9: “Hence since nature is for the sake of something, we must know this cause also. We must explain the why in all the senses of the term, namely, […] and because it is better thus, not without qualification, but with relation to the being of each thing (οὐχ ἁπλῶς, ἀλλὰ τὸ πρὸς τὴν ἑκάστου οὐσίαν). See Ross’ commentary ad loc. and Phys. II 8, 199b15–18 (εἴς τι τέλος).



5 De incessu animalium 5–6



hence from and into four directions (left/right, up/down); now if natural motion from and into these directions is going to happen, a distinction of the locomotive parts according to these four directions is necessary; hence, self-moving animals actually possess their locomotive parts distinguished according to these four directions. Here one may raise an objection. Is it indeed the case that natural selfmotion from and into these four directions requires four corresponding locomotive parts? Why couldn’t there be natural self-motion into the upward and downward direction by an animal that lacks a distinction of upper and lower locomotive parts? The reason why Aristotle seems to think that natural self-motion requires that the self-moving animal’s body somehow mirrors the directions into which it can naturally move itself, I think, may have to do with his theory of animal self-motion. According to that theory, animal self-motion is a process in which the soul moves the body from one place to another (MA 6, 700b9–11; Phys. VIII 4, 254b28–33). In MA 6–11, Aristotle describes how and under what conditions the soul can do that. According to the account there given, the soul can set the body into motion as an unmoved mover.22 In a nutshell, this is what he says: Under the right conditions, the soul – or, more precisely, an act of perception of a desired object – can cause a reaction in the animal body. This reaction to a perceived object of desire consists in certain internal changes, namely the “heatings and chillings” that are concomitant with pleasure and pain and desire (MA 8, 701b33–702a7). These thermic changes can lead to further physiological changes inside the animal body, which on their part can result in mechanical changes (pushing and pulling) and then finally in the movement of the joints of the parts that are instrumental for the movement of the animal as a whole.23 Basically, then, the MA’s account conceives of self-motion as a transitive process in which the soul sets into motion first the inner and then the outer parts of the animal body before the motions of these parts then issue in the locomotion of the animal as a whole. Now the thought would be this. In order to result in natural motions into the four relevant directions, the animal would have to have corresponding bodily articulations, since without such articulations the soul would not be able to naturally cause motion into corresponding directions as a body that lacks an articulation into a given direction could not be naturally caused to move into that direction. 22

For a description of the outline of Aristotle’s theory, see Corcilius–Gregoric 2013: 52–97. For a more extensive description, I refer the reader to Corcilius 2008 and Corcilius–Primavesi 2018. 23 For the details of the physiological changes, see Gregoric–Kuhar 2014.



Part III Interpretative Essays

The last sentence of this section (707a3–5) commits Aristotle to the view that the distinction of left and right has priority over the distinction of up and down. Why is he taking this view? More to the point: Why should the division into front and back not be prior to, or at least as primordial as, the distinction between left and right? Aristotle claims that the distinction of left and right appears as soon as one divides something into two. But, again, why should that be so? I think the answer to this last question lies in the fact that Aristotle is not speaking here about directional distinctions per se but about such distinctions as they appear in self-moving animals. And if that is the case, we can see why the first distinction comes up as soon as there is a division into two. Recall that Aristotle has already argued that every movement originates from the right (IA 4, 705b29– 706a24). Thus, as soon as one is able to draw a distinction within a given movement, this will be a distinction between left and right. The Third Step Since, then, the up and the down, and the right and the left are connected with one another by the same common origin (by which I mean that which controls their motion), it follows that (δεῖ δ’)24 in everything that is going to make motion in each such part, properly, the cause of all the said motions must be arranged in a certain definite position relative to the distances from the origins mentioned before (both those arranged coordinately in pairs as well as those that are arranged in a series); and that the common center is the origin from which the animal’s motions of right and left, and similarly of up and down, originate, and that each has an origin of this kind at a place that is similarly related to each of the origins in the parts described. (IA 6, 707a6–16)

The argument builds on the previous two sections. C3 The upper, the lower, the right, and the left parts of the animal are all linked to one common origin. [C1, C2] C4 Therefore, the motion of one, or of more than one, of these four parts must have its cause (which is their common origin) at some determinate distance from each of these parts. [P2] C5 The common center is the origin of all motions – namely the motions from the left and the right side, and from the upper and lower parts. [P2, C3, C4] 24

I provide the Greek words in brackets because I am departing from the text prepared by Pantelis Golitsis. My reasons for this departure are discussed below.



5 De incessu animalium 5–6



[P5 Nature always does the best from the possibilities for the substance of each kind of animal.] C6 Therefore, the common origin is located in a place that is similarly related to (and hence equidistant from) the upper, the lower, the right, and the left parts. [C1, C4, C5, P5] Before I discuss the argument proper, some remarks on the construction and translation of the passage are in order. The translation and interpretation of the argument is based on a certain reading of the Greek particle δέ in 707a8. I take the δέ as apodotic, that is, as indicating the beginning of the apodosis of the previous causal clause that starts in 707a6 with “since, then (ἐπεὶ οὖν),” and which I tried to render here as “it follows that” (for apodotic δέ see Kühner–Gerth 1904: 276, §532, and, specifically for Aristotle, Denniston 1954: 177ff.25). All other translations interpret the particle δέ (the same would go for the likewise transmitted τε) as in one way or another indicating a further item in the series of conditions that starts in 707a6 and ends only in the beginning of IA 7 (see, in particular, Forster and Kollesch; Louis, who puts a period at the end of IA 6, thinks that the text is irremediably corrupt). However, to interpret the δέ in 707a8 as adding a further item on the list of conditions spoils, I think, the argument. The only exception is Pantelis Golitsis who, in his new edition printed in this volume, adopts the new reading γέ (with manuscript V) instead of δέ or τε. Golitsis apparently also takes the “everything that is going to ….” in 707a8 as the beginning of the apodosis of the ἐπεί-clause, as is done here, albeit without apodotic particle. As far as the argument itself is concerned, we find here again a proof by hypothetical necessity.26 Aristotle demonstrates the fact that animals possess a common origin of their locomotive parts at a location where it is equidistant from each of these parts. I take it that the equidistance to the other locomotive parts makes it the most suitable location possible for the execution of the center’s task of causing these parts to move. Aristotle does not bother to spell this out here. But there is reason to think that he would regard the link between the exertion of its causal function and the equidistance to the locomotive parts as trivial or not in need of justification. After all, it seems that everything falling short of equidistance would require some kind of justification, but not equidistance. Again, the 25

For a further case with a causal protasis (ἐπειδῆ) in Aristotle, see DA III 10, 433b15. Aristotle’s qualification of C3 that the common origin of the directional parts of the animal is the controlling center of its movement, though interesting, does not seem to do any work in the argument. That, it seems to me, was what motivated Bekker in putting the qualification in 706a7–8 within brackets. Most editors since have followed him in that respect.

26



Part III Interpretative Essays

teleological principle according to which nature always does the best from the possibilities for the substance of each kind of animal plays a crucial role in the argument. Without this principle, it would be impossible to derive the fact that animals possess their locomotive center at a place equidistant from the locomotive parts from the premises.

Conclusion If the above interpretation is roughly correct, IA 6 pursues two main goals. The first is to state the reason why animals possess four parts or, more generally, four points of motion, relevant for locomotion distinguished according to left/right, and up/down (C2). The second is to provide an argument for the thesis that there is a center common to the different parts relevant for locomotion (the directional parts) located in a place where it is at equal distance from each of the directional parts (C6). As we have seen, achieving these goals is no trivial task. And there are no other places in the corpus where Aristotle would do precisely this. In MA 8–9, he offers a proof for the existence of an unextended supporting point for animal locomotion, which would support the motion of the limbs (702a21–703b3). That proof includes the claim that this unmoved supporting point is located in the heart. However, the argument is working solely on the basis of mechanical premises (the internal supportingpoint principle). It is, therefore, close to the argument offered here, yet makes a significantly different point. While the MA is concerned with the question of the unmoved starting point of animal locomotion, IA 6 is concerned with the explanation of the locomotive parts of animals. In sum, then, we may say that IA 5–6 prepares the ground for the functional explanation of the parts that are useful for locomotion (that is, for answering the set of questions outlined in the initial agenda of the IA), by approaching the topic from the perspective of the basic bodily architecture of locomotive animals (insofar as they are locomotive). These chapters offer a principled account of that basic architecture of locomotive animals: they have internal bodily articulations according to four of the six cosmic directions, namely left, right, up, and down, and thus also four locomotive parts or points of motion, and they have a common origin of self-motion which is located at the most suitable place at the center of these parts or points. With this, IA 5–6 provides the structural basis for the functional analysis of the workings of the parts that are useful for locomotion in the chapters that follow.

chapter 6

De incessu animalium7–8 Number and Distribution of Feet in Animal Progression Stasinos Stavrianeas

Introduction IA 7 and 8 form a unity inside the IA by taking up the questions that relate to the number of feet necessary or optimal for locomoting animals. In IA 5–6 Aristotle has already painted a picture of the architecture necessary for a locomoting animate body. He now applies his results to specific animal kinds differentiated with respect to the number of their feet. This application results in a hierarchical classification of animal kinds. Furthermore, while IA 5–6 already display Aristotle’s willingness to classify hierarchically the different general kinds of living things, IA 7 completes this classification in a more detailed fashion. Aristotle places blooded animals at the top of the scala naturae. Finally, in IA 7–8 Aristotle employs his two teleological principles, (a) nature does nothing in vain, but (b) always what is best for the being of an animal given the possibilities available for it, in explaining the ways in which the possession of feet is realized in various kinds of animals. These teleological principles reflect, on the one hand, the plasticity of the natures of animal kinds in determining the best possible solution for the constitution of each kind. On the other hand, they also give flexibility to Aristotle’s explanatory project so that he can explain how the structures that are realized in the animal world form a unity, that is, how they obey the principles that govern locomotion.

The Place of IA 7–8 in the Overall Argument IA 6 sheds light on the following question, which is [Q1] in Aristotle’s initial agenda: What are the fewest points by means of which animals move? Additionally, IA 6 defends the claim that the optimal arrangement for animals equipped with the capacity for locomotion is such that a central principle controls their movement by being equidistant from their locomotive parts. IA 7 and 8 apply the results reached in IA 6 to answer 



Part III Interpretative Essays

questions [Q2]–[Q5].1 These questions can be taken to be an attempt to elaborate on the answer given to [Q1] in a more specific way, relative to different animal kinds. Alternatively, they can be taken to be a more concrete set of questions by contrast to the more abstract considerations relating to [Q1]. Thus, IA 7 shows that blooded animals move by means of four points at most, while bloodless animals move by means of more than four [Q2].2 But in order to understand the connection between the properties of being blooded and needing no more than four points for their movement from one place to another, IA 7 builds on the results achieved in IA 6 – namely, the claim that motion in place belongs primarily to animals that move at two or four points. IA 7–8 focus on the particular case of footless animals in order to explain why they are footless, [Q3]. IA 8 then concludes with an argument concerning what is similar between blooded and bloodless animals, namely, the fact that they move, either necessarily or better (viz. more effectively), by using an even number of feet, [Q4] and [Q5].3 Thus, IA 7 and 8 complete the discussion of the first block of questions on the initial agenda of the IA. These are the more general questions put forward in IA 1. They are more general, first, because, as Andrea Falcon notes,4 [Q1], [Q2], and [Q5] concern number of points and not of feet; second, because some of these questions ask for an answer concerning all animals – namely, [Q1], [Q4], and [Q5]; and finally, because some of them deal with the most general differentiations: while [Q2] is concerned with the difference between blooded and bloodless animals, [Q3] is about the distribution of feet in two-footed, four-footed, and many-footed animals. Once this first set of questions is answered, in IA 9 Aristotle 1

For a discussion of the initial agenda of the IA, I refer the reader to the introduction to the volume. Aristotle at 704a10 uses the word σημεῖον, but by that word he must mean the points or parts of the animal body that are in contact with the ground, namely the feet, and not the abstract points by means of which an animal body moves with respect to place (see also Morel 2013: 124). The latter cannot be more than four. 3 One could object that in IA 8 Aristotle speaks of feet and not of points. Thus, one might think, the answer to [Q5] is found at IA 3, 705a19–21, where Aristotle says that what moves always produces the change with respect to place by using at least two parts as its tools. Although in IA 3 Aristotle talks again about locomotive parts as instruments and not of points, in IA 5 he defines foot – a specific kind of locomotive part used for moving on land – as the part that is at a point at which the animal has connection with the ground (706a31–2). This being so, it is reasonable to assume that parts that are used as instruments for locomotion in general are at points of contact with the external medium in which animals move. Therefore, if animals need to move with an even number of instrumental parts, they need to move at an even number of points of contact. 4 I refer the reader to the first interpretative essay in this volume, “The Theoretical Framework and the Beginning of the Actual Investigation” (ch. 3). 2



6 De incessu animalium 7–8



turns to limbs themselves, analyzing their structure either within each kind or across kinds, and in particular explaining the manner of, and the reason for, their bends.5 From this brief review of the argument in IA 7–9 we can conclude that our two chapters, IA 7–8, are not only well placed but are also necessary steps in the overall explanatory strategy of the work. An important thing to note at the outset is that the discussion of fourfooted animals leads Aristotle to deal with an exception, and hence an apparent anomaly in nature: the footlessness of snakes (the most familiar kind of footless animal). In order to show that this exception does not constitute an anomaly Aristotle offers an explanation based on the following teleological principle: nature does nothing in vain, but rather it always aims to achieve the best from among the possibilities for each animal, preserving the proper substance and essence of each of them. (IA 8, 708a9–11)

This is one of the most important passages for the study of how Aristotle employs this principle in his biology. The upshot of the use of this principle is that nature seeks the best possible solution or end for the features of a living kind, but only given certain constraints in each case – constraints that include most pre-eminently the being or essence of the kind in question.6 Hence, the principle points to a prior knowledge of the essence of each kind and, secondly, needs to be supplemented by an understanding of the nature of the possibilities in question.7 Therefore, deservedly, the explanation of the footlessness of snakes has received quite a lot of attention and discussion in the secondary literature, certainly more than any other claim or argument in this portion of the IA. With these introductory remarks in place we can turn to an analysis of the individual chapters.

5

That is, with the second set of questions laid out in IA 1 – namely [Q6] through [Q11]. The essence of a living kind will include specific differentiae, such as being rational for human beings, but also bodily characteristics, such as the shape of the animal (discussed below). The constraints will also include other characteristics that relate to the material constitution of an animal (e.g., being blooded or bloodless), or facts related to its habitat or mode of being (e.g., being a land animal or a water-dweller). For a typology of the relevant constraints see Henry 2013: 225–263. The seminal paper by Jim Lennox (in Lennox 2001b: 205–223) sets the agenda for the discussion of these teleological principles in Leunissen 2010, Henry 2013: 225–263, Gelber 2015: 267–293, Morel 2016: 9–30, and Gottlieb–Sober 2017: 246–271. 7 See Lennox 2001b: 207. 6



Part III Interpretative Essays DE INCESSU ANIMALIUM 7

All Blooded Animals Progress by Means of No More Than Four Points of Contact [IA 7, 707a16–708a8] IA 7 can be divided into four sections: (I) the claim that there is a necessary connection between being blooded and moving by means of four points; (II) evidence that supports this claim; (III) the further claim that footless animals, most notably snakes, also move by means of four points – by using two bends – as well as an explanation of that claim; and finally (IV) a list of footless kinds that move by using bends, focusing primarily on animals that live in water. Motion with Respect to Place Belongs Above All to Blooded Animals [707a16–23] In Section I, Aristotle makes two important claims: [A] no blooded animal can move by means of more than four points; and [B] if an animal naturally moves by means of four points, then it is, necessarily, blooded. It is not immediately clear how the text supports [A], and it is even more obscure how [B] is established. Let us start from the opening claim of the chapter, which must play some supporting role for both [A] and [B]. Aristotle says: It is evident, then, that motion with respect to place belongs either only or above all to those animals that make change with respect to place either by means of two or four points. (IA 7, 707a16–19)

The claim must be evident given what Aristotle has argued for in IA 6, first that only four points are relevant to motion in place, namely right and left, and up and down, excluding the pair front and back (706b33– 707a4), and then that these points must be connected to, and equidistant from, a common principle controlling and coordinating their movement (706b25–28, 707a8–11).8 What precedes in IA 6 thus licenses the claim 8

Thus, in IA 6, Aristotle notes: “everything that is going to make motion in each such part, properly [κατὰ τρόπον], must be arranged in a certain definite position relative to the distances from the origins mentioned before” (IA 6, 707a8–12). For a similar use of the expression κατὰ τρόπον, see PA I 1, 639a5. There, it qualifies the connection between an educated person and his ability to discriminate whether a scientific account is well presented, despite not being a specialist on the science in question. The point in PA I 1 is that this is a distinguishing mark or a defining capacity of the educated person. Similarly here, the distance of the principle from the parts is a distinguishing mark or a defining element of what possesses a capacity for locomotion according to nature in the highest or fullest degree. As Klaus Corcilius puts it (above, ch. 5): “the chapter [sc. IA 6] is entirely



6 De incessu animalium 7–8



that motion in place belongs primarily to animals that move by means of either two or four points. In both cases there is a principle controlling the movement of the limbs by being at the juncture common to these four points.9 The next sentence is crucial for understanding Aristotle’s argument for [A] and [B]. Aristotle argues that: since perhaps this (τοῦτο) happens above all in the case of blooded animals, it is clear that [A] none of the blooded animals can move by means of more than four points, and that [B] if an animal moves naturally by means of just four points, it is necessary for it to be blooded. (IA 7, 707a19–23)

The antecedent of the first sentence supports [A]. However, it is far from clear what exactly “this” (τοῦτο) refers to in our passage. At least two interpretative options are available. The first is that a reference is made to the fact that blooded animals move by means of two or four points. From this it follows that these animals do not move by means of more than four points, which is a factual claim. However, Aristotle’s claim is stronger: these animals cannot move by means of more than four points, which is a normative claim. A way to motivate such a claim is to assume that the teleological principle that “nature always does what is best for the being of an animal given the possibilities available for it” is already at work here. If so, the argument for [A] could be understood along the following lines: 1. Locomotion belongs either only or primarily to those animals that make change with respect to place either by means of two or four points [i.e., this is the best method of locomotion for an animal]. 2. The locomotion of blooded animals occurs by means of two or four points. 3. [Nature does what is best among the possibilities.] 4. Thus, the locomotion of blooded animals cannot take place by means of more than four points. If this argument captures Aristotle’s train of thought, the claim that locomotion belongs primarily to blooded animals depends on the claim that the four points of motion are arranged so as to be equidistant from the devoted to establishing a single theorem … [i.e.] the thesis that there is one common origin of motion in the animal, located in a central place at equal distance from the parts necessary for locomotion.” 9 See chapter 5 of this volume by Klaus Corcilius for a detailed discussion of this passage.



Part III Interpretative Essays

common principle that controls them. Four points of motion equidistant from the principle that controls motion are necessary and sufficient for locomotion. Thus, if an animate body can be structured in this way, it must be structured accordingly: this is the case with blooded animals.10 If this arrangement is the best among the possibilities concerning the structure of the locomotive parts in an animal’s body, it must be so on the basis of some principle of economy. However, this reading does not license claim [B] – namely, that if an animal moves by means of four points, then it must be a blooded animal. In other words, it is not clear what excludes the possibility that there exist bloodless animals equipped with four points of motion. No independent reason is given in our text for this view, and this leads to the idea that we should look for such a justification outside the IA. This is our second interpretative option, which I would like to explore. A justification for claim [B] can be found in Aristotle’s presentation of a hierarchical structure of living beings according to variations in the quality of their natural or vital heat in GA II 1, 732a26ff. Although the classification made there is in terms of the way animals give birth to their offspring, it is clear that the bodily arrangement in blooded animals is a consequence of a more fundamental difference in the quality of their natural heat, with blood being a sign of a greater amount or better quality of heat, thus constituting evidence that the blooded animals are more perfect than the bloodless ones.11 One reason, among others, why some kinds of animals are blooded is that blooded animals, being more perfect and larger for the most part compared to bloodless ones, need a greater amount of heat in order to move their body (GA II 1, 732a19–22). This suggests that blooded animals need the best possible kind of vital heat for locomotion (as well as for other functions). And if so, it is natural to think that they also need the best possible arrangement for their locomotive parts, which (as we have already seen) amounts to using either two or four points of motion. Although there is no explicit hint of such an explanation, the above thoughts may provide a rationale that connects (a) the best structure for locomotion with (b) blooded animals. If, moreover, a reading along these lines is endorsed, then we can see how to justify the biconditional expressed by the conjunction of claims [A] and [B].

10

For Aristotle’s argument that there must be a center which acts as the common origin for the locomotive parts, see Klaus Corcilius’ interpretation of 707a6–16 (ch. 5). 11 The claim that blooded kinds are the most perfect kinds of animals is explicitly made in PA IV 5, 682a34.



6 De incessu animalium 7–8



Evidence that Locomotion Belongs Primarily to Blooded Animals [707a23–b5] Section II offers support to the main claim advanced in Section I. In fact, it introduces some empirical evidence that reinforces the more abstract argument given in the previous section. The whole section focuses on just one such piece of evidence, namely, the fact that blooded animals, if their bodies are divided, cannot continue to live and move, while some bloodless, many-footed animals can, even if only for a limited time. There are at least two questions to consider here: (1) Which claim from the ones made in the previous section is this evidence supposed to support? (2) In what way does the evidence support it? The key for answering both questions is found in lines 707b2–4: And the cause of their living when they are divided is that each of them is constituted as if it were something continuous compounded out of many animals.

Here Aristotle offers his explanation of why some bloodless animals can still live, and even move, once divided. The reason is that these animals enjoy a weaker type of unity compared to animals that cannot survive division – so weak that the division does not compromise their capacity to perform at least some of their characteristic activities. Now, this observation supports the claim that bloodless many-footed animals are less unified compared to blooded two-footed or four-footed animals.12 But one may ask: Why do loosely unified, many-footed animals possess the locomotive capacity in a less perfect way when compared to fully-unified, blooded animals?13 One could, indeed, argue that since the divided parts of the many-footed animals can still move, these animals possess a stronger and more robust capacity for motion in place, one that is not 12

This claim may sound too strong. As the anonymous reader notes, such animals (or plants for that matter) are only potentially many but, when undivided, they are still unified as one animal (or as one plant). Thus, they do have one principle of life until they are divided. (See DA II 2, 413b18– 22.) In Juv., however, when Aristotle revisits the case of divided animals (and plants), he claims that such living beings are more like a concretion of several animals as opposed to animals that are best constituted (i.e., blooded ones) because in the latter their nature is one insofar as it can be (2,468b9–12); Aristotle must mean one both actually and potentially. If the best constitution amounts not only to being a unity under the control of a single soul-principle in actuality, but also to not being able to be potentially more than one such unity, Aristotle must believe that animals that survive their division possess a lower kind of unity. And the reason must be, with respect at least to their locomotive structure, that such animals possess more than just the necessary and sufficient parts for locomotion; they are composed of many four-point locomotive systems. 13 These lines are offered as evidence for the thesis that blooded animals possess the capacity of locomotion in the primary sense (see section I).



Part III Interpretative Essays

compromised by division. Furthermore, many-footed animals can move at least as fast and as efficiently as the four-footed ones. So why is their capacity for locomotion found wanting when these animals are compared to four-footed animals? To shed some light on the criterion that is at work here we need to consider other passages in the biological writings where Aristotle refers to the same phenomenon, namely the capacity of bloodless animals to continue to live when divided. Aristotle returns to the phenomenon of bodily division several times in the biological corpus. The most interesting for our purposes is a passage from PA IV 6, 682b2–9.14 There, Aristotle says that some insects have not just one part where the soul-principle is situated, but potentially several,15 and this fact explains why they continue to live when divided. Yet, Aristotle adds, this deviates from the proper way nature designs animal bodies, for its aim is to design one soul-principle for each animal body and not several, this latter option being only second best (see also Juv. 2, 468b9–12). But why, one may ask, is it better for nature to design a single such principle for each animal? The reason must be that the existence of only one such principle seems to create a dependency of all the parts of an animal’s body on a single source or cause. And this dependency must be signaling a tighter mode of integration for the parts of an animal’s body. The existence of several principles, on the other hand, establishes that parts of the body can enjoy, to some extent, independent existence.16 Furthermore, if we take into account the idea that the soul-principle is the form of an animal’s body, as well as what holds the body together as a unity,17 then a body that possesses several such principles, even if only potentially, is not one in number in every respect. In one way, namely potentially, what we have here are more than one animal sharing the same form (like two distinct specimens of the same kind).18 This is not to 14

Other references to the phenomenon include DA I 5, 411b14; DA II 2, 413b16–21; HA IV 7, 531b30–532a5; Long. 6, 467a18; Juv. 2, 468a25; Resp. 3, 471b20–24 and 17, 478b32–479a7. 15 In DA II 2, where the same phenomenon is discussed, we see that this plurality of principles exists in the bodies of many-footed animals merely potentially for as long as such bodies are united, and becomes actual only once they are divided (DA II, 413b16–19, cf. Resp. 17, 478b32–479a7). 16 Normally the parts of a natural substance cannot survive separation. The reason is that the Aristotelian principle of homonymy rules that when a natural substance or part thereof loses its form or its functional role within the substance, it is the same only in name, and thus not identical to what it was before (see e.g. Mete. IV 12, 389b31–390a2; PA I 1, 641a1). The fact, then, that the divided parts of an insect continue to function properly suggests that they are linked to the form, or formal component, with the only difference being that this form is many in number – as many as the number of the divided parts. 17 For these two claims see respectively DA II 2, 414a17 and DA I 5, 411b7–10. 18 See DA I 5, 411b19–24, where we are told that the parts of the divided living being (a many-footed animal or a plant) do not have one soul in number but only in eidos.



