Aristotle on Logic and Nature (Aristote. Traductions Et Etudes) 9042938374, 9789042938373

Citation preview

ARISTOTE TRADUCTIONS ET ÉTUDES

ARISTOTLE ON LOGIC AND NATURE

EDITED BY

JAN-IVAR LINDÉN

LOUVAIN-LA-NEUVE

PEETERS 2019

ARISTOTLE ON LOGIC AND NATURE

ARISTOTE TRADUCTIONS ET ÉTUDES

ARISTOTLE ON LOGIC AND NATURE

EDITED BY

JAN-IVAR LINDÉN

PEETERS LEUVEN - PARIS - BRISTOL, CT

2019

ARISTOTE TRADUCTION ET ÉTUDES COLLECTION DIRIGÉE PAR P. DESTRÉE ET PUBLIÉE PAR LE

CENTRE DE WULF-MANSION RECHERCHES DE PHILOSOPHIE ANCIENNE ET MÉDIÉVALE À L’INSTITUT SUPÉRIEUR DE PHILOSOPHIE DE L’UNIVERSITÉ CATHOLIQUE DE LOUVAIN

Proceedings of the Helsinki Conference AristotleonLogicandNatureAristotelesüberLogikundNaturAristotesurlalogiqueetlanature (Part of the 2400 Years Anniversary Aristotletoday–Aristotelesheute–Aristoteaujourd’hui, conference series coordinated by the CentreforHistoricalOntology, www.c-h-o.eu.)

A catalogue record for this book is available from the Library of Congress. ISBN 978-90-429-3837-3 eISBN 978-90-429-3828-1 D/2019/0602/10

© 2019, Peeters, Bondgenotenlaan 153, B-3000 Leuven Tous droits de reproduction, de traduction ou d’adaptation, y compris les microfilms, réservés pour tous pays.

TABLE OF CONTENTS

ABBREVIATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3

PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7

Intentions and Directions Tomás CALVO MARTINEZ: Ὄρεξις and Intention . . . . . . . . . . . . . . Charlotta WEIGELT: “The soul is in a way all things”: Aristotle and Internalist Conceptions of Intentionality . . . . . . . . . . . . . . . . . Jan-Ivar LINDÉN: The Ambiguity of Appearance. On Responsivity in Aristotle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

31 45 67

The Role of Argumentation 85 Simo KNUUTTILA: Epistemic Logic in Aristotle . . . . . . . . . . . . . . . Miira TUOMINEN: Reason, Experience, and the Knowledge of the Principles in Aristotle’s PosteriorAnalytics 2.19 . . . . . . . 103 Christoph HORN: Aristotle on adhominem Arguments . . . . . . . . . 139

Reason and Nature in Ethics Thomas TUOZZO: Aristotle on Grasping Moral First Principles . . Jörn MÜLLER:‘Learning by Doing’? Aristotle on the Habituation of Moral Character . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Øyvind RABBÅS: The Functions of Practical Deliberation. . . . . . . Sarah BROADIE: Practical Truth in Aristotle. . . . . . . . . . . . . . . . . .

159 177 211 233

2

TABLE OF CONTENTS

Perspectives Annick STEVENS: Consciousness and Temporality: How Sartre Renews Aristotle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Erwin SONDEREGGER: Aristotle’s Theory of Nature from the Point of View of our Hermeunetical Situation . . . . . . . . . . . . . . . . . 271

ABBREVIATIONS

Categ. Deint. Top. An.pr. An.post. Soph.el. Phys. Decaelo Gen.corr. Meteor. Dean. Parvanat. De sensu Demem. Desomno Deinsomn. Dedivin. Delong. Deiuv. Deresp. Hist.an. Depart.an. Deinc.an. Demotuan. Degen.an. Probl. Met. EN EE MM

Categoriae Deinterpretatione Topica Analyticapriora Analyticaposteriora Desophisticiselenchis Physica Decaelo Degenerationeetcorruptione Meteorologica Deanima Parvanaturalia De sensu et sensibilibus Dememoriaetreminiscentia Desomnoetvigilia Deinsomnis Dedivinitationepersomnium Delongitudineetbrevitatevitae Deiuventute Derespiratione Historiaanimalium Departibusanimalium Deincessuanimalium Demotuanimalium Degenerationeanimalium Problemataphysica Metaphysica Ethicanicomachea Ethicaeudemia Magnamoralia

4 Pol. Rhet. Poet.

ABBREVIATIONS

Politica Rhetorica Poetica

PRESENTATION

The impact of Aristotle cannot be overestimated, covering not only the “first philosophy”, which later was to become “metaphysics”, but several different areas, ranging from ethics and politics to rhetoric and poetry. A special status belongs to the fundamentals of thinking, treated in the logical writings. Another core of Aristotelian philosophy concerns the philosophy of nature — issues of life and soul, natural kinds, animal movement, nature in all its aspects, including the translunar sphere of heavenly bodies. The psychology of Deanimais part of this philosophy of nature, but at the same time includes a noetic sphere, indicating another dimension of human life, which enables true knowledge and truly virtuous actions. These aspects of Aristotelian philosophy are often studied separately. While there are several important works on Aristotelian logic, ethics and psychology, the aim of the current volume is to offer perspectives on the interrelatedness of these domains.

INTRODUCTION Jan-Ivar LINDÉN

1. The Nature of Man and Intentionality There is a strong anthropological element in Aristotle’s philosophy, which expresses how man is dependent on something which is simply given. The zōonlogonechon is an animal, embedded in and interacting with nature, but also with a special relation to something noetic which partially stems from natural experience, but also indicates a special relation to something which seems to be almost divine. In order to articulate and realise this second dimension, every human being however needs to remain animal, i.e. have animal faculties like growth, nutrition, striving, sensibility, imagination, capacity to move etc. There is no possibility of defining the rational human being as a cogito without these functions — as Descartes did in the Meditations—even if the noetic dimension somehow transcends the animality of man. Thinking is a special articulation of something inside the life of a profoundly (in the Aristotelian sense) physical, i. e. natural being. As nature is largely understood from a teleological point of view in Aristotle’s writings, i.e. as a field of tendencies with specific directions, this also means that human life remains influenced by such natural tendencies which have their clearest expression in striving (ὄρεξις). Striving is, in a sense, that which creates movement in life and, interestingly enough, it is directly correlated with a certain finitude of existence and with a corresponding lack (στέρησις) — the human animal thus being directed towards the attracting (το ορεκτον). In a very physical sense a human being is thus directed towards something. In the philosophical tradition direction figures in another context, too, sometimes compatible with the Aristotelian idea of striving directionality, sometimes in a more subjective context, stressing directedness as a

8

J.-I. LINDÉN

significant trait of consciousness. The key word is then intention, which can have a more practical meaning (intend to do something) and a more general one (intention signifying that there is something meant, different from the act of conveying meaning, of being thus directed). The first concept of intentionality is important for pragmatist philosophers like William James, whereas the second one is well known through Husserl’s developments of Brentano (and implicitly of Brentano’s medieval forerunners). It is not easy to harmonise these two strains of intentionality. If directedness originally has something to do with striving, it would mean that a more pragmatic conception would be fundamental. Things are however not that easy, as pragmatism overstresses one aspect of striving, namely action and doing, and in this respect is close to modern forms of voluntarism. Taking into account that Husserl has a theory of passive synthesis, there are also arguments for a certain kinship between Husserl and Aristotle. These aspects are, however, somehow hindered by the Kantian roots in the idea of a Konstitutionsleistung, heavily present in Husserl’s phenomenology. Experience is in this setting a matter of “constituting” intentional objects and thus accords an eminent role to the subject, which is not that evident in an Aristotelian context, stressing given things, phenomena, sensual stimuli and noetic order. In the modern tradition of Bewußtseinsphilosophie the starting point for philosophical analysis is frequently a subjective sphere which is defined as consciousness (cogito), understood as a grounding principle — something which surpasses the occasional remarks Aristotle can have on self-perception.1 In contemporary discussions it is often taken for granted that intentional directedness would even as such be the criterion of consciousness. This is far from evident from an Aristotelian point of view. Direction is, for Aristotle, more to be understood as a characteristic of life, perhaps even of nature as such. It is important to understand how teleology and intentionality are connected. In this context it is regrettable that the superficial early modern critique of teleology still finds echoes especially in popularizing scientific journals, but also among some serious researchers. It is thus necessary to state something about what natural teleology is certainly not in Aristotle. When there is direction in nature, and in an especially obvious way in living nature, this does not imply a planning divine instance which through 1

For example, De an. 425b12-17.

INTRODUCTION

9

preferential choice (προαίρεσις) steers the processes. Striving is instead the decisive function — and Aristotle does not conceive striving as a noetic function, even if striving is omnipresent in the life of psychic beings and thus also manifests itself in our thinking. Thinking and even practical thinking (deliberation, βούλευσις) are closely related to rest, which in human actions means that the agent does not proceed in precipitation, following only his or her drives. As tendencies the entelechies of Aristotle have a direction, even if they do not always reach their goal. Planning, deliberation and thinking are different, concern other aspects of being: in planning and deliberation good ways of arriving at something, in theoretical thinking an ontological dimension which is no more directly related to practical interests. It is almost paradoxically as if the striving attitude in the theoretical realm is contrary to its driving force and has as its main function to affirm the stable, which does not itself move. 2. The Teleological Dilemma There is a problem in the modern critique of teleology. If no teleological direction in nature exists, but only mechanisms without immanent ends, then it was an understandable consequence that radicals already in the 18th century wanted to regard even the human being as a machine (La Mettrie).2 Already Descartes’ famous remark on the animals, which are, according to him, only machines, concerns the human body too. It was only the soul, reinterpreted as pure consciousness, which escaped this destiny. In thus suppressing the animating principle of living bodies (the soul in the Aristotelian sense), it was, in fact, the concept of life which was dismissed. It is thus not only because of spiritualistic exaggerations, that “vitalism” could become a schola non grata. Through the epistemic reduction of the soul, there was no place, or anyhow, no scientifically legitimate place left over for life. For Descartes himself, this seems to be more like an unintended consequence, not something he welcomed. In the later Cartesian tradition quite strange opinions about animals however bring us to more extreme versions3 (which are still 2

La Mettrie (1999). The refusal of Cartesians to regard animals as even sentient beings, was widespread. As an example the following passage of Malebranche on animals may be given: “ils mangent sans plaisir, ils crient sans douleur, ils croissent sans le savoir” and directly related to the topic of orexis, “ils ne désirent rien”, Malebranche (1963),p. 394 (262). 3

10

J.-I. LINDÉN

manifest in contemporary efforts to replace living beings with the constructions of “artificial intelligence” and synthetic biology). The paradox in all this is that none of the early modern philosophers denied that scientists intended to do science, search for truth, have different goals to achieve, sometimes in meticulously preparing experiments. In short: the teleological orientation and capacity of the modern subject was not contested, only the existence of teleological direction in the rest of nature. This was in the 17th century easier to claim, as biology was not the guiding science, but physics. Life was a less helpful concept in the latter kind of mathematical science. When enddirectedness in this way becomes a human privilege, there is a problem which one could call the teleological dilemma. Either the human being cannot be a natural being (as nature is not enddirected) or the human being is a natural being (and there thus exists enddirectedness in nature). The solution given in early modern rationalist philosophy tended to favour the first alternative and — referring to the theological idea of man as imagoDei — accept a quite spiritual definition of man as pure consciousness. The achievement was thus to found the new science on a theological principle. Precisely this was easier in the case of physics. Biology later on began to impose itself with a less mathematical world view (stressing change more than permanent laws) and with a quite naturalistic conception of man. That the quarrel about teleology reappeared in the 19th century was thus unavoidable. The same dilemma — to harmonise the critique of teleology with the existence of a meaningful (teleological) human world — remained. One consequence seemed near: if man is an ape (or a descendant of apes), i.e. a natural being, and the idea of a teleological nature is false, then we have to forget human teleology, too. Some reductionists still adhere to this radical solution. The problem is only, that they then apparently destroy the enddirected foundations of their own scientific endeavour. Other solutions, closer to what has been called emergence or supervenience, stress the possibility of qualitative leaps in nature. Human culture would then have a special status because of such qualitative leaps. It is however hard not to see the shadow of the imagoDei in this new version of a special status, a Sonderstellung. To resume: if human beings are natural beings and nature has to be conceived without any vitalist assumptions (i.e. mechanistically), our meaningful human world has to be reduced to mechanistic principles (the reductionist position). If, on the contrary, meaning is accepted as a real

INTRODUCTION

11

dimension of the natural beings we are, nature must instead be regarded as something, if not quite different from, still anyhow largely surpassing the lawful phenomena of mechanistic science precisely because it allows meaning (science included) to exist.4 Because of this teleological dilemma, it was understandable that early modern philosophers preferred the bifurcation of reality, which separated the subjective realm of meaning from nature, even if they, in this way, tended to lose the concept of life and accept an extremely spiritual version of subjectivity. 3. Is Intentionality a Criterion of Consciousness? Those who criticised the earlier teleology of nature regarded such a teleology as the essential characteristic of the vital conception of the soul. Certainly the definition of the soul as the first entelechy of a living being, offered some reason for this. Teleology is, however, only one aspect of life and it is not necessary to regard every form of vital meaning as a teleological phenomenon. Consciousness could be taken to indicate another form of “meaning”. The word in Aristotle which best corresponds to our awareness or consciousness is aisthanesthai, with strong sensual connotations, but with a quite complicated relation to enddirectedness. When we today speak of meaning, the notion is ambiguous: partly connected with purposefulness, partly with an affective dimension, which can be extremely strong and in conflict with purposeful behaviour. The popular thesis that consciousness could be defined through intentionality is far from evident. From Friedrich Schleiermacher to Antonio Damasio, thus in very different conceptual contexts, the claim has been quite different: that consciousness is more like a certain presence in what happens. If the process in question, what happens, has a direction, it is intentional and it certainly is possible to be conscious of the intentional direction, but consciousness itself is not an intentional state. Consciousness is rather momentaneous, more like a sensation (αἴσθησις) or perhaps a sentiment (feeling) of what is driving us around in the world. There is however a potential in this apperceptive capacity, which is 4 Cf. Galen Strawson on similar issues: Strawson (2006). If the quite ethnocentric opinion that meaningful human culture would be a splendid exception in nature is discarded, the existence of mechanistic science itself would conduct the naturalist to a point of view which stresses the natural reality of the epistemic function as such, perhaps even to panpsychism.

12

J.-I. LINDÉN

important both in theoretical knowledge and practical deliberation: through consciousness we are halting, not advancing, almost resting still, and are through this less enddirected (intentional) attitude capable of understanding more.5 Without its incarnation consciousness has however no possibility to realise itself. Aristotelian psychology always presupposes different psychic levels or functions (vegetative, animal, noetic), whereby the vegetative function is more fundamental from a purely natural point of view with some beings (plants) having only this psychic function and other beings needing it in order to exist, be they rational animals or only animals. The fact that Aristotle seems to conceive consciousness as a sensitive capacity means that several animals have consciousness, not only the zōonlogonechon. One should thus not confound consciousness with logos and nous. Logos as the characteristic of the human animal is a specific function, closely connected with noetic capacities, which allow man to speak, reflect and have knowledge of a theoretical kind (which seems to express a dimension, which is not momentaneous as the consciousness of aisthanesthai, but given without interruption).6 There are, however, some passages in Aristotle, which suggest a different view. In the text of Annick Stevens in this volume some of them figure and offer a contribution to a more problematic discussion of the precise relationship between modern mentalism and Aristotelian psychology. In the practical sphere, the reflective capacities rejoin the general enddirected animal striving and elaborate good ways of achieving goals. Whereas practical intelligence is somehow complicated through the deliberative reflection concerning adequate means, theoretical intelligence is, in a more disinterested way, contemplating what is, which for Aristotle means being one with it (coincidence of act and content).7 Logos is important in both the theoretical and the practical sphere and even if logos presupposes the capacity of aisthanesthai the two are not identical. Consciousness is more like the psychic awareness needed in intelligent functions, which are, however, in contemplative life not primarily intentional and even in practical life only partially intentional, namely insofar as a quite general animal striving is important for the 5 Cf.ἡ νόησις ἔοικεν ἠρεμήσει τινὶ καὶ ἐπιστάσει μᾶλλον ἢ κινήσει, De an. 407a32-33. 6 De an. 430a22-3. 7 De an. 430a3-7.

INTRODUCTION

13

practical agent. From an Aristotelian point of view, it is thus problematic to regard intentionality as a criterion of consciousness — and when philosophers in the footsteps of Brentano are doing this, the inspiration is probably linked to specifically modern preoccupations and a different philosophical anthropology, stressing the foundational role of consciousness in relations between subject and object.8 The philosophical anthropology and ontology of Aristotle are different, certainly stressing a special position of man, but only because of a spiritual affiliation, enabling certain activities, which are not possible for other animals. Not unimportant in this context is the role of practical reason as a certain moderation of temperament and orectic impulses. One could even suggest that this opens up a perspective on the role of logic in human life, logical thinking working as a sort of moderation of conceptual precipitation. 4. Logic and Psychology The stable has a privileged position in Plato as well as in Aristotle and quite generally in the ancient Greek world view. The challenge for Aristotle seems however to be to combine the stable with the changing. Nature and logic somehow express two sides of this project and the manner in which they are closely related is interesting — and untouched by the modern discussion concerning logic and psychology, the Psychologismus-Streit.One could even say that the term “psychologism”, if used in a less modern way, has the advantage of evading the now widespread notion of “naturalism” which tends to confound nature with matter and this in a completely unAristotelian way, which takes no notice of the quite complex concept of matter, which in Aristotle can be both relative, i.e. depending on perspective (bricks being matter for houses, but clay also for bricks) and absolute (indeterminate “prime matter”). The term psychology is also more adequate than the general term nature because it stresses the psychic function as such, which is a principle of life. Noetic processes are possible only in living beings and more precisely in animals with logos— even if nous apparently also has a separable status, independent of striving life. 8 For some references, see for example Caston (1998). The point of departure for Caston is intentionality as “that feature of our mental states in virtue of which they can correctly be said to be ofor aboutsomething or, more generally, possess content”, p. 250.

14

J.-I. LINDÉN

Aristotle offers conceptual possibilities to evince the problems which were frequently associated with “psychologism”. It is evident for Aristotle that logical demonstration (ἀπόδειξις) is not only a normative function, but also an expression of the noetic psuchē. In a way the syllogism can be regarded as a means of showing tendencies of thought — and halting them. In the conclusion the logical process comes to rest (ἠρεμήσει).9 It is as if language would be enddirected even in its particular demonstrative propositions and finding its fulfilment in the stability of the evident. Can this offer us new perspectives on the quarrel about psychologism — a quarrel which was mainly concerned with logic and knowledge, but has a wider significance, too, insofar as it involves the relation between processes (or actions) on the one hand and contents on the other. How are acts and contents related? What is a definition? How do concepts work? Which is the link between a proposition, in its truth value seemingly independent of who is bringing it forth, and a meant utterance? In the quarrel about psychologism (virulent especially in the Germanspeaking world in the late 19th and early 20th century, but by no means thematically limited to this area and period), several camps were involved: as defenders of irreducible logic (and mathematics) mainly Gottlob Frege and Edmund Husserl, as defenders of the psychological perspective on philosophical questions in general and on logic in particular Benno Erdmann, Christoph Sigwart, Wilhelm Wundt and several others. This is perhaps the main constellation, but other currents were involved as well, notably the Neo-Kantian philosophy of Paul Natorp, Wilhelm Windelband and Heinrich Rickert and the Lebensphilosophie, particularly relevant through the hermeneutic psychology of Wilhelm Dilthey.10 The background in English and Scottish associationism and in French equivalents, especially in the so called “ideological” movement, is not unimportant, either.11 9 “The conclusion is somehow an end of the deduction” (πως τέλος, τὸν συλλογισμὸν ἢ τὸ συμπέρασμα), De an. 407a26-27. 10 An interesting text, close to the philosophy of life, even if not that convincing in its critique of Aristotle (p. 77-84) is Palágyi (1903). See especially p. 191-210, where Palágyi gives the relation to the eternal a logical meaning. 11 From an Aristotelian perspective, Husserl is probably more interesting than Frege, because of the emphasized correlation between act and content in phenomenology. The famous text on psychologism is Husserl (1913). A survey of the quarrel for example in

INTRODUCTION

15

This is no place to give a general overview of the PsychologismusStreit, but as the topic is of importance precisely for the relation between logic and nature, it can be of interest to see how Aristotle could confront or dismiss some of the presumptions in this debate. Clear is that there is a platonic strain both in Frege and in the early Husserl, when they defend the independence of logic (and mathematics), i.e. a special ideal status, which no empirical science can determine. One main tenor is that empirical lawfulness is not exact, evident and necessary as is true logical deduction. Another aspect stressed is that the logical is a priori and thus cannot depend on empirical science. There is here an obvious Kantian background, even if the differences should not be underestimated. The Kantian strain is particularly clear in Husserl’s case.Things are however even more complicated as there was also a Neo-Kantian critique of psychologism, which differs from the fundamentally descriptive stance of the Husserlian one. For Neo-Kantians the difference between psychology and logic is a difference between a descriptive and prescriptive sphere, logic belonging to the prescriptive realm, indicating an important dimension of values, the main topic especially for the southwestern Neo-Kantians (the Wertphilosophieof Windelband and Rickert). Finally, there is the position of Dilthey, who develops a psychology, which is not the experimental one en vogue, but which instead stresses “understanding” (Verstehen) as a scientific approach different from “explanation” (Erklären). This of course completely changes the focus in the dispute about psychologism, as it was often taken for granted that psychology would be the emerging experimental psychology, which Windelband called a “natural science of mind” (geistigeNaturwissenschaft).12 If psychology, as Dilthey claims, is more like the special “understanding” access we have to living nature, it does not seem to imply any reduction of logic, or of truth, or of moral obligations. In fact, one could even mean that Dilthey is here quite close to Aristotle, who would certainly not contest that psuchēis the essential characteristic of life. Less evident is, however, that Aristotle’s conception of this life as manifesting evident logical connections, truth and the morally good would be meaningful as such for a philosopher of life. There are here interesting differences between Rath (1994). Concerning the Neo-Kantian context and experimental psychology, see Gundlach (2017). Concerning Brentano, see Brentano (1924). 12 Windelband (1907), pp. 360-2.

16

J.-I. LINDÉN

Nietzsche, Dilthey, Bergson and others, but it is worth remembering that a philosopher, heavily influenced by the philosophy of life, like Max Scheler, in a quite strong manner reappropriates both theoretical and practical Aristotelian topics.13 5. Predication and Judgment In his fourth meditation, Descartes states that every judgment is heterogeneous, made up of a judging act and a content judged. For Descartes, the judging act is volitive, understood as either affirmation or negation, and the content either an imaginative representation stemming from sense perception or something ideal. Truth and error are possible only through judgments, i.e. through a volitive attitude towards (affirmation or negation of) contents. This epistemically reduced notion of volition is later elaborated and clarified by Kant through the concept of spontaneity, which offers the basis not only for the Kantian theory of experience but also for his notion of judgment. Both in Descartes and in Kant the human subject has an eminent position as the spontaneously judging instance, which creates the meaningful space for epistemic claims, which can be true or false. It is this aspect, which later in the Neo-Kantian tradition will conduct to a general philosophy of value, in which also science and knowledge are dependent on a dimension of already presupposed values. 13 A general tendency in the Lebensphilosophie was to stress a concept of nature (quite different from the one in contemporary naturalism), where nature has as its most familiar appearance precisely life, i.e. nature as it expresses itself in ourselves, and in a less familiar way also in other living beings. This renders the notion of subjectivity problematic, as reality is working inside us. The conception is not directly linked to any vitalist idea of intelligent organisation, even if biologists like Hans Driesch gave it such a sense. Driesch (1928). Some currents of philosophy of life were highly critical about teleological assumptions, while other more Aristotelian strains adhered to teleological conceptions. In both cases, there is a potential for a less anthropocentric anthropology, which does not accord to man the role of constituting things in the world. This is evident already on the level of sensual perception and philosophers of life like Dilthey, Nietzsche and Bergson are less prone to regard the spontaneity of a subject as responsible for the determinate character of what appears than both Kantian, Husserlian and associationist writers. Association as a psychological principle presupposes that primary data are plural and also Kantian and Husserlian synthesis is founded on the same idea of an elementary sensual manifold. This atomism of data is exactly what demands association and synthesis. If the originally given is not a manifold but an indeterminate unity, the psychic response must be more like dissociation. Bergson expressed this point in the following way: “C’est l’herbe engénéral qui attire l’herbivore”, Bergson (1990), p. 177. It would be interesting to compare with the Aristotelian notion of krinein.

INTRODUCTION

17

In a way, the faculty of judgment is the place where voluntarist and rationalist motives meet. There is thus a voluntarist aspect of the epistemological tradition, which was however downplayed in several later theories of truth. It even created a scandal when another famous representative of psychological views in philosophy, William James, in an almost Nietzschean manner reminded of the volitive aspects in claiming that truth is a concept like wealth, health and strength.14 He was accused of confounding truth with the psychological state of holding something for true. It is certainly characteristic of this pragmatic point of view, that truth is more like a capacity than like a content or even a judgment. How should we regard Aristotelian predication in this context? In De interpretatione, where Aristotle elaborates the logos apophantikos, he nowhere speaks of will (βοὺλησις), only of denial (ἀπόφασις) and affirmation (κατάφασις). When Descartes identifies these two judging functions with a voluntary act, he has thus modified Aristotle’s theory (a theory he most certainly was acquainted with). It is true that Descartes also mentions the Aristotelian analogy between affirmation and denial on the one hand and fleeing (φεύγει) and pursuing (διώκει) on the other hand, thus suggesting a certain kinship between the judging functions and striving,15 but it seems in Descartes’ case to be only a parenthetical comment, showing erudition more than thematic importance. Aristotelian striving is not the same as Cartesian will. In fact, the Aristotelian concept of will (βούλησις), especially when it is mediated by deliberation (βούλευσις), is very different from modern conceptions, which stress the aspect of pure spontaneity whereas will (when it is not only wish) for Aristotle more seems to be something which moderates spontaneous tendencies through a capacity of holding them back.16 It is 14

James (1975), p. 104. De an. 431a9-10, EN1139a23-25. 16 One has to keep in mind that that the βουλή behind both the concept of will (βούλησις) and deliberation (βούλευσις) means council. Cf. Aubenque (1963), p. 111. When boulēsisis only wish, the noetic mediation is lacking, but wishes concern something which is mostly not given, but only imagined, and thus do not permit any direct actions. There is accordingly in wish something which already through this postponed realisation invites to deliberation — and after the bouleusis the boulēsisis very much like that which commentators want to describe as “rational desire”. See for example Pearson (2012), especially p. 170-198. When Aristotle is connecting deliberation and striving and even mind and striving in expressions such as “striving mind” (ὀρεκτικὸς νοῦς) and “thinking striving” (ὄρεξις διανοητική), this expresses two sides of the preferential choice (προαίρεσις) and must not necessarily imply any specifically noetic kind of orexis, 15

18

J.-I. LINDÉN

more like a complication of striving (complicated striving) than like spontaneous striving itself. What is similar in Aristotle and Descartes (explained by the scholastic education of the latter) is the claim that a specific combination is necessary for something to be true or false. There is thus independent reality, but not independent truth, because truth (as falsehood, too) needs affirmation or denial. This idea of a combination (συμπλοκή, σύνθεσις)17 which renders truth and falsity possible, is a complicated matter in Aristotle and it is not always easy to see how the meaning conveyed by predication — the apophansis, saying something about something (combining logical subject and predicate, onoma and rhēma) relates to the judging attitude of hupolepsis. The fact that predication with its two forms, ascribing something to something (τι κατά τινός) and negating something of something (τι απο τινός) indicates that predication as such is a judging act, either affirming (κατάφασις) or denying (ἀπόφασις). At the same time, there seems to be some kind of meaning affirmed or denied and this meaning could be supposed to be independent of the judging act and attitude.18 The question is somehow to determine if predication is possible without a judging act and attitude. Aristotle seems to stress the act, what indicates a difference between later semantics of propositions and Aristotelian logic. It is as if propositional contents are only possible without judgments, but need the judicative attitude for their actualisation in real utterances.19 Predication is thus always a question of denial or affirmation. For Descartes the judicative relation is between spontaneous will and contents, for Aristotle judgment is more like affirming and denying a tendency in a realm of ordered possibilities. This precisely offers the something which would also be in a certain conflict with the close connection between rest and thinking in Aristotle. In the aforementioned passage Aristotle even closes with the statement “and the principle of this is the human being” (καὶ ἡ τοιαύτη ἀρχὴ ἄνθρωπος, EN1139b4-7), which once more emphasizes the whole incarnated being as the choosing agent. Cf. my contribution “Embedded Deliberating Animals. Aristotle’s Theory of Action” (with some more references) in the forthcoming Proceedings of the Padua conference ThePoliticalPhilosophyofAristotle,Rossitto (2018-19). 17 This synthesisis of course not to confound with the above-mentioned modern synthesis of sensual manifolds. 18 The word Aristotle uses for predicate, rhema,has also a temporal aspect which corresponds to our verbs (“walks”), what can also be rendered in the form “is walking”. The “is” in attributes like “is walking” and “is white” gave later rise to a vast literature on the combining copula. 19 Cf. De an. 429a22-24.

INTRODUCTION

19

analogy between affirmation and denial on the hand and fleeing (φεύγειν) and pursuing (διώκειν) on the other hand. That this latter couple stands for striving functions is evident, but should the analogy between these drives and the judicative function of affirmation and negation also be taken to imply that even epistemic judgments are founded on orexis?Is this latter kind of striving then different from the practically relevant one, or does it fundamentally stem from the same natural root? Is there a specific form of affirmation and denial in the theoretical realm, which has little to do with natural drives — or should we account for the difference between praxis and theory in terms of the judged contents only? Certain is that Aristotle would not want to reduce theory to praxis. In epistēmē the judicative function affirms or denies a content which is not of immediate practical interest, but the question remains: which is the psychic source of the judicative aspect? Is it noēsis or orexis? It would be tempting to link contents to noēsis, but it is harder to see how affirmation and denial could be purely noetic, especially because of the above mentioned analogy with natural drives like fleeing and pursuing which Aristotle himself brings. The question is: does it exist a specifically epistemic kind of orexisor is every orectic function as such a natural drive (even if it can express itself in both theoretical, practical and technical contexts)? Perhaps one could speak of two levels of judgments in epistemic contexts, one which is more like a starting point (“prejudgment”) and another which is mediated through critical thought (judgment,judicium) and somehow corresponds to the decision (προαίρεσις) in the practical sphere, which is what follows when deliberation ends and the agent finally accepts to affirm a real tendency in action, the difference being that epistemic acts do not aim at actions but at theoretical (disinterested) understanding.20 The reality of the epistemic act however depends on its natural incarnation and thus also on striving as judicative attitude. It is in this light understandable that Aristotle in his logical writings, especially in De interpretatione, repeatedly speaks of logical 20 It is interesting to notice that Aristotle in one of the formulations of the so called practical syllogism uses aisthesis, phantasia and nousas alternative instances, offering the presence of what is desirable: “I have to drink, says the desire: this is something to drink says the sensual perception or the [representing] imagination or the mind: and then one drinks” (ποτέον μοι ἡ ἐπιθυμία λεγει· τοδί δὲ ποτόν͵ ἡ αἴσθησις εἶπεν ἢ ἡ φαντασία ἢ ὁ νοῦς· εὐθὺς πίνει), Demotuan.701a32-3 In this case the consequence is an action, an affirmed tendency, and the mediating act a minimal one.

20

J.-I. LINDÉN

processes “in the soul” (ἐν τῇ ψυχή). The soul is a principle of life and thus natural. Interesting is also how Aristotle switches between logical analysis and remarks on physical phenomena. Hupolēpsis is a difficult concept, both to translate and to understand. The common verb behind the more specific Aristotelian concept is hupolambanein, which has several meanings, not only judge, assert, assume, mean, suppose, have an opinion, believe, but also take up, originally take up from underneath, like in the story told by Herodotus about Arion, who was forced to jump into the see, but got the favour of one last song before jumping and thus attracted a dolphin who saved him, comingfromunderneath and taking him up. Werner Theobald has drawn attention to the importance of this conceptual background (even if his interpretation stresses the “saving” reason more than the actualising orexis in the hupolambanein).21 The myth of Arion was apparently well known, also through Plato, and the Aristotelian use of hupolēpsiscan be brought in accordance with the myth in different ways. In the present context, one could stress the dolphin as an animal coming from below, rendering it possible for the singer to live on and use his tongue for the good life, which this organ according to Aristotle can promote when it is human and not only an animal means for nutrition.22 It is however true that some passages suggest that hupolēpsis would be a noetic function. In the Nicomachean Ethics Aristotle says about brutes (θηρία) that “they don’t have the judgment about the universal, but only imagination and memory about particulars” (οὐκ ἔχει καθόλου ὑπόληψιν ἀλλὰ τῶν καθ᾽ ἕκαστα φαντασίαν καὶ μνήμην).23 It is however important to notice that Aristotle does not here speak of hupolēpsis as such, but of a hupolēpsiswhich concerns the universal. It would also be strange not to attribute some kind of judgment to animals if, as Aristotle tells us, pursuing and fleeing are judicative functions. It is certainly true that judgments are important in thinking. The question is, if nousand logos alone permit this. Even when Aristotle links nous to dianoia and hupolēpsis,as is the case in Deanima, the agent is not the nous but the psuchē: “I call mind that by which the soul thinks and judges” (λέγω δὲ νοῦν ᾧ διανοεῖται καὶ ὑπολαμβάνει ἡ ψυχή).24 As I far as I can see, 21 22 23 24

See Theobald (2002), p. 25-37. De an. 420b16-20. EN1147b5-6. De an. 429 a22-24.

INTRODUCTION

21

this does not exclude the interpretation of orexis as a decisive element in judgments. Sometimes hupolēpsisis used as a general notion which covers various forms — “knowledge (ἐπιστήμη), opinion (δόξα), prudence (φρόνησῐς) and their opposites”, something which would fit well with the idea of a judgment which actualises possible contents and permits different manifestations.25 Opinion is often contrasted with knowledge as something without the universal necessity characteristic of epistēmē (which for Aristotle is always true). There are however contingent phenomena and opinion is adequate for this realm of matters of fact which could be otherwise.26 The contrast Aristotle here uses is between doxa and epistēmē, not between hupolēpsisand epistēmē.If hupolēpsisis a function of actualisation, important in both necessarily true universal knowledge and in particular opinion, it seems clear that the above mentioned remark on judgment as a human privilege concerns judgment about the epistemically relevant universals and not every form of hupolēpsis. Aristotle also describes hupolēpsis as something which offers the starting point for argumentation, what could be interpreted as judgment involving a psychic attitude towards a possible content which actualizes this possibility and initiates thinking. Epistemic exercise and involvement is thus important for what we understand and the judicative function somehow has the possibility to transform dispositions into real knowledge. Psychologists in the 19th century stressed the inseparability of these two sides, which has something plausible, but they also tended to a reductionist point of view, overstressing processes to be explained in a modern scientific manner with the means of experimental psychology. This later view has rightly been criticized. Anti-Psychologists in the Fregean vein however in a similar way overstressed the propositional content, and in analytical philosophy it was only through speech act theory and Wittgensteinian conceptions of meaning as use that this tendency later found some corrections. Elisabeth Anscombe’s writings on Aristotle are situated in this same context. Husserl’s position is interesting, because he tries to develop a method of describing the correlation between acts and contents. Through this he later earned the accusation of a “relapse” into psychologism by 25

De an. 427b25-27. In this passage Aristotle however does not follow up the topic. An.post.89a2-3 and 89a passim, which deals with the relation between knowledge and opinion. 26

22

J.-I. LINDÉN

those who tended to regard every intrusion in the ideality of a propositional content as psychologistic. From an Aristotelian point of view the problem in Husserl’s approach is certainly not the crucial role of acts, but the Kantian way of regarding these acts. Phenomenologists sympathetic to Aristotle, however, see possibilities for overcoming these difficulties through Husserl’s remarks on “passive synthesis”. 6. Aristotle’s Non-reductive Psychology Contradiction (ἀντίφασις) in the logosapophantikos stems from incompatible judgments. The logical writings of Aristotle explore the different forms of incompatibility and compatibility. Special attention is given to the form of deduction called syllogism (συλλογισμός), going from premise (πρότασις) to conclusion (συμπέρασμα). The approach of Aristotle is his usual one. He takes a certain field, in this case logical relations and judgments, as his starting point and tries to find out, what is involved. In this way, he expects to shed light on the essence of logical demonstration (ἀπόδειξις). This is a vast field of inquiry and only some aspects are treated in the contributions of this volume. In the present introduction, it is not possible to give a survey neither of the discussion concerning Aristotelian logic nor even of the arguments in the following contributions, but some further remarks related to the perspective logic and nature may be allowed. The question is then how the natural incarnation of logical processes should be understood in general and in relation to enddirectedness in particular. It makes sense to regard orexis as a wide term which designates the drive in living perceiving beings, something which accords also to their noetic capacity a psychic, i.e. natural reality. If striving is what renders a content judicative, then judgment would be what anchors the logos of apophansis in the life of human beings. It would also explain the expression “in the soul” which is used in Aristotle’s description of the logos apophantikos. Could it be that judicative truth is given when the logical animal, i.e. the human being, claims something in accordance with the noetic essence of his soul? One should remember, that the soul according to Aristotle is in a way “all things”, a remark which both states that the soul is not a subjective inner world and that reality is part of us.27 Charlotta 27

De an. 431b21-22.

INTRODUCTION

23

Weigelt gives the topic an interesting discussion in this volume, stressing the non-dualist approach it implies. Concerning intentionality, it means that the soul more is to conceive as an actualization of direction than as a way of establishing intentional relations to something external in somehow “stretching” itself out towards things.28 Weigelt speaks of a “mutual actualisation”. We are dependent of, belong and contribute to reality and psychic life with all its different levels expresses this belonging. In logic the important psychic level is noetic, but noēsis is not independent of the general psychic function. When something noetically essential finds its adequate expression in a true demonstration it is not a subjective matter, but corresponds to a noetic reality which has priority over any personal epistemic undertaking. Such undertakings are successful when they manage to accommodate to this given dimension of noetic reality.29 The Aristotelian critique of Plato’s notion of methexis, which is mainly directed against the role of ideas in Plato, should not obscure the fact that vital belonging to social and natural reality is central to Aristotle’s own conception of the psuchē. Accommodating oneself in reality is also something which generally characterizes ancient Greek thought. It can be a personal endeavour to find one’s place in a larger social context, but it can also mean finding one’s thoughts in the noetic order of reality. This does not imply that there would be no open margins in the cosmos. On the contrary, the whole Aristotelian theory of praxis takes for granted that such margins exist — but, as Tomás Calvo points out, the noēsisnoēseōs is not a practical agent. The main tenor in Calvo’s opening text concerns however the meaning of desire in Aristotle and the conceptual varieties of orexisin particular.My own contribution on perception inside an ontological conception which links orexis to belonging profits from this point of view. Sarah Broadie offers a differentiated analysis of the notion of truth in general and of practical truth in particular. Her defense of practical truth gives an important additional perspective on what judgments are and indicates how the alētheiapraktikēis an Aristotelian answer to the Platonic problem of knowledge as source of ethical behaviour. 28

De mem. 452b9-11. Even in the predicative relation between that of which something is said and what is said of it, Aristotle uses the word huparchein, which can mean precisely belonging to (An.pr.,passim). 29

24

J.-I. LINDÉN

Affirmation and denial are important functions in our finite existence, profoundly marked by striving, and logosalone is not sufficient for judgment. If the desiderative (orectic) element is necessary, it has however different degrees: sometimes directly pushing for action and thus initiating the execution (what Aristotle expresses with the conceptual couple fleeing and pursuing), sometimes in a more moderate form only articulating an attitude like in epistemic affirmation and denial, initiating only thought and not action. This renders Aristotle’s remark on striving as one of the three functions governing also the epistemic thought of the logistikon meaningful (the other two being aisthēsis and nous).30 A singular noetic order of reality in no way excludes different modes of access. This seems to be the main tenor in Aristotle’s recurring stress on the habitual structures in life and also on the habitual modes of thought, the endoxa. The last text in this volume by Erwin Sonderegger deals with possible consequences of such a perspective. Singular access does not imply any relativism, but only the fundamental importance of specific relations. Opinion (δόξα) needs noetic order as its foundation. As endoxa opinions become convincing and in some cases they are true. They can however be true only inside a judging attitude which depends on specific striving. This does not imply that truth would be the same thing as holding something for true, but it means that truth has no meaning outside judgments. Truth is thus dependent on specific access, and specific access dependent on reality, psychic processes and noetic order included. Even the mathematical relations of astronomy (to take an example which from an Aristotelian perspective is less polysemic than physics) must be affirmed in order to be true. This fact does not diminish the reality of astronomical processes. At the same time, precisely this intermediate role of judgment renders it extremely important to study all the means we possess of proving something: conceptual definitions, how we grasp what is general in the particular, make differences between the essential and the accidental, specify deductive principles and valid demonstrations. In this context, the relation between truth and epistemic attitudes is of particular interest. Simo Knuuttila develops these topics in comparing Aristotle’s views with medieval conceptions. Miira Tuominen discusses the relation between the grasp of principles and the possession of simple universals, which seem 30

EN 1139a19-21.