6 De incessu animalium 7–8



say that such animals are not unities. However, they are not unities in the way that blooded animals are, but rather they are potentially a series of animals merely continuous with each other (continuity representing a weaker form of unity). The idea seems to be that for a body to form the highest or best kind of unity it must be dependent on one unique principle (here a soul-principle) for its being alive: unique both in actuality and in potentiality. But the dependency seems to run the other way around as well, namely from the bodily parts to the principle controlling the entire organism. Aristotle has suggested that change of place at two or four points is the primary or best form of locomotion (707a17–18). And this can be taken as meaning that two of four points are necessary and sufficient for locomotion: they determine the smallest set of parts (a fourpoint unit) sufficient for building a locomotive animate system. The bodies of insects, by contrast, are constituted by a series of such fourpoint units of locomotive systems. Hence, their body can still function properly even when one or more of these units are divided.19 Such parts, then, are not necessary for the body to function properly in the first place. If so, the whole in which such parts were participating was not dependent on them, which seems to entail again a looser type of unity for the whole in question. Thus, we can answer the questions raised at the beginning of this section in the following way. First, the examples Aristotle gives are meant to show that blooded animals (both two-footed and four-footed) enjoy a stronger kind of unity between their soul-principle and their locomotive parts. No part of such a substance can continue to be what it is when divided from the whole. And the substance itself cannot survive a division of one or more of its parts without compromising its proper function. This makes for an interdependency of locomotive parts and the principle that controls them in a way that shows, for Aristotle, that in their case nature has achieved the greatest possible kind of unity with respect to a 19

The anonymous reader rightly suggests that one could contest this by pointing out that the parts of a divided insect do not really function in the way that the insect itself does, since they live only for a very short time. If so then they are not really a plurality of different animals, but merely pieces of one animal. Now, indeed Aristotle several times points out that divided animals cannot survive for long, but he specifies that the reason is, crucially, that their divided bodies do not possess the necessary organs (see DA I 5, 411a22–24; Long. 6, 467a20–22; Juv. 2, 468b6–9). Which seems to imply that if they had the necessary organs, they could go on living for much longer. Furthermore, in a passage where Aristotle considers the question how long can the divided parts of insects survive, he remarks that long and many-footed animals can live long enough and move in either direction, presumably because they have the necessary organs for locomotion (HA IV 7, 531b30–532a4), which implies that their locomotive capacity is the least affected by such a division.



Part III Interpretative Essays

body that moves with respect to place.20 Hence, second, Aristotle’s example in this section supports the idea that the bodies of blooded animals are the ones that are constituted most in conformity with nature. Footless Animals Move by Means of Four Points [707b5–27] Section III serves to block a possible counterexample to the claim that blooded animals move by means of (at most) four points. Indeed, a number of blooded animals are footless, and this may give the impression that movement by means of (at most) four points is not necessary for all blooded animals. Aristotle shows, first, that all footless animals move by means of four points (707b5–17), while, second, he proposes an explanation of why that is the case (707b17–27). Our section can be divided naturally into two sub-sections. In the first sub-section, Aristotle establishes that footless animals, like all blooded animals, move by means of four points despite appearances to the contrary. Indeed, they appear to move by means of two points, since they advance by using two bends, one in the front and one in the back (707b14–15). However, these bends themselves considered in their width comprise a right and a left side. Hence, these animals use four points of contact: right and left side of the front bend and right and left side of the back bend. This is not easily observed, argues Aristotle, since the width of the bodies of these animals is narrow. Thus, although they move in exactly the same manner as the four-footed animals, namely by moving first the front right part and then the back left, given that their right and left sides are not easily distinguished, they give the impression that they move by means of two points only, namely front and back (707b15–16). Once Aristotle has set out the four points in the bodies of footless animals – namely, the two sides of each of their bends – he goes on to give, first, an explanation of how they move by means of these bends and, second, a description of the movement of their bends as involving a right and a left part.

20

This claim should not be taken as suggesting that the locomotive capacity is one of the soul-capacities of a living being rather than a capacity of the ensouled being. As Klaus Corcilius points out (personal communication), Aristotle does not refer to any principle of locomotion in terms of soul. However, there is some evidence that the principle of locomotion is included or connected to the soul-capacities of the animal. For instance, in PA II 1, 647a25–29, Aristotle argues that the powers of sensation, locomotion, and nutrition are situated at the same part of the body (cf. Somn. 2, 455b34–456a1). Hence, even though the principle of locomotion is not a soul-principle, it is situated at the center of the body in the primary organ used by the soul-principle of the animal.



6 De incessu animalium 7–8



His explanation relies on the length of their bodies. Because these bodies are very long, these animals advance by bending their bodies. Aristotle does not explain further why length of the body should entail that an animal should move by means of bends.21 Instead, he explains how movement by means of bends works, and how each bend involves a right and a left part by means of an analogy between the movement of the elongated bodies of limbless animals and the movements of tall human beings who bend their bodies backward (λορδοί) to explain how snakes alternate right and left side on each of their bends. The analogy is not immediately clear, but we can understand it if we focus on what Aristotle actually says. Aristotle mentions only two points of the human body, namely the right shoulder and the left hip, and points to the fact that the leaning forward of the right shoulder combined with the leaning backward of the left hip give a shape to their upper body such that the middle is at the same time concave (λορδόν) and convex or hollow (κοῖλον). Now, in the following lines (707b23–25), Aristotle makes the same point with respect to each of the bends of footless animals. By advancing first the right and then the left part of their front bend, these animals alternate concavity and convexity on each of the sides (right and left). When, for instance, snakes advance, this movement makes the right part of the bend concave and at the same time the left part of the bend convex; conversely, when they advance the left part of their front bend, this movement makes their left side concave and their right side convex.22 The alternation between concavity and convexity in each of the two bends of footless animals is a sign for Aristotle that there are four points of motion in them, in exact similarity with the four-footed animals. Kinds of Footless Animals [707b27–708a9] In Section IV, Aristotle lists the animals that move by bending. He focuses primarily on footless animals that live and move in water, for some animals of this group are peculiar in that they possess a pair of fins. 21

A full explanation would need to take into account Aristotle’s explanation of the footlessness of snakes (and all other animals with elongated bodies) in IA 8, 708a9–20 (see my comments on IA 8 below). 22 If this is the point of similarity between tall people and footless animals, then we should not try to find an analogy between the four points of motion in footless creatures and the four corners of the upper portion of the human body, as, for instance, Michael of Ephesus attempts (In IA 150.26– 151.38). What we need to focus on, instead, is merely the double character of the curvature in the human body, which is both concave and convex.



Part III Interpretative Essays

Given that they already have these two points for motion, and given that a blooded animal can move only by means of four points, they should have fewer bends – that is, just one. Let us look at how Aristotle deals with these animals in some detail. Among the animals that live and move on land Aristotle mentions only snakes. He then turns to animals that live and move in water and he mentions eels (αἱ ἐγχέλεις), congers (οἱ γόγγροι), and murenas (αἱ μύραιναι) (see HA II 15, 505b31). He subdivides the class of animals that live and move in water into animals that possess fins and those that do not. The murena is the only animal among those mentioned that does not possess fins. That is why its bodily movement is said to most resemble that of snakes.23 It is interesting that in our passage the murena is judged to be snakelike in its form. This claim suggests that the form of snakes becomes a standard such that other footless animals with elongated bodies are compared to it, and judged to be more or less like it. At PA IV 13, 696a5–6, we are told that all animal kinds that have an elongated body and are more snake-like do not have fins. This is possibly Aristotle’s explanation for the absence of fins in murenas. The less snake-like animals include eels and congers. The possession of fins by both eels and congers is also noted in the HA (I 5, 489b27; II 13, 504b32). Several attributes of these animals are listed in the biological works, and it is worth mentioning two here. First, eels constitute one of the few blooded kinds that are generated spontaneously (HA IV 9, 538a3– 4; HA VI 16, 567a21; GA II 5, 741b1; GA III 11, 762b23–28).24 This looks like an anomalous case in the sense that blooded animals are placed high up in the scala naturae while spontaneously generated organisms occupy the lower end of this classification. Given that their being blooded determines facts concerning their locomotion in the same way in which such facts are determined for animal kinds higher up in the scala naturae, it seems that their being blooded is far more important than their being spontaneously generated. Second, congers (as well as a kind of mullet that will be introduced below) continue to live even when their back is 23

The same observation is made in HA II 1, 498b28–30 and PA IV 13, 696a6. In HA II 13, 504b34 Aristotle suggests that this feature together with the fact that murenas do not have well-articulated gills makes for a difference with other fish. A further difference with other fish is mentioned at HA III 10, 517b7 – namely, that their eggs do not crumble as do those of other fish. This is common also to eels and congers. Murenas resemble snakes in one further respect: in coition they intertwine, belly to belly (HA V 3, 540b2). 24 The context in which the references to eels are made in the GA passages suggest that not enough accurate observation has taken place to confirm this finding. I would like to thank the anonymous reader for pressing me on this point.



6 De incessu animalium 7–8



mutilated (HA VIII [IX] 2, 610b5). This may give one the impression that there is here a violation of the rule observed above that only bloodless animals can continue to live and move when divided. However, in the case of bloodless animals, we were told that both parts continue living, while nothing similar is said of congers. Finally, along with eels and congers Aristotle mentions a further animal kind that moves by bending while also possessing two fins. It is a kind of mullet (γένος τι κεστρέων) that lives in lakes.25 With respect to this kind of fish Aristotle says that, given that it possesses only two fins, it obtains the four points it needs so as to move like all other blooded animals by means of a bending of its body. The way Aristotle phrases this explanation seems to suggest a difference between mullets and the other footless animals that move by bending. For in the case of grey mullets it is the lack of four fins that explains the use of the bending of the body, and not considerations of an optimal solution given the length of the animal’s body, as is the case with snakes. If this is so, then here nature uses bends to compensate for the lack of some locomotive organs (in this case the lack of two fins). Be that as it may, both the mullet and the eels are said to use fewer bends due to the fact that they already have a pair of fins that – given Aristotle’s insistence that blooded animals move with respect to place by means of at most four points – should make the use of more than one bend impossible. Summary IA 7 takes up the results reached in IA 6, concerning the fewest points of contact in an animal’s body necessary for locomotion, and applies it to the extensive kind of blooded animals. The kind of blooded animals is the one that conforms fully to the optimal architectural structure of a locomotive body described earlier, i.e., it possess the minimal number of parts, four points of contact, necessary for locomotion. So, blooded animals display the best possible version of the locomotive capacity that can be realized in nature. The chapter provides some evidence for arguing that the bodies of non-blooded animals are essentially composed of a number of such units of four points of contact. At the same time Aristotle addresses an apparent exception, footless blooded kinds such as snakes, for which he argues that their bodies move in the manner the rest of the blooded animals do, namely by means of four points of contact. 25

More specifically, in the lake of Siphae, as Aristotle notes here as well as in HA II 13, 504b31.



Part III Interpretative Essays

His explanation needs to be supplemented with an explanation of why footless animals do not have feet. IA 8 starts by discussing the lack of this feature in some blooded animals. DE INCESSU ANIMALIUM 8

Why Some Animals Are Footless, and Why Footed Animals Have an Even Number of Feet [IA 8, 708a9–b19] IA 8 deals with three questions and can be naturally divided into three corresponding sections: (I) Aristotle offers a full explanation of the reason why footless animals do not have feet; (II) he explains why all footed animals have an even number of feet by showing that an odd number would make motion impossible; and (III) he qualifies to some extent this latter claim by considering the case of many-footed animals, which do have an even number of feet by nature and yet it is not impossible for them to move if, due to a mutilation, they end up having an odd number. Why Footless Animals Have No Feet [708a9–20] Aristotle’s explanation relies on three main ideas. The first is the teleological principle that nature does nothing in vain but rather does what is best with the possibilities available (cf. IA 2, 704b17). Aristotle’s formulation of this principle consists in fact of a negative part – nature does nothing in vain (NP) – and a positive one – nature realizes the best of the available possibilities (NP*).26 I will elaborate on this point in due course. The second idea is that the available possibilities are limited by the essential nature of the animal in question. The third and final idea is that blooded animals can move by means of four points at most. Before commenting on these three ideas, let us lay down the steps of the proposed explanation: 1. [ESS] A disproportionally long body is part of the essence of snakes. 2. [BLO] Snakes, like all blooded animals can move at most by means of four points. 3. [Tacit premise] Movement by means of four points requires four locomotive organs or limbs. 26

For these two principles, and the introduction of these abbreviations, see Lennox 2001b: 206– 207. For an alternative interpretation attempting to reduce the first principle (NP) to the second (NP*) see Gottlieb–Sober 2017: 250–252. See also Henry 2013: 230, who argues that the two principles work as a unity.



6 De incessu animalium 7–8



4. (from 2, 3) If snakes had feet, they would have at most four. 5. For the motion of a disproportionately long body, four feet would be useless (in vain). 6. [NP] Nature does nothing in vain. 7. (1, 4, 5, 6) Therefore, nature does not create feet for snakes. The first thing to note is that the above argument explains why snakes are footless. Being footless, rather than footed, allows them to move more efficiently. However, for this explanation to hold, Aristotle needs to clarify how exactly the movement of a footless, elongated body is more efficient compared to the imagined case of a footed, elongated body. In fact, Aristotle’s explanation of the movement of footless animals in terms of bends, already discussed in IA 7, may serve for explicating this point. However, the two explanations – namely, the explanation of movement of footless animals by means of bends and the explanation of the footlessness of snakes – seem to be independent. This is because (a) they are concerned with facts at different levels of generality: while the first explanation applies to all blooded animals that are footless, the second applies only to snakes. And thus, (b) they have different explananda. The first explanation deals with the movement by means of bends in blooded animals that are footless. Aristotle shows how this bodily displacement obeys the rule that these animals, precisely because they are blooded, move by means of no more than four points. The second explanation concentrates on the case of snakes. Aristotle shows why these animals are footless.27 Without obscuring the different aims of the two explanations, observations (a) and (b) are less clear-cut than they may seem at first sight, and the two explanations are more tightly related than may be initially thought. Let us address the claim that the two explanations are given at a different level of generality – namely, observation (a). It is doubtful that the explanation of the footlessness of snakes is meant to apply uniquely to snakes. Arguably, it is meant to be extended to all blooded animals with elongated bodies. The evidence for this is the generalization that all blooded animals whose length is disproportionate compared to the nature of the rest of their bodies (such as snakes) cannot have feet (708a14–16). This phrasing suggests that snakes are used as a prominent example of a natural fact that ranges over a much wider domain, namely the domain of the blooded animals with disproportionately elongated bodies. The reason for Aristotle’s explanatory strategy might rely on the 27

Thanks to Andrea Falcon for helping me realize the difference between the two explanations.



Part III Interpretative Essays

assumption that snakes are more familiar to us and hence they can be used as a starting point on which to build a more general explanation.28 Whatever the reason for this strategy, it seems at any rate reasonable to take the two explanations to range, ultimately, over the same set of cases, namely blooded animals with disproportionately elongated bodies. In a way this is what one would expect on conceptual grounds, so to speak. We can see this by turning to (b) – namely, the observation that the two explanations are independent because their explananda are different. The explanation of footlessness turns on the idea that for snakes, and for animals with disproportionately elongated bodies in general, being footed entails a less efficient bodily movement. However, this is a comparative claim, and unless the movement of footless animals with disproportionately elongated bodies is shown to be effective, Aristotle cannot establish that the counterfactual possibility of them being footed entails a less efficient bodily movement. This being the case, to secure the conclusion for the footlessness of snakes, Aristotle will have to rely on what he has already established in IA 7 concerning bodily movement by means of bends. This does not amount to confusing the two explanations. While in IA 7 Aristotle aims to establish that all blooded animals move at four points, his goal in IA 8 is to show, by using snakes as the paradigmatic case, why some blooded animals – namely, all those with disproportionately elongated bodies – are footless. Despite the differences, it is clear that the observations concerning bends will play some role in a full explanation of why the best among the possibilities for snakes and other such animals is to be footless instead of footed. Being footed is pointless for such kinds, only if there is a better possibility for movement – namely, moving by means of bends. Given the above observations, we may try to connect the two explanations. And we may view them as related in at least two ways: (a) The full explanation of the locomotive parts of snakes29 must proceed in two steps, explaining first, by means of (NP), why they are footless, and second, by means of (NP*), why they move by means of two bends. 28

Note that at IA 7, 707b30, where Aristotle lists the animal kinds that move by bends, he qualifies their common form (as animals that move by bends) with the adjective ὀφιώδης (snake-like). There, he uses the qualification ὀφιωδεστέραν, (more) snake-like. At least in this case, it appears that snakes are taken to be a standard against which other kinds of footless animals are studied. For the use of the adjective ὀφιώδης for a similar purpose, see PA IV 13, 696a6–17. 29 In what follows, I will be referring to snakes as the paradigm case of blooded animals with disproportionally elongated bodies. I do not mean that Aristotle’s points apply uniquely to this kind.



6 De incessu animalium 7–8



(b) The explanation of why snakes move by means of bends relies ultimately on (NP*), or the conjunction of (NP) and (NP*), which explains why movement by means of bends is a better alternative for snakes than movement by means of four feet, and ultimately why it is the best possibility for the bodily displacement of snakes. At least three considerations count in favor of option (a). First, this exegetical option takes into account the different roles played by the two versions of the principle that nature does nothing in vain, namely (NP) and (NP*). Second, the two explanations are indeed offered separately (respectively in IA 8 and IA 7). Finally, this reading distinguishes between the two facts to be explained – one negative, namely the absence of feet, and one positive, that is, locomotion by means of bends. Against (a), however, one could raise two objections. First, one could argue that we would expect the two explanations to be presented in that order, first the negative and then the positive explanation, whereas our text presents them in the reverse order. Second, one could point out that the negative explanation on its own, without the support of a positive one, does not seem to be secure, for unless there is an option that allows snakes to move more effectively without feet, it is not clear that their possessing feet would be pointless. To be sure, having a slow-moving snake equipped with feet may be better than having a snake with no feet that cannot move at all. Therefore, unless it is clear that there is a better alternative for the bodily displacement of snakes, the presence of feet in their body would not be pointless. Option (b) can meet this second objection because the absence of feet in snakes is explained by the presence of the alternative possibility that they move by means of bends. Clearly, this is a better alternative for snakes. Moreover, (b) justifies the fact that Aristotle mentions both teleological principles in IA 8 – namely, nature does nothing in vain (NP), and nature realizes the best of the available possibilities (NP*) – although, admittedly, he uses only the first one. In this chapter, Aristotle explains only the absence of feet and falls short of explaining why their moving by means of bends is a better possibility. Thus, I am inclined to read the explanation along the lines of (b). Still, we have to account for the fact that what must be included in the positive explanation (IA 7) precedes the negative explanation (IA 8). This can be done if we observe that the positive explanation is given almost as an aside to the explanation of a slightly different fact, namely that snakes move by means of four points like all blooded animals, despite



Part III Interpretative Essays

appearances to the contrary. Aristotle merely flags that the reason why they have bends is their length, but he is silent concerning the relation between length and possession of bends. This can only be fully understood when he addresses the question of why they are footless. So, it may very well be that in IA 8 we are invited to combine the explanation of the absence of feet, which is again explained by invoking the disproportionate length of the body as the middle term, with the observation that a more effective method of bodily displacement is available for the locomotion of such animals. If so, the explanation of why snakes are footless has to be strengthened, or rephrased, in the following way:30 5. For the motion of a disproportionately long body, four feet would be pointless (in vain). 6. [NP combined with NP*] Nature does nothing in vain, but always realizes the best of the available possibilities. 7. Moving by bends is the best possible mode of bodily displacement for snakes. 8. (1, 4, 5, 6, 7) Therefore, nature does not create feet for snakes, and they move by means of bends instead. Let us now look at the central ideas involved in driving this explanation. Let us focus, in particular, on the claim that the nature or essence of a kind is a constraint on the alternative possibilities among which the creative force of nature realizes the best one. We are told that nature preserves the proper substance of each kind, namely its essence (cf. Phys. II 7, 198b9; DA I 6, 411b23). It is not clear what this proper substance or essence of snakes includes, but Aristotle refers here to at least two facts that are candidates for being parts of it. He mentions (i) their being blooded; and (ii) their possessing a disproportionate length compared to the rest of their body. Let us take these two properties of snakes in turn. Concerning being blooded, there is strong evidence that this property is an essential feature of the animals to which it belongs.31 However, it is not certain that being blooded is included in the essence of snakes. The reason for this is that this property is mentioned as an additional feature, as if it were added to their proper essence. (See, in particular, the ἔτι at IA 30

For steps 1–4 see above. One unequivocal statement of this idea is found in PA IV 5, 678a33–35: the fact that some animals are blooded and some bloodless will be found to be included in the logos which defines their substance (ousia). For the connection between ousia and being blooded see also PA IV 12, 693b6 and PA IV 13, 695b20. For the connection between nature and being blooded see GA I 17, 721a31 and GA III 11, 762b25.

31



6 De incessu animalium 7–8



8, 708a12.) Moreover, one could argue that such an attribute belongs not to the essence of snakes (their proper ousia) but to the wider kind of blooded animals. At least two things could be said with respect to the constraints imposed by the property of being blooded. First, nature in preserving the proper substance of each kind is constrained by facts that are determined by the wider kind or kinds to which it belongs – in this case, the quasi-kind blooded animal. Alternatively, one could propose that the explanation concerns not just snakes but all blooded animals with disproportionately elongated bodies, and being blooded is an essential property of these animals. I am tempted by the first option. For although the explanation is generalized in the subsequent lines over all footless kinds, at this stage Aristotle seems to be talking about his paradigm case, snakes. Secondly, whatever the level of generality at which an explanation is proposed, there will always be attributes of a more general kind that will impose constraints on the possibilities that nature has at hand. If so, the constraints imposed on nature do not only include the essence of a kind but also depend on features that are more widely distributed. And such constraints will have to be explained at the corresponding level of generality. If the above observations concerning the property of being blooded are on the right track, then the features of the essence of snakes that Aristotle must be thinking about include only the dimensions of the animal’s body. Indeed, his reference to “the length” and “the rest of the nature of the body” means that he is willing to talk of the three dimensions of an animal’s body as natures of this body,32 or at any rate that the size and proportionate relations between the dimensions of a body are part of its nature and essence. One may take this reference to the nature of the body as a reference to the material nature of the kind in question as opposed to its formal nature. But if nature explains the notion of proper substance or essence introduced in IA 8, 708a11–12, the reference cannot be to the material nature alone. Rather, it must be a reference to the formal nature, or to the nature of the material substance as a composite of matter and form, although it is not entirely clear how we should take this reference. On the one hand, if essential or formal attributes are attributes that are explanatorily primary, it seems that the dimensions or size of an animal’s body must be taken as attributes pertaining to the formal nature of animals, since they appear as explanatorily primary: they explain the 32

If the dimensions of the body are principles of a body, there is a sense in which they can be said to be natures of the body, or of its size. See Lennox 2009: 187–214.



Part III Interpretative Essays

possession of further attributes. For instance, there is evidence, in the PA and the GA, that the shape of the snakes’ bodies – namely, their length and narrowness – explains further attributes such as the shape of their viscera (PA IV 1, 676b8–9) or their reproductive organs (GA I 6, 718a17– 21).33 If explanatory priority determines the formal attributes of an animal kind, then the dimensions of a body must be taken as part of the formal nature. On the other hand, Aristotle insists that the shape of the body of a composite substance, or of a bodily part, is not sufficient for determining its essence (PA I 1, 640b30–641a2). Its essence is determined by its function, and in the case of an animal by its soul-functions in particular (PA I 1, 641a17–21). By contrast, the shape of a body, or of its bodily parts, is something that a living body and a dead body have in common. (By “dead body” I mean a body that has lost its form and is merely homonymous with a living one.) This could be taken as evidence that shape, dimensions, and size of a body are necessary and indeed explanatorily primary attributes of a living body, a composite of matter and form, but they do not form part of the essence or formal nature of a living body. If this is so, the reference to the proper substance of an animal kind at IA 8, 708a11 could be taken as a reference to the material composite substance. Alternatively, we can take attributes such as the shape, dimensions, and size of a body as merely necessary parts of the animal’s formal nature, and interpret the reference to the proper substance of an animal kind as referring to the formal nature or essence of an animal kind. Finally, let us consider the negative and positive teleological principles, namely the principle that nature does nothing in vain (NP) and the principle that nature realizes the best of the available possibilities (NP*).34 Both principles are introduced in the explanation of the absence of feet in snakes. The negative (weaker) principle is usually introduced in order to explain the absence of a characteristic or part from an animal kind, while the positive (stronger) principle is usually introduced in order to explain the presence of an actual characteristic part in a given animal kind. So these two principles are taken to introduce the thought that the standards that determine the development of animal parts are sensitive to a sort of optimal design.35 There are several questions concerning the 33

Similarly, in the case of the elephant, size and weight are taken as explanatorily primary in explaining further features of the animal kind (PA II 16, 659a26–27). 34 For a discussion of the way these principles are employed in the explanations of limb-bending in IA 12 and leg motion in IA 14 see the contributions by Spyridon Rangos (ch. 10) and Sarah Ruth Jansen (ch. 11) in this volume. 35 The reference to “the best” introduces the idea that animal parts satisfy some standard of perfection that is not absolute but relative to the specific nature of the kind to which they belong. In this



6 De incessu animalium 7–8



introduction and use of these principles that cannot be treated in full here. I am content to flag the main points of disagreement found in the secondary literature, and specifically those that can be illuminated by the use of these principles in IA 8. These questions concern: (1) the exact reference of Aristotle’s use of the term “nature,” i.e., whether it refers to a universal or cosmic nature; (2) whether the teleological principles (NP) and (NP*) are separate principles or are introduced as two parts of a single principle; (3) the reference to “the best” mentioned in the second principle; (4) the range of possibilities nature has at its disposal. Question (1) has been discussed, I believe, in a conclusive way, showing that “nature” here must refer to the specific formal nature of a kind, such as the snake, or at least as quantifying over a number of such specific formal natures when the principle is used generally.36 This agrees with the definition of nature in Phys. II 1 as an internal principle of change.37 I would like to add one additional positive reason for endorsing this interpretation. Aristotle invokes the analogy between craft and nature in Phys. II and elsewhere using several examples from different crafts. This suggests that each particular craft uses its own standards of what is best in light of the goal of the craft.38 By analogy with the case of the crafts, in which there is no universal craft that can determine what is best in each particular one, we should not assume that there is a universal nature that dictates what is best for the different kinds of animals.

sense they can be taken as satisfying some standard of optimal design, and correspondingly the study of such parts must invoke what Henry 2013 labels “optimality reasoning”: reasoning that appeals to some idea of optimal “design” in order to understand why things are the way they are (Henry 2013: 225). See also Lennox 2001b: 205–223; Leunissen 2010, and Andrea Falcon’s essay on IA 1–3 in this volume (ch. 3). 36 Henry 2013: 231. In this volume, Spyridon Rangos and Sarah Jansen follow the same interpretative line. However, Jansen expresses some doubts motivated by her observation that constraints on the possibilities that nature works with are, in some cases, determined at a more general level than the species or even the wider kind the animal belongs to. A compatible view is expressed by morel 2016: 26–28, who argues that the optimality principle entailed by the “nature does nothing in vain” maxim leads to a conception of nature over and above the individual animal natures, one that is governed by a general rule of economy and utility. 37 In Phys. II 1, Aristotle describes things that exist by nature as things that possess an internal principle of change and rest in themselves (Phys. II 1, 192b13–14). A little later, he adds that this formulation determines what “nature” is (Phys. II 1, 192b33). It is not certain that we should take this formulation as a definition in the strict Aristotelian sense. Still, it is an account that determines to some extent what nature is. For a defense of this view see Stavrianeas 2015. For an analysis of the Physics account of nature as a definition, see Kelsey 2003. 38 In the Physics, nature is compared to a doctor and a grammarian (Phys. II 8, 199b33–35) and to a ship-builder (Phys. II 2, 194b5–9). In the biological works, nature is compared to a painter (GA II 6, 743b20–25); a sculptor (GA I 23, 730b24–33); a carpenter (PA I 1, 641a8; GA I 22, 730b19–23; GA II 4, 740b25–741a3); and a housekeeper (GA II 6, 744b16–27).