INTRODUCTION

25

to be preconditions of the epistemic process, not conceptual in the rationalist sense, but anchored in the world as its universal aspect. Knowledge is also in this case intimately linked to something which “comes to rest in the soul”. If nature is partly a constituent of logic, Aristotle never reduces this nature to an object of natural science — as is often the case in later “naturalistic” conceptions, supposing that an objective sphere depending on a certain kind of explanatory natural science would also be the foundation of this science and of knowledge in general. Such a non-transcendental naturalism is not what Aristotle offers, but neither is it a transcendental philosophy in the Kantian sense. Aristotle uses something which could be called transcendental arguments, but with an accessible manifestation of reality instead of the modern subject as the basic condition of possibility of experience. It is not astonishing that the difference between the logical animal uttering something and the ideal content of what is said in the utterance poses a certain problem for Aristotle, who stresses the role of habits and virtues not only in the ethical sphere. On the one hand, it seems important to argue about topics (adrem) and not against persons (adhominem), but still it is a fact that no real discussion would be possible without persons who stand for what they say and have said. One has to subscribe to what one says, take responsibility in one’s responses. In the contribution of Christoph Horn, these questions are given a thorough discussion. The role of dispositions and virtues in the grasp of central meanings is important in several Aristotelian contexts and there is a strong tendency in Aristotle to regard both moral and other first principles, not as “crowning achievement”, but as starting points emerging from habitual being. Like Jessica Moss31, Thomas Tuozzo strongly emphasizes this aspect of an original “sensitivity to value”, established by virtue, but not sufficient for practical wisdom (φρόνησις). The question is similarly present in the contribution of Jörn Müller, which clarifies the relation between poiēsis and praxis and some aspects of how learning is dependent on doing, stressing what Nancy Sherman has called “critical habituation”, but trying to evince any “cognitive overload” of this mnemic function. The problem is similar to the one which guided Félix Ravaisson’s

31

See for example Moss (2011), p. 204-261.

26

J.-I. LINDÉN

famous works on habitual memory.32 Øyvind Rabbås discusses in a related context the problem of “division of labour between character and thought” and proposes a solution where the responsivity in situations enables deliberation not only to concern the means for achieving preexisting ends (Aristotle’s explicit statement about bouleusis), but also to “scrutinize” the ends. The issue is important because of its implications for the position of human reason in the ethical sphere: how much is rationally guided and how much given in some other way, through second nature and education? For moderns this is a decisive question because of our loss of the ontological confidence that there is something good in the world order, a confidence which was for Aristotle intimately related to a different, less interventionist anthropology — even if action is of course also for Aristotle a central category. There seems however to exist an important difference between practical quite relative goals (σκοπόι) and necessarily presupposed ends (τέλοι), the difference being that goals are also means, while ends are presuppositions of goals and means, but not themselves means. LITERATURE AUBENQUE, Pierre (1963), La prudence chez Aristote, Paris: Presses universitaires de France. BERGSON, Henri (1990), Matière et mémoire, Paris: Presses universitaires de France. BRENTANO, Franz (1924), Psychologie vom empirischen Standpunkt, I Band, Leipzig: Meiner. CASTON, Victor (1998), “Aristotle and the Problem of Intentionality” in PhilosophyandPhenomenologicalResearch, vol. LVIII, no 2, June 1998. DRIESCH, Hans, PhilosophiedesOrganischen.Gifford-Vorlesungen,gehaltenan der Universität Aberdeen in den Jahren 1907-1908, Leipzig: Quelle & Meyer. GUNDLACH, Horst (2017), WilhelmWindelbandunddiePsychologie.DasFach Philosophie und die Wissenschaft Psychologie im Deutschen Kaiserreich, Heidelberg: Heidelberg University Publishing 2017. JAMES, William (1975), Pragmatism. A New Name for Some Old Ways of Thinking in TheWorksofWilliamJames, Cambridge, Mass. and London: Harvard University Press. 32 Ravaisson 1984 (1. ed. 1838). See also his Essai sur la métaphysique d’Aristote, Ravaisson 1996 (1. ed. 1837).

INTRODUCTION

27

HUSSERL, Edmund (1913), LogischeUntersuchungenI,Prolegomenazurreinen Logik,Halle: Niemeyer. LA METTRIE, Julien Offray de (1999), L’hommemachine,Paris: Denoël (1. ed. 1748). MALEBRANCHE, Nicolas de (1963), OeuvresdeMalebranche,tome II, Recherche delavérité(éd. G. Rodis-Lewis), Paris: Vrin. MOSS, Jessica (2011), “Virtue Makes the Goal Right”: Virtue and Phronesis in Aristotle’s Ethics” in Phronesis 56/2011, p. 204-261. STRAWSON, Galen (2006), “Realistic Monism: Why Physicalism entails Panpsychism” in ConsciousnessanditsPlaceinNature(eds. G. Strawson et al.), Exeter & Charlottesville: Imprint Academic. PALÁGYI, Melchior (1903), Die Logik auf dem Scheidewege, Berlin: C. A. Schwetschke und Sohn. PEARSON, Gilles (2012), AristotleonDesire,Cambridge: Cambridge university press. RATH, Matthias (1994), DerPsychologismus-StreitinderdeutschenPhilosophie, Freiburg/München: Alber. RAVAISSON, Félix 1984, Del’habitude, Paris: Fayard (1. ed. 1838). RAVAISSON, Félix (1996), Essai sur la métaphysique d’Aristote, tome I, Hildesheim, Zürich et New York: Georg Olms Verlag (1. ed. 1837). ROSSITTO, Cristina (2018-19), Proceedings of the Padua conference Thepolitical philosophyofAristotle(ed. C. Rossitto),forthcoming. THEOBALD, Werner (2002), “Spuren des Mythos in der aristotelischen Theorie der Erkenntnis. ‘Hypolepsis’ bei Aristoteles” in Archiv für Begriffsgeschichte44 (2002). WINDELBAND, Wilhelm (1907), Präludien.AufsätzeundRedenzurEinleitungin diePhilosophie,Tübingen:J. C. B. Mohr (Paul Siebeck)

INTENTIONS AND DIRECTIONS

ὌΡΕΞΙΣ AND INTENTION Tomás CALVO MARTINEZ

In what follows, I will offer some considerations about the central role played by orexis in Aristotle’s anthropology and, more generally, in his understanding of the life and behaviour of living beings endowed with consciousness. Before entering into the content of my contribution I would like to make two preliminary remarks. Firstly, I want to state that along my paper, in order to avoid any misunderstandings, I am only and always using the noun ‘desire’ to translate the Greek word ὄρεξις, as well as the verb ‘to desire’ to translate the Greek verb ὀρέγομαι-ὀρέγεσθαι. Secondly, I want to remark that I think ― and I will try to show ― that, strictly speaking, orexis is not a phenomenon associated with life in general. It does not belong to every natural living being. It specifically belongs to sentient living beings. Properly understood, orexis is a kind of intention which permeates all the activities of sentient natural beings.1 1 We have not to forget that, when philosophers and historians of philosophy quote Aristotle as an antecedent of the contemporary idea of “intentionality”, they do not refer specifically to Aristotle’s notion of orexis. As is known, the phenomenological idea of “intentionality” brings us back ― through Brentano ― to the medieval notion of intentio and inexistentiaintentionalis. And the medieval theory of intentiones, in its turn, becomes associated with Aristotle’s conception of the soul as the place of forms (τόπος εἰδῶν: De an. III 4, 429a27-29), as well as his complementary statement that the soul “is in a way all things” (πῶς πάντα) (ibid.). To which one could add the Aristotelian qualification of psychic phenomena as essentially relative (πρός τι). However, as scholars like J-I Lindén have remarked, I think that any attempt to connect or to compare the Aristotelian psychology with contemporary phenomenology has to take Aristotle’s notion of ὄρεξις into account. Although from a phenomenological point of view, we have to acknowledge that Aristotle’s attitude is the one described as the “natural attitude”, since it is not consciousness which “constitutes” or grounds nature. On the contrary, it is the real nature (φύσις) which “constitutes” or grounds our consciousness. (Cf. Lindén (2011-2012), pp. 339-52).

32

T. CALVO MARTINEZ

1. Ὄρεξις and the Different Kinds of Soul According to his permanent will to systematize and to classify phenomena, Aristotle appears particularly interested in classifying the various vital functions, as well as the different parts or kinds of soul. Surely, I think, he felt strongly dissatisfied with the Platonic classification of the kinds of soul, and particularly with Plato’s tripartite division of the soul into νοῦς, θυμός and ἐπιθυμία. In De anima III, 9, we have the following text: “The problem at once presents itself, in what sense we are to speak of parts of the soul and how many there are. For in a sense there seem to be an indefinite number and not only those which some people say when they distinguish the calculative (τὸ λογιστικόν), the passionate (τὸ θυμικόν), and the appetitive (τὸ ἐπιθυμητικόν), or according to others the rational (τὸ λόγον ἔχον) and the irrational (τὸ ἄλογον) parts; for taking into account the differences by which they distinguish these parts, there clearly will be other parts too, far more distinctly separated from one another than these, those we have just mentioned: the nutritive (τὸ θρεπτικόν), which belongs both to plants and to all animals, and the sensitive (τὸ αἰσθητικόν), which could not easily be classed as either irrational or rational; further the imaginative (τὸ φανταστικόν), which is, in its essence, different from all, although it is very hard to say with which of the others it is the same or not the same, if we are to posit separate parts in the soul; and lastly the desiderative (τὸ ὀρεκτικόν), which would seem to be different from all both in definition and in potentiality. And it would be absurd to break up the last-mentioned faculty: for wish (βούλησις) is found in the calculative part and appetite (ἐπιθυμία) and passion (θυμός) in the irrational; and ifthesoulistripartitetherewillbedesire ineachpart (ἐν ἑκάστῳ ἔσται ὄρεξις).” (De an. III, 9, 432a22-b7)2

If we attentively read this passage, we will find, I think, that the whole of it contains a critical remark against the Platonic-Academic classification or classifications, against the one that distinguishes three parts or functions, or even three souls (νοῦς, θυμός and ἐπιθυμία), and also against the other one that distinguishes only two parts, the rational and the irrational ones (τὸ λόγον ἔχον and τὸ ἄλογον). Actually, the tripartite classification can be easily adapted to the bipartite one by incorporating passion (θυμός) and appetite (ἐπιθυμία) into the irrational part. Even more, from a strictly biological point of view one should also include the nutritive function (τὸ θρεπτικόν) into the irrational part. 2 I have used the English translation of DeAnima by D. W. Hamlyn and by J. A. Smith, as well as the translation of Physics by R. P. Hardy and R. K. Gaye. However, I have introduced several changes in the passages I quote from both works.

ὌΡΞΙΣ AND INTENTION

33

By doing so, the bipartite division into the rational and the irrational parts would become definitely enriched. In fact, Aristotle himself resorts to this classification in the Nicomachean Ethics I, 13, 1102a26ff with the explicit indication that it comes from the “Exoteric Treatises”.3 And, once it has been enriched this way, the bipartite classification seems to become both useful and sufficient within the ethical context of the study of virtues, although it continues to be insufficient to afford a full explanation of the human nature from a psychological and anthropological point of view. In my opinion, Aristotle’s main objection and criticism of the Platonic conception of the soul is directed against the tripartite classification of nous (reason), thumos (passion) and epithumia(appetite) in so far as thisdivision seemstobebasedontwodifferentandalternativecriteria. Thus, its first term,nous, belongs to the field of cognition, is a kind of knowledge among others, while the other two words,thumos and epithumia, do not belong to the field of cognition at all. They rather belong to the sphere of emotions and desires. Now, as is known, the logical rules for classification demand that all the members of a division belong to the same field, be it the field of cognition or the field of desires without mixing and confusing them. Thus, if one tried to maintain a reasonable coherence, one could propose the three following levels and terms in the field of cognition: aisthēsis (τὸ αἰσθητικόν),phantasia (τὸ φανταστικόν) and nous (τὸ λογιστικόν). So far, so good. But, what does happen when it comes to desire? In the field of desire Aristotle becomes confronted with a dual problem. In the first place, we find that in the enumeration of the different kinds of desire, there is an emptybox: actually, once we have taken the word nous away, we are left with epithumiaandthumos. So, we lack a word, a technical term to name the rational desire. As is known, Aristotle solved this problem by resorting to the word boulēsis. This way, the semantic field of desire becomes definitely systematized by Aristotle with the triad epithumia,thumos and boulēsis.4 3 This bipartite division into the rational and the irrational parts of the soul appears explicitly attributed to Plato in MM I, 1, 1182a23. 4 The tripartite classification of ὄρεξις explicitly appears, for example, in the following texts: 1. “... desire comprises appetite, passion and wish: ὄρεξις μὲν γὰρ ἐπιθυμία καὶ θυμὸς καὶ βούλησις” (De an. II 3, 414b2). 2. “... desire is divided into three forms ― wish, passion and appetite: ἀλλὰ μὴν ἡ ὄρεξις εἰς τρία διαιρεῖται, εἰς βούλησιν καὶ θυμὸν καὶ ἐπιθυμίαν” (EE II, VII, 1223α26).

34

T. CALVO MARTINEZ

The second problem Aristotle had to deal with concerns the lack of a generic term to include all these three different kinds of desire. In this case, Aristotle solved the problem by introducing the word orexis. I’m strongly inclined to think that it was Aristotle himself, who created and coined the Greek word orexis. As a noun, this word does not appear in Plato at all. As far as I know, the word ὄρεξις did not exist ― or, at least, it cannot be found in written texts ― before Aristotle.5 No doubt, there was the corresponding verb oregomai. Plato himself uses it in its broadest meaning as “to desire”, “to strive for”. As is well known, oregomaicarries the idea of moving towards something: to stretch out trying to get something for oneself. It points, therefore, to anactiveandintended movementtowardssomething. And this meaning brings us to the notions of desiring and wishing. In addition to the more or less literal usages and applications of the verb oregomai (such as desiring or wishing) we find a very special usage of it in Plato’s Phaedo. It appears in the well-known passage where Socrates remarks that physical substances don’t ever reach the perfection of Forms. Thus, two physical worldly things which are taken to be equal never will be found to be perfectly equal. Only the Form Equality ― Equality itself ― is truly and fully equal. Nevertheless, physical beings strive to be like the Forms. And to express this tendency Plato uses the 3. “... desire has three forms ― appetite, passion, wish: ὀρέξεως δ’ ἐστὶν εἴδη τρία, ἐπιθυμία θυμὸς βούλησις” (MM 1 12 2). 4. “Wish, however, passion and appetite, all they are forms of desire: βούλησις δὲ καὶ θυμὸς καὶ ἐπιθυμία πάντα ὄρεξις” (Demotuan. 700b22). 5 After having already finished this text, I had the opportunity to know M. Nussbaum’s suggestion that it was, “very probably”, Aristotle who created the noun ὄρεξις. Nussbaum (1983). I feel inclined to agree with her on this. However, the textual situation is as follows: (1) There are two testimonies, one on Empedocles (fg. 95), and the other one on Philolaus (fg. 27) where we find the noun ὄρεξις, but they do not contain literal quotations. (2) It also appears twice in the Definitions, in those of βούλησις (ὄρεξις μετὰ λόγου κατὰ φύσιν, 413c8) and φιλοσοφία (τῆς τῶν ὄντων ἀεὶ ἐπιστήμης ὄρεξις, 414b8), but this does not prove that its usage predates Aristotle. (3) Finally, ὄρεξις can also be found in two ethical fragments of Democritus, preserved by Stobaeus: (a) χρημάτων ὄρεξις, ἢν μὴ ὁρίζηται κόρωι, πενίης ἐσχάτης πολλὸν χαλεπωτέρη· μέζονες γὰρ ὀρέξεις μέζονας ἐνδείας ποιεῦσιν (fg. 219), and (b) ἢν μὴ πολλῶν ἐπιθυμέηις, τὰ ὀλίγα τοι πολλὰ δόξει· σμικρὰ γὰρ ὄρεξις πενίην ἰσοσθενέα πλούτωι ποιέει (fg. 284). But this could be taken as an evidence of a post-Aristotelian redaction of these ethical fragments attributed to Democritus (this is M. Nussbaum’s suggestion). However, in both cases its usage is strictly limited to only one specific kind of desire, to greed, to the desire for money. Thus, even if the noun had not been coined by Aristotle himself, he certainly was the first to use it in its most general meaning, as a generic term for “desire” as such.

ὌΡΞΙΣ AND INTENTION

35

verb oregomai: “all of them are striving to be like the Equal but fall short of it” (ὀρέγεται μὲν πάντα ταῦτα εἶναι οἷον τὸ ἴσον, ἔχει δὲ ἐνδεεστέρως, Phaedo, 75A). This Platonic remark that natural beings strive towards their own substantial perfection will be assumed by Aristotle in his notion of immanentteleology. And it opens the way for Aristotle himself towards a broad notion of “striving for” and “desiring”: I mean the broadest meaning of oregomai, in so far as all natural beings (not only living beings, but all natural beings) are endowed with a natural tendency, an ἔμφυτος ὁρμή towards their own good and perfection. This ὁρμή appears and acts in natural movements, in those movements which are not forced. Hence the well-known Aristotelian distinction between “being moved” violently, that is, against one’s own nature (βίᾳ, παρὰ τὴν φύσιν), and moving according to one’s own natural tendencies (κατὰ φύσιν καὶ τὴν ὁρμήν).6 This brings me further to two relevant remarks. (1) In the first place, one can say that desire, orexis, as it appears in animals and human beings, is their specific way to perform this universal tendency to the good. Aristotle very often refers to this universal natural tendency. However, I want to pay special attention to two passages. One of them belongs to the text in the Physics where Aristotle deals with matter and form as constitutive principles of natural beings: “For given that there is something divine, good, and worthstrivingafter (θείου καὶ ἀγαθοῦ καὶ ἐφετοῦ), we hold that there are two other principles, the one contrary to it, the other such as of its own nature to strive afteritanddesireit (ἐφίεσθαι καὶ ὀρέγεσθαι). But the consequence of their view [the view prevalent in the academy] is that the contrary desires its own destruction. Yet the form cannot strive after itself (αὐτὸ αὑτοῦ οἷόν τε ἐφίεσθαι), for it is not defective; nor can the contrary aim at it, for contraries are mutually destructive. Actually, what aims at the form is matter, as the female at the male and the ugly at the beautiful — only the ugly or the female not in itself but accidentally.” (Physics I, 9,192a16-23).

The second passage I want to present belongs to Degenerationeetcorruptione II 10. There Aristotle tries to explain why it is that the process 6 The words κατὰ φύσιν καὶ τὴν ὁρμήν obviously constitute a hendiadys which points to the “natural tendencies” of things.

36

T. CALVO MARTINEZ

of generation and corruption is eternal, without an end. His argumentation begins with the following words: “As we have said, coming-to-be and passing-away are continuous, and will never fail owing to the cause we said. And this is fully reasonable. For in all things, as we affirm, naturealwaysdesiresthebetter (ἐπεὶ γὰρ ἐν ἅπασιν ἀεὶ τοῦ βελτίονος ὀρέγεσθαί φαμεν τὴν φύσιν).” (Gen. Corr. II ,10, 336b25-28).

(2) The second point I want to stress after reading these two passages is that Aristotle never uses the noun orexis ― never, I insist ― when referring to this universal natural tendency to the good. The word orexis doesn’t appear in any of the two texts I have presented. Like Plato, he uses the verb oregomai, indeed. And together with oregomai, in the quoted passage of Physics he also uses the verb ephiomai. He says: ἐφίεσθαι καὶ ὀρέγεσθαι. And he also uses the adjective ephetos. To my mind, all this points to the fact that the noun orexis ― strictly understood as desire ― was deliberately coined by Aristotle to name those tendencies that properly and exclusively belong to animals and human beings, i.e.livingbeingsendowedwithsensualfaculties. As far as insentient living beings ― plants ― are concerned, Aristotle’s attitude seems to be somehow ambiguous. Surely, in so far as they are natural beings, they have the corresponding ἔμφυτος ὁρμή. However, could the natural tendencies of plants be qualified as orexis in the strictest Aristotelian meaning of this word? I know only one passage where Aristotle uses the word ὄρεξις (again, I mean the noun, not the verb) in connection with life as such. This passage can be found in the Eudemian Ethicswhere Aristotle criticizes the identification of the Good with the One. He explicitly refers to the argument based on the assumption that numbers strive after the one, somehow aim at the unity (ἡ ἀπόδειξις ὅτι τὸ ἓν αὐτὸ τὸ ἀγαθόν, ὅτι οἱ ἀριθμοὶ ἐφίενται). And he says: “On the other hand, it is also a misleading way to demonstrate that unity is the Good Itself to say that numbers aim at it (ἐφίενται); for no one says distinctly how they aimat it (ἐφίενται), but the statement remains wholly unqualified. And how can one suppose that there is desire where there is nolife? (καὶ ὄρεξιν εἶναι πῶς ἄν τις ὑπολάβοι ἐν οἷς ζωὴ μὴ ὑπάρχει;” (EE I 8, 1218a24-30).