Part III Interpretative Essays

With respect to question (2), Jim Lennox has argued that we have two principles functioning independently, or at least working at different stages of the explanations proposed (one explaining the absence of an organ, namely (NP), and another explaining the presence of an organ, namely (NP*)). Devin Henry, on the other hand, contends that we have a single principle. Henry seems to be right that the two principles are used together in IA 8, for, as we have seen, although the negative and the positive principles play different roles in the explanation, they must be employed together to understand the absence of feet in snakes and other snake-like animals. Again, we should not overlook that the two explanations have a different target. In IA 7 the target is to explain how footless, blooded animals move in the same way as all other blooded animals, despite appearances to the contrary. In IA 8, the target is to explain why some kinds are footless, taking snakes as a paradigmatic case. However, a full explanation of why such kinds are footless needs to refer to material from both IA 7 and IA 8. And this is probably the reason why the teleological principles (NP) and (NP*) are mentioned together as a unity in the opening lines of IA 8. This does not mean that the facts that each part of the principle explains are not different. (NP) explains the absence of feet, while (NP*) explains the presence of bends. The explanation of these two facts must be treated as two stages in the explanation of the absence of feet in snakes and in all snake-like animals. But these two facts are not independent from each other, since we cannot understand why having feet is pointless if we do not realize that their movement by means of bends is indeed a better and more efficient means of bodily displacement. So this suggests that (NP) and (NP*), at least in IA 8, function as a unified principle.39 The standard of goodness – the “best” in question (3) – must be conceived, as already stressed by Allan Gotthelf, not as an independent standard of goodness, but always as relating to the life or being of the animal.40 Our discussion of snakes and other disproportionately elongated creatures seems to confirm this result. However, as we will see below, there seem to be some general constraints concerning movement with respect to place in general, or progression in particular, that set 39

Note, however, that Aristotle refers to the two principles in the plural in IA 2, 704b12–18. The two principles are implicitly introduced in tandem, but in a different wording alluding to what is useful and what is useless, in IA 11, 710b32–711a2. At the end of IA 11, the two principles are again implicitly employed when Aristotle points out that nature does nothing contrary to nature but rather what accords with nature (711a2–7). See Rangos’ contribution in this volume (ch. 9) on the relation between the two principles. 40 Gotthelf 1988: 113–139; cf. Henry 2013: 235.



6 De incessu animalium 7–8



some standards independent of the specific nature of an animal kind. For instance, all footed animals need to have an even number of feet. Furthermore, as proposed in IA 3, all animals that move in place, either by jumping or by progression, need two instrumental parts in order to do so: one compressing and the other being compressed.41 This suggests that the “best” of the possibilities is not only determined by the particular nature of the kind in question but also by certain general principles that are necessary constraints applying more widely. So, let us turn to question (4) concerning the determination of the range of possibilities. Nature works with possibilities defined by several constraints that, ultimately, can all be traced back to the essence of the animal in question.42 In Aristotle’s explanation of the absence of feet in snakes and other blooded animals with elongated bodies, their proper nature – namely the elongated shape of their body – is a first constraint, relating to the particular formal nature of the kind. Now, as suggested above, the attribute of being blooded adds a further constraint. While this further constraint is set and explained at a different level of generality, it, too, belongs to the formal nature of the kind in question. The difference is that this constraint does not work at the level of the particular nature of snakes, but at the level of the wider kind. And it will have to be explained at that level. The same goes for the constraint that all moving animals must move by means of an even number of instrumental parts (IA 3, 705a19–21; 8, 708a21). The existence of constraints at such a high level of generality may give the impression that the “nature” that realizes the best of the possibilities must similarly be placed over and above the particular or the generic natures that we find at lower levels of generality. And although this does not entail a cosmic nature responsible for arranging the structure of animal bodies, it seems to leave room for such a hypothesis.43 If so, what Aristotle says in the IA does not count against the existence of a concept of universal nature. However, on the other hand, the fact that there are constraints at some high level of generality does not force us to assume that there is an operative nature at the same level. The natures of 41

See the discussion by Andrea Falcon of this principle as an application of the more general principle concerning motion and change in nature requiring an active and a passive element in every case of change (ch. 3 in this volume). 42 See Leunissen 2010: 129; Henry 2013: 237–238. Henry provides a detailed classification of the constraints delimiting the range of possibilities that nature works with. 43 Sarah Jansen points out (ch. 10 in this volume) that “‘nature’ (in the teleological principle) is not simply shorthand for particular essences, sub-kinds, or genera but, in some cases, encompasses broad biological functions that different kinds of animals share.” This use points, she argues, to a cosmic unity requiring a cosmic nature responsible for this unity, however thinly construed. For a similar interpretative line, see morel 2016: 26–28.



Part III Interpretative Essays

the various kinds may suffice for explaining the claim that nature does nothing in vain. In other words, the question concerning the level of generality of constraints does not determine the level of generality at which we have to introduce candidates for the role of nature. And IA does not seem to require the introduction of any natures other than those of the species and wider kinds of animals it mentions. Why Animals That Progress on Land Possess an Even Number of Feet [708a21–b4] Section II deals with the question of why it is necessary for animals that progress on land to possess an even number of feet. The claim can be formulated positively (footed animals must have an even number of feet) or negatively (footed animals cannot have an odd number of feet). In the course of the discussion of this question, Aristotle offers an argument for both the positive and the negative claim. However, the negative claim does not hold universally for all footed animals. The reason is that some footed animals, that is, the many-footed animals, can move with an odd number of feet (as the movement of mutilated many-footed animals actually shows). In their case, it is just more difficult to move with an odd number of feet. Hence, the negative claim must be qualified in the following way: it is either impossible or more difficult for footed animals to move with an odd number of feet. The positive claim must be qualified in an equivalent way: it is either necessary or better for footed animals to possess an even number of feet. Aristotle’s treatment of this topic begins from the case of animals that change or move in place by means of jumping.44 More precisely, he refers first to animals that move only by jumping and then to animals that move by jumping but also progress on land by means of feet.45 At first sight, this whole discussion (708a22–26) seems to be parenthetical.46 So let us see what Aristotle is trying to establish by first quoting the passage: 44

In IA 3 Aristotle considers jumping as one of the two possible ways in which an animal can displace its body, the other way being a movement part by part, as for instance in walking (3, 705a3– 6); and see Falcon’s comments on the importance of the consideration of the case of jumping in establishing the thesis that any type of bodily displacement requires the articulation of parts (ch. 3). 45 Aristotle first uses µεταβολὴ κατὰ τόπον (change in place) as a more general expression referring to the bodily displacement of all animals that move from one place to another (i.e., to those that jump and those that both jump and walk), and then uses πορεία (progression) to refer to the capacity of some animals to move by means of feet. For Aristotle’s careful use of these terms, see the interpretative essays by Andrea Falcon (ch. 3) and Klaus Corcilius (ch. 5). 46 Indeed, Michael of Ephesus does not comment on these lines.



6 De incessu animalium 7–8



For (γάρ), [1] all those animals that make their own change with respect to place by using only jumping do not need feet at all for this type of movement. However, [2] as for those that use jumping, though this kind of motion is not sufficient for them as they also need progression, it is clear that [3a] for some of them it is better in this way [sc. by having an even number of feet], while [3b] for others it would be altogether impossible to progress otherwise.

The γάρ introducing the first clause of the passage suggests that what follows explains the initial phrase of the chapter, that all animals with feet possess an even number. However, the remarks in [1], [2], and [3] do not directly show why an even number of feet is necessary for some animals. Rather, they seem to specify what kind of animals, among those that move in place, need feet for their movement. Aristotle distinguishes between animals that move by means of jumping from animals that not only move by jumping but also use a different mode of bodily displacement. Aristotle labels the latter “progression.” The animals in this second class are those that, strictly speaking, need feet; they are the only ones that appear to need an even number of locomotive organs. So in this remark, which should be read as introductory rather than parenthetical, Aristotle tries to clarify that his central claim does not apply to all animals that move from one place to another;47 rather, it applies to a specific kind of animals: the animals that move by means of what Aristotle labels “progression.”48 Aristotle’s main goal in the IA is the explanation of this kind of bodily displacement, and it is with respect to this type of movement that having an even number of locomotive organs is either necessary or best. However, as in IA 3, Aristotle needs to place his chosen focus, namely animals that engage in progression, in the context of the wider category of animals that move with respect to place, and this latter class includes not only animals that progress but also animals that move by jumping 47

In PA IV 6, 682a33–b3, Aristotle claims that insects that move by jumping, e.g., locusts and fleas, possess an even number of feet. Why does he not repeat this observation here? One reason may be that the fact that they do have an even number of feet for moving does not entail that they need an even number, as they would need were they animals that progress on land. 48 Aristotle seems to be using the term “progression” in a general way that applies not only to animals that progress on land but also to animals that progress in other media, e.g., water (see IA 3 and the remarks ad locum by Andrea Falcon). However, the claim introduced in the current section concerns feet, that is, the locomotive organs suitable for progression on land. Yet, Aristotle does not add the qualification here. So, either we have to understand that the qualification is implicitly assumed, and read “progression” as meaning “progression on land,” or we have to assume that what holds of animals that progress on land (they need an even number of feet) holds mutatis mutandis of all progressing animals (they all need an even number of locomotive organs).



Part III Interpretative Essays

– i.e., animals that move their whole body all at once. To return to the above passage, the animals that fall under the scope of the claim concerning the even number of feet are those mentioned in clause [2]. The concluding sentence, the apodosis [3], must refer to animals that both jump and progress on land by feet. This sentence holds that it is better for some of these animals to have an even number of feet, while for others it would be impossible to progress without them.49 Michael of Ephesus understands this sentence as referring to manyfooted animals – clause [3a] – and to either four-footed or two-footed animals – clause [3b].50 Indeed, this reading works well with the rest of the chapter. As we will see, Aristotle first establishes the claim that it is necessary for blooded animals to move with an even number of feet – that is, it is impossible for them to move with an odd number. This corresponds to animals under the scope of [3b]. Subsequently, he shows that for bloodless animals it is merely better to move with an even number of feet, since it is not impossible for them to move with an odd number of feet. At this point, we can turn to the two kinds of animals that progress on land by means of feet. Unlike jumping, which is the motion of the whole body all at once, progression takes place part by part – where “part” means locomotive part. In walking, the parts complete opposite functions: one part moves while another is at rest and supports the weight of the animal. Thus, for 49

But see the French translations of this passage by Pierre Louis and Pierre-Marie Morel. In both translations the two options range over the same case. Louis 1973: “il est evident que la marche se fait mieux avec des pieds en nombre pair, tandis qu’autrement elle est absolument impossible.” Morel 2013: “il est clair qu’il est préférable qu’ils possèdent les pieds en nombre pair, tandis qu’autrement il leur serait tout à fait impossible de marcher.” To begin with, this reading of the μέν-δέ clause is strange. Moreover, both scholars endorse Jaeger’s insertion of ἄλλως in line 26, which is unnecessary and is not found in Michael of Ephesus. Finally, it would be odd for Aristotle to claim that the presence of a part is both for the better and is necessary (necessary in the sense that its absence would make a function impossible). According to the PA I 1, 640a36–37, these are alternative ways of explaining the presence of a bodily part. The anonymous reader points out that the two types of explanation often go together (for instance, the explanation of respiration at PA I 1, 642a31–b2, of shedding of horns at PA III 2, 663b12–14, and of female births at GA IV 3, 767b8–15). Indeed, in such cases the explanandum happens both out of necessity (e.g., in the case of female specimens the semen is unable to master the maternal contribution) and for the better (specimens of both sexes are needed for the reproduction of human beings). However, the case we discuss here is slightly different: if it is merely better for many-footed animals to have an even number of feet, then it is because their locomotion is not impossible with an odd number. In the previous cases we have both a material causal chain (necessity) and a final causal chain (the good) converging to the same result. Here we have a result that is either a necessary and sufficient condition for locomotion or merely sufficient and optimal. (This does not mean that if it is necessary for a two-footed or footed animal to have an even number of feet, this is not also something that happens because it is better.) See the same disjunction between what is necessary and what is for the better in line 711b30, with the comments by Spyridon Rangos in ch. 9. 50 Michael of Ephesus, In IA 151.17–20.



6 De incessu animalium 7–8



animals that move part by part, their locomotive parts must come in pairs that alternate in moving and supporting the weight of the animal. This seems to be the positive argument, or the argument for the positive claim, in this section. But Aristotle reaches this conclusion only after rehearsing an argument for the negative claim. This argument shows the impossibility of progressing on land with an odd number of feet.51 Indeed, he needs such an argument to show not just that it is good or better for animals to walk with an even number of feet, but for the stronger claim that it is impossible, at least for some kinds of animals, to walk without an even number of feet. The argument for the negative claim is that an animal with either one foot or three feet would have no adequate support as it progresses on land. While an animal with a single foot would fall right away, an animal with three feet would have support only in one of the two opposed pairs, which would result in a fall when the animal tries to move forward.52 Now, it turns out that this latter claim is not quite accurate for a four-footed animal, which even if it were mutilated could manage to move by limping. But we may accept Aristotle’s point, given that limping in such a case would be much more accentuated when compared to the case of a mutilated many-footed animal. The latter case will be the focus of the next section. In fact, Aristotle wants to emphasize the difference between these two groups of animals by claiming that an even number of feet is necessary for two-footed or fourfooted (blooded) animals, but it is present merely for the better in the many-footed (bloodless) animals. Before turning to the case of many-footed animals, however, we need to stress one last point concerning the necessary presence of an even number of feet in one kind of animal and the presence for the better in another kind of animal. Both conclusions can be established only if we assume the operation of the two teleological principles introduced in Section I: that nature does nothing in vain (NP), but always realizes the best of the possible alternatives (NP*). In the case of blooded animals, an odd number of feet would be pointless, because the animal would have locomotive organs for walking without being able to do so. In the case of many-footed animals, the animal would be able to progress forward, but At this stage of his argument, Aristotle switches from talking about πορεύεσθαι (progressing) to talking about βαδίζειν (708b1). I try to capture this by speaking of “walking.” When in IA 14 Aristotle answers the question why animals move their feet diagonally he claims that if animals were to move with both their right or left legs at the same time they would fall, presumably because he thinks it impossible to find balance on only the left or right row of feet (712b1–2). See Jansen’s discussion in chapter 10 of this volume.

51

52



Part III Interpretative Essays

its walking would be seriously compromised. So if nature does nothing in vain but always realizes the best of the available possibilities, and if an odd or an even number of feet are equally feasible alternatives for footed animals, an even number must be the alternative realized in nature. The Case of the Many-Footed Animals [708b4–19] This final section expands on the second claim made in the previous section: for many-footed animals,53 it is (a) merely better, and (b) not necessary to possess an even number of feet for progressing on land. Aristotle does so by first observing that an even number of feet is not necessary for progressing on land (establishing (b)), and then by showing that an even number of feet is better than an odd number (establishing (a)). With respect to many-footed animals that lack a limb by mutilation, Aristotle observes that they are still able to progress on land (πορεύεσθαι),54 but that they do so imperfectly. He says that they are still able to move forward because they possess such a great number of feet on each side. By this he may mean that the right or left side that lacks one foot still possesses a sufficient number of feet so that the animal is able to compensate for the loss. What happens in this case is that the rest of the feet on the same side drag the mutilated limb along. And this makes for their imperfect manner of bodily displacement. Thus, Aristotle says that this mode of locomotion is a dragging (ἔφελξις), which is not a proper mode of progressing forward.55 Having shown that it is possible for the many-footed animals to move with an odd number of feet, Aristotle shows that it is better for them to possess an even number of feet (claim (a) above). Given that nature realizes the best of the available possibilities, establishing (a) explains why many-footed animals possess an even number of feet. Having an even number of feet is a better solution than having an odd number of feet, since the animals in question will possess an equal number of limbs on each side, and thus will be able to distribute the weight when moving between their right and left 53

Aristotle mentions the centipede, which elsewhere he acknowledges to branch into two kinds: one that lives on land and one that lives in the sea (HA II 14, 505b13). All centipedes possess an even number of feet. See HA I 4, 489b21–23. Moreover, they all continue to live and move when divided (HA IV 7, 532a5; Resp. 3, 471b23). 54 This supports claim (b) above. 55 Jaeger deletes the words ἀλλ’ οὐ βάδισις (708b10), which are also missing in Michael’s text. If his motivation is that this claim seems to deny that mutilated many-footed animals walk, while it was earlier asserted, the deletion is unnecessary. Aristotle may be denying not the claim that mutilated many-footed animals walk or progress forward, but rather the claim that the mutilated foot walks; indeed, it is merely dragged along.



6 De incessu animalium 7–8



side. By contrast, having an odd number of feet instead would entail that one of the two sides will have an empty space in one of its two rows of legs, and thus the animal will progress by oscillating on one side or the other. The discussion of the case of many-footed animals concludes the topic of the necessary number of locomotive organs for both blooded and nonblooded animals. Aristotle will now move on to examine the limbs themselves, by focusing on their articulation and bending. Summary IA 8 answers two basic questions. The first, which is part of [Q3] in the initial agenda of the IA – why some animals are footless – continues the discussion of footless animals that move by bending in IA 7. Aristotle explains why there exist such animals, justifying the absence of feet by means of his principle that nature does nothing in vain, but always the best from among the possibilities available to the nature of the organism in question. The principle manages to explain successfully an exceptional case in the animal kingdom, namely footless animals. Hence, it proves to be explanatorily powerful and, as a result, it also gains in plausibility. The second question, which is [Q4] in the original list of IA 1, deals with the question why all footed animals have an even number of feet. Aristotle gives in fact a double answer. (a) On the one hand, two-footed and fourfooted animals could not progress at all with an odd number of feet. Thus, if they had an odd, instead of an even, number of feet they would not be able to walk or stand, and, one may add, the possession of feet would be in vain. (b) Many-footed animals, on the other hand, although they are able to progress even with an odd number of feet, are better off with an even number. And since the latter possibility is better, and, one may add, nature realizes the best of the possibilities for the being of the animal in question, their nature equips them with an even number. In this way we can see how Aristotle’s teleological principles ((NP) and (NP*)) may be implicitly at work in answering [Q4], and how they can provide a basis for syllogisms that can derive the necessity of the presence of particular features in animal kinds.

chapter 7

De incessu animalium 9 Aristotle’s Mathematical Kinesiology: The Case of Bending Christopher Frey

Introduction The focus of IA 9 is bending’s role in animal locomotion. The claim that initiates the chapter concerns the relationship between bending and rest. Aristotle argues that if nothing were at rest, neither bending nor straightening could occur. But the majority of the chapter concerns the relationship between bending and locomotion. Aristotle defends the claim that there could be no walking, swimming, flying, or any other variety of animal locomotion without bending. That locomotion involves bending may seem obvious on perceptual grounds. But the import of Aristotle’s discussion resides more in the form of argument he employs in support of the conclusion – namely, a narrowly geometrical argument – than in the conclusion itself. It is important, nevertheless, to note that if Aristotle is able to establish both of the chapter’s principal claims, they imply, by a hypothetical syllogism, a central Aristotelian tenet, namely, that there could be no animal locomotion if nothing were at rest. I will break up the chapter into four sections. Section 1 comprises Aristotle’s definition of bending and his argument that bending requires a point at rest. In section 2, Aristotle provides a geometrical argument for the claim that animal locomotion requires bending. To properly interpret this argument, we must broach several methodological issues. Most notably, we will discuss the role that principles of pure mathematics can play in explanations of physical phenomena. The role geometry plays in this argument is quite unlike the appeals to geometry one finds elsewhere in the biological works. I will argue that Aristotle’s explanation I would like to thank Andrea Falcon and Monte Johnson for their comments on earlier drafts, and Jim Lennox for pointing me toward several helpful passages. Thanks are also due to the other participants of the 2016 De incessu workshop at the University of Patras for their many helpful questions and suggestions. All translations are my own unless otherwise stated.





7 De incessu animalium 9



belongs neither to pure geometry nor to observational biological science, but rather to an applied mathematical science I call mathematical kinesiology. Section 3 discusses bending at joints other than the knee and provides a second geometrical depiction of animal locomotion. Section 4 extends Aristotle’s findings to animals that crawl, swim, fly, and “ooze.” I aim to better understand how Aristotle conceives bending, to explore and evaluate the ways he appeals to bending in his explanations of animal locomotion, and to situate the methodology on display within the subtly interrelated network of sciences Aristotle countenances.

Bending and Rest [IA 9, 708b21–26] Aristotle defines bending (κάμψις) as “the change from what is straight to what is curved or angled” (708b22–23). Movements in the contrary directions, from what is curved or angled back to what is straight, are not instances of bending, but rather of straightening. Elsewhere, Aristotle defines bending as “a change to the convex or the concave without a change in the length” (Mete. IV 9, 386a1–2). This alternative definition differs in two ways from the definition offered in the IA, but is, despite these differences, ultimately compatible with it. First, the Meteorology’s definition mentions only changes from what is straight to what is curved (i.e., convex or concave). It is clear that the IA considers changes to what is angled instances of bending; though snakes, caterpillars, birds, and many fish move, either entirely or in part, by curving their limbs and bodies, most animals, especially those that walk, move by bending their limbs at joints. But the purpose of Mete. IV 8–9 is to catalog, explain, and provide examples of the passive capacities that differentiate homoeomerous natural bodies. Changes from what is straight to what is angled do not appear elsewhere in this discussion, so, given how comprehensive Aristotle intends the list of passive capacities to be, it is safe to conclude that Aristotle considers such changes instances of bending.1 Second, the Meteorology’s definition restricts bending to changes that do not involve a change in the length of that which bends. Though Aristotle’s definition in Aristotle’s list comprises eighteen pairs of passive capacities: (i) capable or incapable of solidification, (ii) meltable or unmeltable, (iii) softenable or unsoftenable by heat, (iv) softenable or unsoftenable by water, (v) bendable or unbendable, (vi) breakable or unbreakable, (vii) capable or incapable of fragmentation, (viii) impressible or unimpressible, (ix) mouldable or unmouldable, (x) squeezable or unsqueezable, (xi) tractile or non-tractile, (xii) malleable or non-malleable, (xiii) fissile or non-fissile, (xiv) cuttable or uncuttable, (xv) viscous or friable, (xvi) compressible or incompressible, (xvii) combustible or incombustible, and (xviii) capable or incapable of giving off fumes (Mete. IV 8, 385a14–18).

1



Part III Interpretative Essays

the IA does not mention this feature of bending, the claim that the lengths of limbs neither increase nor decrease during locomotion plays an important role in the arguments Aristotle goes on to make. So Aristotle’s considered view, which he employs in IA and is compatible with his other discussions, is that bending is a change from what is straight to either what is curved (i.e., concave or convex) or what is angled without a change in the length of that which bends. Aristotle’s argument that bending and straightening require that something be at rest is quite brief. That if nothing were at rest no bending or straightening could occur, is evident from what follows. For bending is the change from what is straight to what is curved or angled, straightening is the change from either of these to what is straight. In all such changes, the bending or straightening must necessarily be relative to one point. (IA 9, 708b21–26)

Beyond the definition of bending, the only substantive claim Aristotle makes is that both bending and straightening are necessarily relative to one point. It is this single point that is at rest. It is an open question what this resting point is. There are at least two ways to bend, say, a straight pipe, into an arc. Either someone can hold one of the endpoints fixed and exert pressure on the other endpoint, or one can exert pressure on both of the endpoints simultaneously. If the arc that results from the second kind of bending is a circular arc, there are two options for what the point at rest could be (Figure 7.1a). It could be either the midpoint of the pipe or the theoretical center point, i.e., the focus, which determines the arc by being equidistant from every point the arc comprises. But the theoretical option seems incorrect for two reasons. First, there is no single point that determines other varieties of arc – elliptical arcs are determined by a pair of foci and parabolic arcs are determined by a point, its focus, and a line, its directrix (Figures 7.1b and c). Second, if the result of bending is neither convex nor concave, but is an angle, the point at rest is clearly the angle’s vertex (Figure 7.1d). In the discussion that follows, Aristotle treats the bending of limbs in footed animals to be the principal case. And when a limb, say, a leg, bends, the point at rest is in the limb’s central joint, viz. the knee. That this is so is evident from Aristotle’s other discussions of the movements of limbs. Aristotle is clear that, if one of the parts of an animal be moved, another must be at rest, and this is the purpose of their joints; animals use joints like a center, and the whole member, in which the joint is, becomes both one and two, both



7 De incessu animalium 9

(a) circular arc

(b) elliptical arc

(c) parabolic arc

(d ) angled line



Figure 7.1  Possible resting points for a bending straight and bent, changing potentially and actually by reason of the joint. (MA 1, 698a17–21, trans. Farquharson)2

If bending that results in an arc is analogous, the point at rest would be the arc’s apex or vertex.

Aristotle’s Geometrical Argument [IA 9, 708b26–709a7] Having shown that bending requires a point at rest, Aristotle goes on to offer a remarkable geometrical argument for the claim that animal locomotion requires bending. Moreover, without bending there could not be walking or swimming or flying. The reason is that, since footed animals stand and take their weight alternately on one or the other of their opposite legs, as one leg strides forward the other must necessarily be bent. For the opposite legs are naturally of equal length, and the one that is under the weight must be a kind of perpendicular at right angles to the ground. When, then, one leg strides forward, it becomes the hypotenuse of a right-angled triangle. Its 2

Strictly speaking, the point at rest is not the joint as a whole, but a point within the joint. Aristotle says that “something initiates motion instrumentally when the starting point and the end point are the same, for instance in a hinge joint; for here the convex is the end point and the concave the starting point (for which reason the latter is at rest and the former is moved), and though differing in account, they are inseparable in magnitude. For all things are moved by pushing and pulling; consequently, it is necessary, just as in the case of a circle, for something to remain fixed and for the motion to begin from there” (DA III 10, 433b21–27, trans. Shields, slightly modified; cf. Meta. VII 16, 1040b9–14). For a discussion of the several types of joint, see PA II 9, 654b15–23; HA I 15, 493b30–31; and De spir. 7, 484b22–26.



Part III Interpretative Essays square then is equal to the square on the other side together with the square on the base. But since the legs are equal, the one at rest must bend either at the knee or, in any kneeless animal that walks, at some other joint. (IA 9, 708b26–709a4)

This argument warrants an extended treatment. I will analyze the argument, highlight how unique this sort of geometrical argument is within Aristotle’s biological works, and situate the argument within Aristotle’s more general accounts of geometrical, physical, and biological explanation. The Argument Though Aristotle’s stated conclusion is that walking, swimming, and flying require bending, the argument only discusses footed animals that walk. Aristotle does ultimately establish that swimming and flying (as well as other varieties of animal locomotion) require bending, but he does not extend this argument’s results to other ways of progressing until 709a24. The initial premise that footed animals alternate their weight upon their opposite legs is more subtle than it may initially appear. We can see what it means to take one’s weight on a leg by looking at a passage in the Mechanica. Mech. 30 concerns what one must do to rise from a seated position.3 It is impossible to stand up from a chair if one keeps both one’s lower legs and one’s back perpendicular to the ground (just try!). In this position, one’s weight, that is, one’s center of gravity, is not upon one’s feet; if the chair were removed one would fall backwards onto the ground (Figure 7.2a). To rise, one must form two acute angles by leaning one’s head forward and moving one’s feet inward toward the body. Only then will the seated individual be “at right angles to the ground” where this means that he will “have his head in the same line as his feet” (Mech. 30, 857b29–31). That is, it is only when a line that is roughly perpendicular to the ground can be drawn through the head and the feet that one will be in a position to take one’s weight on one’s feet and stand (Figure 7.2b).4 3

Though scholars disagree about Mechanica’s authenticity, the explanation present in chapter 30 contains the very same technical terminology that Aristotle employs in the geometrical argument under consideration and, as we will see, its meaning in the former clearly sheds light on the latter usage. 4 “Why is it that when people rise from a sitting position, they always do so by making an acute angle between the thigh and the lower leg and between the chest and the thigh, otherwise they cannot rise? […] Let A be the head, AB the line of the chest, BC the thigh, and CD the lower leg. Then AB, the line of the chest, is at right angles to the thigh, and the thigh at right angles to the lower leg, when a man is seated in this way. In this position, then, a man cannot rise; but to do so



7 De incessu animalium 9



A

B

G

C

D (a) weight not on feet

C

B

F (b) weight on feet

Figure 7.2  Rising from a seated position In his geometrical argument, Aristotle assumes that walkers take their entire weight on one leg at a time. Consistent with the discussion of rising from a seated position, whichever leg “is under the weight must be a kind of perpendicular at right angles to the ground” (708b31–32). That is, the walker’s center of gravity will remain over the trailing leg. If the leg that extends forward remains straight, the walker’s legs will form a right triangle with the ground. The lead leg will be the hypotenuse of this triangle and the trail leg upon which the walker’s weight rests will be the side of the triangle perpendicular to the ground. So if the length of one’s leading leg is five units and the walker steps forward three units, the length from the walker’s hip to the ground must be four units (Figure 7.3b). This is simply an application of the Pythagorean theorem to the triangle the walker’s legs form. But a typical walker’s legs are of equal length, and when one walks, the length of one’s legs neither increases nor decreases. So the only way for the distance from the hip to the ground to be four units is for the trailing leg to bend at the knee (Figure 7.3c). Aristotle’s geometrical argument does not depend directly upon observable phenomena. But he does appeal to such phenomena to support his conclusion. Though incomplete in all presently available manuscripts, he must bend the leg and place the feet at a point under the head. This will be the case if CD be moved to CF [and if AB be moved to GB], and the result will be that he can rise immediately, and he will have his head and his feet in the same straight line; and CF will form an acute angle with BC [and GB will form an acute angle with BC]” (Mech. 30, 857b21–858a2, trans. Forster with ­additions).

Part III Interpretative Essays



5

5

(a)

5

4

5

106°

3

3

(b)

(c)

Figure 7.3  Bending the trail leg when walking Michael of Ephesus provides a full characterization of the evidence to which Aristotle appeals (Michael of Ephesus, In IA 154.18–155.7) and his characterization allows us to fill in what is missing in Aristotle’s text as follows: This is shown by the following fact: if a human being were to walk on the ground along a wall the line described would not be straight but zigzag, because it would go lower when the human being bends and higher when it stands and raises itself. (IA 9, 709a4–7)

When a walker’s legs are farthest apart, the head is lower than the position it occupies when the walker’s legs are together. As the walker alternates between these orientations, the head descends and rises.5 If walkers move as Aristotle’s geometrical argument depicts, the line it describes would be zigzag (Figure 7.4a). In actual fact, the line it describes is a series of curves (Figure 7.4b). 5

Michael of Ephesus supplies even more evidence: “If someone were to walk along a wall as high as his eye-level, and someone else the same height were to stand on the other side of the wall, he would not see continuously the top of the head of the one walking, but when the walker extended his leg, he would not see the top of his head because of his becoming shorter, but when the advanced foot is pulled up, he would see the head because it is lifted and gets up to its height again” (Michael of Ephesus, In IA 155.1–7, trans. Preus).



7 De incessu animalium 9

(a) zigzag



(b) series of curves

Figure 7.4  The position of the head when walking Geometrical Argumentation in the Biological Works Aristotle’s natural philosophy is replete with appeals to mathematical theorems, including those of geometry.6 But almost all appeals to geometry within the biological works belong to one of three classes: (i) analogy, (ii) illustration, and (iii) non-geometrical description of observable morphology. Analogical appeals to geometry do not involve direct applications of geometrical claims to observable phenomena. Instead, they aim to show that the relationships in which various biologically relevant items stand resemble, in important respects, those that obtain among geometrical relata. For example, in DA II 3, Aristotle argues that the way in which a lower soul is present in the soul of a comparatively sophisticated organism, e.g., the way in which a nutritive soul is present in an animal’s perceptual soul, is analogous to the way in which a simple figure is present in a comparatively complex figure, for instance the way in which a triangle is present in a quadrangle (DA II 3, 414b20–415a1). But souls do not possess any other features of figures and even the one respect in which souls and figures are analogous is not realized identically in their respective subjects. Moreover, in analogical arguments of this sort, the conclusions are not themselves established geometrically. Illustrative appeals to geometry do involve the application of geometry to observable phenomena. But these applications are heuristic and can be eliminated without altering the non-geometrical arguments they serve to illuminate. For example, in MA 1, Aristotle argues that when a limb bends at a joint, one point within the joint is moved and another point within the joint remains at rest. He elucidates this conclusion by saying that this is “just as would happen if, on a diameter, AD were to remain at rest and B moved so as to bring about AC” (MA 1, 698a22–24; Figure 7.5). This geometrical illustration is helpful, but we can ultimately dispense with it, because the non-geometrical arguments Aristotle provides before 6

See Hussey 1991: 213–242 for a discussion of several examples.



Part III Interpretative Essays C

D

A

B

Figure 7.5  The geometrical argument in MA he states the claim illuminated by the illustration sufficiently establish it as a conclusion.7 Appeals to geometry within descriptions of observable morphology involve genuinely explanatory applications of geometrical concepts. But these uses do not involve properly geometrical arguments. For example, in PA IV 5, Aristotle concludes that sea urchins must have five “ova,” five teeth, and five stomachs, and the fact that their bodies are spherical is an ineliminable premise in his argument. But the explanatory role that the sphericity of a sea urchin’s body plays in Aristotle’s argument does not reside in any narrowly geometrical features of spheres. Instead, their sphericity matters because Aristotle maintains that the distribution of objects on spherical bodies must be balanced (PA IV 5, 680b4–681a3). The argument in IA 9 does not fit comfortably into any of these familiar patterns. Its conclusion, like the conclusions of the arguments we have just discussed, concerns the proper subject matter of biology insofar as the argument purports to establish a general truth about the physiology and activity of footed animals. But Aristotle’s argument about bending differs from the others in being a genuine explanation that conforms to 7

That Aristotle views this geometrical example heuristically is further confirmed by his insistence that he is not considering the illustrative circle’s center to be realized physically (and thereby “potentially and actually now one, now divided”), but is only considering it mathematically (and thereby “indivisible in every respect”) (MA 1, 698a26–28; cf. Mem. 1, 450a1–6). MA contains two other illustrative appeals to geometry, in chapters 9 and 11. The best candidate for a genuinely geometrical argument occurs in chapter 7. Aristotle argues that small qualitative changes due to perception can have large consequences for bodily movement because the parts of the body involved in motion stand to the perceptual organ, the heart, as the circumference of a circle stands to its center, and “it is not hard to see that a small change occurring at the center makes great and numerous changes at the circumference, just as by shifting the rudder a hair’s breadth you get a wide deviation at the prow” (MA 7, 701b24–28, trans. Farquharson; cf. Phys. VIII 4, 254b28–30, Mech. 848a10–19). But though Aristotle could employ this theorem concerning circular motion in a non-illustrative way, it is not obvious that its presence in this passage (and its application to ships) is anything but illustrative or analogical.



7 De incessu animalium 9



the methodological prescriptions of the Posterior Analytics.8 In addition, though the definition of bending is not a principle of geometry, the argument’s major premise, which employs the Pythagorean theorem, unequivocally is. The argument’s explanatory force depends upon the truth of this narrowly geometrical principle and its presence is, in contrast to the other appeals to geometry we have discussed, ineliminable. Before turning to a more thorough exploration of the positive role geometrical principles play in Aristotle’s argument about bending, we should mention a related respect in which this argument differs from the other geometrical appeals one finds in the biological works. That is, the argument’s premises are barely tethered to observable phenomena. The only physiological assumptions Aristotle makes are that walkers (i) do not progress by moving all of their limbs at once; (ii) progress in a way that does not involve their falling down; (iii) have limbs that do not change their length when they progress; and (iv) have limbs of equal length. The first pair of assumptions is arguably included in the account of walking, the third is included in the account of bending, and the last could be dispensed with, though doing so would require a more complex argument for the same conclusion. Moreover, though Aristotle goes on to cite observable phenomena in support of the argument’s conclusion, the argument itself depicts a gait that is wildly at variance with what anyone can easily observe. It involves no bending at the ankle or the hip. The lead leg remains completely straight. And it requires that one squat over one’s trail leg (Figure 7.6). In his notes to Michael of Ephesus’ commentary, Anthony Preus remarks upon the gulf between the gait Aristotle’s geometrical argument depicts and the observable facts. He says, “it is more than a little odd that Aristotle argues this way, since the facts of walking are not that difficult to observe, and in fact the Greek artists represent walking, running, and other gaits in men and animals (horses, dogs, and so on) with 8

The demonstration is as follows. [P1] Length of perpendicular side is n 2 − m 2 belongs to Right triangle constructed upon perpendicular line, a, of length n with hypotenuse of length n, parallel side of length m, and perpendicular side a line segment of a. [P2] Right triangle constructed upon perpendicular line, a, of length n with hypotenuse of length n, parallel side of length m, and perpendicular side a line segment of a belongs to The legs of a footed-animal with limbs of length n progressing a distance of length m by walking. [C1] Length of perpendicular side is n 2 − m 2 belongs to The legs of a footed-animal with limbs of length n progressing a distance of length m by walking. Given Aristotle’s definition of bending, his conclusion follows trivially from C1. [C2] Bending belongs to The legs of a footed-animal with limbs of length n progressing a distance of length m by walking.



Part III Interpretative Essays

Figure 7.6  Walking like Aristotle Copyright © Robert Crumb, 1968. I’d like to thank the artist for kindly granting me permission to use this illustration.

considerable accuracy” (Preus 1981: 64). Are we to believe that Aristotle is unaware of what walking actually looks like? It seems clear, given how obvious it is that actual walkers do not look the way Aristotle’s argument depicts, that Aristotle is not attempting to capture the observable manner in which walkers progress. Presumably this divergence is a feature, not a bug, of the kind of geometrical argument Aristotle gives. But what, precisely, is Aristotle doing in this argument and how are we to understand geometrical arguments for biological conclusions more generally? Geometrical Arguments for Biological Conclusions According to Aristotle, the sciences are autonomous. That is, the principles of each science concern the per se attributes of the objects that belong to their proprietary domains, e.g., arithmetic’s subject genus is number and geometry’s subject genus is spatial magnitude. Aristotle maintains a proscription against arguments whose premises and conclusions concern objects that belong to the subject genera of distinct sciences.9 So one cannot use the principles of one science, say arithmetic, to prove claims about the proper objects of another science, say, geometry. If 9

APo I 7. A nice discussion of Aristotle’s injunction against such “kind-crossing” is Hankinson 2005: 23–54.



7 De incessu animalium 9



this proscription were completely general, it would exclude the very possibility of geometrical arguments for biological conclusions. But Aristotle allows for an important class of exceptions. If the subject genus of one science “comes under another” (θάτερον ὑπὸ θάτερον), then the former science is said to be subordinate to the latter. In such cases, it is possible (and often necessary) to use the superordinate science’s principles as premises in the subordinate science’s arguments. For example, mathematical optics is subordinate to geometry, mechanics is subordinate to stereometry (i.e., solid geometry), and mathematical harmonics is subordinate to arithmetic.10 In each of these cases, claims whose proper home is pure mathematics play explanatory roles in applied fields of inquiry. The applied mathematical sciences that are subordinate to the several fields of pure mathematics also have observational sciences subordinate to them. For example, observational optics (which Aristotle says includes iridology – “the study of the rainbow”) is subordinate to mathematical optics, and acoustical harmonics is subordinate to mathematical harmonics.11 The purpose of these observational sciences is to collect the facts that the applied mathematical sciences purport to explain. So Aristotle seems to accept that some of the sciences are organized into tripartite hierarchies.12 In such hierarchies, applied mathematical science occupies a position intermediate between entirely mathematical inquiries and entirely observational inquiries. But there is a sense in which all three of a given hierarchy’s sciences study the same objects. Pure mathematics studies the geometrical properties that natural bodies possess, but does so in a way that prescinds entirely from their natural realization. Aristotle says that “natural bodies have planes, volumes, lines, and points,” and makes it clear that, in 10

See APo I 7, 75b13–20; I 9, 76a16–25; I 13, 78b37–79a16; and Meta. XIII 3, 1078a5–21. Cf. footnote 17 for some reservations about the subordination of mechanics to stereometry. 11 APo I 13, 78b35–79a13. Interestingly, when Aristotle actually discusses rainbows and other phenomena that result from the reflection of optical rays, his arguments are paradigmatic instances of applied mathematical science. He uses geometrical principles to explain the observable, physically realized, geometrical features of optical phenomena, e.g., that a halo is always a circle or a segment thereof and that a rainbow’s arc is never greater than a semi-circle (Mete. III 3, 372b34–373a19; III 5, 375b16–377a11). Johnson 2015: 163–186 offers a thorough analysis of these arguments and, on Johnson’s interpretation, Aristotle’s argument for the perfect circularity of lunar and solar halos is a geometrical argument for a meteorological conclusion that mirrors, in many important respects, this essay’s interpretation of the way in which the argument under consideration is a geometrical argument for a biological conclusion. 12 The claim that subordinate sciences are arranged triadically is not entirely uncontroversial. Hankinson, Lennox, and McKirahan discuss several difficulties that attend this interpretation (Hankinson 2005: 47–50, Lennox 1986: 42–44, and McKirahan 1978: 213–215).



Part III Interpretative Essays

addition to the student of nature, “the mathematician also studies these things, but not as they are limits of natural bodies, nor does he study accidents as accidents of such bodies.”13 Aristotle summarizes the relationship between pure mathematics and applied mathematical science succinctly when he asserts that “geometry studies natural lines, but not as natural, whereas optics studies mathematical lines, but as natural, not as mathematical.”14 Both applied mathematical science and observational science study the same properties of natural bodies that pure mathematics studies, but unlike pure mathematics, they study them as natural. But observational science does not employ mathematical principles in its explanations and is thereby unable to establish why the geometrical facts it determines to hold of natural bodies are as it observes them to be. When it comes to naturally realized geometrical properties, “it is for the observational scientists to know the fact (τὸ μὲν ὅτι) and for the [applied] mathematical scientists to know the reason why (τὸ δὲ διότι).”15 So applied mathematical science focuses on the mathematical properties that we perceive to be realized per se in natural bodies and considers them both as the mathematical properties they are and as the properties of the bodies that possess them. Insofar as the properties are mathematical, one can apply a relevant class of general mathematical principles to them. Insofar as these properties are naturally realized, the application of these mathematical principles can explain observable facts about the objects that possess them.16 Given this framework, where are we to situate Aristotle’s geometrical argument in IA 9? It appears to be a clear instance of applied mathematical science. The argument’s object is the triangularity of a walker’s legs, but it does not consider it in abstraction; the argument considers the triangularity only insofar as it is realized naturally in a hypothetical walker. Moreover, the argument applies a general principle of geometry, the Pythagorean theorem, and it is this principle that explains various facts about the legs of walkers. The observationally-oriented biologist or 13

Phys. II 2, 193b24–25 and 31–34. Phys. II 2, 194a9–11; cf. Meta. XIII 3, 1078a2–4. Lennox captures this dual character when he notes that in the explanations of applied mathematical sciences “[t]he middle term picks out the description of the natural object in virtue of which it has a certain mathematical property; that property is a per se property of a natural kind qua being a mathematical kind” (Lennox 1986: 41). 15 APo I 13, 79a3–4; cf. APr I 30, 46a19–21. 16 This is, of course, only the barest sketch of Aristotle’s account of the applied mathematical sciences and of scientific subordination more generally. Distelzweig 2013: 85–105, Hankinson 2005: 23–34, Johnson 2015: 163–186, Lennox 1986: 29–51, and McKirahan 1978: 197–220 include especially rich discussions of these and related matters. 14



7 De incessu animalium 9



student of physiology can tell us that a walker’s legs have certain mathematical properties, but they cannot use the presence of these mathematical properties to explain any other facts they observe to be true of walkers. The facts Aristotle’s geometrical argument explains hold of all walkers’ legs, but they do not hold because they are properties of legs. These facts hold because the legs of walkers, when they progress, are ­triangular. Only someone engaged in the relevant applied geometrical ­science will be in a position to explain these facts properly. Let us call this applied mathematical science mathematical kinesiology.17 The arguments of mathematical kinesiology employ principles of pure geometry to explain facts that have the naturally realized geometrical properties of progressing animals as their cause. These arguments needn’t depict the manner in which walkers typically progress. Aristotle is not attempting to describe how walking must be realized physiologically. Rather, the purpose of Aristotle’s argument is to place constraints on any physiological realization of walking. He proceeds as he does because it allows him to employ readily recognized theorems about right triangles. But his result can be extended easily to gaits that involve the movement of the walker’s center of gravity to a position not directly over his trailing foot, to gaits that involve bending at the ankle and the hip, to gaits in which the lead leg bends, and even to the gaits of animals without knees. 17

It is unclear how mathematical kinesiology is related to the other sciences Aristotle recognizes. Given that Aristotle’s definition of bending is not a principle proprietary to mathematical kinesiology – as we have seen in this chapter (above, “Bending and Rest”), Aristotle defines bending and employs this definition in explanations of inanimate bodies in Mete. IV as well – it is likely that mathematical kinesiology is a branch of a more comprehensive science. A plausible suggestion is that it is a branch of kinematics, which is itself a branch of mechanics. But this proposal is not without its difficulties. First, Monte Johnson defines a mechanical explanation as “a demonstration proper to the science of mechanics (primarily of simple machines, such as the screw, lever, pulley, pump, etc.)” and defines a mechanistic explanation as “a demonstration directly modeled on a mechanical explanation (for example, a biomechanical explanation)” (Johnson 2017: 127). Aristotle’s geometrical argument in IA 9 is not modeled on explanations of simple machines and so, strictly speaking, is neither a mechanical nor a mechanistic explanation. On the other hand, arguments that would presumably belong to mathematical kinesiology, such as the argument we discussed concerning how people arise from a sitting position (Figure 7.2), belong to the Mechanica. Second, if mathematical kinesiology is ultimately a branch of mechanics it is unclear to which pure mathematical science it is subordinate. Aristotle says that mechanics is subordinate to stereometry (APo I 13, 79a1), but elsewhere he says that it is subordinate to geometry (APo I 9, 76a24). The arguments Aristotle gives in IA 9 employ principles of planar geometry, not solid geometry. But this would be evidence for mathematical kinesiology being a branch of a different applied mathematical science as much as it would be evidence for mechanics being subordinate to geometry. Perhaps Aristotle follows Plato and maintains that stereometry is in some sense posterior to geometry (Rep. VII, 528 B–E). Though this would, in a sense, dissolve the issue, whether Aristotle does view geometry as prior (and precisely how he would conceive this priority) is far from clear.



Part III Interpretative Essays

Other Joints and a Second Geometrical Argument [IA 9, 709a8–24] Aristotle already extends the results of his geometrical argument to animals that walk without bending at the knee when he says that, “since the legs are equal, the one at rest must bend either at the knee or, in any kneeless animal that walks, at some other joint” (709a2–4). In the chapter’s next section, he provides several concrete examples of animals that walk in just this way. It is, however, possible to move even if the leg has no bend, as when children crawl. (This is the old account of the movement of elephants, but it is untrue.) Such a crawling movement involves a bending in the shoulders or the hips. But nothing could progress upright in this way continuously and safely, but would only move like men in the wrestling schools who convey themselves forward through the dust on their knees. (IA 9, 709a8–14)

Aristotle rejects elephants as an example of such locomotion; though their forelimbs bend outward at the knee, they do in fact have knees (Figure 7.7).18 But the other examples suffice to show such locomotion is possible. What is it though, to bend at the hip?

Figure 7.7  The articulation of elephants All walking requires some bending at the hip, namely, the bending which advances the lead leg forward. But the bending that is relevant to Aristotle’s argument would be a bending that effectively increases the length of the lead leg or effectively shortens the length of the trail leg. 18

At PA II 16, 659a29, Aristotle says that elephants’ legs bend with difficulty and that their primary purpose is support. Elsewhere, however, he reiterates his claim that they can in fact bend at the knee (HA II 1, 498a8–13).



7 De incessu animalium 9



Only then would walking conform to the geometrical constraints that Aristotle has shown govern walking. Neither limb actually lengthens or shortens. But one can bend at the hip in a way that changes the distance between one’s feet and the center of one’s hip (Figure 7.8).

B

C

E

E

A

D

A

D

Figure 7.8  Bending at the hip If one bends both legs at the hip, using the center of one’s hip as a fulcrum, the foot of the leg extending from the lower side of the hip (AE) will be farther away from this center point than the foot of the leg extending from the higher side of the hip (DE). So if one bends one’s hips in this way, the two changes in length can satisfy Aristotle’s geometrical constraints. At this point, Aristotle returns to the geometry of walking. He provides a second geometrical argument that depicts walking in a way that more closely resembles how we perceive it to occur. But this second argument follows a recapitulation of his first. The reason is that the upper portion of the body is big, so the leg must be long; consequently, there must be a bending. Since a standing position is perpendicular , if that which moves forward does not bend, it will either fall as the right angle becomes less, or else it will not advance at all. If one leg is at right angles and the other is advanced, the latter will be at once equal and greater; it will be equal to the leg at rest and also to the hypotenuse of the right-angled triangle. Therefore, that which goes forward must bend, and while bending one, extend the other leg, and incline forward at the same time and make a stride and remain above the perpendicular; for the legs form an isosceles triangle and the head goes lower when it is perpendicular to the triangle’s base. (IA 9, 709a14–24)

Part III Interpretative Essays



In this discussion, the walker begins not by advancing their lead leg, but by leaning forward (Figure 7.9b). If someone standing upright begins to lean forward, their center of gravity will no longer be over either of their feet. If they lean forward too far, they will fall onto their face. A

B

B

(a) standing upright

(b) leaning forward

E

A

C

B (c) lead leg advances

D

B

(d ) stable isosceles triangle

Figure 7.9  Moving forward with bending In order to prevent falling over, the walker must advance one leg and extend the trail leg, presumably by bending at the ankle, in order to continue forward momentum and facilitate the placement of the lead foot on the ground (Figure 7.9c). Aristotle does not say why the lead leg must bend. It is not required by his geometrical constraints. But it is required in order to overcome the friction between the lead foot and the ground.



7 De incessu animalium 9



Without bending at the knee or hip, moving one’s leg forward can only be accomplished with great effort (Figure 7.10a). At a minimum, one must shift one’s weight to the trail leg. But such shifts of weight involve bending at the hip. The need to bend the lead leg is only exacerbated if one is leaning forward. For in this inclined position, moving one’s lead leg without bending it would initially move it downward into the ground (Figure 7.10b).

(a)

(b)

Figure 7.10  Moving forward without bending At midstride, with one’s lead leg now on the ground, the legs form an isosceles triangle (Figure 7.9d). The walker’s weight is not above either of the feet. But a line can be drawn from the head to the midpoint of the isosceles triangle’s base that is perpendicular to the ground, and having one’s center of gravity midway between one’s feet is a stable position. This depiction also conforms to the observable phenomena to which Aristotle appeals in support of his first geometrical argument. For the walker’s head will be at a lower position when his legs form an isosceles triangle than it is when he stands upright.

Other Varieties of Animal Locomotion [IA 9, 709a24–b19] Aristotle concludes the chapter by extending the result that walking requires bending to other varieties of animal locomotion. He begins with the locomotion of footless animals. Some footless animals advance by undulations (this happens in two ways: for some, the bending is upon the ground, such as with snakes, while for



Part III Interpretative Essays others it is up and down, such as with caterpillars) and undulation is bending. Others move by oozing, such as what are called earthworms and leeches. (IA 9, 709a24–29)

The three cases he discusses are (i) undulations in which what bends remains on the ground, (ii) undulations in which the bending lifts part of the animal off the ground, and (iii) oozing (ἰλύσπασις).19 This last variety of locomotion is curious. Aristotle says, “For these advance with one part leading the way, and then drawing the remainder of the body to them, and in this way they change from place to place” (709a29–31). Both earthworms and leeches are segmented, and this has led many to model this form of locomotion on either telescopic extension, in which the segments would overlap when the body is drawn together, or on concertina-like extension and compression.20 Neither of these suggestions is factually correct. We now know that earthworms move by means of peristalsis, that is, by alternating between (i) radially symmetrical contractions of circular muscles that wrap around each of their body’s segments, and (ii) contractions of longitudinal muscles that extend lengthwise through their cylindrical hydrostatic bodies. The worms first anchor their body’s posterior to the ground with small hair-like bristles called setae and contract the circular muscles located in their body’s anterior (Figure 7.11a and b). These contractions compress the anterior segments thereby pushing the fluid in their hydrostatic skeleton forward and elongating their anterior. The earthworm then anchors its body’s anterior to the ground and contracts its anterior’s longitudinal muscles (Figure 7.11c). The contraction of their longitudinal muscles reverses the process and makes their body’s anterior thicker and shorter thereby drawing the central segments of its body forward. By cascading these contractions along the length of their bodies, earthworms are able to progress anterogradely (Figures 7.11d and e). Though neither the contraction of circular muscles nor the contraction of longitudinal muscles involves bending as Aristotle defines it, there is at least one respect in which this is immaterial to our interpretation. For Aristotle would have rejected the contemporary account of earthworm The neologism ἰλύσπασις is derived from ἱλύς which means “mud” or “slime.” Though I ultimately disagree, Farquharson 1912 at least makes a reasonable choice when he translates ἰλύσπασις as “telescopic action” and notes the parallel to concertinas. Farquharson rightly criticizes the LS translation “wriggling,” “which is not a worm’s normal movement.” For the same reason, I maintain that the LSJ translation, “crawling,” is incorrect, though both Forster 1937 and Louis 1973 follow this recommendation.