As we can see, Aristotle remarks that ― properly speaking ― only living beings can have orexis. However, does this mean that actually there is orexis in every living being, plants included? Surely not, from a logical point of view. No doubt, according to this Aristotelian statement life is

ὌΡΞΙΣ AND INTENTION

37

a necessary condition for the existence of desire. But it doesn’t seem to be a sufficient condition. 2. Orexis and Praxis Properly speaking, then, orexis/desire only belongs to sentient living beings, that is, to human beings and other animals. This fact becomes definitively confirmed by the Aristotelian tripartite division of the soul, in so far as orexis is a generic notion whose species are epithumia,thumosand boulēsis, and only animals can have these kinds of tendencies.7 From a psychological and anthropological perspective, the most important point, to my mind, is the essential connection between orexis and praxis, that is, the structural link between orexis and animal behaviour. Actually, when Aristotle wants to explain the basic structures of action, he does it in biological and psychological treatises like Deanima and De motuanimalium. To my mind, this means that for Aristotle animal behaviour is the appropriate framework and the suitable perspective in order to explain the structure of human praxis. In Demotuanimalium Aristotle asks what are the principles and the causes of the movement of animals, and he finally states that there are basically two: cognition and desire (orexis). Thus, animal ― as well as human ― behaviour takes place and develops as the result of an interaction of cognition and desire.8 7 An explicit textual evidence for this can be found in the following passage (partly quoted above, n. 4.1): “Of the psychic powers which have been mentioned, some kinds of living beings, as we have said, possess them all, others some of them, others only one. Those we have mentioned are the powers of nutrition, sense-perception, desire, movement in respect of place, and the power of thinking. Plantshavethepowerofnutritiononly (ὑπάρχει δὲ τοῖς μὲν φυτοῖς τὸ θρεπτικὸν μόνον), while other living beings have both this and senseperception. And if they have the power of sense-perception, then they have also the power of desiring: for desire comprises appetite, passion, and wish. Now, all animals have at least one of the senses, viz. touch, and whatever has sense-perception experiences pleasure and pain and therefore there are pleasant and painful objects for it, and where these are present, there is also appetite, for appetite is a desire for that which is pleasant” (De an. II 3, 414a29). 8 Aristotle’s explanation begins with a somehow careless and incomplete enumeration of different causes: “We see that animals are moved by thinking, imagination, choice, wish, and appetite”. However, he immediately reduces them all to two main causes, cognition and desire: “However,allthesearereducibletothoughtanddesire (ταῦτα δὲ πάντα ἀνάγεται εἰς νοῦν καὶ ὄρεξιν). For both imagination and sensation occupy the

38

T. CALVO MARTINEZ

This issue has often been discussed and analysed. However, I would like to stress the two following points. 1. In the first place, one can underline that ― according to Aristotle ― the articulation of cognition and desire is structurally the same for both, animals and humans, although the corresponding processes are much more complex in human beings. Thus, at the level of human cognition, reason and deliberation (bouleusis) come to the scene. As for human desire, it also reaches a higher level, which is the level of preferential choice (prohairesis). At its highest level, then, the human articulation of cognition and orexis becomes an interplay of deliberation and choice which characterizes and defines the human being as such, the human essence and nature. For as Aristotle explicitly states in the Nicomachean Ethics, preferential choice ― that is, prohairesis― is “eitherdesiringreasonorreasoning desire (διὸ ἢ ὀρεκτικὸς νοῦς ἡ προαίρεσις ἢ ὄρεξις διανοητική)”. To which Aristotle immediately adds that “thisprinciple is a human being” (καὶ ἡ τοιαύτη ἀρχὴ ἄνθρωπος) (VI 2, 1139a17-20, a30-b5). 2. In the second place, I want to underline that in this articulation of animal and human praxis the desiring faculty ― or desiring power ― appears as a principle which produces movement while being itself moved. These are Aristotle’s own words in De anima III 10: “There are three things, (1) that which moves, (2) that by means of which it does so, and (3) that which is moved. But the expression ‘that which moves’ can refer to two different kinds of movers: that which is itself unmoved and that which at once moves and is moved. That which is unmoved is the practical good (τὸ πρακτὸν ἀγαθόν), and that which at once moves and is moved is the power of desire (τὸ ὀρεκτικόν) (for that which is moved is moved insofar as it desires, and desire as actual is a kind of movement (καὶ ἡ ὄρεξις κίνησίς τίς ἐστιν, ἡ ἐνεργείᾳ), while that which is moved is the animal.” (433b13-19).

same place as thought, since all three are faculties of discrimination, though differing accordingtothedistinctionswehaveproposedelsewhere.Wish,however,passion,and appetite,alltheyareformsofdesire.” (Demotuan. 700b17-23) Obviously, in this passage the word νοῦς is used analogically, as a generic term embracing any kind of cognition, imagination and sensation included, in so far as they are capacities to “discriminate” (κριτικά), as Aristotle literally says.

ὌΡΞΙΣ AND INTENTION

39

Two Further Remarks Concerning this Passage. (a) One can point to the need to distinguish thedesiringfaculty (i.e. desire as a disposition) from actualdesire.The desiring faculty is named toorektikon by Aristotle, and it is identical with the natural tendency towards one’s own good, the tendency to that which incidentally can be presented by cognition as something good, in so far as “eachseeksitsowngood” (ἕκαστον γὰρ ἰδίου ἀγαθοῦ ὀρέγεται). This faculty or tendency becomes actualized by its object, by “the practical good” (τὸ πρακτὸν ἀγαθόν), and its actualization is orexis, theactualdesire (ἡ ὄρεξις, ἡ ἐνεργείᾳ). Thus, praxis as an “intentional” behaviour carries on the intention itself of actual desire. (b) And, since desire becomes actualized by cognition, my second remark once more points to the fact that there can be orexisonly if there is cognition. And as we have already seen (and will realize later), we should specify that there is orexis only if there is sensiblecognition. 3. Orexis and Consciousness Up to this moment I have considered the interaction and the connections between orexis, praxis, sensitive knowledge and consciousness. In order to reach a better understanding of the structural links between these psychic phenomena, and particularly a better understanding of the nature of orexis, it will be useful to look at the Aristotelian conception of consciousness. Needless to say, in Aristotle’s thought the phenomenon of consciousness has not the philosophical import and place it will be later given in modern and contemporary thought by philosophers like Descartes or Husserl. Nevertheless, we find in his works some statements which seem to contain an interesting and unambiguous description of consciousness. Let’s look at them. In Deanima Aristotle raises the following question: “Since weperceive (αἰσθανόμεθα) that we see and hear, it must either be by sight that one perceives (αἰσθάνεσθαι) that one sees or by another .” (III 2, 425b12ff).

Here the starting point for Aristotle is the obvious fact that we not only see, but we also “perceive” that we see, we not only hear, but we also

40

T. CALVO MARTINEZ

“perceive” that we hear, etc. Now, Aristotle uses the verb aisthánomai to name this consciousness of our own psychic states and activities. And this verb specifically refers to the sensitive knowledge or perception. And as is known, Aristotle attributes consciousness to the commonsenseor sensitivity. For according to his psychological theory, sensitivity is one and the same, although it appears organically diversified.9 In the NicomacheanEthics (IX 9) there is a parallel passage which is more explicit and wholly consistent with the one I have already quoted. The immediate context is the analysis of friendship and the conception of a friend as an alterego. Within this framework Aristotle remarks that (a) our life is activity which mainly consists of perceiving and thinking, (b) that living is something good and pleasant, (c) and that we feel the pleasure of living in so far as weareconscious that we exist and we are living. And Aristotle says about this consciousness: “and if he who sees perceives (αἰσθάνεται) that he sees, and he who hears, that he hears, and he who walks, that he walks, and in the case of all other activities similarly there is something which perceives that we are doing something (ἔστι τι τὸ αἰσθανόμενον ὅτι ἐνεργοῦμεν), so that if we perceive, that we perceive, and if we think, that we think; and that we perceive or think that we exist (for existence is either perceiving or thinking) [...]” (IX 9, 1170a29-33).

Aristotle refers to self-consciousness, to our being conscious that we act and do something (ὅτι ἐνεργοῦμεν) as a consciousness of our own praxis and life. This consciousness has the following characteristics. (1) In the first place, the consciousness we have of our own activity is essentiallylinkedwithsensibility, not with thought. As I have already remarked, the verb used in both texts by Aristotle is aisthanomai which means perceiving and refers to sensitivity, to sensitive 9 The consciousness of our own perceptions is explicitly attributed to the common sense by Aristotle in the following passage of the treatise Desomnoetvigilia: “For every sense there is something specific and also something common; specific, like, e.g., seeing is to the sense of sight, hearing to the auditory sense, and so on to each of the other senses; but thereisalsoacommonfacultywhichaccompaniesthemall,in virtuewhereofeveryoneperceivesthatheseesandhears (ἔστι δέ τις καὶ κοινὴ δύναμις ἀκολουθοῦσα πάσαις, ᾗ καὶ ὅτι ὁρᾷ καὶ ἀκούει αἰσθάνεται), for it is not by sight that one sees that he sees; and it is not by taste, or sight, or both together that one discerns, and is able to discern, that sweet is different from white, but by a certain part which is common to all the senses; for there is one sensitivity, and the controlling sensory organ is one (ἔστι μὲν γὰρ μία αἴσθησις, καὶ τὸ κύριον αἰσθητήριον ἕν).” (455a12-27)

ὌΡΞΙΣ AND INTENTION

41

knowledge. Even our consciousness that we are thinking, not that there is thinking, but that it is our thinking, I mean, the consciousness that it is me who thinks, depends on sensitivity: it cannot be separated from our ― from my ― own images and reminiscences. (2) In the second place, we cannot forget that sensitivity, as the faculty of self-consciousness, cannot be separated from the body. Actually, it is an organic,bodilyconsciousness. In the quoted text we read that “he who sees perceives (αἰσθάνεται) that he sees, and he who hears, that he hears, and hewhowalks,thathewalks.” Thus, according to Aristotle, our consciousness that we are walking, the consciousness of our bodily movements in general, is at the same level as the consciousness of those other activities that will be considered exclusively “mental” after Descartes. (3) Finally, and in the third place, since consciousness is connected with sensitivity, we have to acknowledge that it properly belongstothe level of animality. The place for consciousness is not thought, at leastinthecaseofhumanconsciousness. (Concerning the possibility of a divine, purely intellectual consciousness, see below.) Sensitivity,corporality,animality. These three features of our consciousness clearly show the distance and difference between Aristotelianism and Cartesianism. Actually, according to Aristotle, we shouldn’t say Cogito ergo sum. We should rather state: Sentio ergo sum: I am, and I am sensitively conscious that I am. 4. God and Human Being Let’s finally go a step further. The Aristotelian conception of consciousness has been often and extensively discussed in contemporary thought and it usually has been interpreted according to the description I have tried to propose.10 In any case, we surely may state that orexis, praxis and sensitive consciousness 10 The specific issue of consciousness in Aristotle’s thought has been often and abundantly discussed in contemporary scholarship, mainly in connection with topics related to philosophy of mind. It would be useless to add a complete bibliography. However, let me explicitly mention Ch. Kahn’s article “Sensation and Consciousness in Aristotle´s Psychology” (Kahn 1979), since I do still substantially agree with his interpretation. (On consciousness in Stoic philosophy, cf. Long 1991, where one can find useful bibliographical references). I myself have dealt with these topics in Calvo Martinez (1997).

42

T. CALVO MARTINEZ

are essentially linked with each other in Aristotle’s anthropological thought. These essential, structural links established, I would like to end with a complementary reference to the divine being which ― as is well known ― is characterized by Aristotle as νόησις νοήσεως νόησις, as a “thinking which is a thinking of thinking” (Met. XII 9, 1074b33-35). What to say about it as compared with human consciousness, praxis and desire? Let me give just a short sketch on this issue by introducing the three following points. (1) In the first place, it is obvious that inGod,thereisnoorexisatall. Orexis ― desire ―, as we have seen, is a psychic tendency to get something good for oneself. It essentially implies lacking something. Let’s say it once more: when the intention of desire becomes actualized, it initiates the appropriate praxis in order to get the desired good. (2) In the second place, since there is no orexis in the divine being, there cannotbepraxisiniteither. In this connection, it is worth noting that Aristotle uses the word energeia to define and explain the divine activity (for example, in Met. XII 7), but he never uses the word praxis or the verb prattein in this “theological” context. No doubt, the divine thought is energeia, but Aristotle does not consider it as any kind of praxis. Why? On this specific point several considerations could be brought into play. However, I will propose only one comment which is more specifically connected with the issue of this contribution.11 Let me underline that the notion of praxis ― as well as the meaning of the verb prattein― include two features which could not be applied to the divine act of thinking at all. (a) On the one hand, the notion of praxis connotesdesire, it connotes orexis as the mover which moves the animal to act. Therefore, as I said, if there is no orexis, there cannot be praxis either. (b) On the other hand, the notion of praxis connotesareferencetoasubject, to someone who acts, the one who performs the action. And again, it seems that no reference to a subject can be attributed to the divine being, since it is not characterized as someone who thinks, but as pure activity of thinking. Neither is 11

Cf. among others, Natali (1993).

ὌΡΞΙΣ AND INTENTION

43

it characterized as nous, but as noēsis, that is, not as a faculty of thinking, but as the activity itself of thinking. (3) Finally, since the Aristotelian divine entity has no orexis, no praxis, no aisthēsis, should we conclude that there is no consciousness in it either? No doubt, we are here confronted with a difficult question which would demand thorough argumentation and interpretation of the Aristotelian texts. I will thus only finish with the suggestion that the divine noēsis does not have consciousness, in so far as consciousness in human beings appears to be associated with sensitivity, corporality and animality. As a hermeneutical device, I suggest drawing a distinction between selfknowledgeandself-consciousness. Let self-knowledge mean the knowledge of what one is, of what I am. Thus, self-knowledge belongs to the field of essence, while self-consciousness belongs to the field of existence, it is not knowing what I am, it is rather knowing that I am. Thus, self-consciousness is always consciousness of one’s own individuality, it is knowing that itismyself who knows or thinks, who acts and does this or that. No doubt, a “thought of thought” is knowledge of what it is, since it is thought of itself — something which does not however imply the knowledge of oneself as an individual existence. LITERATURE CALVO MARTINEZ, Tomàs (1997), “El sujeto en el pensamiento clásico griego” in Sanfélix Vidarte, V. (ed.) Las identidades del sujeto, Valencia: Pretextos, pp. 59-72. KAHN, Charles (1979), “Sensation and Consciousness in Aristotle´s Psychology” in Barnes, J., Schofield, M. & Sorabji, R. (eds.), Articles on Aristotle 4. PsychologyandAesthetics, London: Duckworth, pp. 1-31. LINDÉN, Jan-Ivar (2011-12), “Intentionnalité et perception: une esquisse aristotélicienne” in χώρα* REAM, 9-10, 2011-2012, pp. 339-52. LONG, Antony Arthur (1991), “Representation and the Self in Stoicism” in Everson, S. (ed.) Companions to ancient thought 2: Psychology, Cambridge: Cambridge University Press, pp. 102-120. NATALI, Carlo (1993), “Attività di Dio e attività dell’uomo nella Metafisica di Aristotele” in RivistadiFilosofianeo-scholastica 2-4, 1993, pp. 324-51. NUSSBAUM, Martha (1983), “The ‘Common Explanation’ of Animal Motion” in Moraux, P. & Wiesner, J. (eds.) Zweifelhaftes im Corpus Aristotelicum, Berlin: de Gruyter.

“THE SOUL IS IN A WAY ALL THINGS”: ARISTOTLE AND INTERNALIST CONCEPTIONS OF INTENTIONALITY Charlotta WEIGELT

When Parmenides declared it an impossibility either to think or to speak about that which is not,1 he introduced the problem of intentionality into philosophy, albeit in a somewhat roundabout way. When Plato in the Sophist tried to respond to Parmenides’ challenge, he saw no other way but to reinterpret not-being in terms of difference, so that he could preserve his predecessor’s conviction that it is impossible to refer to what is not in an absolute sense, while at the same time making room for the possibility of speaking and thinking about that which is not in a specific sense, since on the interpretation proposed by Plato (or the Eleatic stranger), not being this or that simply means being different from that which one is not (257b3-4). With this interpretation, the Sophistoffers both a corroboration and a correction of the claim made earlier by Socrates in the Theaetetus,the dialogue immediately preceding the Sophist, namely, thatacts like believing, seeing, hearing and touching must be of something that is in order to be real: if you believe something that is not, you are in fact not believing anything at all (188e6-189b2). But on the basis of the reinterpretation of nothingness launched in the Sophist, it becomes possible to affirm that in order for something to be a λόγος, a meaningful statement, it is enough that the statement is about something at all, whether existent or not. A meaningful statement, in other words, simply has the form λόγος τινός (262e6-7). Though Parmenides’ and Plato’s worries about nothingness are certainly very much a part of the contemporary discussion of intentionality, 1 Fr. B2.7-8: “for neither could you know what is not (for it cannot be accomplished), nor could you declare it.” Translation Graham (2010), p. 213. The reference follows Diels (1956).

46

C. WEIGELT

the latter exhibits a decisive change of emphasis in relation to its ancient prehistory. Today, the basic problem concerns not primarily how it might be possible to intend non-existing objects, but rather how it is possible to intend existing objects to begin with. As McDowell puts it in Mind andWorld: modern philosophy suffers from a peculiar anxiety, originally aroused by Descartes, which is “an inchoately felt threat that a way of thinking we find ourselves falling into leaves minds simply out of touch with the rest of reality.”2 McDowell’s diagnosis, which I believe is correct, is that the anxiety in question is rooted in a conception of mind and world as ontologically distinct, that is to say, as two different entities, not to say substances, each endowed with a nature of its own, but which may, under the proper circumstances, take up a relation to one another. This relation is, presumably, intentionality. If one conceives of intentionality in this way, as an objective relation between mind and world, the question naturally arises as to which of the relata is responsible for the intentional relation. When referring to an object, is the content of my mind, in virtue of which I am able precisely to refer to or intend the object in question, determined primarily by factors internal to me as a cognizing subject, or is it determined either exclusively or at least in part by the way the world actually is? The former position is commonly labelled “internalism”, the latter “externalism”. Externalism is usually centred on the claim that intentionality is somehow founded or “supervenient” upon causality.3 In its more extreme version, it reduces intentionality to a causal transaction in nature, so that there would really be nothing more to be said about our subjective experiences, for example the sensation of cold, than that our bodies are affected in certain ways, in this case so that they are chilled. As Aristotle notes in Deanima, however, plants too can become cold, but this does not mean that they are capable of feeling cold (424a32-b2).4

2

McDowell (1996), p. xiii. In a recent anthology on externalism, this position is described as follows (with respect to perception): “In perceptual thought, who or what is being thought about will be determined in part by one’s causal-perceptual relations to one’s environment; and (there will be cases in which) what properties one ascribes to that item in thought is determined in part by the ‘external’ properties with which one oneself has been in causalperceptual contact in the past,” Goldberg (2015), p. 3. 4 I have used Hicks’ edition of DeAnima (1907), and also his translation, if nothing else is indicated. 3

“THE SOUL IS IN A WAY ALL THINGS”

47

In any case, this kind of explanation of intentionality gets into trouble when faced with what seems to be a plain fact, in spite of what Parmenides said, that we do also intend non-existing objects, as when my youngest son expects to get a visit from the tooth fairy, every time he has lost one of his baby teeth. We might then be inclined to think that in a case like this, the intentional object is not an extra-mental but an intramental entity, like an image. And when we realize that, even in ordinary perception, we see any given thing, not simply as it is in itself, but necessarily from a certain, one-sided perspective, we might want to conclude with Hume that, even in cases where we had absolutely no reason to doubt that we were genuinely perceiving an object, in reality it was “nothing but its image, which was present to the mind.”5 When Franz Brentano in Psychologie vom empirischen Standpunkt introduced the term intentionality in a modern context, in order to achieve a criterion for psychical as distinct from physical phenomena, he too defended a representational theory of intentionality, though not so strong a version as Hume: “Every psychical phenomenon is characterized by what the scholastics of the Middle Ages used to call the intentional (or mental) in-existence of an object, and which we might call — although with expressions that are not entirely unambiguous — the relation to a content, the direction towards [auf] an object (which is here not to be understood as something real [eine Realität], or the immanent objecthood [Gegenständlichkeit]. Each contains something as an object [Objekt] within itself, although not in the same way. In presentation something is presented, in judgment something is accepted or rejected, in love loved, in hate hated, in desire desired, and so on.”6

In a note to this passage, Brentano claims that “already Aristotle” had spoken of this “psychical inhabitation [Einwohnung]”, namely in De anima. There might indeed seem to be good reasons for thinking that Aristotle belongs within the internalist and, more precisely, the representationalist camp, though the standard reference in this respect is perhaps not De anima, but rather the opening passage of De interpretatione, where it is stated that spoken language is a symbol (σύμβολον) of the affections, παθήματα, of the soul, which in their turn are ὁμοιώματαof things (16a3-8). It is perhaps not immediately clear how ὁμοιώματα should be understood, and so the translator of the Loeb edition of 5 6

Hume (1997), p. 152. Brentano (1924), pp. 124-125. My translation.

48

C. WEIGELT

De intepretatione, apparently in an act of desperation, translated this expression as “representations or likenesses, images, copies”.7 Be that as it may, what Aristotle seems to suggest here is that when we speak about things, we refer, not directly to these things themselves, but rather to our representations of them.8 In setting out to discuss Aristotle in relation to internalist conceptions of intentionality, it is not my aim to take a stand in the debate between internalism and externalism, not even to sort it out properly. That would require another paper, I believe, considering that the labels “internalism” and “externalism” are used in a variety of different senses and also with respect to different philosophical issues.9 Moreover, as far as I can see, Aristotle does not really have a problem of intentionality either in the modern or in the Greek — that is, Parminedean and Platonic — sense.10 He is not worried that it may turn out to be impossible to explain how the subject is able to “reach out” or “hook on” to the supposedly objective world, nor does he find it mysterious that we are able to think or speak about non-existing objects.11 But to judge from the content and overall argument of De Anima, Aristotle nonetheless considers it an important task to clarify the nature of the relation between the soul and its world; and when exploring the soul’s various activities, notably sensation and thought, he provides us with something that is at least an outline of a theory of intentionality. This being said, I take it to be one of the merits of Aristotle’s account of the soul in Deanima, which shall be my focus in what follows, that it invites us to reconsider the very distinction between the internal and the external, and to ask ourselves whether it is possible to demarcate sharply that which is in the mind from what belongs to the world. The basic claim of De anima, that the soul is a kind of receptor of forms, implies, or that is at least my suggestion, that Aristotle does not conceive of the soul and the world as two ontologically distinct entities but rather as two aspects or dimensions of one and the same reality. This is not to 7

Tredennick, in Cooke and Tredennick (1983), p. 115. The tradition of ascribing representationalism to Aristotle has been aptly summarized by Caston (1998). 9 For a clear and nuanced exposition of this problematic, see Gallagher and Zahavi (2008), pp. 121-126. For an informative account of the historical background to the debate between internalism and externalism, see Kornblith (2001), pp. 1-7. 10 For an opposing view, see Caston (1998). 11 Met. 1003b10: “Hence, we even say that what is not is not-being.” My translation. 8

“THE SOUL IS IN A WAY ALL THINGS”

49

say that he would be unable to distinguish between the soul on the one hand, and the things with which the soul is concerned on the other, which he clearly is not, but that this distinction must be understood within a more comprehensive unity that is constituted by the teleological structure of reality, according to which the soul constitutes the actuality of the otherwise merely potential intelligibility of the world.12 And the place where to begin an investigation of that unity is the soul. If this is granted, then Aristotle, at least in his overall approach, comes pretty close to a version of internalism that we find in phenomenology, perhaps most clearly in Husserl.13 Internalism in this sense is founded on the conviction that intentionality is an irreducible and intrinsic feature of the mind (and so cannot be reduced to a causal transaction in nature, since it is not an objective relation between mind and world to begin with), which consequently must be explored from within. The point of this approach, however, is not to deny that the perceiving and cognizing subject is directed towards the world itself (rather than towards some representation of it), but to emphasize that the very idea of the “world itself” can only be made sense of in relation to an apprehending subject, to whom the world appears, but who is also in its turn claimed by this very world. The kind of internalism which is endorsed by Husserl, and which I think is foreshadowed in Aristotle, does thus not entail representationalism, nor is it incompatible with externalism with respect to mental content, if by this we understand the view that in experience we are open to or take in the world such as it is, and not simply as we expect it to be.14 It is just that, in order to reach the world, we do not have to transcend the soul, because the nature of the soul itself is precisely to be self-transcending. This, I think, is actually implied in the very concept of intentionality, at least when properly understood, which is to say, as the directedness of the 12 By “teleology” I shall first of all understand the view that every activity or process must be conceived as an actualization (ἐνέργεια) of a potentiality (δύναμις). The fact that such an actualization essentially involves the achievement of an end (τέλος), will be less important to my argument. 13 Husserl worked on his understanding of intentionality more or less throughout his career, but see in particular his fifth logical investigation (1993), Ch. 2, and (1992), §§ 36-37. 14 Internalism, at least in epistemology, is often associated with Cartesian dualism, but the kind of internalist conception of intentionality that I wish to attribute to Husserl (and, in a manner, also to Aristotle) is on the contrary an attempt to overcome the dualistic conception of mind and world.

50

C. WEIGELT

soul. And I believe that something at least not very far from this claim may be attributed to Aristotle’s statement that the soul is in a way all things: ἡ ψυχὴ τὰ ὄντα πώς ἐστιν (De an. 431b21). So, more relevant to our conception of intentionality than the distinction between the internal and the external is perhaps the distinction between the first and the third person, or between being a subject for the world and being an object in the world. And to Aristotle, these are just two sides of one and the same coin. As far as I can see, however, there is a tendency in current scholarship to assume that intentionality as conceived by Aristotle must ultimately be explained in terms of the causal impact exerted by the world upon the soul.15 In what follows, I shall try to show that such a reduction of the realm of subjective experience to its alleged objective foundation is not warranted on the basis of Aristotle’s text, but that another reading is possible. The stated aim of Deanima, however, is not to throw light on the nature of intentionality or in general on subjectivity, but to give an account of the soul as a principle or ἀρχή of life. In the words of Aristotle: “It would seem, too, that an acquaintance with the subject [the soul] contributes greatly to the whole domain of truth and, more particularly, to the study of nature, the soul being virtually the principle of all animal life” (402a4-7). The task must therefore be to investigate the nature and essential being, φύσις and οὐσία, of the soul (402a7-8). Aristotle thus approaches the study of the soul as a piece of natural philosophy, though the discovery of the peculiar nature of the soul will motivate an inquiry that differs in important respects from his other works on natural philosophy. It is an essential feature of animal life that it does not simply sustain its existence within the world of nature but also has a perceptual awareness of its natural surroundings; and in the case of man, this perception also includes awareness of a second order, that is to say, an awareness of himself as a sentient and thinking being (425b12-20, 429b5-9). As regards the method of investigation to be followed here, Aristotle accordingly observes that he or she who wishes to get clear about the 15 Caston (1998) is a particularly clear example of this tendency, but see also Magee (2000), Marmodoro (2012), Scaltas (1996), Shields (2016), pp. xxxiv-xxxix, Sorabji (1974). It should be mentioned, however, that Sorabji later has developed a more nuanced view of Aristotle’s conception of intentionality (1992).

“THE SOUL IS IN A WAY ALL THINGS”

51

nature of the soul must be both a φυσικόςand a διαλεκτικός, where the former would merely be interested in the material conditions of the soul’s affections (defining anger, for example, by reference to the ferment of the blood), and the latter in their ends or functions (defining anger as desire for retaliation) (403a29-b2). It seems, in other words, that Aristotle is here advocating something like Davidson’s anomalous monism: one and the same event can be regarded both as mental and as physical, and thus be explained by reference either to reasons or to causes — in the modern sense of that term.16 To Aristotle, however, the situation is a bit more complex than that, since the soul (or the realm of subjectivity) is nothing but the fulfilment of the body. In his own words, the soul is the first actuality (ἐντελέχεια) of the body, the power of the body to exercise its functions, as for example perception (412a27-28).17 Or with another formulation: if an axe were a natural body, its very axeity or being (τὸ πελέκει εἶναι) would be its soul (412b11-13). Clearly, the being of or for an axe necessarily belongs to precisely an axe: it would not make much sense on its own, as it were. Analogously, the soul is the soul of a body. Not even thinking, Aristotle observes, can take place without the body, provided that it involves imagination, φαντασία (403a8-10), though the faculty of thought, νοῦς, is not the fulfilment of any particular organ but, as it seems, rather of man as a whole, and for that reason, the intellect may to some extent be investigated in separation from the body (429a24-25). In the third chapter of the first book of De anima, Aristotle devotes some space to a critique of the view, advocated by Plato among others, that the soul is in motion. The broader implication of this critique is that the soul must not be mistaken for a “thing” or substance at all: it is not a magnitude and so does not occupy any place. “For how can it possibly think if it be a magnitude?” (407a10-11). Further, if the soul is not subject 16

Davidson (2001), Ch. 11. Aristotle illustrates the distinction between (first) potentiality, first actuality (which is likewise a kind of second potentiality) and second actuality in the following way: any man is, just in virtue of being human, potentially knowledgeable within the field of grammar, but when he studies grammar he acquires a capacity or potentiality for being knowledgeable in this respect in a stronger sense. He possesses knowledge of grammar and can make use of it whenever he wants to (unless hindered in some way). This is, thus, the first form of actuality: the presence of a power or capacity. But when actually exercising his knowledge of grammar, he is knowledgeable in an even stronger sense, so this is the second, and supreme, form of actuality (De an. 417a21-b2). 17

52

C. WEIGELT

to motion, then, clearly, it cannot be subject to causal impact, either. But if the soul is not a concrete substance but an ἐντελέχεια, or as Aristotle will later say, a λόγος and a δύναμις, an order and a power, it is not entirely easy to see what sense it could make to apply the distinction between the internal and the external to the soul.18 To say that something is either within or outside the soul is to use a language of spatial metaphors, which is hardly proper to the soul. Not that Aristotle is completely capable of refraining from such language, but in that respect, contemporary philosophy of mind is no better off either, as far as I can see. In the opening passages of Deanima (402a23-b10), Aristotle lists a host of questions that must be settled right at the beginning of the inquiry, all of which concern the nature of the soul: what kind it belongs to, what its mode of being is, and so on. But then he introduces a further query that will give De anima its specific direction, for it concerns the question whether one should first investigate the parts of the soul or their functions (ἔργα), for example: should we first inquire into the process of thinking, τὸ νοεῖν, or into that which thinks, ὁ νοῦν? In case we opt for the former alternative (which is what Aristotle will do), the question arises as to whether we, when exploring the activities of the soul, should turn our attention first to the objects, for example τὸ νοητόν, and only then to the faculty of thinking, νοῦς, to which the objects of thought are correlated (402b10-16). And the answer that Aristotle eventually will give is that we should begin with the objects (415a2-22). Right at the beginning of Deanima, then, the basic concepts of intentionality are articulated, and as we have just seen, Aristotle is convinced that the investigation of the soul should focus on the activities, the ways in which the faculties or powers of the soul are enacted. If we do so, we will immediately also encounter the world, for when analysing our psychical activities, we cannot avoid the objects towards which the former are directed, because this directedness will turn out to be the very essence of the soul. Let us now turn to Aristotle’s account of perception. Perception seems quite obviously to be of things “out there” in the world, so that the senses 18 At 414a13-14, he says that the soul is λόγος and εἶδος, at 414a27 that it is ἐντελέχεια and λόγος (and here we should keep in mind that the actuality in question is of the first kind and thus a power or capacity), and at 424a26-28, Aristotle explains that neither sensitivity (or the being of the faculty of sensation, τὸ αἰσθητικῷ εἶναι) nor the sense (αἴσθησις) is a magnitude but a λόγος and a δύναμις.

“THE SOUL IS IN A WAY ALL THINGS”

53

“produce no sensation without external sensibles” (417a4).19 From this fact, Aristotle concludes that, in itself, the faculty of sense perception (τὸ αἰσθητικόν) exists only as a potentiality or as a capacity (417a6-7). But remember that this is potentiality in the second, qualified sense: it is an “actual” capacity for which it is natural to be actualized, which in fact will be actualized if nothing prevents it, as Aristotle says about δύναμις in general in the Physics (254a34-b1). This intrinsic tendency towards activity could be described as a directedness belonging to the soul, that is to say, as its structure of intentionality. Or, as Aristotle puts it in De memoria: thought does not need to stretch itself out in order to reach its object (452b9-11).The faculty of sense perception, however, needs to be activated, as it were, by encountering such things as can be perceived, just as fuel does not burn by itself but needs something to make it burn (417a7-9). How are we to understand this idea? I do not think it has to be interpreted in terms of causal interaction. Or more precisely, the causal story cannot tell us what it is like to have a perceptual experience, from the point of view of the perceiving subject, but only what is required by way of material conditions in order for something like this to occur. In this case, I think that part of Aristotle’s point is simply that sensation needs the presence of sensible objects, so that there is something there for us to see, hear, and so on. This is at least how I understand the following passage: “And to have actual sensation corresponds to exercise of knowledge. There is this difference, however, that in the one case that which brings forth [τὰ ποιητικά] the activity is external, as, for instance, the objects of sight, hearing and the other senses. The reason is that actual sensation is always of particulars, while knowledge is of universals: and these universals are, in a manner, in the soul itself. Hence it is in our power to think whenever we want to, but to perceive is not in our power: for the presence of the sensible object is necessary.” (417b18-26)20

Particular objects cannot be in the soul, for obvious reasons, since they are material. In order to perceive them, we need them to show themselves to us. A common objection to causal explanations of intentionality is that the causal conditions of, for example, perception are highly complex, so that 19 20

ἄνευ τῶν ἔξω οὐ ποιοῦσιν αἴσθησιν; Hicks’ translation has been slightly modified. Hicks’ translation has been slightly modified.

54

C. WEIGELT

these conditions do not by themselves pick out the intentional object.21 We do not find any argument to this effect in Aristotle, but in his analysis of causality in the Physics,he entertains an idea that is not entirely unrelated to this, namely, that the kinds of causal impact a thing is susceptible to is prescribed by its nature.22 The reason why the grass grows after having been fertilized, exposed to sunlight and watered is that its nature is such as to be receptive to such treatment (if we wants a human being to grow we must take other measures). The moral of this example with respect to our current issue is that we have to get clear about the nature of the soul before we can take a stand on the question concerning the possibility and nature of causal relations in the psychical domain. And what Aristotle in effect shows is that the soul could not be affected by things if it were not by itself or according to its nature open to or directed towards things, which is thus how I understand its being a δύναμις in the sense of πρώτη ἐντελέχεια. But as noted earlier, the soul itself can hardly be involved in causal relations, since it is not a material substance. It is different with the sense-organs, of course, but what about the senses? Seeing, for example, requires a sense-organ, but is itself a faculty of the soul. This faculty is exercised as an activity, and the activities that are motions, κινήσεις, involve change: something is acted upon by something else, so that it becomes like it (ὅμοιον) in a specific regard (417a1820), as when a sick person becomes well again after having taken medicine. When she is cured, she is “like” the medicine, in that both of them are now healthy, though obviously not in the same sense (cf. Met. 1003a33b6).23 But in actual sensation, our faculty of perception is actualized in such a way as to be preserved and fulfilled, Aristotle notes, so that there is no transition from one state to its opposite (417b2-7). We may note in 21

See, for example, Mohanty (1981), p. 715. More precisely, it is implied in the very definition of motion (κίνησις) as the actualization of the potential precisely as potential (Phys. 201a10-11, 27-29), that the causes of motion serve to “help” the being that is in a mode of potentiality to move or evolve in accordance with its own nature. See also Aristotle’s discussion of the relation between agent and patient in Phys. III.3 23 The meaning of this idea has been the subject of much debate in the literature, as some scholars, beginning with Sorabji (1974), have interpreted it literally, as if Aristotle’s point would be that, when seeing red, for example, the eye actually becomes red itself. This interpretation, though recently defended (with some reservations) by Rapp (2001), has been criticized in a variety of different ways, which I will not go into here; instead, I refer the reader to the survey of the debate in Marmodoro (2014), pp. 111-118. For references, see also above, note 15. 22

“THE SOUL IS IN A WAY ALL THINGS”

55

passing that Aristotle explains the absence of change in perception by comparing it to a builder who does not change when he builds (417b8-9),24 which might strike us as a somewhat strange comparison, since it suggests that sensation is like actively operating on a given material, namely the objects of sensation. In book II.5, Aristotle introduces his idea that the sensitive faculty is in potentiality such as the sensible object is in actuality, which in book II.12 is further specified in the following way: “[S]ense is that which is receptive of sensible forms without their matter, as wax receives the imprint of the signet-ring without the iron or gold of which it is made: it takes the imprint which is of gold or bronze, but not as gold or bronze. And similarly sense as relative to each sensible is acted upon by that which possesses colour, flavour or sound, not in so far as each of those sensibles is called a particular thing, but in so far as it is of a certain kind and in respect of its structure [κατὰ τὸν λόγον].”25 (424a17-24)

First, the simile with the wax which receives an imprint does not seem to be particularly suitable for explaining what goes on in the soul when we perceive things, since the soul, including its faculty of sensation, is not changed by the object of perception, but only activated by it. By contrast, the sense-organ may very well be said to receive the form of the thing as an imprint; it becomes informed, so to speak. But for the sense to receive the form is to perceive the object as being of a certain kind. That is why plants can become cold but not feel cold, that is to say, because they are incapable of receiving form (424a32-b3). As far as I can see, then, Aristotle, somewhat confusingly one might think, uses one and the same model, reception of form, to account for what is going on both on the subjective and on the objective level, so to speak, that is so say, to account for what it means to apprehend something as something, as well as for what is involved in the material or bodily process.26 But this 24 Actually, in that particular context, Aristotle is speaking about understanding (φρονεῖν), but it is clear from the broader context that he is not making any difference between cognition and reception in this respect (see in particular 417b18-19), that is to say, as regards the absence of change. 25 Hicks’ translation has been slightly modified. 26 Brentano (1967/1867), pp. 79-81, is one of few who is careful in making this distinction. As noted by Sorabji (1992), several of those who have criticized the so-called literalist reading of Aristotle’s idea that the patient becomes “like” (ὅμοιον) the agent as it receives the form of the latter (cf. above, note 23), nonetheless hold on to the basic assumption of the literalist position, namely, that reception of form is a physiological

56

C. WEIGELT

is in perfect agreement with teleology, as Aristotle understands it: it is an articulation of both the ontological structure of the world and the structure of our subjective experience of this world.27 And a few lines later, Aristotle comments on the relation between these two levels: the power to apprehend a particular object as being of a certain form does certainly belong to the sense-organ, but the being of the sense-organ is different from that of the power, insofar as neither sensitivity (or the being of the faculty of sensation), nor the sense, is a magnitude but a λόγος and a power of the organ (424a24-28). So the power of sensation is a capacity, a δύναμις,in relation to the particular objects, whose forms it grasps. But as such a capacity, it is not completely empty or naked, because as we have seen, the universals, that is, the forms, are in a way in the soul. And to account for this, the subjective conditions of experience, is part of the point of the distinction Aristotle later will make between the active and the passive intellect, where the latter “becomes all things” (430a14-15), whereas the former “makes all things, like a kind of disposition [ἕξις], such as light. For in a manner light, too, converts colours which are potential into actual colours”

process (p. 210). See, for example, Magee (2000), p. 327, who explicitly makes this claim. But if one fails to realize that Aristotle with this notion aims at something that goes on both in the subjective domain and on the physiological level, one runs the risk of reducing the subjective experience (perceiving something as something) to its causal presuppositions. See, for example, Caston (1998), who in an attempt to rescue Aristotle from representationalism of a Humean kind appears to do away with the subjective dimension of experience altogether, as he explains intentionality in causal terms, as a transmission of information from the objects represented by us, “that are preserved in the representations themselves, in virtue of which they are about those objects” (p. 28). Similarly, Scaltas (1996) has described the reception of form as a selection of information (p. 26) and in terms of “encoding” (p. 32). But this seems to me to attribute what is a subjective achievement (receiving and encoding information) to a level (the physiological) where it does not properly belong and therefore is hard to make sense of. On the other hand, I cannot find the functionalist reading proposed by Nussbaum and Putnam (1992) as an alternative to materialist reductionism (p. 27) altogether satisfactory either, because insisting that the soul be understood as the functional organization of the body does not by itself warrant a shift from a third-person to a first-person perspective on the soul’s activities: it does not tell us what sensation or awareness is like. 27 Briefly put, given the universal scope of teleology, as conceived by Aristotle, it is both the case that the processes of nature actually are for the sake of the ends that complete the processes and bring them to a close, and that we, in observing this teleological order of nature, ourselves are reaching out to and so are governed by some particular end, as the objective and moment of completion of our observation.

“THE SOUL IS IN A WAY ALL THINGS”

57

(430a15-17).28 If we take these two intellects as just two aspects of one and the same faculty, we could say that even though our intellect is receptive of forms (just like the soul), this receptivity is at the same time a kind of spontaneity: a capacity to throw conceptual light on the world, whose structure we then apprehend. This fits well with Aristotle’s description of νοῦςas “the form of forms”, and of sensation as “the form of sensibles” (432a2-3). Maybe this is also the reason why Aristotle considers it proper to compare the activity of sensation to a builder processing his material in order to produce a building, as we saw above. At least under normal circumstances, the intentional object in perception is what Aristotle identifies as what is perceived by virtue of concurrence (κατὰ συμβηβεκός), like Diares’ son. The object of our sensation in a narrow sense is simply something white, but we perceive this white thing as Diares’ son (418a20-23). Here, the reception of form involves apprehending something as something, τι κατά τινος. Aristotle explains:“But, as the perception by sight of the proper object of sight is true, whereas in the question whether the white object is a man or not, perception by sight is not always true, so it is with immaterial objects” (430b29-31).29 The structure of τι κατά τινος suggests that it is not the form that is the intentional object; and we have already seen Aristotle emphasizing that perception is of particulars. So there is nothing here, as far as I can see, that is suggestive of a representational theory of perception or of intentionality in general; that is to say, it seems unnecessary to assume that Aristotle’s forms would be some kind of mediators between us and the world.30 Rather, they appear to constitute the as-structure of perceptual awareness.