19

20



7 De incessu animalium 9

A

B



C

D

(a) initial position

A

B

C

D

(b) anterior circular muscles contract

A

B

C

D

(c) anterior longitudinal muscles and central circular muscles contract

A

B

C

D

(d) central longitudinal muscles and posterior circular muscles contract

A

B

C

D

(e) anterior longitudinal muscles contract to final position

Figure 7.11  Diagrammatic representation of worm movement locomotion altogether because he infamously fails to appreciate that muscles play any significant role in bodily movement and instead attributes this role to sinews (νεῦρα).21 Nevertheless, this case remains problematic even on Aristotle’s mistaken view. First, sinews effect movement “through contracting and relaxing,” so their operation is not unlike that of muscular contraction.22 Second, and more important, if Aristotle’s 21

Aristotle thought that muscles were a type of soft flesh with two main functions: (i) protecting other parts of the body, and (ii) being a medium for the sense of touch. For a thorough and convincing discussion of Aristotle’s views on muscles and sinews, see Gregoric–Kuhar 2014. 22 PA III 4, 666b14–15: διὰ τοῦ ἕλκειν καὶ ἀνιέναι (cf. MA 7, 701b9–10).



Part III Interpretative Essays

understanding of oozing conforms even minimally to what is observable, then it will primarily involve the sides of the worm’s body becoming concave and then straightening.23 But each of these changes also involves a change in the body’s length. So even on Aristotle’s preferred account, oozing seems to be a case in which locomotion occurs principally by parts of the body changing their size or length, and this prevents these bodily changes from being instances of bending. Consequently, this appears to be an example of animal locomotion that does not involve, let alone require, bending.24 The two varieties of undulation Aristotle describes are not saddled with the same difficulties; they both require bending. For, it is clear that, if the two lines they form were not greater than the one, movement would be impossible for undulating animals. The reason is that, when the bend is extended, they would not have made any advance, if it subtended an equal line; as it actually is, the line is longer when it is extended, and then this part stays still and draws up the remainder. In all the aforementioned changes, that which moves advances by first extending itself straight and then by curving itself; it straightens itself with its leading part and curves itself in the parts which follow. (IA 9, 709a31–b7)

The two lines Aristotle mentions are the lines between the endpoints of the curve that undulating animals form and the vertex of this curve (Figure 7.12). Aristotle notes what should be obvious: the length of the sum of these lines, AB + BC, is greater than the length of the straight line between the curve’s endpoints, AC. One could interpret Aristotle as making a more substantive claim, namely, that each of the lines AB and BC, not their sum, must be greater than AC. On this second reading, if undulating animals are to progress, the angle or curve they produce must be particularly acute. But this is not necessitated by any geometrical 23

Though unlikely to have been written by Aristotle, the Problemata provides some evidence that Aristotle would countenance these changes in the sides of worms being instances of bending insofar as it asserts that hot water is a cause of wrinkles and describes wrinkles as a bending of the skin (Prob. XXIV 7, 936b10–12). 24 This kind of locomotion does, however, require that there be at least one fixed point at rest. Setae fix the worm’s posterior when the anterior contracts and advances and similarly fix the worm’s anterior when the worm’s posterior contracts and is brought forward to meet the rest of the body. So without some part of the body being at rest, earthworms cannot progress. Though Aristotle often allows for exceptions to his principles, especially in the biological works, this example is problematic insofar as (a) Aristotle cannot abandon his definition of bending without significantly undermining his previous arguments, and (b) the placement, without comment, of oozing among other varieties of animal locomotion he explicitly maintains involve bending makes it unlikely that he considers it exceptional.



7 De incessu animalium 9



B

A

C

Figure 7.12  Geometrical properties of undulation considerations, nor is it needed to show that undulation is bending, and Aristotle fails to supply any reason for thinking that undulation must (or even does) occur in this way. When an undulator’s body bends, the anterior end point remains fixed and draws the posterior toward it. When the body straightens, the posterior end point remains fixed and the anterior is extended forward. Aristotle’s final remarks concern animals that jump, fly, and swim. All jumping animals as well make a bend in the lower part of the body, and jump in this manner. So too flying and swimming animals progress, the one straightening and then bending their wings to fly, the other their fins to swim. Some of the latter have four fins and others, those with a longer shape, for instance eels, have two. These move by substituting a bending of the rest of their body for the missing pair of fins, as we have already said. Flat fish use two fins, and the flat part of their body instead of the second pair of fins. Really flat fish, like the ray, produce their swimming with the actual fins and with the outer periphery of their body, alternately bending and straightening. (IA 9, 709b7–19)25

According to Aristotle, the locomotion of birds and fish is largely analogous. Just as birds bend their wings to fly, so fish bend their fins to swim. And just as birds use their tails like the rudder of a ship to direct their flight (cf. IA 10, 710a1–4), so fish use their tails to direct their swimming. 25

Aristotle makes similar divisions among swimmers in the HA: “Of swimming creatures that have no feet, some have fins, as fishes: and of these some have four fins, two above on the back, two below on the belly, as the gilt-head and the bass; some have two only – to wit, such as are exceedingly long and smooth, as the eel and the conger; some have none at all, as the muraena and others that use the sea just as snakes use the dry ground – and snakes swim in water in just the same way. Of the selachia some have no fins, such as those that are flat and long-tailed, as the ray and the sting-ray, but these fishes swim by means of their flat bodies” (HA I 5, 489b24–32, trans. Thompson; cf. PA IV 13). On bending in birds, see also HA II 1, 498a29–31. Note that there are no explanations offered in this passage; it comprises only descriptions of swimmers’ features and points to some of the ways in which these features are correlated. It is, consequently, an example of observational science whose purpose is to know the fact (τὸ μὲν ὅτι) rather than to know the reason why (τὸ δὲ διότι).



Part III Interpretative Essays

As a matter of fact, the way most fish swim is exactly the opposite of what this second analogy claims. Movement in the tail causes the forward momentum of most fish and the principal function of the pectoral and ventral fins is to direct this motion. But despite these inaccuracies, jumping, flying, and swimming conform to Aristotle’s thesis that animal locomotion requires bending.

chapter 8

De incessu animalium 10–11 Flight and Two-Footedness Timothy Clarke

Introduction In IA 10, Aristotle continues his discussion of the role of bending and straightening in the different kinds of animal locomotion. This discussion began near the start of IA 9, with the claim that bending is necessary for progression on land,1 swimming, and flying (708b26–27). The remainder of IA 9 focused mostly on progression on land. First Aristotle gave a detailed explanation of why effective walking requires bending at the knee. He then discussed other kinds of locomotion on land, such as the motion of snakes and caterpillars. In IA 10, he now focuses on flying and swimming. (The discussion of flying and swimming actually started at the end of IA 9, at 709b9–19. In this respect the traditional chapter break is not in the ideal place.) In order to fly or to swim, an animal must advance part of its body into the external body – either the air or the water – which it then uses to move itself forward. Aristotle adds that there is a further way in which bending and straightening are involved in flight: birds bend their tails in order to maintain direction. There will be more to say about bending and straightening in the following chapters. In IA 12–13 and 15–17, Aristotle will address a series of questions about the directionality of limb-bending – for instance, why humans and birds bend their legs in opposite directions.2 Before tackling these questions, however, he returns in IA 11 to another of his main topics: the explanation of the number of feet in different kinds of animal. By the time we get to IA 11, Aristotle has already addressed this topic at length. He has given a general argument to show that blooded animals can move by no more than four points of motion.3 (In this context, a “point,” I would like to thank the participants at the conference in Patras for helpful discussion. Special thanks to Andrea Falcon, Christopher Frey, and Sean Kelsey. 1 In contrast to Frey (this volume, ch. 7), I take πορεία at 708b27 to refer to progression on land in general, and not just to walking. 2 Cf. IA 1, 704a17–18. 3 For this conclusion, see IA 7, 707a20–21.





Part III Interpretative Essays

σημεῖον, refers to a body part which makes contact with the ground, water, or air, and which the animal uses to move itself from place to place. Examples are feet, fins, and wings.) He has also explained why all footed animals have an even number of feet, and why snakes do not have any feet at all.4 However, there remains the question of why some blooded animals have only two feet. In IA 11, we get Aristotle’s answer to this question, with his explanations of the two-footedness of humans and birds. D E I N C E S S U A N I M A L I U M 10

I divide IA 10 into four sections. The opening section (709b20–26) deals with a potential objection to the theory of points of motion developed in IA 6–7. The second section (709b26–710a1) resumes the previous chapter’s discussion of the role of bending and straightening in locomotion, focusing specifically on flying and swimming. The third section extends the discussion of flight (710a1–15), offering a teleological explanation of the bird’s tail. The final section (710a15–b4) gives explanations of further aspects of avian morphology.

Birds and the Four-Point Theory of Locomotion [IA 10, 709b20–26] Back in IA 6–7, Aristotle had developed a general theory of the number of points by which animals move. The present chapter begins by considering a potential objection to that theory. If the theory is supposed to be that all blooded animals move by means of four points,5 then birds might seem to present a counterexample: But someone might perhaps be puzzled as to how birds move by means of four points, either when flying or when progressing on land, thinking that we said that all the blooded animals move by means of four points. But actually we did not say that; but rather that they do not move by more than four . However, birds would not be able to fly if their legs were taken away, nor progress on land if their wings were taken away, since nor even could a human being walk if it were not moving its shoulders a little. (IA 10, 709b20–26)

Birds are blooded animals, and yet they only seem to use two points when flying (their two wings). They also seem to use only two points 4

Both of these questions are addressed in IA 8. As is implied by IA 1, 704a11–12.

5



8 De incessu animalium 10–11



when moving on land (their two feet). Aristotle does not mention swimming, but web-footed birds seem to use only two points when swimming, too. So it seems that, regardless of how they are moving, birds move by only two points at a time. The answer to the objection is in two parts. Aristotle starts by clarifying the conclusion arrived at in IA 7, which (contra the imagined objector) was not that all blooded animals move by exactly four points, but that they do not move by more than four.6 This clarification is enough on its own to defuse the objection. If it should turn out that certain blooded animals move by fewer than four points, this is not a problem for the earlier argument, once we get clear on that argument’s conclusion. After making this initial response to the objection, Aristotle then goes on to claim that, in any case, birds actually do move by four points. They use their legs when flying, and their wings when walking, just as we humans move our shoulders (and arms) when we walk.7 One may wonder whether there is any blooded animal that moves by just two points. The likeliest candidates would seem to be bipeds (birds and humans), snakes, and two-finned and finless fish (eels, rays, and so on). Over the course of IA 7–10, Aristotle argues that each of these animals actually moves by four points.8 So his position appears to be this. The general argument of IA 6–7 establishes that blooded animals can move by no more than four points. But this general argument can be supplemented by the investigation of particular cases, which shows that all blooded animals move by exactly four points.9

Bending and Straightening in Flying and Swimming [IA 10, 709b26–710a1] Aristotle now resumes his discussion of the role of bending and straightening in locomotion: But all, at any rate, make their change by means of bending and straightening, as was said. For all go forward into what is beneath them, up to a point, that is, into what yields. (IA 10, 709b26–28) 6

This corresponds to IA 7, 707a20–21: “none of the blooded animals can move by means of more than four points.” Cf. also 707b5–7 (“the animals that are constituted according to nature in the highest degree move naturally by means of two or four points”) and IA 8, 708a13–14. 7 Cf. IA 5, 706a28–29. 8 On snakes and fish, see IA 7, 707b7–708a8; cf. IA 9, 709b11–19. 9 Cf. also HA I 5, 490a26–32 for a clear statement of the view that blooded animals move by exactly four points.



Part III Interpretative Essays

These two sentences have caused serious trouble for interpreters. On the face of it, the first sentence appears to say that all animals progress by bending and straightening.10 This, of course, seems true, at least if we exclude those animals that do not “progress” at all.11 The problem is that the explanation offered for this claim seems obviously false: “For all go forward into what is beneath them, up to a point, that is, into what yields.” The thing that is “beneath” the animal, the ὑποκείμενον, is the underlying body that the animal presses against so as to move itself forward.12 The problem is that not all animals move by going forward into this underlying body. Think of a mouse walking across a stone floor. In what sense is the mouse going forward into the stone? Previous translators have tried to alleviate this difficulty by rendering Aristotle’s προέρχεται εἰς as “go forward upon” rather than “go forward into.”13 This seems to give a more satisfactory sense, but it is a strained translation of the Greek. If Aristotle had wanted to say that all animals go forward upon the underlying body, we should expect ἐπί and not εἰς.14 A different approach is taken by Farquharson, who speculates that a crucial participle may have dropped out of the text, and translates accordingly: “all things progress by pressing upon what being beneath them up to a point gives way…”15 Neither of these approaches seems very appealing. Instead, I want to propose a different solution. The problem arises, I suggest, from a mistaken assumption about the scope of the two occurrences of “all” (πάντα and ἅπαντα) at 709b26 and 27. It is typically assumed that Aristotle here means to refer to all animals. But an alternative reading is available. I want to suggest that he is specifically taking up the discussion of flying and swimming that began at IA 9, 709b9. So when he says at 709b26–27 that “all” move by bending and straightening, he is not making a claim 10

If we interpret the sentence in this way, then the back-reference is most likely to IA 9, 708b26–27: “without bending there could not be progression on land, swimming, or flying.” For this view, see Kollesch 1985: 127 and Louis 1973: 160, n. 8. 11 For stationary animals see e.g. DA I 5, 410b18–20, HA I 1, 487b6–9, and PA IV 7, 683b4–9. Another exception may be those animals which move by “oozing” (ἰλύσπασις), such as earthworms. While Aristotle regards oozing as involving bending and straightening, there is a question as to whether he is right about this. For discussion, see Frey’s essay in this volume (ch. 7). 12 Cf. IA 3, 705a7–8: “what moves always changes [sc. in place] by supporting itself against what is beneath it.” 13 Forster 1937: “they all progress upon that which, being beneath them…”; Louis 1973: “tous progressent sur le substrat”; Preus 1981: “everything proceeds on a substrate”; Kollesch 1985: “alle bewegen sich auf dem Substrat.” 14 Cf. IA 3, 705a8–12. 15 See Farquharson’s note ad locum. Cf. also Morel’s translation, which follows Farquharson’s: “tous avancent en s’appuyant sur un support.”



8 De incessu animalium 10–11



about all animals; rather he is making a more restricted claim about all of the animals currently under discussion, that is to say, all flying and swimming animals.16 If this is right, the explanatory sentence at 709b27–28 no longer presents a problem. In the case of flying and swimming animals, the underlying body (the ὑποκείμενον) is the air or the water. All swimming and flying animals progress by advancing part of their body into (εἰς) this external body, which yields to the animal,17 and is then used as a support. Given that flying and swimming work like this, it follows that the bodies of flyers and swimmers need to be able to bend and straighten in the appropriate way. In the next lines, we can see Aristotle as drawing exactly this consequence: So that even if the bending does not come about in another part, nevertheless it is necessary for it to come about at least from where the origin is – the origin of the membranous wing18 in the case of whole-winged animals, of the feathered wing in the case of birds, and of the analogous part in the others, such as fish. And for some, like snakes, the origin of the bending is in the joints of the body. (IA 10, 709b28–710a1)

In flying and swimming, the bending must arise from the “origin” of the body part that advances into the air or water. Take a bird’s wing, for example: the bird advances its wing into the air, which it then uses to propel itself. For this to be possible, the wing needs to be flexible at its origin, that is, at the point at which it attaches to the rest of the bird’s body. Aristotle’s choice of examples bears out my proposal that he is talking specifically about flying and swimming animals in this section. He starts by mentioning the two main kinds of flying animals: “whole-winged” animals and birds.19 He then moves on to swimming animals, mention16

The back-reference at 709b26 (“as was said”) is therefore to IA 9, 709b9–19. Cf. IA 15, 713a13–15 for the characterization of air and water as “yielding” to the animal as it moves. Aristotle here uses two different words for “wing,” πτερόν and πτέρυξ, the former for the wings of whole-winged animals, the latter for birds’ wings. This usage is not fixed: he will also use πτερόν for birds’ wings (see e.g. 710a25, b32), and πτέρυξ to refer to wings in general (see e.g. 709b10). Usually I use “wing” to translate both words, but in this passage I use two different (over-)translations – “membranous wing” for πτερόν and “feathered wing” for πτέρυξ – to mark the difference in the Greek. 19 Aristotle distinguishes between “whole-winged” and “split-winged” animals (cf. PA IV 12, 692b13– 15; APo. II 13, 96b38–97a1). Birds’ wings are “split” in the sense that they are composed of discrete feathers. Insects’ wings, by contrast, consist of continuous membranes, and so are “whole” (cf. PA IV 6, 682b17–20). I assume that, strictly speaking, the class of “whole-winged” animals also includes bats (cf. PA IV 13, 697b10–11), although when Aristotle talks about “whole-winged” animals in the IA he seems to be thinking primarily of winged insects.

17 18



Part III Interpretative Essays

ing fish, the fins of which he takes to work in an analogous way to birds’ wings.20 Finally, he mentions snakes. At first glance, the reference to snakes may be thought to present difficulties for my proposal: snakes were earlier classified as land animals (IA 7, 707b27–28). However, I take it that Aristotle is here thinking of snakes qua swimmers. Unlike fish, snakes have no fins that they can extend into the water. Rather, they swim just as they move on land, by bending and straightening their bodies (cf. 708a1–2). So, in the case of snakes too, it is necessary for their being able to swim that they be able to bend and straighten their bodies; the difference is that, in their case, the bending and straightening does not take place at the origin of some locomotive limb, but arises from the vertebral joints.21

A Further Role for Flexion in Flight [IA 10, 710a1–15] By this point, Aristotle has defended his earlier claim that bending is necessary for progression on land, swimming, and flying. But there is more to be said about the role of flexion in locomotion. So far we have been concentrating on the flexion that is required for forward propulsion. In the next section, Aristotle considers a different way in which bending and straightening are involved in flight: And the tail in winged animals is for the purpose of keeping course in flight, just like the rudders in ships. And it is necessary for these too to bend at the point of attachment. This is precisely why certain animals do not move in a straight course, namely whole-winged animals and those among the split-winged animals whose tails are not naturally suited for the aforementioned use, such as peacocks and domestic fowl, and generally speaking the birds that are not flyers. For of the whole-winged animals in general, none has a tail, so that they are borne along just like a rudderless ship, and each of them collides with whatever it happens upon; in a similar way both the sheath-winged animals, like dung beetles and cockchafers, and the non-sheathed ones, like bees and wasps. And in the birds that are not flyers the tail is useless, for example, in purple gallinules22 and herons and all the swimming birds. But instead of the tail they fly by ­stretching out their feet, and they use their legs instead of their tail for the purpose of keeping course in flight. (IA 10, 710a1–15) 20

See IA 9, 709b9–11 and 18, 714b3–7. Other animals swim in a similar way, for example eels, which have a “snakelike” form: IA 7, 707b27–30. 22 For the identification of this bird (ὁ πορφυρίων) as the purple gallinule, see Arnott 2007: 197–198. 21



8 De incessu animalium 10–11



The section begins (at 710a1–4) by offering a teleological explanation of the bird’s tail:23 it is for the purpose of keeping course in flight. In the first of several nautical similes in this chapter, Aristotle compares birds’ tails to ships’ rudders. Just as rudders must be able to bend and straighten in order to keep ships on their course, so tails have to bend and straighten for this purpose in birds.24 This claim about the function of the tail is supported by considering two types of flying animal that lack a properly-functioning tail: (i) wholewinged animals, specifically winged insects, and (ii) birds whose tail is “useless.” The supposed inability of winged insects (dung beetles, cockchafers, bees, wasps, and so on) to fly in a straight line is explained by the fact that they do not have a tail which they can bend and straighten to control their motion.25 The rudder analogy is invoked again: lacking a tail, these insects move like ships without rudders, and tend as a result to collide with things in their path. (Note that the point here is not that winged insects are completely unable to direct themselves; a bee can obviously direct itself from the hive to the rosebush and back again. The point is about control: unlike birds, insects do not fly in a straight line.) This supports the earlier proposal about the function of birds’ tails. By considering flying animals which lack any tail at all, and reflecting on how their flight differs from that of birds, we will be able to support our claim about the function of the tail in birds. Additional support for this claim comes from considering those birds that are “not flyers” (τὰ μὴ πτητικά). The terminology here is a little confusing, but as the examples make clear, Aristotle does not mean to refer to flightless birds (such as ostriches) but to birds which, although they can fly, live around water or have a “terrestrial” (ἐπίγειος) way of life.26 Such birds typically lack large tails,27 and accordingly are less able to use their tails to keep a straight course when they fly. To compensate, 23

Aristotle thinks that similar things can be said about the function of the tailfin in fish: see PA IV 9, 685b21–23 and cf. IA 18, 714b7. 24 When Aristotle says that “it is necessary for these too (καὶ ταῦτα) to bend at the point of attachment,” I take “these” to refer to the tails of birds. He may mean that, like rudders, tails too must bend at the point of attachment; alternatively, he may mean that, like wings, tails too must bend at the point of attachment. 25 Cf. HA IV 7, 532a24–25. 26 See PA IV 12, 694a6–8. By “way of life” (βίος), Aristotle refers to a cluster of facts about how an animal lives: where it lives, what it eats, how it gets its food, and so on. He often appeals to an animal’s way of life in explaining its bodily features: the idea is that an animal possesses certain bodily features because these features contribute to its characteristic way of life. For more on this mode of explanation in Aristotle’s biology, see Lennox 2010: 239–258 and Gelber 2015: 267–293. 27 An exception to this is the peacock; Aristotle will return to this exception below (see footnote 29).



Part III Interpretative Essays

they stretch out their legs behind them, using their legs as other birds use their tail.28 This phenomenon again supports Aristotle’s claim about the function of the tail.

Birds and Winged Insects: Further Comparison [IA 10, 710a15–b4] In the final section of the chapter, Aristotle extends his comparison of birds and winged insects. The focus now is on differences in the speed and strength of their flight. These differences are again explained in terms of differences in their morphology: And the flight of whole-winged animals is slow and weak, due to the fact that the nature of their wings is not proportionate to the weight of their body, but instead this is great whereas their wings are small and weak. So, just as if a cargo ship were to try to make its voyage using oars, so in this way do these animals use flight. And also the weakness both of the wings themselves and of their outgrowth contributes something to what we mentioned. Among birds, the peacock’s tail is sometimes useless because of its size, and sometimes of no benefit because of the shedding.29 But birds are in the opposite condition to the whole-winged animals with regard to the nature of their wings, and especially the swiftest flyers among them. Such are the crook-taloned birds. For swiftness of flight is useful to their way of life. And the remaining parts of their body seem to conform to their proper30 movement: in all of them, their head is small, their neck is not thick, and they have a strong and sharp breast – sharp so as to be forceful, as if it were the prow of a lembos-type ship, and strong by virtue of the nature of its flesh, so that it is able to push away the air hitting against it, and does this easily and not laboriously. And the rear parts are light and come together again to a taper, so that they follow after the front parts without dragging the air because of their breadth. (IA 10, 710a15–b4)

Insects are relatively slow and weak flyers. The explanation for this is that they have relatively small wings and a relatively heavy body. Again, a nautical analogy is used: such animals are compared to a heavy cargo ship 28

On this phenomenon, cf. HA II 12, 504a33–35: birds with a large tail “fly holding their feet close to their belly; but the small-tailed birds fly with them stretched out.” Cf. also PA IV 12, 694b20–25. 29 This sentence (710a22–24) relates back to the previous discussion of birds’ tails, and is somewhat out of place in the present section. In the previous section, Aristotle was discussing birds with “useless” tails, and in particular small-tailed birds (such as herons) which have to use their legs instead of their tail to keep course when airborne. The peacock is another example of a bird with a tail that is useless for flying, but its uselessness is differently explained. The train of the male peacock is too big to be used in flight, and is annually shed. 30 Reading οἰκείαν. Jaeger emends to ὠκεῖαν (“quick”).



8 De incessu animalium 10–11



powered only by oars. Birds, by contrast, have relatively small, light bodies, and large, powerful wings. Particularly good examples are birds of prey (“crook-taloned birds”). Their way of life – the fact that they are hunters of fast-moving prey – requires them to be able to fly at great speeds.31 This in turn requires that they have particularly large and strong wings relative to the size of their bodies. Aristotle notes that other features of these birds’ bodies also contribute to their capacity for swift flight: a small head, a narrow neck, a strong and pointed breast, and light and tapered rear parts.32 This time the analogy is a lembos-type ship, a light and fast galley with a sharply pointed prow.33 This final section of IA 10 should be seen as an appendix to the foregoing discussion of the role of bending and straightening in flight. It naturally follows on from the previous section, where Aristotle had compared the flight of birds and insects in order to defend his teleological explanation of the bird’s tail. He now extends this comparison to offer explanations of further features of birds, namely their powerful wings and light, aerodynamic bodies. While these features were not explicitly mentioned in the agenda at the start of the IA, they clearly fall within the scope of the treatise’s inquiry into “the parts that are useful to animals for motion with respect to place” (IA 1, 704a4–5). We should recall that the initial list of agenda items in IA 1 was not supposed to be exhaustive; as 704b8–9 indicates, Aristotle also intends to explain other “related” (συγγενῆ) facts. D E I N C E S S U A N I M A L I U M 11

Having completed his general account of the role of bending and straightening in animal locomotion, Aristotle returns to the topic of the number of feet. In IA 11, he gives explanations of the two-footedness of humans and birds, as well as explaining some further related facts about these animals’ bodies. The chapter thus provides a partial answer to the sixth question on the agenda in IA 1: “why a human being and a bird are two-footed whereas fish are footless” (704a16–17). (The chapter does not provide a complete answer to that question, because it does not explain why fish are footless. This fact is explained later on, at IA 18, 714a20–b2.) The chapter can be divided into three sections. In the first section (710b5–17), Aristotle explains why humans are two-footed and why they 31

Cf. PA IV 12, 693b28–694a6. Cf. PA IV 12, 693b16–19. 33 See Casson 1971: 162–163. 32



Part III Interpretative Essays

have relatively large and heavy lower bodies. In the second section (710b17–30), he gives a contrasting explanation of the two-footedness of birds. In the final section (710b30–711a7), he explains why birds lack an upright posture comparable to that of humans.