28 Hicks’ translation has been slightly modified. Shields (2016) translates ἕξις as ”positive state,” and Seidl (1971), p. 119, translates it as Haltung, suggesting that the analogy with the light is intended to depict the active intellect in its first, “habitual” actuality, to be distinguished from its active operation, that is, second actuality (when it actually “makes all things”). Since this habitual actuality is equally second potentiality, the power or capacity to throw light on the world, the active intellect must, thus, not be identified with the intellect in its actuality (nor should the passive intellect be identified with the intellect in potentiality), as has been relatively common in the scholarly tradition. 29 Hicks’ translation has been slightly modified. 30 This is also the view taken by Esfeld (2000), p. 330, who does thus not think that Aristotle endorses a representationalist theory of thought and perception.

58

C. WEIGELT

At this point, it might be worthwhile to recall that the contemporary discussion of intentionality displays a strong focus on the possibility and nature of individuation of mental content: what is it that lets us intend or refer to a particular object?31 But it seems that, to Aristotle, it is simply obvious not only that there are particular things, but also that the world we encounter is first of all a world of concrete, particular substances.32 Therefore, his query rather concerns our ability to apprehend not simply particular things, but things of a specific kind, that is to say, things that are not simply τόδεbut τόδε τι.And the passage I quoted above, where Aristotle states that the reception of forms involves apprehending a thing as being of a certain quality, apparently confirms this. Accordingly, the analysis of intentionality must account for the role and nature of conceptual structure in the soul’s apprehension of the world.33 To return to the question concerning representationalism in Aristotle, one might think that this gets more complicated when we turn to φαντασία, the power of imagination, which precisely seems to enable us to intend things, notably absent ones, by providing us with images of them. The concept of φαντασία is no doubt a highly complex issue in Aristotle, and I will only give a few comments on it here. When introducing it in De anima III.3, Aristotle begins by distinguishing it from both thinking and perception, observing, however, that imagination does not occur without perception, just as belief does not occur without imagination (427b14-16). But whereas we are free to think whatever we like (“for we can bring forth an object before our eyes”),34 we cannot believe just anything, since when believing we take a stand on the truth of the 31

For a recent overview of the emergence of this problematic, see Sen (2015), pp. 12ff. That it is obvious does not mean that Aristotle does not recognize it as an important task to explain what it is that is responsible for the particularity of the concrete, individual substance; on the contrary, this is one of the central tasks of the discussion of οὐσία in the Metaphysics. As far as I can see, however, the hylomorphic analysis developed to this end in book Ζ never succeeds in doing that. Indeed, the view of Frede and Patzig (1988), that Aristotle identifies the οὐσία of a particular thing with the latter’s individual form, is far from widely accepted today. 33 If confronted with the question concerning the conditions of individuation of reference, however, I think that Aristotle could reply by pointing to his analysis of φρόνησις as a capacity to grasp and respond to the particular circumstances of action in the NicomacheanEthics(VI.6, 1141b8-23), and in fact also to the Physics, where he tries to show how the now (τὸ νῦν) is that unique point of view from where we receive the world (IV.10-11). 34 427b18-19. Hicks’ translation has been slightly modified. 32

“THE SOUL IS IN A WAY ALL THINGS”

59

matter, for which reason we are also affected emotionally by our beliefs: if we believe that something is truly terrifying, then we will also be afraid (427b20-23). “But when we are under the influence of imagination we are no more affected than if we saw in a picture the objects which inspire terror or confidence” (427b23-24). This passage may seem to suggest that in imagination, what we are directed towards is a kind of picture of something, rather than towards the thing itself (and that is why we are not emotionally affected in the same way when we imagine things as when we actually have them before us). In Dememoria, however, Aristotle makes clear that this is not how imagination should be understood. Here he claims, precisely as in De anima, that thinking is not possible without φαντασία, and he illustrates this with an example borrowed from geometrical demonstration: “though we do not make any use of the fact that the quantity in the triangle is determinate, we nevertheless draw it determinate in quantity. So likewise when one thinks, although one does not think a quantity, one envisages a quantity [τίθεται πρὸ ὀμμάτων ποσόν], but one does not think of it as a quantity” (450a2-5).35 Aristotle thus makes a distinction between the object of our thought and the φάντασμα or visualization of the object that we use as a help, or perhaps cannot help making use of, when we think. The latter presents the former in a certain way, but is not the intentional object, because the object by its very nature transcends our presentation or image of it. So when imagining something fearful, the fearful object presents itself to us precisely as a fearful object, though one that is imagined. The imaginative character, in other words, is part of the object’s presence or givenness. I thus take φαντασία to constitute a mode of presentation.36 Similarly, when discussing the nature of memory, Aristotle argues that when remembering, it is the past thing or event that constitutes the object of our act of remembering, not the image that makes this event present 35

My translation. My interpretation of φαντασία is not so far from the one defended by Nussbaum (1978), pp. 221-269, who similarly emphasizes that this concept is intended to capture a mode of givenness or the “as what” of the appearance of things. Nussbaum has been criticized for trying to make Aristotle into a proponent of antirealism, e.g. by Davidson (1991). See also Caston (1998), p. 280, and Moss (2012), p. 54, who read Nussbaum in a similar way, arguing that she ties the concept of φαντασία to that of an interpretative act. As I have already indicated, however, I think that her basic intuition with respect to φαντασία is perfectly sound, as long as it is not taken to warrant a dualism of appearance (or “interpretative act”) and reality. 36

60

C. WEIGELT

to us, because in that case, memory would not concern the past but the present. So the role of φαντασίαin this context is to make present something precisely as past, it lets us picture it, but what we remember is not a picture but an event, as presented to us by the φάντασμα. We can of course make an image into an intentional object, as when we reflect upon our very memory as a representation of the past, but then we have changed our perspective. Leaving the question of representationalism aside, one might think that there is in fact something much more problematic going on in Aristotle: the positing of a pre-established harmony between the soul and the world. For apparently, Aristotle’s implicit answer to the question of what determines mental content is that it is both mind and world: in both sensation and thought, we receive forms, but these are in a way already in our soul, so that the soul simply finds in the world what it has itself already put there in advance.37 And this harmony seems to exist both at the lowest and the highest level of human cognition, that is to say, both with respect to the perception of the objects that are “proper” to each of the senses, like colour to sight, and sound to hearing, and so on — and in Aristotle’s idea of the pure intellect which, when contemplating the formal dimension of reality, thinks nothing but itself as the place of the forms (429a27-29). Let us begin with the former case.Myles Burnyeat (1992) once complained that one of the reasons why we no longer can take Aristotle’s theory of mind seriously is that he conceives of the “secondary” qualities as objectively existing out there in the world for us to simply receive. But this does in fact not seem to be true. Aristotle explains: “On this point the earlier natural philosophers were in error, when they supposed that without seeing there was neither white, nor black, and without tasting no flavour. Their statement is in one sense true, in another false. For the terms sensation and sensible thing are ambiguous. When they mean the actual sensation and the actual sensible thing, the statement holds good: when they mean potential sensation and potential sensible, this is not the case. But our predecessors used terms without distinguishing their meaning.” (De an. 426a20-26) 37 So conceived, Aristotle’s conception of the soul would ultimately be governed by his theological ideal, which centers on the idea of a seamless unity of divine thinking with its object, of νοῦς and τὸ νοητόν. Not everyone has found such a notion problematic, however. To Plotinus, the problem with Aristotle in this respect is, on the contrary, that he did in fact not manage to completely eliminate the duality of thinking and the object of thought; cf. Nyvlt (2012), Ch. 5.

“THE SOUL IS IN A WAY ALL THINGS”

61

In perception, it is not only we who move from potentiality to actuality, namely as we exercise our powers of sensation, but there is a parallel transition going on in the things themselves: they move from being potentially to actually intelligible. “For, just as acting and being acted upon are in the subject acted upon and not in the agent, so also the actuality of the sensible object and that of the sensitive faculty will be in the percipient subject” (426a9-11). When the soul encounters the world, there is thus a kind of mutual actualization, not to say constitution of our knowledge of the world and of the world itself as intelligible. This is also suggested by the Metaphysics Book M.9: knowledge of universals is as such potential knowledge, and it is only as actualized, when it directs itself towards a specific object, that it receives a determinate content.38 In this transition from potentiality to actuality, there is at least the possibility of friction, and thus of falsity (as well as of truth), as Aristotle also remarks in Deanima, when commenting on the as-structure of perception of particular objects, just as we saw above. “Every statement of something with respect to something [τι κατά τινος], like an affirmative statement, is true or false” (430b26-27).39 As far as the intellect is concerned, it is no doubt true that Aristotle with his notion of thinking thinking itself envisages a complete reconciliation of mind and world, where we no longer apprehend things as being of a certain form but contemplate the forms themselves, which are in a way already within us. When the intellect has become each several thing it thinks, it has also acquired the capacity to think itself, Aristotle remarks (429b5-9), and “the intellect is itself intelligible [νοητός] in just the same way as the objects of thought [τὰ νοητά]. For where the objects are immaterial that which thinks and that which is thought are the same” (430a3-4).40 At this level, falsity is excluded; it is simply a matter of having or not having the concepts in question, as Aristotle makes clear in the Metaphysics (1051b30ff.). It seems impossible, in other words, to 38 In his attempt to saddle Aristotle with a version of direct realism, Esfeld (2000), p. 334, claims that what matters in this respect is that, on Aristotle’s view, the world has a conceptual structure, not whether the mind is purely receptive in relation to this structure or also involves spontaneity. But if I am correct that the spontaneity of our cognitive faculties amounts to a capacity to actualize an otherwise merely potential conceptual structure of the world, then Aristotle’s position is perhaps not so much a direct realism as rather a “subtle realism”, to borrow an expression of Marmodoro (2014), e.g. pp. 98-102, 141. 39 My translation. 40 Hicks’ translation has been modified.

62

C. WEIGELT

misunderstand the forms themselves, taking them as something they are not: “The process of thinking indivisible wholes belong to a sphere from which falsehood is excluded” (430a26-27). At the same time, however, νοῦςis, according to Deanima, “nothing at all actually before it thinks” (429a24), and even when it has become that which it thinks, it remains a capacity (the latter is, thus, never exhausted), though not in the same sense as before, since now it has acquired knowledge (429a8-9). This claim, I propose, should be read together with Aristotle’s statement that the universals are “in a way” (πώς) in the soul (417b23-24): strictly speaking, universals are not that kind of entities that can have a location, nor can anything be in our intellect, since it is not a magnitude. What positively can be said about it is simply that it is a capacity or faculty. The forms are potentially within its reach in much the same way as writing is “in a tablet which has nothing actually written upon it” (430a1-2). Moreover, even if Aristotle stresses that the intellect in itself is eternal, divine, and so on, he also, as we have seen, emphasizes that we normally think as embodied beings, which means that even when we are thinking in the abstract, we need the faculty of imagination, φαντασία, the power to make things present to us. “But, since apart from sensible magnitudes there is not a thing, as it would seem, independently existent, it is in the sensible forms that the intelligible objects exist, both the abstractions of mathematics, as they are called, and all the qualities and attributes of sensible things. And for this reason, as without sensation a man would not learn or understand anything, so at the very time when he is actually thinking he must think with the help of images.”41 (432a3-9)

To the extent that our intellect is bound up with sensitivity, we necessarily actualize our universal conceptual powers in the direction of something particular, which means that there is never a complete harmony between thinking and the object thought, but the latter, even when it is an abstract form, is always taken as something, which is to say that it appears to us, to our power of imagination, in a specific way. That is at least one possible way of interpreting Aristotle. In Deanima, Aristotle outlines his idea of a teleological unity of mind and world, according to which these two are mutually related to one another as potentiality and actuality: on the one hand our soul is a capacity for receiving forms, and in that sense it is in potentiality what its object is 41

Hicks’ translation has been slightly modified.

“THE SOUL IS IN A WAY ALL THINGS”

63

in actuality, but on the other hand, to receive a form, and thus to apprehend something as something, is to actualize the otherwise merely potential intelligibility of the object, which actualization has its subjective conditions, insofar as the forms are in a way in the soul, as potential conceptual powers that need to encounter a specific object, be it concrete or abstract, in order to be activated. We might think that Aristotle’s “solution” to the modern problem of intentionality, the fear that we are stuck in a way of thinking that leaves mind out of touch with reality, because it focuses too much on the distinction between what is internal and external to the mind, comes at a very high prize, that it is nothing but full-fledged idealism. But at least when it comes to Aristotle’s account of perception, or in general of our apprehension of particular objects, where our conceptual powers are not necessarily confirmed but may also be challenged by the object that presents itself to us, precisely in virtue of its particularity, I think he offers us a pretty attractive view of how our dealings with the world involves receptivity and spontaneity, so that we neither are standing naked before the world, nor just project our concepts upon it.

LITERATURE BRENTANO, Franz (1924), PsychologievomempirischenStandpunkt.ErsterBand (ed. O. Kraus) Hamburg: Meiner. BRENTANO, Franz (1967/1867), Die Psychologie des Aristoteles, Darmstadt: Wissenschaftliche Buchgesellschaft. BURNYEAT, Myles F. (1992), “Is an Aristotelian Philosophy of Mind Still Credible?” in Nussbaum, M. & Oksenberg, A. (eds.) EssaysonAristotle’s De Anima, Oxford: Clarendon Press, pp. 15-26. CASTON, Victor (1998), “Aristotle and the Problem of Intentionality” in PhilosophyandPhenomenologicalResearch 58.2, pp. 249-298. COOKE, Harold P. & Tredennick, Hugh, transl. (1983), (The Loeb Classical Library),Aristotle,vol.I:TheCategories,OnInterpretation,PriorAnalytics, Cambridge: Harvard University Press. DAVIDSON, Donald (2001), Essays on Actions and Events, Oxford: Clarendon Press. DAVIDSON, Jack D. (1991), “Appearances, Antirealism, and Aristotle” in PhilosophicalStudies 63.2, pp. 147-166. DIELS, Hermann (1956), Die Fragmente der Vorsokratiker I-II (ed. Walther Kranz), Berlin: Weidmannsche Buchhandlung. ESFELD, Michael (2000), “Aristotle’s Direct Realism in De Anima” in The ReviewofMetaphysics 54.2, pp. 321-336.

64

C. WEIGELT

FREDE, Michael & PATZIG, Günther (1988), Aristoteles,“MetaphysikZ.”Text, ÜbersetzungundKommentar, München: Beck. GALLAGHER, SHAUN & ZAHAVI, DAN (2008), The Phenomenological Mind: an Introduction to Philosophy of Mind and Cognitive Science, London: Routledge. GOLDBERG, SANFORD C., (ed.) (2015), Externalism,Self-knowledgeandScepticism,Northwestern University, Illinois: Cambridge University Press. GRAHAM, Daniel W. (2010), TheTextsofEarlyGreekPhilosophy1, Cambridge: Cambridge University Press. HICKS, R. D. (1907), AristotelesDeAnima, Cambridge: Cambridge University Press. HUME, David (1997), Enquiries Concerning Human Understanding and Concerning the Principles of Morals, Oxford: Clarendon Press (third ed., reprinted from the 1777 ed.). HUSSERL, Edmund (1992), IdeenzueinerreinenPhänomenologieundphänomenologischenPhilosophie, Hamburg: Meiner. HUSSERL, Edmund (1993), Logische Untersuchungen: Untersuchungen zur PhänomenologieundTheoriederErkenntnisII/1, Tübingen: Niemeyer. KORNBLITH, Hilary (ed.) (2001), Epistemology: Internalism and Externalism, Oxford: Blackwell. MAGEE, Joseph M. (2000), “Sense Organs and the Activity of Sensation in Aristotle” in Phronesis 45.4, pp. 306-330. MARMODORO, Anna (2014), Aristotle on Perceiving Objects, Oxford: Oxford University Press. MCDOWELL, John (1996), Mind and World, Cambridge: Harvard University Press. MOHANTY, J. N. (1981), “Intentionality and Noema” in TheJournalofPhilosophy 78.11, pp. 706-717. MOSS, Jessica (2012), AristotleontheApparentGood:Perception,Phantasia, Thought,&Desire,Oxford: Oxford University Press. NUSSBAUM, Martha C. (1978), Aristotle’sDe Motu Animalium, Princeton, NJ: Princeton University Press. NUSSBAUM, Martha C. & Putnam, Hilary (1992), “Changing Aristotle’s Mind” in Nussbaum, M. & Oksenberg, A. (eds.) EssaysonAristotle’sDe Anima, Oxford: Clarendon Press, pp. 27-56. NYVLT, Mark J. (2012), Aristotle and Plotinus on the Intellect, Lanham, MA: Lexington. RAPP, Christof (2001), “Intentionalität und phantasia bei Aristoteles” in Perler, D. (ed.) Ancient and Medieval Theories of Intentionality, Leiden: Brill, pp. 63-96. SCALTAS, Theodore (1996), “Biological Matter and Perceptual Powers in Aristotle’s DeAnima” in Topoi 15.1, pp. 25-37. SEIDL, Horst (1971), DerBegriffdesIntellekts(νοῦς)beiAristoteles, Meisenheim am Glan: Hain. SEN, Madhucchanda (2015), Externalism and the Mental, New Delhi: Oxford University Press.

“THE SOUL IS IN A WAY ALL THINGS”

65

SHIELDS, Christopher (2016), Aristotle. De Anima. Translated with an IntroductionandaCommentary,Oxford: Oxford University Press. SORABJI, Richard (1974), “Body and Soul in Aristotle” in Philosophy 49.187, pp. 63-89. SORABJI, Richard (1992), “Intentionality and Physiological Processes: Aristotle’s Theory of Sense-Perception” in Nussbaum, M. & Oksenberg, A. (eds.) EssaysonAristotle’sDe Anima, Oxford: Clarendon Press, pp. 195-225.

THE AMBIGUITY OF APPEARANCE. ON RESPONSIVITY IN ARISTOTLE Jan-Ivar LINDÉN ὅταν δ’ ἐνεργῇ τὸ δυνάμενον ἀκούειν καὶ ψοφῇ τὸ δυνάμενον ψοφεῖν, τότε ἡ κατ’ ἐνέργειαν ἀκοὴ ἅμα γίνεται καὶ ὁ κατ’ ἐνέργειαν ψόφος



“But when that which has the power of hearing is exercising its power, and that which can sound is sounding, then the active hearing and the active sound occur together…” De anima 425b30-426a1

Appearance is for Aristotle crucial. It stands for how reality shows itself (φαίνεισθαι).1 The appearing world is however phenomenal in another way than the phenomena of Immanuel Kant, which are constituted by a subject through a synthetic activity which renders indeterminate sensual manifolds perceivable. When Kant uses the Aristotelian concepts of form and matter, they have a new meaning: form standing for subjective presuppositions (forms of the sensual intuition, “Formen der Anschauung”, but also categorial forms of the intellect, “Verstandesbegriffe”) and matter for sensual material, what transforms the Aristotelian hylemorphic scheme of real things into a subjective one. Things thus for Kant become knowable objects, because they are adapted to subjective faculties. For Aristotle, on the contrary, the world is simply given. There 1 The main focus in twhat follows can be understood as a thematically oriented interpretation of certain passages in Aristotle concerning sensation, imagination and perception. The literature is vaste. I mention only some texts with relevance in the present context: Barnes, Schofield & Sorabji (1979), Herzberg (2011) and Nussbaum & Oksenberg (1992). Sometimes the interpretations create problems in projecting modern views on Aristotle. As Charles Kahn puts it: ”The possibility of such contradictory interpretations arises from the fact that Aristotle does not speak the language of traditional dualism. He does not contrast the mental with the physical”, Kahn in Barnes, Schofield & Sorabji (1979), p. 44.