The Two-Footedness of Human Beings [IA 11, 710b5–17] Human beings have distinctive bodily proportions: our lower bodies are larger and heavier (in relative terms) than the lower bodies of other footed animals. In Aristotle’s view, this is directly connected to our twofootedness. He explains both of these features – our two-footedness and our large lower bodies – as necessary consequences of the fact that we are the only animals that walk upright: But why it is necessary for the animal that is to walk upright to be twofooted, and for it to have the upper parts of its body lighter, and the parts set below these heavier, is evident. For only if organized like this would it be able to carry itself easily. This is why a human being, being the only upright animal, has legs that – in proportion to the upper parts of its body – are the biggest and strongest of the footed animals. (IA 11, 710b5–11)

An “upright” (ὀρθόν) animal is one whose body, when standing and walking, is in a straight line perpendicular to the ground. While other animals are upright at certain times and in certain situations,34 human beings are the only properly “upright” animal. We already know that, as blooded animals, humans cannot have more than four locomotive limbs. This follows from the fact that blooded animals cannot move by means of more than four points. Further, in IA 8 it was argued that all footed animals must have an even number of feet. This means that, for humans, the only options are two feet or four. Now, it is obviously hard to see how a properly upright animal could walk effectively if it had to use four legs to do this. Given that humans are upright animals, then, two legs are better than four. This is why nature, acting for the best, endows humans with two legs and two feet. Further, an upright biped needs to have relatively large and strong legs, since the weight of its body will be borne on only two supports.35 Aristotle does not here offer an explanation of why humans are upright. However, he does not consider this a brute fact. He gives a 34

For example, bears (HA VIII 5, 594b15–16) and hedgehogs (HA V 2, 540a3–4; GA I 5, 717b29–31). The latter point is also made at PA IV 10, 689b5–15; cf. also 690a27–30 on human beings’ large feet.

35



8 De incessu animalium 10–11



teleological explanation of human posture in the PA, where he suggests that an upright posture is a requirement for an animal with a divine and rational nature. The idea is that uprightness is necessary if we are to be able to think and reason effectively. If we were not upright, then too much weight would be pressing down on the heart, and this would impede our perceptual and cognitive faculties. So, given that we are essentially rational animals, it follows that we are also upright (PA IV 10, 686a27–32).36 In the next lines of the IA, Aristotle goes on to defend his explanation of why humans have heavy lower bodies. He does so by making an observation about infant development: And what happens in the case of little children also makes this evident. For they are unable to walk upright, because of the fact that they are all dwarfish, and have the upper parts of their body bigger and stronger in proportion to the lower parts. But as they get older the lower parts undergo more growth, up to the point at which they acquire the proper size, and then they produce with their bodies upright walking. (710b12–17)

At first, children are unable to walk upright. This is because they are “dwarfish”; that is to say, they have relatively large upper bodies.37 It is only when their legs become bigger and stronger that they start to walk.38 This observation is evidence for the causal connection between our ability to walk upright and the size and strength of our legs. Aristotle takes this as support for the explanation he has just offered for why adult humans have large and heavy lower bodies. A human being is an animal that walks upright on two feet, and doing this well requires large and strong legs.

The Two-Footedness of Birds [IA 11, 710b17–30] Because birds are not upright in the way that humans are, the explanation of avian two-footedness must be different. Here Aristotle appeals to two peculiarities of birds’ bodies – their weight distribution and (what he regards as) their special “ischium”: But birds, being light, are two-footed because of the fact that their weight is situated at the back, just as people make bronze horses that have raised up their front legs. But most of all the cause of their being able to stand, 36

These lines are quoted and discussed by Klaus Corcilius in his essay on IA 5–6 (above, ch. 5). For further commentary cf. also Lennox 2001c: 317–318 and Gregoric 2007: 92–93. An alternative, materialistic explanation of human posture is given in PA II 7. 37 On Aristotle’s terminology of “dwarfishness,” see Lennox 2001c: 318–319. 38 For this observation cf. also PA IV 10, 686b6–12.



Part III Interpretative Essays despite being two-footed, is their having an ischium like a thigh and so large that they seem to have two thighs, one in the leg before the joint, and one going toward this part from the rump. But it is not a thigh, but rather an ischium. If it were not so large, a bird would not be two-footed. For just as with humans and four-footed animals, the thigh and the rest of the leg would begin immediately from the ischium, being short. Then their whole body would be inclined too far forward. But as it is, being long, the ischium stretches up to beneath the middle of the belly, so that from here the legs, having been put under as a support, carry the whole body. (IA 11, 710b17–30)

Aristotle’s first explanation of the two-footedness of birds is that their weight is distributed more toward the rear of their bodies. If their weight were evenly distributed, or distributed more toward the front, they would be unable to stand on two feet without falling forward. The analogy is with bronze statues of horses rising up on their hind legs. Such statues would often have an additional support placed underneath the belly, but Aristotle must have in mind statues with counterweighted haunches instead of belly supports. Birds have a similar weight distribution to these bronze horses, and this is one reason why birds can stand and walk on only two feet.39 A second reason is that they have a special “ischium.” Aristotle is in fact thinking here of the bird’s thigh, which he misidentifies as its ischium.40 Despite the misidentification, he is right to say that the length and position of this bone contribute to the bird’s stability.41 If it were shorter, then the legs would be situated more toward the back of the animal, rather than at its center. The bird would then need two forelegs as additional supports, and so would have to be a quadruped and not a biped.42 Given the bird’s elongated “ischium,” however, additional legs are not needed. Aristotle suggests that this is the main reason why birds are able to stand on two feet. We might initially be puzzled as to how this counts as an explanation of why birds are two-footed. In effect, Aristotle has told us how it is that birds can function with just two feet. But the fact that an animal can function with a bodily feature F does not yet explain why it actually has bodily feature F. So has he neglected to give us the promised explanation of the two-footedness of birds? 39

I am grateful to Andrew Stewart for discussion of these lines. He says that it is “like a thigh,” but denies that it really is a thigh. Cf. HA II 12, 503b35–504a3. 41 On the stabilizing function of the bird’s “ischium,” see also PA IV 12, 695a1–b13. 42 Cf. PA IV 10, 686a32–b2 and IV 12, 695a6–7. In these passages Aristotle explains the forelimbs of quadrupeds by saying that they are for the purpose of supporting the weight of the animal’s body. 40



8 De incessu animalium 10–11



In response to this worry, I think it is important to notice the participial phrase in the first sentence of the present section: “But birds, being light, are two-footed because of the fact that their weight is situated at the back” (710b17–19). Why does Aristotle mention birds’ lightness here? Why is this relevant to their being two-footed? I suggest that we can unpack the idea as follows. Aristotle is presupposing that birds are by nature flyers.43 One consequence of this is that they need to be as light as possible. (Recall the comparative discussion of birds and winged insects at the end of IA 10: one reason why birds of prey are such effective flyers is that, relative to the size of their wings, their bodies are small and light – contrast a bee, say.) Other things being equal, the fewer legs an animal has, the lighter it will be. And so, given that a bird can function with just two legs, nature will actually endow it with two legs and not more. This is the ideal arrangement; additional legs would be suboptimal, but nature always acts for the best.44 Aristotle has now explained why humans and birds are two-footed. To summarize, human two-footedness is explained as a necessary consequence of our distinctive upright posture, which (as the PA tells us) is itself a consequence of our rationality. The explanation of avian two-footedness is different, given that birds are not upright in the same way. Aristotle here appeals to two bodily features that allow birds to function effectively with just two feet, namely their weight distribution and their long “ischium.” Given that birds need to be as light as possible, and so to have as few legs as possible, consistently with their having feet,45 these features serve to explain why birds have only two feet, and not more.

43

See e.g. PA IV 12, 693b13: “the ability to fly is in the substance (οὐσία) of the bird.” An alternative way for Aristotle to explain the two-footedness of birds would be to appeal to the four-point theory of locomotion. Blooded animals cannot move by more than four points. However, birds need two wings in order to be able to fly. So, if birds were to have more than two feet, they would accordingly move by more than four points – which would conflict with their nature as blooded animals. Aristotle argues in this way at PA IV 12, 693b5–15: “They [sc. birds] are two-footed of necessity; for the substance of the bird is that of the blooded animals, but at the same time that of the winged animals, and blooded animals do not move by more than four points. Accordingly, the attached parts are four – as in the other locomotive land-dwellers, so too in the birds. But four arms and legs are present in the one group, while in the birds, instead of forelimbs or arms, wings are a common feature; and in virtue of these they are able to stretch out, and the ability to fly is in the substance of the bird. So it remains for them to be, of necessity, twofooted; for in this way they will move, with their wings, by means of four points” (trans. Lennox, with minor modifications). 45 For the point that birds need to have feet, see IA 18, 714a21–22: they “cannot always remain up in the air, and so necessarily have feet.” 44



Part III Interpretative Essays

The Posture of Birds [IA 11, 710b30–711a7] In the next section, Aristotle argues that the foregoing considerations also explain why birds are not upright, or at least not upright in the way that humans are: And it is also clear from these things that it is not possible for the bird to be upright in the way that the human is. For, as birds now have their body, the nature of the wings is useful to them, but it would be useless to them were they upright, just as people paint Erotes with wings. (IA 11, 710b30–711a2)

This explanation appeals to the fact that if birds were upright – if their bodies resembled those of painted Eros figures – then their wings would be useless.46 The actual, non-upright posture of birds means that their wings are useful. What is the connection between birds’ posture and the usefulness or uselessness of their wings? This is not immediately obvious. One possible answer is offered by Michael of Ephesus, who suggests that the idea is that if birds were upright, then their wings would “drag on the ground … and in doing so would hinder them from walking well” (In IA 159.26–29). However, while this would no doubt be a design flaw of the wings of these hypothetical upright birds, it is not obvious why it would make their wings “useless,” as opposed to merely cumbersome. An alternative interpretation is therefore needed. We know from the first section of IA 11 that a properly upright animal must have large and heavy legs – this is necessary if such an animal is going to be able to carry itself effectively. However, for birds to be effective flyers, their bodies need to be light (see 710b18). In particular, birds need to have light lower bodies, since effective flight requires powerful wings and thus a relatively large and heavy upper body. (Again, recall the earlier comparison of birds and insects at IA 10, 710a15–b4.) Now, if birds were upright in the way that humans are, they would need to have relatively large and heavy lower bodies. This would make their wings useless; they would be unable to get off the ground. This is why birds are not upright. An upright posture would conflict with the bird’s essential nature as a flying animal. This interpretation is confirmed, I suggest, by how Aristotle goes on to explain his winged Erotes analogy: For, at the same time, it is evident from what has been said that neither a human nor anything else of this kind of form can be winged, not only 46

Aristotle is referring to the kinds of Eros figures we see in Classical vase-painting, which are typically winged youths. The representation of Eros as a winged infant becomes common in the Hellenistic period.



8 De incessu animalium 10–11



because it will then move by means of more than four points, despite being blooded, but also because the possession of wings would be useless to them in moving naturally. But nature does nothing contrary to nature. (IA 11, 711a2–7)

In these final lines of the chapter, Aristotle gives two reasons for why nothing human-like in form could have wings, and hence for why in reality there are no animals like the Eros figures we see in paintings. His first reason is that such animals would be blooded animals that move by more than four points: two arms, two feet, and two wings. This would conflict with the earlier claim that blooded animals can move by at most four points. The second reason is that if winged humans (or winged human-like animals) did exist, then their wings would be useless – comparable to the wings of our hypothetical upright birds. Aristotle puts this by saying that the possession of wings would be “useless to them [sc. the winged human beings] in moving naturally.” I suggest that the thought here is that if an animal has wings, then we can infer that one of its “natural motions” is flight, given that wings are for the sake of flying. But given their large and heavy lower bodies, the wings of these hypothetical winged humans would be useless for flying. So their wings would be “useless to them in moving naturally.” This means that these creatures would have bodies that are “contrary to nature” (παρὰ φύσιν), in the sense of impeding them from engaging in their natural motion qua winged animals. This conflicts with the principle that “nature does nothing contrary to nature.” This is why there are no such animals in real life. So the same factors which explain why birds are not upright also explain why there are no winged humans or human-like animals. An animal with wings could not possibly be upright, because uprightness requires a relatively heavy lower body, and such a lower body would render the wings unfit for purpose. I would like to conclude by considering the teleological principle on which this final argument is based: “nature does nothing contrary to nature.” This is Aristotle’s only explicit appeal to this principle. It should be distinguished from the principle that “nature does nothing in vain.” In IA 2, Aristotle had identified the latter as one of the main principles that will guide his inquiry. As stated there, the “nothing in vain” principle says that nature does nothing in vain but always what is best from among the possibilities for the substance (οὐσία) of each kind of animal (704b15–17).47 47

This is Andrea Falcon’s translation. There are two ways of understanding the syntax of the sentence. One option is to take τῇ οὐσίᾳ with τῶν ἐνδεχομένων, and as limiting the range of the possibilities



Part III Interpretative Essays

For any given animal, nature can endow that animal with a range of possible bodily features.48 Given the kind of animal it is, some of these features will suit it better, some worse. The “nothing in vain” principle says that nature always acts so as to bring about the best possible outcome for the animal, given the kind of animal that it is. As we saw above, this idea plays a crucial (albeit implicit) role in Aristotle’s explanations of the twofootedness of humans and birds. Aristotle does not spell out what he means when he says that “nature does nothing contrary to nature.” However, the context suggests the following interpretation: nature never endows an animal with bodily features that would impede it from engaging in its natural activities (where its “natural activities” are those activities that are determined by its essence, that is, by the kind of animal that it is).49 If this interpretation is correct, then while the two principles are distinct, they are nevertheless closely related. If an animal were to be endowed with bodily features that impede it from engaging in its natural activities, this would presumably be a suboptimal outcome for the animal. (This is assuming, of course, that there will always be a possible outcome in which the animal is not impeded from engaging in its natural activities.) Hence, if nature always brings about the best possible outcome from among the available options, it follows that it will do nothing “contrary to nature”: it will not endow the animal with features that impede it from engaging in its natural activities. So even though the two principles are distinct, the “nothing contrary to nature” principle is subordinate to, and entailed by, the “nothing in vain” principle. Thus, here at the end of IA 11, Aristotle is not appealing to a wholly new explanatory principle. Rather, he continues to be guided by the principles he laid down at the start of the treatise. referred to. Alternatively, one can take τῇ οὐσίᾳ with τὸ ἄριστον: “what is best for the substance.” For the latter construal see e.g. Lennox 2001b: 206. Cf. also IA 8, 708a9–12 and 12, 711a18. 48 On the question of how to understand the “nature” that is the subject of this principle, see especially Lennox 2001a: 182–204. On Lennox’s interpretation, Aristotle is speaking generally about the formal natures of individual organisms. 49 So understood, the principle is far from trivial. Contrast Preus 1981: 177.

chapter 9

De incessu animalium 12–13 Limb-Bending and Natural Teleology Spyridon Rangos

Introduction According to the introductory chapter of IA, the work as a whole is divided into two parts. In the first part (IA 2–10), the general questions about progression [QQ1–5] are explored and answered, while in the second part (IA 11–19) Aristotle deals with more specific questions [QQ6– 11], including some (such as those encoded as [Q*] and [Q**] in Sarah Jansen’s chapter, and the classification of possible combinations of ­limb-bending in IA 13) that have not been explicitly announced or prefigured in IA 1. IA 12–13, in particular, explore the question of the bending of bodily parts responsible for locomotion in blooded animals that progress on land by means of feet. Aristotle’s main aim in these chapters is to explain why limb-bending in footed and blooded animals happens the way it does. Aristotle assumes that the precise way limb-bending occurs in those animals needs some explanation. The question (or rather set of questions) about the different ways in which animals bend their limbs in order to progress has been announced in IA 1 ([QQ7–10]). Although neither there (IA 1) nor here (IA 12) does Aristotle explicitly say why this is an interesting question (or rather set of questions) to raise, it is not hard to see why he thinks it is. As a matter of observation, animals progress in many different ways. Some of them walk, others crawl, others jump, others fly, and still others swim. Even within each of these categories, there is variation. Not all walking animals walk by means of the same number of feet. Some are four-footed, others are two-footed, and still others are many-footed. Moreover, all four-footed animals (with the notable exception of elephants, of which Aristotle is well aware; IA 13, 712a11, cf. HA II 1, 498a5) bend their I am very grateful to the editors of this volume as well as to the anonymous reader for their most helpful comments, corrections, and suggestions for the improvement of this essay. Needless to add, I take full responsibility for errors and inadequacies that may remain.





Part III Interpretative Essays

front and back legs in opposite directions (forward and backward, respectively), and of two-footed animals human beings bend their legs in one way (i.e., forward) while birds do so in another (i.e., backward). Such a variety of actual ways of walking and, more particularly, such a variety of actual ways of limb-bending seems to be in need of an explanation. One might claim that actual limb-bending is the result of pure chance. Empedocles, for instance, would have explained the fact that there are differences in limb-bending by saying that the limbs of the relevant animals bend the way they do because they have been accidentally joined together in a way that allows or necessitates this particular bending in the present world-arrangement; in other world-arrangements they might have been, or will possibly be, joined together in other ways so that other ways of limb-bending would then be actual (DK 31 B 57–62; cf. Aristotle, DC III 2, 300b25–31, Phys. II 4, 196a20–24). In Empedocles’ view, although the joining of animal parts is the result of the connecting power of Love, this does not mean that Love joins them the way it does with a set purpose in mind. All imaginable combinations are equally possible and all were, are, or will be equally unintended. Some of them may be fitter for survival and some less fit or not fit at all (cf. Aristotle, Phys. II 8, 198b29–32). But since Love does not have any specific concerns apart from the overarching concern to bring all things together and thus extend the application of her power as far and long as possible, the survival of the individual or the species is not among her concerns (not to mention concerns about the best kind of life for each individual or ­species). One might think that an appeal to chance is a reasonable assumption to make about the actual state of affairs with respect to animal limb-bending. In Aristotle’s time people actually made such assumptions and proposed corresponding explanations crucially based on the notion of chance. Aristotle is well aware of them. He writes, for instance, “Empedocles claims that the majority of animal parts have been made by chance” (Phys. II 4, 196a23–24). In Aristotle’s mind, to invoke chance in order to explain what occurs always or for the most part in nature is not to explain but to explain away. For to answer a “Why is it so?” question with a “Because it chances to be so” answer is to evade the issue at hand. Aristotle believes that in what occurs always or for the most part there must be a reason that is valid, and a corresponding cause that is operative, always or for the most part. In Aristotle’s mind chance cannot be such a reason or cause: an appeal to chance testifies, rather, to the absence of a reason or cause (in the strict sense of the term). For Aristotle, chance is an accidental, not a proper, kind of cause (Phys. II 5,



9 De incessu animalium 12–13



197a12–14), and its specific domain is limited to human affairs where choices play a crucial role in determining what is made on purpose and what not (Phys. II 6, 197b1–13, 20–22). Therefore, there is no chance in nature. However, there is something that is the equivalent of chance in nature, and this other accidental cause that is to be found in nature is called “spontaneity” (Phys. II 6, 197b13–20, 32–35). But the very notion of spontaneity as devised by Aristotle implies a whole domain of natural facts that have to be explained in a way that does not involve spontaneity, since spontaneity is meant precisely to account for the exceptions to the rule (where the rule is understood as what occurs always or for the most part in nature) and, therefore, for those rather rare natural events (or things in nature) that come to be for no specific reason. In Aristotle’s view, fixed natural patterns, including the fitness of animal parts for their function, cannot be adequately accounted for by means of spontaneity (Phys. II 8, 198b10–199a32). And since limb-bending belongs to the machinery whereby animals normally progress, it cannot be adequately accounted for by means of what other thinkers might call “chance” and Aristotle calls “spontaneity.” The reasons Aristotle seeks to establish in the chapters I am going to discuss are teleological, and the corresponding causes are final. Aristotle is not primarily interested in what we might call, perhaps a bit anachronistically (cf. Mech. 1, 847a18–19), the “mechanics” of limb-bending, i.e., in what should be materially involved in joints, bones, sinews, and flesh if the relevant parts are to do their job well. As the beginning of the treatise indicates (IA 1, 704a5–6), his interest lies in answering a διὰ τίνα αἰτίαν (“Due to which cause?”) question, understood as a τίνος ἕνεκεν (“For the sake of what?”) investigation. In IA 12–13 Aristotle seeks to establish the reasons or final causes that explain why it is best for the relevant kinds of animals to bend their limbs (both legs and arms or wings) the way they do. And he provides some specific arguments meant to show the advantages of the actual kinds of limb-bending in footed and blooded animals over other imaginable possibilities. Those arguments, and their precise wording, will be analyzed from a close-reading perspective in the main part of this chapter. Aristotle’s explanation of limb-bending in footed and blooded animals is based on his famous “Nature does nothing in vain” principle. Since the extended formulation of the teleological principle is mentioned three times in the rather limited size of IA as a whole (in IA 2, 8, and 12, plus a fourth time in a presumably condensed form at the end of IA 11), and the modern bibliography on the issue is steadily growing, a discussion of



Part III Interpretative Essays

what is actually involved in Aristotle’s principle is relevant here. Three distinct questions, or sets of questions, will be raised in this essay: (i) whether Aristotle has in mind two related principles or just one principle formulated in two different versions; (ii) whether the principle says something about how nature actually works (and is, therefore, an ontological principle that can be used as a major premise in demonstrative syllogisms) or is a merely heuristic device for detecting the presence or absence of relevant traits (and is, therefore, an epistemological assumption or working hypothesis for the zoologist to make without an actual correspondence to the operations of nature itself ); and (iii) whether the nature referred to in the principle is specific or global, how it manages to achieve its ends, and what are its limitations. According to a modern suggestion that has initiated a very interesting discussion in recent years, Aristotle uses the “negative” formulation “Nature does nothing in vain” (abbreviated as NP, according to Lennox’s initial suggestion) in order to account for the absence of traits in particular kinds of animals, and the “positive” formulation “Nature does nothing in vain but always what is best from the available possibilities” (abbreviated, again by Lennox, as NP*) to account for traits that are present.1 This might imply that Aristotle operates with two distinct principles or, at least, two distinct versions of one principle. But the distinction is not water-proof. There is at least one instance in the Aristotelian corpus in which NP is invoked to account for a present trait, and at least another instance in which NP* is invoked to account for the absence of a trait. It seems more reasonable to assume, then, that NP is the short or condensed version of a principle which, in its full formulation, is always meant to be understood as NP*. To claim that nature does nothing in vain would, on this interpretation, imply that nature does always what is best from the available possibilities, given the circumstances. Aristotle claims that his teleological principle is an assumption based on observation of what happens in nature (GA V 8, 788b20–22). What this view seems to imply, I claim, is that the teleological principle is, to begin with, an assumption made from similar cases of what is better known to us, namely from particular empirical observations perhaps initially limited to individual animals and directly similar species. As an epistemological assumption the principle can be used as an heuristic tool for the determination of causally relevant features in other, less similar, 1

Lennox 2001b: 205–223.



9 De incessu animalium 12–13



species. But the principle is not only an epistemological assumption. Aristotle seems to think that this epistemological assumption not only can be generalized by induction but also should be promoted to the status of an ontological principle that is better known by/to nature. As a principle of nature sensu stricto the teleological principle can be used (I argue against a strictly epistemological interpretation and in line with what other scholars have maintained) as a major premise in scientific demonstrations meant to reflect the way nature operates. In agreement with mainstream interpretations (including Stasinos Stavrianeas’ treatment of the subject in the present volume), I think that the nature referred to in NP* is not a universal or global nature that works with a view to the best coordination of parts in the universe as a whole but a shorthand for the particular natures of all different kinds of animals. Its limitations, I submit, are mainly internal to the indivisible species under investigation, and they are determined by the genus and sub-kinds to which it belongs. Aristotle occasionally considers the possibility of an external constraint imposed on nature by the scarcity or quality of matter that it has to work with in order to achieve its ends. But in most cases its constraints are internal and they ultimately amount to the essence in which nature operates in each particular animal according to its kind. In her contribution to the present volume (ch. 10), Sarah Jansen argues that IA 14–15 undermines the mainstream interpretation in laying emphasis on essence, and that a cross-kinds interpretation of the teleological principle should be also allowed, at least in some cases. But I am not convinced that she is right. For one thing, the teleological principle is not explicitly invoked by Aristotle in the chapters she analyzes. For another, if large groupings of animals that do not form genuine kinds like those mentioned in IA 14–15 share a certain feature, such as the diagonal movement of their legs referred to in those chapters, this does not necessarily mean that the teleological principle applies in a crosskinds fashion. To be more precise, in IA 14, 712a28–b5 Aristotle lays out the logical possibilities – (i) the front legs move first and the back legs follow, (ii) the right legs move first and the left legs follow, (iii) front and back legs move diagonally – in order to discredit option (i) as physically difficult (since it would be a kind of continuous jumping) and option (ii) as physically impossible (since the animal would lose its balance and fall down). With the elimination, for some good reasons, of options (i) and (ii) it necessarily follows (ἀνάγκη) that the only available option is (iii). But the



Part III Interpretative Essays

necessity of option (iii) is predicated on factors that do not pertain to natural kinds as such but to physical reality and what we would call kinetics. If such factors play a role in determining animal progression, as they seem to do, they must be inscribed, I suggest, in the essences of all natural kinds and as such be part of the most comprehensive genus of “animal” to which all the animals referred to in IA 14–15 obviously belong. If that is a reasonable assumption to make, it would follow that, even if we consider the teleological principle to be implicitly present in the argumentation of IA 14–15, those chapters do not in fact provide us with any counterexamples to the rule which says that the factors constraining nature are determined by the genera and sub-kinds to which an animal belongs, and ultimately by its essence. D E I N C E S S U A N I M A L I U M 12

IA 12 begins with a brief summary of the conclusions reached in IA 9, then reminds the reader of the claims made in the introductory chapter of the treatise (IA 1) about the differences with respect to limb-bending between (i) human beings and birds and (ii) human beings and fourfooted animals, and finally provides some arguments meant to explain teleologically actual limb-bending in humans and four-footed animals. The chapter can be easily subdivided into the following parts: [1] An introduction (711a8–19) that repeats claims already made and thus paves the way to the explanations of the actual limb-bending of animals; [2] A section (711a20–b6) with two distinct arguments ([2.1] and [2.2] below) explaining why the forward bending of front legs (in the case of four-footed animals) and of legs (in the case of human beings) is necessary or at least preferable; [3] A section (711b6–12) that explains why human beings bend their limbs the way they do; [4] A section (711b12–32) that explains, with the use of two new arguments ([4.1] and [4.2] below), why live-bearing, four-footed animals bend their pairs of legs in opposite directions. The chapter as a whole picks up and answers [Q8] and [Q9] on the initial agenda of the IA, namely “Why do human beings bend their legs and arms in opposite directions?” (704a20–22), and “Why do the four-footed animals that are live-bearing bend their legs in the opposite way to human beings and also in opposite ways with respect to themselves?”



9 De incessu animalium 12–13



(704a22–b5).2 [Q7], “Why do human beings and birds, being both twofooted, bend their legs in opposite directions?” (704a17–20) – a question that is likely to arise in the reader’s mind given the way Aristotle speaks in the first argument of section [2] – is postponed for treatment in IA 15. All three questions have been announced in the introductory chapter of the treatise and are of the same form (the “Why is it so?” form); they ask for explanations of facts (IA 1, 704a17–b10).