68

J.-I. LINDÉN

is no sceptical problem concerning the “external world” and no necessity of proving its existence.2 The problem is instead the essential one of understanding what is given when something is given, in other words to better understand the familiar or even sometimes to understand how something is in itself (kath’auto). The faculty which lets imagelike contents appear, phantasia, however creates some problems for Aristotle as also for later commentators and philosophers. It is not even clear if we should translate phantasia with imagination or not. A later tradition tended to regard imagination as a capacity to offer mental pictures which not only figure in dreams and fantasies, but also in sensual perception as representations which somehow resemble things. Descartes explicitly argues against such a theory of perception and the main problem for him was the idea of a perceptive imagination (imaginatio) founded on similarity. He attributed this for him false view to Aristotelian psychology and favoured an Augustinian solution with modern adjustments. However, it is not at all clear that Aristotle would regard the phantasmata as images resembling things. When he brings up this topic for the first time in Deanima (414b16), he is quite cautious. He adds that it is unclear (αδηλον) how we should understand the phantasia and postpones the question for the moment. In the last book of Deanima,he restates that it is extremely difficult either to identify the phantasiawith, or to distinguish it from, other functions of the soul.3 It is evident that the topic is a problematic one already for Aristotle himself. In the following, I will try to sketch an interpretation, which sees this faculty — which I hesitate to translate with imagination — as a mediating function between sensation and striving (ὄρεξις). In Deanima,Aristotle tells us that the striving function of the soul (ορεκτικόν) is not different from the perceiving one (αἰσθητικόν), even if their being (το εἶναι) is different. He even adds the faculty of avoidance (φευκτικὸν) as an identical function with different “being”. There are living creatures without phantasia, like the plants, which are tendential in a way similar to animal striving, but for which Aristotle generally uses other expressions than orexis.4 Striving in the narrower sense, linked to movement and sensation, are those beings (i.e. animals), 2 3 4

Phys. 193a3-9, Met. 1010b33-1011a2. De an. 432b1-4. See the contribution of Tomás Calvo above, pp. 31-43.

THE AMBIGUITY OF APPEARANCE

69

where imagelike contents (φἀντασματα) play an eminent role. In other words, the same tendential function has different modes. It is thus quite important to take into account how striving and perception are interrelated. It is in this context also obvious that there is an emotional component which is active in animal striving — as it is the case when the animal tries to approach something pleasurable or avoid something painful. Emotional and sensual affection are thus intertwined. A phenomenon is something which can show itself and when this is actually happening there is not only a phenomenon, but also an appearance, a perceived phenomenon. What does sensual perception add to the phenomenon and how is this linked to the question of intentionality? The sensual faculty (αἰσθητικόν) fulfils its function when it exposes itself to the influence of aisthēta. Aristotle uses aisthēton for a perceivable thing, but as it seems, also for sense dependent aspects of the things — like colours for sight and sound for hearing. There are also qualities which are common to the different senses and interestingly enough the list given — movement, rest, number, shape and size (κίνησις,ἐρημίᾱ, ἀριθμὸς,σχήμα,μέγεθος) — resembles what was later called primary qualities.5 Concerning these common qualities, which became that fundamental as “primary qualities” in the early modern tradition, Aristotle notes that they are more prone to incite error than the sense-specific idia.6 This shows how radical the shift in the early modern tradition was and also the confidence Aristotle had in the reality of the sensual qualities, which the moderns tended to regard as secondary. In his critique of Democritus, Aristotle also argues against the role the atomistic school attributed to the common qualities, which the atomists however linked to one single sense, touch. There would certainly be a lot to say about this. In fact, the materialist tendency to stress the sense which seems to be in most direct contact with the corpuscular sphere, has as its corollary the possible manipulation of the tactile dimension of reality. Aristotle, on the contrary, regards all senses and the corresponding sensual qualities as important in their specificity and tactile qualities can thus in no way be regarded as ontologically primary (even if tactile exposition is fundamental for living beings because they expose us more than “distant” senses). If the “materialist” tendency of the atomists is one possibility, there is 5 6

De an. 418a18-19. A somewhat different list is given in Desensu 442b5-6. De an. 428b18.

70

J.-I. LINDÉN

also the “idealist” version of the Pythagoreans, in which the sensual world becomes a matter of mathematical relations. Aristotle criticizes both strains of thought7 and precisely this parallel critique is of interest today, with modern materialist ideologies tending to regard their basic matter as something measurable. The Aristotelian theory of perception offers us other possibilities. Aristotle also speaks of a third group of aisthēta, the indirectly perceived, for example when something of a certain colour (incidentally) appears as a person we know.8 This last group implies a certain attribution of identity, which is already doxastic. The common qualities, too, seem to be marked by an intellectual component, and would thus not be pure affections like the sense specific qualities. In this context, one has to recall that the possibility of error comes through combination (σύνθεσις). Because of this the sensations of specific qualities are even described as “always true and characteristic of all animals”, whereas it is possible “to think falsely”.9 Discursive thought is, however, dependent on the conceptually combining faculty of logos, which can certainly conceive something to be true, but precisely because it is attributing predicates to something and demonstrates through an examination which aims at excluding error — even if not always succeeding. This kind of thought has neither the incontestable character of aisthēsis nor the immediacy of opinion (δόξα), even if it presupposes both of them. As epistēmē and nous it has, however, its own way of being incontestable. According to Aristotle the process of sensation somehow means that perceiver and perceived coincide, are the same even if with “different being”.10 This leads him to a theory of sensual qualities which emit movement and is capable of influencing perception. Qualities like colour, sound, tactile properties, smell and flavour are real aspects of things and create effects in the corresponding five senses. In other words: there is something in the world triggering sensations. Aristotle is very explicit about this: the sensual qualities exist also when they are not perceived, 7 Against the Pythagoreans Aristotle argues: The aisthēton “cannot be something out of mathematical abstractions” (οὐ γὰρ δὴ ἔκ γε τῶν μαθηματικῶν), Desensu 445b14-15. 8 De an. 418a20-24. 9 ἡ μὲν γὰρ αἴσθησις τῶν ἰδίων ἀεὶ ἀληθής, καὶ πᾶσιν ὑπάρχει τοῖς ζῴοις, διανοεῖσθαι δ’ ἐνδέχεται καὶ ψευδῶς, καὶ οὐδενὶ ὑπάρχει ᾧ μὴ καὶ λόγος, De an. 427 b 11-14. This is a quite general concept of ‘truth’ in Aristotle, which is characterized by the impossibility of error. Truth has also a more strict sense, linked to judgment and noetic capacity. 10 De an. 425 b 26-27, 426 a 15-17.

THE AMBIGUITY OF APPEARANCE

71

because they still have their dunamis, a word which in this case could be rendered by power.11 Concerning matters of translation there is also a difficulty with the word aisthēton, which is not well rendered by “perceived object”. Objects seem to presuppose objectifying activity — and thus would rather be results than beginnings. For Aristotle sensual perception is however something starting from the perceived. “It seems obvious that the sensually perceived makes a real sensation out of an only possible one: then it [the aisthēton] is neither undergoing anything nor is it changing” (φαίνεται δὲ τὸ μὲν αἰσθητὸν ἐκ δυνάμει ὄντος τοῦ αἰσθητικοῦ ἐνεργείᾳ ποιοῦν· οὐ γὰρ πάσχει οὐδ’ ἀλλοιοῦται).12 The perceived is more like a sensual power of things, initiating experience13 and the perceiving animal does something in response to this initial stimulation. (Aristotle can, in this respect, be regarded as a precursor of biosemiotics.)14 Both in De anima and De sensu we are occasionally told that the aisthēta are “outside” (ἔξωθεν), but it is not completely clear how this externality should be understood. One tendency is to describe an influence from things on the organism, involving a medium like air or water and sensual organs like eye and ear, which receive impulses from outside. Both the things and the organs have a spatial size (μέγεθος), but even when Aristotle elaborates his view with concepts of this kind, he remains faithful to his basic idea of the soul as an unextended principle of life. It is this psychic principle which renders worldly influences and organic responses meaningful, and the Democritean view of a spatial transmission of small corpuscular images (εἴδωλα) is certainly not a satisfying explanation for him. (All examples which could suggest an eidolon, moving through the air or another medium in order to arrive in an equally material sensible substrate are analogies which should not be taken literally.)15 It is thus not surprising to find the remark that sense and 11

De an. 426a21-26. De an. 431a4-6. 13 De an. 426b8 ff. 14 Modern naturalism has however often a reductionist character which would be foreign to Aristotle. Concerning the natural condition of man, see Labarrière (2005). 15 Like the one about the signet ring in wax, De an. 424a19-20. Cf. also 435a. Neither is Aristotle adhering to the modern idea of a “veil of perceptions”, reduplicating the perceived things. He simply speaks about what is happening in perceptual situations where there is a thing perceived with sensual qualities and an animal capable of sensually taking notice of this thing. 12

72

J.-I. LINDÉN

sensibility cannot have size, but instead has to be understood as a logos tis, a kind of good proportion.16 (Some translators use ratio.) This leads to the second more ontological tendency in Aristotle’s theory of perception, in which the outside has another meaning, rather related to the unextended soul. It is said that the aisthēta are “particular and external” (τὰ αἰσθητὰ τῶν καθ’ ἕκαστα καὶ τῶν ἔξωθεν)17, but the contrast made between what is outside and insidethesoul suggests that the aisthēta are external sensibles, because they are not yet assimilated by the soul.18 Through our vital constitution we are involved in the world as practical animals, acquainted with things and our soul is in its perceiving function “like a hand”.19 In sensual perception we are however not using things in a manual way, because the sensually perceived things are presented to us as imagelike contents (φἀντασματα)“without matter” (ἄνευ ὕλης).20These contents are especially important for the animal with its locomotive capacities and there is certainly in movement a kind of difference between the outside and the inside as all other things are not moving in the same way as the animal itself, but this form of spatial animality is related to a multitude of places for a psychic being and does not imply that sensual perception could be regarded as spatial transmission. What at first sight seems to be a movement from outside into the organism, is in ontological terms more like the realisation of potentialities in the phenomenal world. The perceiving animal assimilates the world through its responses and, in this case, the outside is more like the nonassimilated heteron. The otherness of the perceived is then something triggering reactions which become responsive through the animal effort to render influence meaningful. Aristotle uses the concepts of similarity and dissimilarity, present in earlier Greek philosophy of nature, to express this in the following way: an aisthēton is a particular, from which the sensually perceiving animal receives something which in the beginning 16

De an. 424a27-28. De an. 417b27-28. How the status of the external has to be understood in the last analysis is not the topic neither in Deanima nor in Desensu, but in the Physics and in the MetaphysicsAristotle has several things to say about these matters — and none of them seem to confirm an Aristotelian equivalent to the modern notion of the external world. 18 De an. 417b16-18. 19 De an. 432a1-2. 20 De an. 432a10-11. 17

THE AMBIGUITY OF APPEARANCE

73

is unlike, but then is assimilated and becomes alike. Aristotle speaks of what is “saved”.21 The faculty of sensation is sensitive to different aspects of the world and can discern what these aspects are all about. Belonging to the world is for Aristotle always a primordial fact. The change from potentiality to actuality in the sensual faculty is initiated by a similar process in the world, where a sensible thing creates movement under certain conditions, but without the sensible thing itself necessarily moving at all. As far as concerns the sensual faculty on the other hand, it is almost as if it would become active in order to become passive, expose itself to insisting aisthēta. The Aristotelian remark that the soul in a wayisthe beings, the things it knows or perceives is, in this context, understandable.22 How influence happens is however not as clear, but obviously related to the Aristotelian idea of things as informed matter. In sensual perception, the form of the thing is somehow transferred to the perceiving animal, but as we are told, without the matter. The fact that a thing is an aisthēton means that it carries several sensual qualities, which then in sensual perception contribute to the transmission of the formal characteristics of the thing. Through the qualities we discern what the thing is. It is not very plausible to ascribe the view to Aristotle, that the soul would become like a plant when it through its sensual faculty perceives a plant. The plant follows its own formal development in evolving precisely as a plant and the instructive form of the plant moves of course nowhere when the plant is perceived. The realisation of the form of the perceived thing is not identical with the realisation of the form which constitutes the perceiving animal — and in the animal form an essential component is striving (ὄρεξις). Sensation alone does not render a phenomenon accessible neither to action nor to conceptual thought. It is here that the mediating function of the imagelike contents come into play. Sometimes Aristotle stresses that phantasmata exist separately from the sensual affections, the aisthēmata, but mostly the phantasmata are described in their close connection with aisthēsis. There is however the double possibility that phantasia“comes about either through thought or through sense-perception” (γίνεται ἢ διʹ 21

soteria;De an. 417 b3. The famous “the soul is in a sense all existing things” (ἡ ψυχὴ τὰ ὄντα πώς ἐστι πάντα), De an. 431b21-22. Cf. the contribution of Charlotta Weigelt above, especially pp. 52-53. 22

74

J.-I. LINDÉN

νοήσεως ἢ δι αίσθήσεος).23 Imagination not always being dependent on sensation is an important fact — and this explains the crucial role phantasiacan have both in our theoretical and practical life. In sensual experience, the imagelike contents of phantasia are the functions, which give a sort of stable content24 to the fluid sensual affections, provoked by real aspects or things in the world. This stability is however no sign of ontological primacy, as the stable in other contexts can be both in Aristotle and in the Platonic tradition. The phantasm stands for a relative stability in movement, which often turns out as wrong and thus has not the incontestable character of sensation. Because of this the phantasmatahave a certain independence and express an interpretation, which can be erroneous. The phantasms are like an elementary content, which then in thought can be elaborated in order to find what is really stable in the appearance. Like so often in Aristotle we also in this case go from the obviously given to its essential traits or, as Aristotle puts it in Deanima, from the obvious but vague to the better conceived (κατὰ τὸν λόγον γνωριμώτερον).25 It is in this context understandable that thought always has to involve phantasmata, which are conceptually clarified through the intellect. Even more evident is the role of phantasia in practical life and action.26 In the second book of Deanima, it is said that sensation necessarily implies phantasia and — what is interesting in relation to intentionality — also striving. It is the pleasure animals want to attain and the pain they want to flee, which create both striving and movement — and apparently this is not without sensual significance. The general context of an Aristotelian theory of perceptive intentionality — if there is such a thing — must lie in the fact that perceiving (human and non-human) animals are teleologically oriented, i.e. striving towards something. One could call this an orectic intentionality. How does striving influence sensual perception? What is the relation 23

Demotuan.702a19. De an. 429a4-5. 25 De an. 413a12. 26 Cf. Pierre Destrée on akrasia as “lack of phantasia logistikē”, Destrée (2007), p. 161 in the collection of essays on AkrasiainGreekPhilosophy, Bobonich & Destrée (2007), with contributions of Thomas Brickhouse, Nicholas Smith, Christopher Rowe, Chris Bobonich, Christopher Schields, Roslyn Weiss, Gabriela Roxana Carone, LouisAndré Dorion, Pierre Destrée, Marco Zingano, David Charles, Jean-Baptiste Gourinat, Ricardo Salles and Lloyd Gerson. 24

THE AMBIGUITY OF APPEARANCE

75

between aisthēmata,phantasmataand orexis? Does Aristotle perhaps have a successor in Husserl and his theory of intentional acts and contents? Such a connection could find some evidence in the role Franz Brentano, with his explicitly Aristotelian bent, had in the early stages of Husserl’s phenomenology. One could even claim that Husserl is closer to Aristotle than Brentano, because two main, as it seems to me, quite non-Aristotelian aspects of Brentano’s view are less present in Husserl’s phenomenology: the idea of the purely internal character of intentions and Brentano’s mentalist refusal to accept any unconscious aspects of experience. As already Carl Gustav Carus saw in his work Psyche from 1846, the Aristotelian soul is to a large extent unconscious.27 Concerning Husserl, the problem is however that his phenomenology is closely linked to the Kantian idea of synthesis and subjectively constituted objects. If there is a difference between act and content in Aristotle, it seems that it must have a different meaning than both the Brentanian and the Husserlian one. It is in this context that I use the notion of orectic intentionality.28 The sensitive psychic function, the aisthētikon, is closely connected with movement (κίνησις), which can mean both spatial movement (κίνησις κατα τοπὸν or φορά) and change (ἀλλοίωσις). Insofar as every natural realisation of potentiality means change, life itself as realisation (ἐντελέχεια) stands for change and accordingly also sensation must be marked by this fact. In animals the soul is however different from the only nourishing and reproductive psychic function (similarly present in plants) in that it generally also comports spatial movement.29 It is thus evident that sensations in such animals (like human beings) must vary during a perceptual act insofar as there are on-going kinetic processes — change, but often also spatial movement. 27 Carus (1964). Cf. Lindén (2011/2012), and for a thematic development of similar topics, Lindén (2017).There are passages in Aristotle where he seems to mean a constant awareness, such as Desensu448a, but he then speaks about the temporal process of actual perception. This does not contradict the existence of vitally dominant unconscious psychic drives. 28 This form of intentionality seems to play a certain role in some pragmatist accounts. See especially James (1981). A phenomenological account of vital intentionality is found in Barbaras (2003). 29 De an. 414b16-19, 415a6-7, 415b21-23, Desensu 436b18-21. In Demotu, Aristotle speaks of several moving faculties—dianoia,phantasia,prohairesis,boulēsis,epithumia —but then resumes these as nousand orexis.Demotuan. 700b17-19.

76

J.-I. LINDÉN

Both the perceiving animal and the perceived phenomenon or thing have essential formal characteristics. In perception these two meet and the perception itself happens through a transmission, which is not possible without a medium (μεταξύ). What is transmitted is somehow the form of the perceived thing, but in a particular way, which is marked by the fact that the perceiving animal, too, has its formal ends and strives according to these. The appearing form of the thing thus must appear inside a psychic constellation, which expresses a different form than the form of the thing perceived. The transmission itself is also due to the (colourful, smelling, sounding) sensual qualities, offering a sort of replacement for the lost hyletic basis of the form, which in the thing as such carries its existence but somehow gets dematerialized in sensation. As forms can be actually given only when they are carried by something underlying (hupokeimenon), also the sensually given forms need something carrying them, in this case affections (aisthēmata). These affections can be understood as reactions to sense-specific influence. Aristotle even poses the question if sensual qualities of one kind can influence sensual faculties of another kind (smell influencing seeing, colours influencing the olfactive sense etc.)30 — what shows how important the idea of sensual forces in surrounding nature is for him (even if his answer is negative). Man is a responsive being in the middle of nature. The role of the phantasma is to coordinate the sensual influence of things with the formal and final characteristics of the perceiving and moving animal. Sensual perception concerns the way of real phenomena to appear in the striving life of a being with ends different than the ends of the phenomena perceived. The phantasm seems to express something of our desires (— Freud was quite Aristotelian in acknowledging this —), but it would not be true to say that the phantasm for Aristotle is only an expression of the desires of the perceiving animal. We normally do not perceive the contents we want to, but quite to the contrary, the importance of sensual perception lies exactly in the fact that there is otherness in sensual experience (πάντα δὲ τῷ δι’ ἑτέρου αἰσθάνεσθαι ποιεῖ τὴν αἴσθησιν).31 A sense is also a discerning function which discloses aspects of the perceived world, κρίνει τὰ αἰσθητά.32 Some human senses are more 30 31 32

De an. 424b. De an. 435a16-17. De an. 424a6.