Introduction [IA 12, 711a8–19] On the whole, there are six claims made in this introduction to IA 12 that are said to have been previously stated. The first two refer back to IA 9, while the remaining four refer back to IA 1: (i) Blooded and footed animals could not step forward if they had no bending in their legs, shoulders, or hips (711a8–10 referring back to IA 9, 708b26–709a24). In other words, all blooded and footed animals must have joints in their legs, shoulders, and hips in order to progress. (ii) In order for a bending to function as a bending, one part of the relevant limb must be at rest while another such part must move (711a10–11 referring back to IA 9, 708b21–26). (iii) Though both human beings and birds are two-footed, they bend their legs in opposite directions (711a11–13 referring back to IA 1, 704a17–20). (iv) Four-footed animals bend their pairs of legs in opposite directions (711a13 referring back to IA 1, 704a22–b5) – namely, the front legs in convex fashion and the back legs in concave fashion. (v) Four-footed animals bend their front legs in a direction opposite to that in which human beings bend their arms, and they bend their rear legs in a direction opposite to that in which human beings bend their legs (711a13–17 referring back to IA 1, 704a20–b5). (vi) Birds bend their legs in the same way as four-footed animals bend their (back) legs (711a17; not explicitly mentioned before but implied in IA 1, 704a19–20, 704b4–5). Aristotle argues extensively for the first two claims in IA 9. The most relevant passage for (i) is the following: Moreover, without bending there could not be walking or swimming or flying. The reason is that, since footed animals stand and take their weight 2

For more on the initial agenda of the IA, I refer the reader to the general introduction to the volume (ch. 1), as well as to the first essay (ch. 3), by Andrea Falcon.



Part III Interpretative Essays alternately on one or the other of their opposite legs, as one leg strides forward the other must necessarily be bent. (IA 9, 708b26–30)3

In this passage, Aristotle begins his discussion by making a claim about progression quite generally rather than a specific claim about walking, as he does at the beginning of IA 12, but walking is explicitly mentioned. Moreover, Aristotle speaks of all footed animals rather than blooded animals equipped with feet, as he does at the beginning of IA 12. But if this rule of bending applies to all footed animals, it must also apply to blooded animals equipped with feet, as at the outset of IA 12. The most relevant passage for claim (ii) is the following: That if nothing were at rest no bending or straightening could occur, is evident from what follows. For bending is the change from what is straight to what is curved or angled, straightening is the change from either of these to what is straight. In all such changes, the bending or straightening must necessarily be relative to one point. (IA 9, 708b21–26)

When at the outset of IA 12 Aristotle says that “birds also do the same” (711a17), he presumably means that they bend their legs in the way the four-footed animals do with respect to their back legs. This is made explicit in IA 15, especially 712b34–713a2. Since claims (iii)–(vi) have been announced in IA 1 but have not been treated in any other part of the treatise so far, we may safely assume that Aristotle refers us back to the introductory chapter. By resuming some of the agenda of the introductory chapter Aristotle is here distinguishing what already has been adequately treated from what is presently going to be explained with recourse to causes. In the present section, the cause given for what is the case in nature according to claims (iii)–(vi) is the general proposition that nature does nothing in vain. As we proceed, more specific arguments will be given in their favor. But all the subsequent arguments will depend, in one way or another, on the validity of the teleological principle. It is, therefore, advisable to say a few things about Aristotle’s teleological principle itself. In IA, the claim that nature does nothing in vain but everything with a view to achieve the best from among the available possibilities (this is the strong version of the well-known stock phrase, which is equivalent to NP*; see below) is first announced as a universal principle for the proper 3

Except for excerpts from IA, where I follow the joint translation provided by the contributors of this volume, translations of other Greek texts are mine.



9 De incessu animalium 12–13



investigation of nature (ἀρχὴ δὲ τῆς σκέψεως ὑποθεμένοις οἷς εἰώθαμεν χρῆσθαι πολλάκις πρὸς τὴν μέθοδον τὴν φυσικήν) in IA 2, 704b15–17. It is then invoked in IA 8, 708a9–12 as the cause (αἴτιον) of the absence of feet in snakes, and it is given again in IA 12 as the cause (αἴτιον) of the differences in limb-bending across four-footed animals, human beings, and birds. Although there is verbal variation in the three instances, the intended meaning is clearly the same.4 In the present context, nature is made into a craftsman by means of the marked verb δημιουργεῖ – variously rendered as “crafts,” “fashions,” “creates” – instead of the unmarked, and rather common, ποιεῖ – usually translated as “does” or “makes” – used in the other two instances. It is worth noticing that, although δημιουργεῖ/δημιουργήσασα is a rather common predicate of nature (see, for instance, PA I 5, 645a9; PA II 9, 654b31; GA I 23, 730b31; GA II 6, 743b23; and GA III 5, 755a20), this is the sole instance in the whole Aristotelian corpus that the stock phrase “nature does nothing in vain” occurs with this particular verb. The closest parallel to the present usage comes from GA I 23, 731a24, where Aristotle, while discussing sexual differentiation in animals as opposed to plants, writes: καὶ ταῦτα πάντα εὐλόγως ἡ φύσις δημιουργεῖ (“and nature reasonably fashions all these things”). Various metaphors and similes are used in other contexts: nature is said to contrive (μηχανᾶται) means to counterbalance excesses (PA II 7, 652a31–33); to resemble a molder rather than a joiner in that she rearranges the available parts rather than introducing new ones (GA I 22, 730b29–32); to use instruments like a craftsman (GA II 4, 740b25–34); to act like a painter who first makes a drawing and then adds colors to the artwork (GA II 6, 743b20–25); to resemble a good housekeeper who does not dispose of possibly useful materials (GA II 6, 744b16–17); to will (βούλεται) things that she is unable to accomplish with precision due to the indeterminacy of matter (GA IV 10, 778a4–9). But all these metaphors and similes should not lead us to the conclusion that Aristotle intends, here or anywhere else, an intelligent and purposeful agent like Mother Nature that works with a view to the best result in each case. As is clear from other treatises as well as Aristotle’s general ontology, “nature” is the collective name for all the various natures or essences of 4

The reader may compare the treatment of the claim that nature does nothing in vain that I offer here with Stavrianeas’ treatment of the same subject in his reading of IA 8. Although there are variations in emphasis between the two approaches, Stavrianeas and I are in agreement on most, if not all, points.



Part III Interpretative Essays

the different, living or non-living, kinds of things in the sublunary world.5 Aristotle’s “nature” is primarily a generalization over individual natures. This becomes evident also from the qualifications and restrictions that attach to the workings of nature in IA 2 and IA 8 (but not IA 12). The best achieved by nature in IA 2, 704b16 is explicitly said to be the best “of each kind (γένος) of animal” and the available possibilities from which nature is going to choose the best are those that pertain to its “substance” (οὐσία). IA 8, 708a11–12 is even more explicit in claiming that nature, by aiming to achieve the best among the available possibilities, preserves (διασώζουσαν) the proper substance (οὐσία) of each (sc. kind or animal) and its own essence (τὸ τί ἦν αὐτῷ εἶναι). Especially in the domain of Aristotelian biology, where ecological concerns (including the notion of an ecosystem) are conspicuously absent,6 Aristotle does not seem to have seriously envisaged the possibility of an overarching, collective Nature that acts as a purposeful agent and arranges things with a view to the goodness of the whole. Aristotle’s nature is a normative notion, partly reached through experience and partly through hypothetical reasoning (GA V 8, 788b20–23), that explains why things are always or for the most part the way they are (Phys. II 8, 198b34–36; GA I 19, 727b29– 30). Although Aristotle states (IA 11, 711a7) that “nature does nothing contrary to nature,” he admits that some animals conform with nature in a higher degree than others (IA 5, 706b3–10; IA 7, 707b6–7; IA 18, 714a6–8; IA 19, 714b10–19). Since the better arrangement is always more in conformity with nature (IA 2, 704b17–18), what is in consummate conformity with nature must be the perfect individual or species. Jim Lennox has suggested that the principle that nature does nothing in vain appears in the Aristotelian corpus in two distinct versions: a weak one and a strong one.7 Since his analysis is helpful for a precise understanding of what is involved in Aristotle’s principle and its role in 5

This is, I take it, the mainstream view on the issue and I happily subscribe to it. Pierre-Marie Morel (in Morel 2016: 9–30) has recently defended it anew with special reference to the IA. But other interpretations are possible. Pamela Huby (in Huby 1991: 163), for instance, writes that “there is a mass of evidence that Aristotle treated nature as a universal force, sometimes alongside god.” More recently, David Sedley (in Sedley 2010: 18–29) has made a strong case for what he calls “Global Teleology” in Aristotle’s universe, a view that entails the notion of a global or cosmic Nature, “causally governed by the Unmoved Mover,” which is fashioned on the model of the general and the army mentioned in “the vital culminating chapter of Aristotle’s theology” (Meta. XII 10, 1075a11–25). At this juncture, it is worth quoting Sedley’s concluding sentence of the paper as a whole: “It seems then that any natural collective system composed of discrete natural substances, be it an army, a household, a city or a world, has as its ‘nature’ its own complex functionality, this being, irreducibly, an end over and above the individual functionality of its various components.” 6 Henry 2013: 258. 7 Lennox 2001b: 205–223; cf. Johnson 2005: 80–82.



9 De incessu animalium 12–13



scientific explanation, it is worth spending some time to illustrate his interpretation and then see the objections it has provoked. The weak version of the “nature does nothing in vain” principle (encoded as NP by Lennox) states simply that nature does nothing in vain, while the strong version (his NP*) supplements this “negative” claim with the “positive” one that nature does what is best from the available options. According to Lennox, “NP is used as a premise in an explanation for the absence of a feature, which if present, would be superfluous” (author’s italics).8 The absence in question refers not to any absent feature whatsoever of a given species – for absent properties in a given species are almost countless – but rather to a specific feature that similar (in some relevant respects) species possess and for whose absence in the species under investigation one might reasonably raise a “Why not?” question. To be more precise, NP is used as a major premise in a syllogism of the form: • For a given species A to have a certain feature p would be in vain. • Nature does nothing in vain. ___________________________ • Nature does not produce A with p. For instance (the example stems from IA 11, 711a1–7): • (Minor premise 1.1) For a winged animal to be upright would be in vain (for in that case wings would not be put into proper use) or, alternatively • (Minor premise 1.2) For an upright animal to have wings would be in vain (for in that case wings would not be put into proper use). • (Major premise) Nature does nothing in vain. ___________________________ • (Conclusion 1) Nature does not produce an upright animal with wings. or, alternatively, • (Conclusion 2) Nature does not produce a winged animal which is upright. This is supposed to be the proper Aristotelian answer to the questions “Why aren’t birds upright?” or “Why aren’t there any upright birds?” and “Why don’t human beings have wings?” A corollary of the conclusion is 8

Lennox 2001b: 214.



Part III Interpretative Essays

that the winged cupids produced by artists are merely figments of human imagination, the sort of things that could never be produced by nature. The strong version of the Aristotelian principle, NP*, does not explain just the absence of something, as NP does. In NP* the negative claim that nature does nothing in vain is supplemented by the positive claim that nature does everything with a view to what is best from the available possibilities. As such NP* seems to be used as a major premise in syllogisms meant to reach conclusions about why animals have the features they have. This is Lennox’s contention. The general form of such syllogisms is the following: • For a given species A to have a certain feature p is best. • p is possible for A. • Nature does what is best for each kind of thing from the available possibilities. _______________________________ • A has p by nature. What is possible for a given species is determined by what the genus and the sub-kinds to which the species belongs allow.9 The available possibilities are, therefore, determined by the internal constraints that the genus and sub-kinds put on the species. An exception to this rule is the indeterminacy of matter, which may indeed act as an external constraining factor, according to Aristotle (cf. GA IV 10, 778a4–9). In those cases, the tokens of the species are less well-determined than their essential nature demands. In most cases, however, the constraints that nature has to face come from the generic origins of the species and as such they are internal to it. For instance, since it is in the nature of blooded animals to progress by means of four points, it is not possible for human beings, being blooded, to have wings; for having wings would increase the number of their points of motion from four to six, which is an option closed by the very nature of the genus of blooded animals. But we must add, I think, that the genus and the sub-kinds to which a given species belongs are, ultimately, essential and intrinsic determinations of the species itself (as Aristotle himself says in dealing with definitions in Meta. VII 12) rather than independent or external factors that pose constraints on the range of available possibilities with respect to the species in question.

9

Cf. Gotthelf 2012a: 173, who thinks that “the genos of animal as such” also plays a role in the determination of the range of possibilities that are available.



9 De incessu animalium 12–13



Lennox’s interpretation has not passed unchallenged. Devin Henry, for example, has noticed that in at least one instance (GA II 5, 741b4–5) NP is used to account for the presence of an attribute (the existence of sexes in animals) whereas in another – and this comes from our treatise at IA 8, 708a9–20 – NP* is used to account for the absence of a feature (here the absence of feet in snakes).10 Henry concludes that, for all practical purposes, Aristotle approaches NP and NP* not as two separate, though interrelated, principles (or distinct versions of a principle) invoked in different explanatory contexts, but as a single unified principle – and I think that he is right. Rather than dividing Aristotle’s principle into two versions that play distinct explanatory roles in different contexts we should approach it as one and the same principle, namely NP*, that has two verbal formulations: a full and an abbreviated one. This is shown not only by the fact that NP is repeated in NP* as its first part but also by the intended meaning of NP itself. Even when using NP alone Aristotle does not seem to aim at a limited praise of nature according to which, although it is admitted that nature has provided animals with no superfluous traits, it is also allowed that it, like a parsimonious mother, may not have provided them with everything it could provide them with. This is indicated by the formulation of the teleological principle in GA V 8, 788b21–22 where Aristotle writes that “nature does not fall short of its task nor does it make anything in vain from among the possibilities for each thing” (οὔτ’ ἐλλείπουσαν οὔτε μάταιον οὐθὲν ποιοῦσαν τῶν ἐνδεχομένων περὶ ἕκαστον). Here the standard positive claim that nature does the best (as in the second part of NP*) is substituted by the “negative” claim that nature does not fall short of its task, and the standard “negative” claim that nature does nothing in vain (i.e., NP) is combined with the qualification “from the possibilities for each thing” which we standardly find in the so-called “positive” part of NP*. It seems, then, that in Aristotle’s mind the negative claim of NP implies the positive claim of NP*, and “nature does nothing in vain” is a shorthand for “nature does the best given the circumstances (including both internal and external constraining factors) in which she is forced to operate.” Mariska Leunissen has criticized Lennox for claiming that the principle that nature does nothing in vain is part and parcel of Aristotle’s demonstrative science.11 In her view, this principle is, rather, a heuristic device meant to identify “causally relevant features when these are not immedi10

Henry 2013: 230; cf. Leunissen 2010: 130. Leunissen 2010: 119–135.

11



Part III Interpretative Essays

ately discernible.”12 “These features,” in turn, “are to be picked out in the ultimate, syllogistic explanation.”13 Both NP and NP*, whether distinct formulations of a single unified principle or discrete articulations of two separate principles, are clearly universal hypotheses based on empirical observation. In GA V 8, 788b20–22 Aristotle writes: “we assume, and assume from what we see, that nature does not fall short of its task nor does it make anything in vain from among the possibilities for each thing” (τὴν φύσιν ὑποτιθέμεθα, ἐξ ὧν ὁρῶμεν ὑποτιθέμενοι, οὔτ’ ἐλλείπουσαν οὔτε μάταιον οὐθὲν ποιοῦσαν τῶν ἐνδεχομένων περὶ ἕκαστον). Aristotle does not seem to have begun his investigations with a preconceived assumption about how nature works but rather to have discovered the teleological aspirations of nature in the course of his empirical study of animals. Once this universal assumption is reached and promoted to the role of a well-established explanatory principle in the domain of biology, it is possible for the assumption to function as a premise in a syllogism that seeks to explain either the absence or the presence of a certain part in a given animal species.14 To think of NP and NP* as heuristic tools for the determination of the causally relevant features, and to think of them as premises in scientific demonstrations, are not incompatible uses of the teleological principle. What is better known to us, namely particular observations about the presence or absence of traits in individual animals, is inductively generalized to the presence or absence of traits in species or genera of animals, and the conclusion is reached that in all those cases nature does nothing in vain. Once reached from many similar cases of inductive reasoning, the conclusion is promoted to the status of a teleological principle that is assumed to be better known to/by nature, and as such it can be used both as an established epistemological assumption for the determination of causally relevant traits in other species and as a major premise in scientific demonstrations meant to reflect the way nature in fact operates. Though helpful in making us sharply aware of what is involved in Aristotle’s invocation of the principle that nature does nothing in vain, the views of Lennox and Leunissen are not so far apart as they might seem at first sight. It is more difficult to determine precisely what Aristotle thinks is best for a given species. It seems that environmental factors (such as whether the animal under investigation lives in water, on earth, in subterranean holes, or in the air), physical considerations (such as the length of its 12

Leunissen 2010: 112. Leunissen 2010: 121. 14 Cf. Henry 2013: 250. 13



9 De incessu animalium 12–13



body or the material of its skin), and nutritive characteristics (such as the kind and position of the appropriate food) all play crucial roles in Aristotle’s determination of the best. But the single most important factor in the determination of what is best for a given species is the very essence of the species in question in which, I submit, all other factors can be subsumed and in which they are in fact included. In PA IV 10, 687a5–24 (an often-quoted passage), Aristotle says that since human beings are upright animals, nature provides them with arms and hands instead of front legs and feet. He then criticizes Anaxagoras for maintaining that human beings are the most intelligent (φρονιμώτατον) animals because they have hands. For Aristotle, this is a reversal of the causal sequence. He claims that human beings have hands because they are the most intelligent animals, and not the other way round. The reason he gives is the strong version of the “nature doing nothing in vain” principle, namely NP*. What Aristotle seems to have in mind is this: (i) it is part of the essence of human beings to be the most intelligent animals; (ii) their intelligence permits human beings to use hands in the best possible way – for example, for the creation and manipulation of tools, for hands are, as it were, the tools par excellence for the handling of tools; (iii) nothing prevents nature from providing human beings with hands; therefore, (iv) nature provides human beings with hands. Aristotle’s critique of Anaxagoras entails that a bodily part is best for a given species if and only if it allows the species in question to fully actualize the potential of its own, particular kind of life, as that potential is determined by the very essence of the species in question and may be derived, ultimately, from the definition of that essence. If this is stricto sensu so, it follows that the environmental factors, physical considerations, and nutritive characteristics mentioned above would have to be subsumed into, and indeed derived from, the essence of the species in question. In other words, Aristotelian natures are independent of environmental factors not in the sense that they are not well adapted to their natural environment, but rather in the sense that they are not regulated by the environment. It is the essence of the species that determines in which environment its members live, and not vice versa.15 15

For an interesting discussion about why Aristotle does not explicitly address the question concerning the fit between animals and their habitats, see Gelber 2015: 267–293. In her opinion, neither a Global Teleology view nor an Adaptation Theory are very plausible candidates to account for Aristotle’s belief that animals suit well the places in which they live. She, rather, thinks that “Aristotle considers habitat to be partially constitutive of the capacities that comprise a kind’s



Part III Interpretative Essays

We should also perhaps add that Aristotle does not always seem to subscribe to the principle of economy in determining what is best. The principle of economy states that if a required function of an animal species may be accomplished by means of either one organ or more than one, it is best for the species in question to have just one such organ rather than a combination of more than one; but if the required function may not be accomplished by means of a single organ, it is best for the species in question to have as many organs as needed for the accomplishment of the said function. A case in point is Aristotle’s explanation of the multiple stomachs of animals such as the camel, the sheep, the goat, and the cow, which are said to need multiple stomachs because of the scarcity of their teeth (PA III 10, 674b5–17).16 Wouldn’t it be better for them to have more teeth than multiple stomachs? Unless he contradicts himself, Aristotle does not seem to think so. To be sure, he allows for some species, such as the mole (HA I 9, 491b26–34; IV 7, 532a2–12), the lobster (PA IV 8, 684a32–b1), the flat fish (IA 18, 714a6–7), the seal and the bat (IA 19, 714b10–14; PA IV 13, 697b1– 13), to be handicapped by nature (cf. HA VIII 2, 589b29).17 The reasons for Aristotle’s attribution of abnormality to whole species vary from case to case. But as a general rule Aristotle assumes that in their case nature is regularly prevented from doing the best possible during the generation process by some external restriction – much like craftsmen may be prevented from doing the best possible due to an inherent defect or limitation that the material they have to work with imposes on their craftwork (cf. Phys. II 8, 199a31–b4). A similar explanation for the presence of essence, and not merely an external or enabling condition under which an animal can exercise its essential capacities” and, therefore, that “there is no need to ask why an animal is well suited to the habitat in which it lives if its habitat is already included in what it is to be an animal of that kind” (Gelber 2015: 279). I fully agree with the reason given for Aristotle’s silence on the issue of the fit between an animal and its habitat. I am a bit more skeptical about the former claim. If Aristotle thinks that habitat is included in an animal’s essence because it partially determines what it is to be an animal of this kind, the environmental factor would assume a causal priority over some of the essential capacities of each species in a way that might be called “non-evolutionary Darwinism” (cf. O’Rourke 2016: ch. 7). I would think rather that in Aristotle’s mind an animal’s essence includes an index pointing to the type of environment in which that particular kind of animal will thrive. 16 Cf. Gotthelf 2012a: 170–171. 17 Witt 2012: 83–106 is the best treatment of the subject. It is worth noting, however, that Witt does not seem to be aware of the cases specifically mentioned in IA (i.e., flat fish and bat, cf. Witt 2012: 84, n. 4). Johnson 2005: 188–210, while discussing the application of Aristotle’s teleology to specific normal and abnormal cases of living individuals, fails to discuss the natural abnormality, as it were, of some specific species like those mentioned in IA. We may also note that the relevant chapters of IA do not figure at all in the Index of Texts and Commentaries of the book. The same neglect appears in the Index Locorum offered in Leunissen 2010. A convincing account of the “natural” failures of nature is provided in Stavrianeas 2018: 51–71.



9 De incessu animalium 12–13



­ ultiple stomachs in the animals that have them is not readily available m since Aristotle thinks that the scarcity of teeth is a good teleological reason for the presence of multiple stomachs: under the circumstances, nature is doing the best possible. In any case, it seems that Ockham’s razor is not Aristotle’s overriding concern. Nature, while doing the best possible, seems to be exuberant: it shows a tendency for variation rather than economy pure and simple.18 Even a quick look at IA as a whole, where a great variety of actual modes of animal progression is exhibited and teleologically explained, may convince the reader of the exuberance and generosity contained in Aristotle’s concept of nature. One might object that in all those cases nature not only does the best from the available possibilities but also the best in the most economical way. But Aristotle never explicitly ascribes such a limitation to the mode of nature’s operation and I doubt that he implies it in his teleological principle. The occasionally roundabout ways in which nature seems to achieve its ends, as in the case of multiple stomachs, might remind us of the way a river finds its course, and thus determines its bank, in the particular landscape in which it is forced to flow: in a sense the course is economical indeed, but it is rarely a straight line as in artificial channels. The stock phrase “nature does nothing in vain but always what is best from among the available possibilities for each kind” raises the further question about the role and precise function of Aristotle’s teleology in natural science.19 The question cannot be discussed here in any detail. However, we can safely discredit an older view according to which, when the appropriate circumstances are fulfilled (when nutrition from menstrual fluids is available), an immaterial agent within the sperm guides the development of the embryo in a way analogous to the way a sculptor 18

Cooper 1982: 205–206 denies that Aristotle ever explicitly considers the theory of maximal, or at any rate great, variety that I attribute to him. Cooper is right insofar as explicit theory is concerned. However, from the way Aristotle treats the wonderful variety of organs for similar or selfsame functions distributed among different plant and animal species one may deduce a view about the exuberance of nature as an implicit presupposition for the kind of explanations that Aristotle actually provides. Cooper also seems to be right in claiming that the question as to whether the world exhibits maximal or less than maximal variety or richness can only be answered by means of an independent standard of what may count as maximal variety, and Aristotle’s metaphysics offers no such independent standard. I fully agree. But to claim that Aristotle’s nature, as empirically approached by him, is exuberant and shows a tendency for variation is not to claim that the richness achieved by it is maximal: it just is what it is. In the same vein, I disagree with Leunissen 2010: 124, who claims that “Nature’s actions in giving parts to animals are always economical.” The evidence she provides is not, to my mind, conclusive. What she calls “principles of economical assignment” (Leunissen 2010: 124, n. 31) are present in Aristotle’s explanations of the number of animal parts in some cases but absent in others: the explanation of multiple stomachs in camel, goat, sheep, and cow is exemplary of his non-commitment to the principle of economy. 19 Cooper 1982: 197–222; Sedley 2010: 5–29.



Part III Interpretative Essays

shapes his statue, the main difference being that while the sculptor is external to his work the agent that guides the sperm is internal to it and rules its development. This is clearly not the way Aristotle thinks of the nature of a particular species achieving its end since there is no textual indication in this direction, apart from the often used analogy of craft and nature and the implicit role of the craftsman in it.20 But to think that, since the craftsman is an agent, a similar agent must be operative in the midst of each and every nature would be to stretch the analogy too far. Moreover, to assume an immaterial agent as responsible for teleological arrangements would be to collapse the difference between an efficient and a final cause. Two other lines of interpretation proposed in recent scholarly literature seem to be more promising. According to the first, and rather widespread, interpretation, Aristotle’s teleology represents not a real causal sequence of events, say, from conception to maturity, but, rather, an (a posteriori?) explanation of that sequence in terms that render it intelligible. The problem with this view is twofold. First, it turns into a non-causal explanation what Aristotle clearly, no matter how clearminded or confused he may have been on this issue, intended to be a causal explanation. Second, it does not say how the teleological explanation manages to perform its explanatory role if there is nothing in the real nature of things that corresponds to it. For if there is such a thing in the real nature of things, then this thing must be the cause that renders the teleological explanation adequate as an explanation.21 These and similar worries have given rise to another interpretation of Aristotle’s teleology whose main propounders have been John Cooper, Allan Gotthelf, and David Balme.22 According to one version of this view, the development of a living organism is not the result of a sum of actualizations of element-potentials the identification of which includes no mention of the form of the mature organism, but is in fact the actualization primarily of a single potential for an organism of that form, an actualization which incorporates many element-potentials, but is not reducible to them. (Gotthelf 2012a: 11; author’s emphasis) 20

Contrary to mainstream interpretations that take the craft analogy merely as a pedagogical tool, both Sedley 2010: 11–18 and Witt 2015: 88–106 argue for the cogency of Aristotle’s analogy, which they claim should be construed as disregarding the agency (and necessary deliberation) of the craftsman. If they are right, as I think they are, then the craft analogy is even less meant to indicate anything like an immaterial agent in natural processes. 21 Cf. Cooper 1982: 215: “The recent tendency to explicate and defend Aristotelian teleology exclusively by appeal to essentially epistemological considerations leaves out of account this crucial fact about Aristotle’s theory, that he grounds his teleological explanations thus in the very nature of things.” 22 Balme 1987b: 275–285; Cooper 1982: 197–220; Cooper 1987: 243–274; Gotthelf 2012a: 3–44.



9 De incessu animalium 12–13



What this view maintains is that the actualization of all the elementpotentials of the seed of a given species is irreducibly governed by the form of the species in question, which leads the process of development to its end because it (the form) has, already from the beginning, the potential to that end and the innate tendency to actualize it if the circumstances are favorable. In other words, the form contained in the seed is dynamic in a very well-determined way and manages to subsume all the elementpotentials that are necessary to its actualization under the rule of its own well-determined dynamism. What is important in this view is that the actualization of the form-potential is not reducible to the element-potentials that are necessary for it, but rather contains something over and above them all (for instance, precise rules about the sequence of their actualization). This is what is sometimes referred to as “the natural state” to which an organism develops, and the so-called “natural state model” provides a plausible interpretation of Aristotle’s teleological principle.23

Why the Forward Bending of Front Legs is Necessary or Preferable [IA 12, 711a20–b6] This section begins with a long, and rather convoluted, sentence consisting basically of one main clause and four subordinate ones.24 The main clause comes at the end of the sentence beginning with the Greek words δῆλον ὡς ἀναγκαῖον (“it is evidently necessary”) at lines 711a24–25 and states the conclusion. The first of the subordinate clauses (πᾶσιν ὅσοις […] τοῖν σκελοῖν) determines the population scope for the application of the claims to follow. The other three subordinate clauses pick up ἐπεί at the beginning of the sentence and are related by means of a μὲν–δὲ–δὲ construction. They make one claim each. The three claims are: (i) The standing leg bears the weight of the animal; (ii) At the same time the leading foot bears no weight; (iii) As progression continues, the weight of the animal falls on the leg that formerly bore no weight. The conclusion is simple and straightforward: the leg that was formerly bent must become straight while the leading foot does not lose contact with the ground. In this stretch of text, Aristotle builds on the mechanics 23

Gottlieb–Sober 2017: 247–252. I say “basically” because, from a strictly grammatical point of view, the main clause consists of just one word (δῆλον); what follows (ὡς ἀναγκαῖον) introduces a relative clause depending on δῆλον (sc. ἐστί). Hence, strictly speaking, the number of subordinate clauses of this sentence amounts to five.

24



Part III Interpretative Essays

of progression put forward in IA 9.25 What he means to say is that since the weight of the animal is borne by each leg alternately, each leg must be able to bend and then become straight again. But this presupposes a bending in the leg. And the question now arises, which is the main focus of the whole chapter, as to whether the bending of the leading leg, and consequently of both legs since they alternate in being leading legs, should be forward or backward in relation to the direction of the overall movement of the animal. I have already indicated that Aristotle offers two distinct arguments to explain why forward bending is preferable. The first argument, [2.1], goes as follows: if the bending of the leading leg is forward, namely if the bending of the knee faces the direction of the overall movement of the animal, it is possible for alternation of the weight-bearing legs to happen; but if the leading leg is bent backward, namely if the bending of the knee faces the opposite direction to that of the overall movement of the animal, it is impossible for alternation of weight-bearing legs to happen. But why is this so? It is because, Aristotle continues, in the latter case the body of the animal would have to stretch backward whereas in the former case the body of the animal would have to stretch forward, namely in the same direction as the direction of the overall progression. What Aristotle actually means by these claims is not very clear. It seems that Aristotle’s tacit premise in [2.1] is that the direction of bending determines the direction of the whole body’s stretch, and the whole body’s stretch determines the direction of the movement. But if this is what he actually means, then the backward bending of birds’ legs would make their forward progression impossible, which is actually an unwanted conclusion. We shall come back to this problem after we have analyzed the second argument. The second argument, [2.2], is less obscure. In a nutshell, Aristotle’s idea is that if the bending of the knee were backward, then for progression to take place a contrary, i.e., forward, movement should be involved. Aristotle does not here say that this is impossible, as he does in the much stronger claim of [2.1], and it is, we may add, what actually happens in the case of a bird progressing by means of legs. But he seems to mean that since a forward bending of the knee avoids the contrary movements necessarily involved in a backward bending of the knee, such a bending is preferable. The problem with the two-movements option is that it requires simultaneous movements in contrary directions, a kind of unnecessary 25

I refer the reader to the excellently illustrated analysis of IA 9 by Christopher Frey (this volume, ch. 7).



9 De incessu animalium 12–13



complexity involving the opposition of forward and backward. The best that nature crafts from the available options is, if possible, the least complicated solution to the problem in view – the problem of progression by means of feet in our case. This seems to be Aristotle’s tacit premise here. From the way this stretch of text begins, it is clear that all the arguments offered here – the argument in the long introductory sentence as well as the following two arguments [2.1] and [2.2] – are meant to apply (i) to two-footed animals alone rather than to all footed animals; and (ii) to two-footed animals quite generally rather than to human beings alone. This can be firmly established by the dual form τοῖν σκελοῖν at 711a21 and the emphatic πᾶσιν ὅσοις at 711a20. We may also regard as significant the general word τὸ ζῷον used in line 711a27. The same idea about the population scope of the main argument of the whole chapter is also implied later on at 711b6–9 when Aristotle writes “the human being, because it is two-footed and naturally makes change with respect to place by means of legs, bends its legs forward due to the stated cause.” Clearly, what Aristotle means is this: we have put forward two distinct arguments whereby it has been shown that it is impossible for two-footed animals quite generally to bend their legs backward; it follows that since human beings are twofooted our conclusion will also apply to them. But if this is how Aristotle’s train of thought unfolds, as it apparently does, we have to face a serious difficulty. Since Aristotle’s two arguments are meant to apply to all two-footed animals, then the progression on land of a bird will turn out to be, against all empirical evidence, simply impossible. And this is clearly an unwanted conclusion, since (i) birds bend their legs backward, and (ii) they are able to move forward by means of feet. Can we save Aristotle from such a contradiction? We can do so only on the assumption that his wording is extremely careless. The population scope of the initial claim that all two-footed animals bend their leading legs forward cannot be limited to human beings alone since the expressions used by Aristotle are very emphatic, and they are obviously meant to apply to more than one kind of animal. Moreover, if we assume that Aristotle meant, without saying it explicitly, only those two-footed animals that are also upright, we would face the difficulty that for Aristotle no other upright, two-footed animal existed beside the human being. Hence, the population scope of the initial claim cannot be limited in such a way as to include human beings but exclude birds. It seems that Aristotle, while writing this piece, had his mind firmly focused on the way human beings walk and, without realizing it, made a more general claim about all two-footed animals.



Part III Interpretative Essays

Since the problem is created not only by the unlimited scope of the assumption in lines 711a20–21 but also by the repeatedly mentioned locutions “toward the front” (εἰς τοὔμπροσθεν, or εἰς τὸ ἔμπροσθεν) and “toward the back” (εἰς τοὔπισθεν, or εἰς τὸ ὄπισθεν) in 711a27–b6 – which clearly appear to be synonymous with the earlier expressions “toward the concave” (ἐπὶ τὸ κυρτόν) and “toward the convex” (ἐπὶ τὰ κοῖλα/τὸ κοῖλον), respectively – one might think that we can solve the difficulty if we manage to interpret these locutions differently in the present context. One way to do so would be to think that the words ἔμπροσθεν and ὄπισθεν, rather than referring to the direction of the bending, refer to the direction of the bent part. For it is evident that both human beings and birds move forward their already bent legs no matter whether the bend faces the front or the back of the animal. But such an interpretation would stretch Aristotle’s text too much. For it is clear that the two arguments we are considering deal with bending and examine the reasons why bending occurs the way it does. If they were meant to give us reasons why some moving part of the animal must move forward if the animal as a whole is to move forward, they would seem to be quite inappropriate at this juncture, given how the chapter begins, and rather superfluous in themselves. To be sure, we find the idea of a biped’s forward movement formulated by “toward the front” (εἰς τὸ πρόσθεν) at 711a22. But this particular meaning is required by both the syntax and the context there, and creates no problem for the interpreter. In the following lines, by contrast, if we were to interpret εἰς τὸ ἔμπροσθεν and εἰς τὸ ὄπισθεν as referring to the direction of the movement, the two arguments we are considering would make no sense at all. In general, the expressions “toward the front” (εἰς τὸ ἔμπροσθεν), “toward the concave” (ἐπὶ τὸ κυρτόν), and “toward the circumference” (ἐπὶ τὴν περιφέρειαν) are all used as synonymous in the present chapter. Michael of Ephesus (In IA 161.6–9) rightly takes “forward bending” to be synonymous with “convex bending.” He also writes (with a view to lines 711a29–31): He [sc. Aristotle] calls “brought forward” (προενεχθέν), and in general “bringing forward” (προένεξις), the forward movement, which is the way animals naturally move – for all animals with the exception of crabs naturally move forward – and “brought back” (ἀνενεχθέν) and “bringing back” (ἀνένεξις) the backward movement, which is the way no animal by nature moves.26 Michael of Ephesus, In IA 161.12–15: λέγει δὲ προενεχθὲν καὶ ὅλως προένεξιν τὴν εἰς τὸ πρόσω πορείαν, ἣν καὶ πεφύκασι κινεῖσθαι τὰ ζῷα (πάντα γὰρ πλὴν τοῦ καρκίνου εἰς τὸ πρόσω πέφυκε κινεῖσθαι) ἀνενεχθὲν δὲ καὶ ἀνένεξιν τὴν εἰς τὸ ὄπισθεν, ἣν οὐδὲν κατὰ φύσιν κινεῖται.

26



9 De incessu animalium 12–13



In other words, Michael understands the last sentence of [2.1] to refer to the direction of the whole animal’s movement – which may be what Aristotle indeed intended, although this is clearly not what we would expect from the long introductory sentence of [2]. But Michael does not explain why the contrary directions of a knee’s bending and body’s movement, which is what actually happens in the case of birds, create an “impossible” situation as Aristotle writes (εἰς τοὔπισθεν δ’ ἀδύνατον). On the whole, Michael does not see the problem we are trying to address (In IA 160.22–162.8). He, rather, explains the dual τοῖν σκελοῖν with the plural διὰ τῶν σκελῶν (In IA 160.28) and makes it clear that arguments [2.1] and [2.2] apply to human legs and the front legs of fourfooted animals (In IA 161.7–8) but not to two-footed animals, including birds, quite generally. His interpretation may capture the true sense of Aristotle’s intended meaning but certainly not its (contradictory) ­formulation. In conclusion, on the level of intended meaning as distinct from, if not opposed to, the level of verbal formulation, we may be in a position to save Aristotle from contradiction provided that we follow Michael of Ephesus’ interpretation, which is – we must note – a free paraphrase of Aristotle’s intentions (as understood by Michael) rather than an explanation of the text’s own wording.

Why Human Beings Bend Their Limbs in the Way They Do [IA 12, 711b6–12] As Aristotle turns to the case of human walking, he writes that human beings bend their legs in the way they do for the “stated cause.” The cause he has in mind seems to be the conclusion of the first argument, [2.1], alone – namely, that it is impossible for forward movement to occur if the bending of the legs is backward. The “stated cause” cannot refer also to the conclusion of the second argument, [2.2], since this conclusion allows for the co-presence of a backward bending of the knee and a forward walking. The fact that human arms are bent in concave fashion is explained with respect to the employment of hands quite generally and in particular for gathering food. Aristotle emphasizes the use of hands for gathering food presumably because, contrary to most animals, human beings do not directly use their mouth to get their food; rather, they use their hands to bring food to their mouth. In view of PA IV 10, 687a5–24, we might think that Aristotle considers the hand (as the tool par excellence, the tool of tools) to be specific to the human nature and to have been



Part III Interpretative Essays

given to human beings because they are the most intelligent animals.27 However, from HA II 8, 502a16–18, we learn that some animals, such as monkeys of different kinds, lie halfway between human beings and fourfooted animals (ἐπαμφοτερίζει) and resemble both groups of animals in some respects (cf. PA IV 10, 689b31–32). One respect in which monkeys resemble human beings is that monkeys have hands (with nails, fingers, and all the other anatomical details) that are similar to human hands, except that theirs are more beastlike (HA II 8, 502b3–4: πρὸς δὲ τούτοις χεῖρας καὶ δακτύλους καὶ ὄνυχας ὁμοίους ἀνθρώπῳ, πλὴν πάντα ταῦτα ἐπὶ τὸ θηριωδέστερον). What Aristotle seems to imply there is that the perfect condition of the hand as such is the one we find in human beings, who are able to create and manipulate a great variety of tools for their own benefit using their hands. Other animal kinds may only imitate (cf. HA II 8, 502b9: μιμούμενον) the human hand in a way appropriate to their beastly nature. Now, if having perfect and not beastly hands is one of the many aspects of what it is for an animal to be human, it follows that the bending of the arm of such an animal, which allows the hand to be properly used, should be the way it is. If the opposite were the case, the perfect hand would be useless or functionally restricted, and nature, if not hindered, does not provide creatures with useless or functionally restricted parts. That is also the reason why monkeys bend their arms like human beings (HA II 8, 502a35–b3). The functional optimality of the human hand requires a concave rather than convex bending of the arm on which the hand depends for its movements.

Why Live-Bearing Four-Footed Animals Bend Their Pairs of Legs in Opposite Directions [IA 12, 711b12–32] In the first sentence of this section, the omission of the second definite article τὰ in the Greek locution τὰ δὲ τετράποδα καὶ ζῳοτόκα seems to be required by the context. It is not the sum total of the classes of fourfooted animals (τὰ τετράποδα) on the one hand, and live-bearing ones (τὰ ζῳοτόκα) on the other that is intended here, but rather the subclass of the four-footed animals that are live-bearing. This is made clear by the separate treatment of egg-laying animals that are four-footed in IA 15, 713a15–25. According to Aristotle’s distinction of the so-called greatest kinds (μέγιστα γένη) of animals advanced in HA I 6, 490b7–491a6 – a 27

Lennox 2001c: 320–321.



9 De incessu animalium 12–13



distinction that arguably follows popular taxonomy but goes well beyond it – live-bearing four-footed animals and egg-laying four-footed animals are among the main categories that get recognized.28 Aristotle explains the forward bending of the front legs in live-bearing four-footed animals by referring us back to the cause of leg-bending in the case of human beings. This is neither uncommon nor surprising. On the contrary, Aristotle quite often takes the human being to be the paradigmatic case of animal, and usually his explanations of biological features in other animals begin from considerations of what pertains to human beings. This happens for two reasons: the human body and its functions are sometimes said to be better known to us (HA I 6, 491a19–23; PA II 10, 656a9–10), so that the exploration naturally begins from there; but, more importantly, the human race is regarded as the most perfect animal in the sublunary region (GA II 4, 737b25–27; cf. Pol. I 8, 1256b15–22), a race that alone or most of all partakes of the divine order (PA II 10, 656a7–8) and that possesses a body that, because of its erect position, exhibits the parts, both external and internal, in their natural order (PA II 10, 656a10–13);29 it follows that an investigation of what pertains to the human body provides the standards by which other animal bodies may be studied. Aristotle’s tendency to begin his explanations with the most perfect or complete or integrated kind in a particular domain is quite widespread. Andrea Falcon, for instance, has convincingly argued that, according to the available evidence of the Aristotelian corpus, the study of plants (of which no single specimen in the form of a self-contained treatise has survived from Aristotle’s own hand) was meant to follow, rather than precede, the study of animals. The reason for what might seem to be a reversal of the expected sequence from the simpler to the more complex cases of soul (as DA II–III would prompt us to anticipate) is that priority is given to the fullest and most perfect developments. In GA II–III, for example, the analysis of the types of animal reproduction begins with human beings (GA II 4), continues with other live-bearing animals (GA II 4–8), then extends to blooded egg-laying animals (GA III 1–8), proceeds to bloodless egg-laying animals, including insects (GA III 8–10), to terminate with stationary animals resembling plants (GA III 11).30 We may note in passing that this method of starting with the perfect (in the 28

Gotthelf 2012c: 293–306. Lennox 2001b: 222–223. 30 Falcon 2015: 80–81. 29



Part III Interpretative Essays

sense of the most complete or integrated) is not restricted to biology.31 The Poetics, for instance, famously begins the exploration of poetry’s salient features with a structural analysis of tragedy, the most complex and serious complete poetic genre in Aristotle’s eyes: Some parts are the same in both tragedy and epic poetry and some are peculiar to tragedy; hence, whoever is a judge of good and bad tragedy is also a judge of epic poetry; for the parts of epic poetry exist in tragedy but the parts of tragedy do not all of them exist in epic poetry. (Poet. 5, 1449b16–20)

A similar line of thought would prompt Aristotle to believe that the perfect knower of the functioning of the human body will be best equipped to study, understand, and judge the operations of other animal bodies as well. For Aristotle the human body is a model for the bodies of other animals.32 Back to our passage. Aristotle explains the forward bending of the front legs in live-bearing, four-footed animals by referring us back to the cause of such leg-bending in the case of human beings. However, he has explained the forward bending of human legs by referring us back to section 711a8–b6, and in particular to [2.1]. Aristotle seems to think that this section has provided arguments for the explanation of the general rule of leg-bending, while the subsequent stretch of text is meant to consist in the applications of the general rule to the case of human beings and livebearing, four-footed animals. The two arguments that Aristotle offers in these sections – [4.1] and [4.2] – are counterfactual. They prompt us to visualize the fictional situation of live-bearing four-footed animals with legs that bend in the direction contrary to what is empirically the case. In this sense, they both consist in thought experiments. Argument [4.1] takes the case of front legs first. If those legs were bent backward, Aristotle contends, the knee would come up against the belly, the consequence being that the feet would not be allowed to move far off 31

“First things first” is Aristotle’s stock phrase whereby he states his determination to begin the proper investigation of the subject at hand with some prior things (GA II 4, 737b25: ἀπὸ τῶν πρώτων ἀρκτέον πρῶτον; PA I 5, 646a4: ἀρξάμενοι, καθάπερ διωρίσαμεν, πρῶτον ἀπὸ τῶν πρώτων; PA II 10, 655b28–29: νῦν δὲ λέγωμεν οἷον ἀπ᾽ ἀρχῆς πάλιν, ἀρξάμενοι πρῶτον ἀπὸ τῶν πρώτων; EE I 7, 1217a18–19: λέγωμεν ἀρξάμενοι πρῶτον ἀπὸ τῶν πρώτων; Poet. 1, 1447a12: λέγωμεν ἀρξάμενοι κατὰ φύσιν πρῶτον ἀπὸ τῶν πρώτων). Although the precise notion of priority that he intends in each particular case can only be determined from the context, it is clear that ontological priority plays a very prominent role in his thinking and motivates the repeated use of this stock phrase in various milieus. 32 Carbone 2011: 105–138.



9 De incessu animalium 12–13



the ground and the steps thus made would as a result have to be short. The progression of the animal would then be restricted, inflexible with respect to obstacles, and slow. A similar consideration is made with respect to the back legs. Were they bent forward, they would come up against the belly, and the consequences would have been exactly the same with respect to walking. As things stand, Aristotle clearly implies, the optimal function of the legs is warranted. Notice that the entire argument is given in the syntactical mode of the second-class condition, that of assumed unreality, which states what is contrary to the way things actually are (ἂν ἐμετεώριζον, ἂν εἶχε, ἂν ἐγίγνετο). Argument [4.2] provides an additional reason for the goodness of things as they actually are. This argument has to do with rearing and nursing. Four-footed females would be seriously hampered in their ability to nurse their offspring since the available room below their belly would be severely limited by the bending of their legs. The result of such a situation would be that those female animals could not properly feed their young (or protect them under their belly) and move from one place to another at the same time. In the case of egg-laying and four-footed animals, which are said to bend both front and back legs in a forward fashion (HA II 1, 498a3–16; cf. IA 15, 713a15–25), no such restriction applies since they neither nurse nor protect their young but also because they bend their limbs slightly sideways with respect to their body (μικρὸν εἰς τὸ πλάγιον παρεγκλίνοντα). It is important to note that here Aristotle uses an argument about what is beneficial for female live-bearing animals and their young in order to reach a conclusion about the bodily parts responsible for walking in both female and male members of the species. As Connell has observed, “in many instances in the biological treatises, what it is to be a certain type of animal is closely connected to the female.”33 This observation redresses the balance of a rather widespread misconception according to which Aristotle always considers the female sex to be a handicapped male. If what is beneficial to female members of a species determines animal traits of both sexes, such as the way they move, it would follow that the general requirements of the specifically feminine functions are, occasionally or as a rule, inscribed in a kind’s essence on a par with the corresponding requirements of the specifically masculine functions. It is worth highlighting the disjunction between what is necessary and what is better in line 711b30 (ἀναγκαῖον ἢ βέλτιόν γε) with recourse to a 33

Connell 2016: 290.



Part III Interpretative Essays

passage from the Generation of Animals. According to GA I 4, 717a15–16, “nature does everything either because it is necessary or because it is better” (εἰ δὴ πᾶν ἡ φύσις ἢ διὰ τὸ ἀναγκαῖον ποιεῖ ἢ διὰ τὸ βέλτιον, κἂν τοῦτο τὸ μόριον [sc. testes] εἴη διὰ τούτων θάτερον). Here the disjunction appears to be exclusive. A certain part or feature of an animal’s body is necessary if the animal in question cannot perform a function necessary for its being the animal that it is (namely, a function prescribed by its own essence) without it. The necessity involved is hypothetical necessity. By contrast, a certain part or feature of an animal’s body is the way it is because it is better for it to be so if and only if the function corresponding to this part or feature is improved in some relevant aspect by the presence of this part or feature although the function in question could also be performed by a different part or feature. The “necessary” of GA I 4 seems to pick out a certain part or feature that is hypothetically necessary for the performance of a given function, while the “better” seems to pick out a different part or feature that allows for the improved performance of the same function. However, if we are to examine how precisely the “better” responsible for this improved performance will come to pass, we will be led to the conclusion that the “better” is also hypothetically necessary, that is, necessary for the function in its improved state or necessary for the function if the function in question is to be improved. It follows that the disjunction of GA I 4 is not between the hypothetically necessary and the better; it is, rather, between two cases of hypothetical necessity, that is, between the hypothetically necessary for the mere presence and operation of a given function and the hypotthetically necessary for the improved performance of this same function (or any closely related function). The distinction of GA I 4 appears also in our passage, although in a more cryptic and complicated manner. Aristotle says that the outward bending of the legs in the case of nursing animals is (hypothetically) necessary for the proper functioning of nursing or, at any rate (γε), a better arrangement for this function than the inward bending. Why is Aristotle downgrading the better to a lower status than the necessary? Presumably because the hypothetical necessity involved in the better makes the outward bending of the legs less strictly necessary than the case would be if nursing were altogether impossible without it. As already noted, the previous argument [4.1] in its entirety is given in the syntactical mode of the second-class condition. Aristotle’s contention there was that the function of walking would be seriously impaired with an inward bending of the legs. In [4.2], by contrast, the claim is milder: what makes the outward bending of the legs a better arrangement – Aristotle’s presumably



9 De incessu animalium 12–13



preferred option – is not so much that nursing would be seriously endangered without it as it is the fact that the space below the belly of the mother created thereby also provides shelter for the young. The better arrangement envisaged here is not only one that optimizes a certain function (nursing) but also one that allows for some other, related, function (protection of the young) to materialize. Without the outward bending of the legs nursing would still be possible (though not easily so while walking) but simultaneous protection of the young would not. Once again, we are led to realize that things as they stand are arranged by the natures of the corresponding animals in the best possible way. DE INCESSU ANIMALIUM 13

The chapter can be subdivided into two parts. In the first part (712a1–13), Aristotle lays out abstractly the logical possibilities of leg-bending in the case of four-footed animals, and then assigns natural kinds to the logical categories thus produced. In the second part (712a13–22), Aristotle states that the bending of arms and legs in the case of human beings follows an alternate pattern throughout. The chapter as a whole is a bit of a surprise since it does not answer any question announced in IA 1; nor does it provide any reason or explanation (unless what I have called “the principle of alternate bending” is such an explanation). Rather, it schematizes the empirical results about leg-bending reached so far, and provides an additional remark on the bending of human limbs – namely, that it occurs in contrary directions in arms and legs throughout all the bends. IA 13 seems to be an appendix to IA 12 that was not initially designed to be part of the treatise but occurred to Aristotle as an afterthought upon the completion of IA 12. The diagrammatic presentation of the logical possibilities of its first part has the ring of an educational device meant for instruction and memorization, since it highlights the common rule of limb-bending in all four-footed live-bearing animals and the exceptional position occupied by elephants (perhaps due to their size) and humans (due to their unique uprightness) in nature. The principle of the alternate limb-bending of humans implied in the second part of the chapter has a similar ring: it seems to be a rule of thumb for the principal manners in which human limbs are bent rather than an accurate description of all available directions of bending. That is perhaps the reason why, if strictly approached, the principle of alternate bending cannot accommodate all the motions available to human limbs (and to arms, in particular).



Part III Interpretative Essays

Diagrammatic Presentation of Logical Possibilities [IA 13, 712a1–13] In the first part of the chapter, the unstated but readily implied assumption is that the two members of the pairs of front and back legs are both bent in the same way. The opposite possibility is not considered, presumably because it would create a very unstable situation in walking, perhaps comparable to that of mutilated bloodless animals encountered in IA 8, 708b4–19. We may recall that IA 8, 708a21–b4, has independently established that all footed animals possess an even number of legs. The possibilities laid out in the first part are obviously explained by means of a diagram, but no such diagram is to be found in the manuscripts. Two options can be envisaged: either (i) a diagram originally accompanied Aristotle’s own text but was subsequently lost in the manuscript tradition, or (ii) a diagram drawn on a teaching board in Aristotle’s lecture room was taken for granted when Aristotle composed this text. I think that the second option is more likely to represent historical fact since the diagrams to be found in the manuscripts of other Aristotelian works date mainly from Byzantine times.34 In any case, mentioning Α, Β, Γ, and Δ would make sense only if such a diagram is presupposed. It follows that a diagram such as that which accompanies modern editions and translations of IA was part of Aristotle’s intentions. In Figure 9.1, the direction of movement is from right to left ( ). The left pair of arrows indicates front legs, the right pair refers to rear legs. Since no other options than the combinations produced by the forward–backward contrariety of bending are envisaged in this diagram, it follows that those egg-laying four-footed animals which bend their legs sideways, as Aristotle admits in IA 15, 713a15–25, are not meant to be included in this diagram. Aristotle claims that neither two-footed nor four-footed animals bend their legs as in Α or Β. (To be sure, in HA II 1, 498a3–16, Aristotle says that four-footed egg-laying animals bend both front and back legs forward. That would make them fall under category Β. But if what Aristotle has primarily in mind here are live-rearing fourfooted animals, those egg-laying animals referred to in HA II 1 do not find a place in our diagram.) Aristotle then claims that all four-footed animals bend their legs as in Γ with the exception of elephants, which bend their legs as in Δ, much as human beings bend their arms and legs. The same point, though duly constricted to live-bearing four-footed 34

For diagrams in modern editions of Aristotle’s Mechanics as compared with the evidence of the manuscripts see van Leeuwen 2016: 139–157.



9 De incessu animalium 12–13 A.


rear legs







front legs

rear legs

∆.

>