THE AMBIGUITY OF APPEARANCE

77

discriminating than others, according to Aristotle sight, hearing and touch for example more than smell. Finite senses become aroused in particular situations because they are receptive to the otherness of the non-assimilated aspects of the world, which are (still) “external” to the soul. The sensual reaction is an exposition to influence (πάσχειν), but when the sensual process has been traversed, it means reduction of otherness. This assimilation of unfamiliar aspects continues throughout animal life, which as such is dependent on aisthēsis and its never ending responses to changing particular circumstances. Life is, in a way, the constant exposure to otherness. Sensual reaction is thus responsive and integrates new aspects of the world in the familiar psychic organisation with its inherent own ends and aims. Keeping the teleological orientation (the striving) of the animal and the influences coming from its sensual exposure together is the role of phantasia. An animal has to cope with things, lots of things, quite different things, which are there and trigger reactions in the experience. All these things are certainly not desired ones, but they appear for an animal with plenty of desires. The phantasm is the mediating function, which gives a relatively stable content to the sensual presence of dematerialized forms of things on the one hand and the psychic direction of the perceiving animal on the other hand. There is some space for this mediation because of the non-immediate character of sensual experience, what Aristotle explains with the medium (μεταξύ) necessary for sensual transmission. In DeanimaAristotle states that sensing presupposes a medium.33 One of Aristotle’s examples of the necessary medium concerns sight. The eye can only see if the seen is not placed directly on its surface.34 In this context, it is said that even touch has the flesh (σάρξ) as its medium. (Aristotle uses the example of a fabric placed on the skin and the possibility to feel through the fabric in order to clarify the medial role of

33

Concerning the temporal aspects of the medium, see also Desensu 446a-b. De an. 419a12-13. In the Greek-German Meiner edition there is here a translation error. The sentence ἐὰν γάρ τις θῇ τὸ ἔχον χρῶμα ἐπ’ αὐτὴν τὴν ὄψιν, οὐκ ὄψεται is rendered without the negation ouk. The sentence thus becomes “Wenn nämlich jemand das farbige Objekt auf das Auge legt, so wird er es sehen” and not as it should be: “Wenn nämlich jemand das farbige Objekt auf das Auge legt, so wird er es nicht sehen.” The translation of the other relevant passage is however correct: 423b20-22. Aristoteles Über dieSeele.Mit Einleitung, Übersetzung (nach W. Theiler) und Kommentar herausgegeben von Horst Seidl, Felix Meiner Verlag, Hamburg 1995. 34

78

J.-I. LINDÉN

flesh.35) Thus, even if we can make the distinction between distant senses (seeing, hearing and smelling) and senses through direct contact (taste and touch), this does not seem to mean that there would be no medium at all in touch and taste. As Aristotle later remarks, the main difference is rather that distant senses can have their organs destroyed through extreme sensual influence, whereas such extreme influence in the case of direct senses destroys not the organs, but the whole animal. Sensual exposition is potentially dangerous and, in some cases, exposition can even be fatal.36 The life of plants is thus particularly vulnerable as plants cannot move away and do not even have the sensual faculty in the same multiple way as animals. If there is some kind of sensation in plants — which is suggested by Aristotle’s comparison of the animal mouth with the roots of a plant37 — it is a direct one which assimilates matter, thus an only quasi-sensual faculty perhaps somehow analogous to taste and touch. Plants do not have the mean (μὴ ἔχειν μεσότητα) which according to Aristotle is necessary for the sensual reception of forms and the discerning capacity (τὸ γὰρ μέσον κριτικόν).38 What we can perceive is not too loud, not too silent, not too dark, not too bright, not too low and not too high in pitch, not too bitter and not too sweet…39 It is, however, also true that the sensation is under receptive influence (τὸ γὰρ αἰσθάνεσθαι πάσχειν τι ἐστίν).40 What exposes the animal to the world seems to be an affection which has its first impulse in the deviation from the mean.41 In the beginning there is an aisthēton which is not alike, which arouses reactions, and then little by little is saved for the future of the perceiving animal when it is getting integrated in its psychic constitution.42 35

De an. 423a2-4. De an. 435b4-19. The extreme destroys the sensual organ. Aristotle in this context seems to speak abut the form constituting the organ, a logostis,De an. 424a31. Such a logosevery perceiving zōonseems to have, not only the zōonlogonechon. 37 De an. 424a6. 38 De an. 424b2, 424a7. 39 De an. 422a20-33, 424a2-6. 40 De an. 424a1. 41 Maine de Biran will later develop a theory of sensation, which takes this passivity of affection as its starting point in order to develop the difference between sensation and perception. Maine de Biran (1987). 42 Cf. “Nothing produces or suffers an effect except in so far as it contains some element of contrariety” (ᾗ δʹ ὑπάρχει ἐναντιότης ἐν ἑκάστῳ͵ ταύτῃ πάντα καί ποιοῦσι καί πάχουσιν), Desensu 441b14-15. 36

THE AMBIGUITY OF APPEARANCE

79

I try to summarize: Aristotle stresses the interrelatedness of sense perception and the animal faculty of movement. He also tells us that the imagination is an essential part of how we perceive and that it contributes to our striving. The in itself unmoved orekton(i.e. what we desire) creates movement when it is given as thought or phantasma. The self-movement, characteristic of animal beings, depends on striving and striving generally presupposes the phantasia which renders the desired present. Imagelike contents can figure without sensation (in dreams, but also when we are thinking of something without corresponding sensations) and such phantasms thus constitute a sort of intermediary dimension which can trigger movement and striving even when no sensual perception is going on. As Aristotle says, phantasia can be working, not only in perception, but also in rational planning (λογῐσμός), which ends up in the execution of thoughtdirected acts. He even speaks of a “deliberating imagination” — phantasiabouleutikē— an expression which, however, suggests an internal tension similar to that of the bouleutikēorexis,the deliberating striving in the Nicomacheanethics.43 Deliberation is exactly something which retains us from doing, gives us time to reflect — and thus as such is not striving. If the phantasm has such a function, too, it seems to be in a certain contradiction with its capacity of inciting action. One could perhaps say that precisely this tension between a potentially inciting and only contemplating tendency is essential to the mediating function of thephantasia.44 This is exactly what allows the phantasmata to become important in human action as steering imagelike contents, permitting an initiative in natural processes. Striving seems to be essential in Aristotelian theory of perception, but also the judging faculty of affirmation and negation remains rooted in orexis: “when something is pleasurable or painful, the animal is driven towards or flees it, as it is also the case in affirmation and negation” (ὅταν δὲ ἡδὺ ἢ λυπηρόν, οἷον καταφᾶσα ἢ ἀποφᾶσα διώκει ἢ φεύγει).45 From the third book of Deanima,Descartes will later in his fourth meditation take the example of how the sun appears46and use it for a quite a 43

De an. 434a6-8,EN1113a. Concerning the contemplative function, see especially De an. 427b23-25. 45 De an. 431a9-11. 46 De an. 428b3-5. Similarly in Desensu448b15-17. Concerning the Cartesian version, see the fourth meditation in Descartes (1996), especially IV.11. See also the chapters concerning Descartes in Schwarz (1934). 44

80

J.-I. LINDÉN

different theory of judgment. He can, in this new context, not accord any particular importance to the parallels between striving and judging (even if he parenthetically mentions them). Both Aristotle and Descartes claim that the false in the case of the sun appearing big as a hand (or a foot), does not lie in the appearance as such, but in the affirmation of the appearing size as a real one. For Descartes, the heterogeneous character of judgment resides in the voluntary affirmation or negation of a content which is independent of our will and in this way Descartes suggests a radical subjective freedom of will, which even motivates the description of false judgment (error) as a “sin” (falloretpecco).47 In the same context, Descartes explicitly links the infinite human will (and not the finite intellect) to man’s character of ImagoDei. Because of our infinite quasi divine will, we tend to affirm more than we understand, finite as is our intelligence. Aristotle would agree that there is no judging assumption (ὑπόληψις) without a content,48 and he describes the content as an expression of phantasia, but he has no similarly radical notion of will. In some passages he even suggests that the doxastic function is so closely connected with belief, that we choose our opinions to a lesser extent than our phantasms. We see something in a direct way as either true or false and in the practical sphere this is also accompanied by emotional reactions.49 This difference reflects a deeper one, which concerns — to use the phrase of Max Scheler — the position of man in the cosmos. For Aristotle the necessary condition of experience, action and knowledge is that we belong to reality. Descartes, mainly interested in a new form of objective knowledge, looks for an “archimedic point” outside reality and regards this non-belonging as the condition of possibility of real science — but as the sixth meditation, the correspondence and the work on passions show, he then has great difficulties to render his epistemological project practically convincing. The problem is however not only one regarding our practical life, but the fact that Descartes and Cartesian modernity has no ontology of life. The quite strange interpretations of Aristotelian teleology, still figuring especially in several popularizing scientific works, reflect this in a striking manner. A mentalist version of 47 48 49

Descartes (1996), IV.9. De an. 427b15-17. De an. 428a20-24.

THE AMBIGUITY OF APPEARANCE

81

intentionality is taken for granted in human existence and intentionality then restricted to this sphere because of the same mentalist presuppositions. Intentionality becomes a criterion or even the criterion of consciousness. Nothing of this is evident if the most obvious expression of teleology is striving, which as such is neither planning nor particularly conscious, but still directed. Human intentionality is thus in no way the proprium of man, but on the contrary, that which makes him similar to living beings in general and to other animals in particular.

LITERATURE BARBARAS, Renaud (2003), Vie et intentionnalité. Recherches phénomenologiques, Paris: Vrin. BARNES, J., SCHOFIELD M. & SORABJI, R. (eds.) (1979), ArticlesonAristotle4. PsychologyandAesthetics, London: Duckworth. BOBONICH, C. & DESTRÉE, P. (eds.) (2007), AkrasiainGreekPhilosophy.From SocratestoPlotinus, Leiden – Boston: Brill. CALVO MARTINEZ, Tomás (2018): “Ὄρεξις and Intentionality” in the present volume. CARUS, Carl Gustav (1964), Psyche. Zur Entwicklungsgeschichte der Seele, Darmstadt: WBG. DESCARTES, René (1996), Meditationes de prima philosophia, Oeuvres de DescartesVII(éd.Adam et Tannery) (French version: OeuvresdeDescartes IX), Paris: Vrin. DESTRÉE, Pierre (2007), “Aristotle on the Causes of Akrasia”in Bobonich, C. & Destrée, P. (eds.), AkrasiainGreekPhilosophy.FromSocratestoPlotinus, Leiden – Boston: Brill., pp. 139-165. HERZBERG, Stephan (2011), WahrnehmungundWissenbeiAristoteles:zurepistemologischenFunktionderWahrnehmung, Berlin: de Gruyter. JAMES, William (1981), ThePrinciplesofPsychology, Cambridge, Mass.: Harvard University Press. LABARRIÈRE, Jean-Louis (2005), Laconditionanimale.ÉtudessurAristoteetles Stoïciens,Paris: Vrin. LINDÉN, Jan-Ivar (2011-12), “Intentionnalité et perception: une esquisse aristotélicienne” in Chôra 9-10,2011/2012,pp. 339-52. LINDÉN, Jan-Ivar (2017), “Apperception and Experience. Some Ontological Perspectives” in Trópos.JournalofHermeneuticsandPhilosophicalCriticism1 (X)/2017, pp. 53-67. MAINE DE BIRAN, Pierre (1987), Mémoiressurl’influencedel’habitude, Oeuvres, tome II (éd. Gilbert Romeyer-Dherbey), Paris: Vrin. NUSSBAUM, M. & OKSENBERG, A. (eds.) (1992), EssaysonAristotle’sDe Anima, Oxford: Clarendon Press,

82

J.-I. LINDÉN

SCHWARZ, Balduin (1934), Der Irrtum in der Philosophie. Untersuchungen überdasWesen,dieFormenunddiepsychologischeGenesedesIrrtums imBereichederPhilosophie,miteinemÜberblicküberdieGeschichteder Irrtumsproblematik in der abendländischen Philosophie. Münster: Aschendorff. WEIGELT, Charlotta (2018): “’The soul is in a way all things’ — Aristotle and Internalist Conceptions of Intentionality” in the present volume.

THE ROLE OF ARGUMENTATION

EPISTEMIC LOGIC IN ARISTOTLE Simo KNUUTTILA

Epistemic logic as part of logical theory was introduced into late medieval logic when many authors added various extended modalities to their treatises on modal logic. The most discussed of these was the notion of knowledge, the logic of which was treated as analogous to that of necessity. A monograph by Ivan Boh as well as some further recent studies deal with various aspects of medieval developments and more will probably appear because this is an intriguing part of the history of logic.1 While the general formal approach to the study of the notions of knowledge and doubt was a medieval invention based on combining it with the principles of alethic modal logic, medieval authors found many texts in Aristotle which they regarded as relevant in this context. These were Topics VIII which influenced the medieval theory of disputation logic known as obligations logic, Posterioranalytics I.1 and Prioranalytics II.21, both of which deal with the famous question of whether the universal knowledge of the properties of triangles allows one to know the same about a singular triangle which one has not even seen and some related questions, and the Sophistici elenchi example of knowing Coriscus and knowing something about him without knowing that it is he. Apart from the more general approach in the Topics VIII, these texts pertained to the relationship between identificatory and nonidentificatory knowledge which was considered an important issue by medieval logicians. I will take a look at these texts, extensively studied in medieval epistemic logic, and consider two questions relevant to evaluating the nature of Aristotle’s epistemic logic and its relation to medieval developments. 1

Boh (1993); Boh (2000), Panaccio (2012), Knuuttila (2015).

86

S. KNUUTTILA

The first is the question of deductive closure with respect to knowledge or logical omniscience, as it is called in contemporary discussions, and the second is that of the relationship between universal and particular identificatory knowledge. Speaking about Aristotle’s epistemic logic is complicated by the fact that Aristotle did not develop any special theory of the logical properties of epistemic notions. Nevertheless, his insights on this topic can be gathered from the discussions in which he makes use of certain rules for the logical behaviour of these notions, thus offering the potential to see analogies with later theories in which such models are explicated. What I aim to do in this paper is to compare Aristotle’s formulations with some systematic later insights without anachronist constructions. The situation here is the same as with the inference rules for propositional modal logic which Aristotle employs without explicating in his syllogistic theory. (See An. pr.I.15.) 1. The Question of Epistemic Closure In 1963, Roderick Chisholm published a review of Hintikka’s book KnowledgeandBelief (1962) which introduced the theory of epistemic logic into the philosophical logic of the last century.2 Chisholm attended to some similarities and differences between Hintikka’s theory and the fourteenth-century views in Pseudo-Scotus’s treatise on Prioranalytics, arguing that the medieval author did not accept the deductive closure with respect to knowledge and regarded this as a view more reasonable than Hintikka’s thesis that one’s knowing that p implies that one knows the truths which are logically implied by p. Many others shared Chisholm’s scepticism about Hintikka’s position of logical omniscience, as it was called by Chisholm, as well as about the so-called KK -thesis implied by the logical omniscience that if Socrates knows that p, he knows that he knows that p.3 Let us reconsider Chisholm’s view of the absence of logical omniscience in medieval epistemic logic and the role of this principle in Aristotelian sources. In PriorAnalytics II.21, Aristotle takes it for granted that one cannot simultaneously have knowledge and ignorance about the same thing in

2 3

Chisholm (1963); Hintikka (1962). Knuuttila (2018).

EPISTEMIC LOGIC IN ARISTOTLE

87

the same respect. There were many fourteenth-century book chapters on epistemic logic with the title Descireetdubitare or to defend the same thesis.4 PriorAnalytics II.21 is part of the discussion of evaluating various dialectical arguments from the syllogistic point of view. Aristotle’s question is that since it is clearly impossible to know that p and not to know that p at the same time, why are people apparently holding such incompatible epistemic attitudes in syllogistic contexts? Explaining this phenomenon is the objective of Aristotle’s relative long chapter. Aristotle first formulates two syllogistic examples of cases which seem to undermine the plausibility of the principle that one cannot know and be ignorant about the same thing at the same time. The first one runs as follows: It is possible for the same thing to belong to several things primarily and for someone to fail to notice one of these and think the term belongs to none of it, but to know that it belongs to another one. For let A belong to B and to C according to themselves, and these likewise to every D. Now, if someone thinks that A belongs to every B and this to every D, but that A belongs to no C and this to every D, then he will have both knowledge and ignorance about the same thing in the same respect. (PriorAnalyticsII.21, 66b20-26).5 It is assumed that knowing the premises of a Barbara syllogism implies that one knows the conclusion as well. Knowing the premises allows one to know the conclusion even if one does not form a separate act of knowing the conclusion. Knowing the premises is sufficient for knowing the conclusion in a logical sense, so to say. Furthermore, it is proposed in the example that someone at the same time assumes the premises of a Celarent syllogism with a different middle term but with the same major and minor terms as in the first syllogism. Accepting these premises is taken to imply that one mistakenly accepts a conclusion which is incompatible with the first one, so that one knows that every A is D and believes that

4 See, for example, William Heytesbury, De scire et dubitare, ch. 2 of his Regulae solvendisophismata (Venice, Bonetus Locatellus, 1494); translated in William Heytesbury (1988); Paul of Venice, Logica Magna I, Tractatus de scire et dubitare, edited with an English translation by P. Clarke, Paul of Venice (1981). 5 All translations from Aristotle’s Prioranalytics are from Smith (1989).

88

S. KNUUTTILA

no A is D. Modifying the analytic table in Nico Strobach’s commentary, this can be described as follows:6 Truth

Epistemic attitude

1. AaB

1. K(AaB)

2. AeC

2. B(AeC)

3. BaD

3. K(BaD)

4. CaD

4. K(CaD)

[5. AaD]

[5. K(AaD), 6. B(AeD)]

In this paragraph, Aristotle seems to employ a deductive closure with respect to epistemic attitudes in attributing the knowledge of the conclusion of a Barbara syllogism to a subject who knows its premises. In the same way, he argues that if a person accepts the two premises of a Celarent syllogism, he or she can be said to accept a conclusion incompatible with that of the first syllogism. Accepting the Celarent premises while knowing the Barbara premises is impossible from the logical point of view, but is psychologically possible, as Aristotle explains below. According to Aristotle’s view of contradictory beliefs analysed in Metaphysics IV, the psychological possibility of error does not extend to consciously maintaining an inconsistent set of propositions. If there are examples which seem to imply such inconsistency, as in the opening part of PriorAnalytics II.21, a special explanation is needed. This is the objective of that chapter.7 Aristotle’s way of looking at knowledge and inference shows some similarity to the principle of epistemic closure or epistemic omniscience in Hintikka’s Knowledge and Belief, according to which the subject of the epistemic attitude of knowledge knows whatever logically follows from what he or she knows and, correspondingly, refutes what is logically incompatible with it. Hintikka stresses that these rules do not correspond to the everyday use of epistemic notions. They belong to the level of fully rational agents with the knowledge of the logical implications of what they know and what is incompatible with these.8 While 6 7 8

Strobach (2015), p. 459. See also the commentaries in Smith (1989), p. 213, and Strobach (2016), pp. 164-166. Hintikka (1962), pp. 19-21, 30-38.

EPISTEMIC LOGIC IN ARISTOTLE

89

some philosophers interpreted this as an idealized analysis, Hintikka preferred to characterize his approach as the conceptual analysis of the notion of knowledge taken in a strong sense.9 Aristotle seems to hold an analogous view of knowledge and syllogistic logic. In the text quoted above, he argues that it is not possible to think about the premises in the way described because the conclusions are mutually exclusive. Knowing the conclusion from the known syllogistic premises excludes the belief of any proposition the acceptance of which would imply that one believes the conclusion of a syllogism which denies that which one knows. (For Aristotle’s views of the relationship between knowledge, belief, and opinion and their veridicality, see section 2 below.) Aristotle’s second example of the violation of the principle that one cannot simultaneously have knowledge and ignorance about the same thing is simpler, the terms of the syllogisms belonging to the same series. One knows the premises of a Barbara syllogism and believes the contrary of the conclusion: Next, this may happen if someone should be in error about terms from the same series. For example, if A belongs to B, this belongs to C, and C to D, but he believes that A belongs to every B and, again, to no C, for he will at the same time both know that it belongs and believe it does not. Based on these premises, then, would he be doing anything but claiming not to believe that very thing which he knows? For in a way he knows that A belongs to C by means of B (that is, as we know the particular by the universal knowledge); consequently, what he knows in a way, that he also claims not to believe at all, which is impossible. (Prior AnalyticsII.21, 66b26-34.) This can also be illustrated by modifying a table in Strobach’s commentary:10

9 10

Truth

Epistemic attitude

1. AaB

1. K(AaB)

2. BaC

2. K(BaC)

3. [AaC]

[3. K(AaC), B(AeC)]

Hintikka (1969), Ch. 1. Strobach (2015), p. 473.

90

S. KNUUTTILA

In these two examples, the syllogistic conclusion is said to be known because of the knowledge of the premises, without the conclusion being formulated separately. Not stating the conclusion plays an important role in the rules for competitive dialectical games in the Topics: This would be best achieved if one were to establish by reasoning not only the necessary premises but also some of those which are requisite for obtaining them. Further, you should not state the conclusions, but establish them by reasoning at the same time at a later stage, for then you would keep the answerer as far as possible from the original proposition. To put the matter generally, he who wishes to conceal his purpose while eliciting answers should frame his questions in such a way that when the whole argument has been the subject of questions and he has stated the conclusion, it should still be asked ‘Why is it so?’11 Aristotle’s answer to the original question of how people might have knowledge and uncertainty about the same object refers to a difference between the modes in considering the relevant propositions. While it is impossible to consciously accept the premises with opposite conclusions in the first example, it is possible that one accepts them separately, without considering them together. Similarly, one may accept the opposite of a conclusion without combining it with the premises. Aristotle formulates the famous example that one knows that all mules are sterile and then sees a particular mule and believes that it is a mule and in foal ((67a3038). This is meant to shed light on an error resulting from not considering the relevant premises together, but it is associated with a difference between universal and particular knowledge which Aristotle explains, before coming to the mule example, by considering a case in which one knows that every triangle has interior angles equal to two right angles without knowing this about a particular triangle (67a11-21). I shall return to this example in the next section. Aristotle assumes in An.pr. II.21 that human agents do not accept two fully understood opposite propositions at the same time and that the conclusions from syllogistic premises which they know can be regarded as known by them. In a dialectical context, this means that the answerer ought to answer positively the question of whether he or she knows the proposition which in fact follows from the premises known by the 11

Top. VIII.1, 156a10-15.

EPISTEMIC LOGIC IN ARISTOTLE

91

answerer. When people do not answer in this way, they deviate from normative rationality. One reason for bad answers may be that the reasoning power of the intellect itself does not work properly, but this possibility is not discussed at all in An. pr. II.21. Aristotle is interested in other errors which he associates with the notion of suntheorein (contemplating together). In the first text quoted above, not contemplating together is something like operating with the premises of one syllogism in one diagram and those of another syllogism in another diagram. Suntheorein is more than simply having these configurations in mind. The error which follows from not considering the premises together typically derives from not attending to their mutual relevance. This neglect does not allow one to know and doubt the same at the same time but, say, doubt a proposition which one should grant and would grant if one paid attention to its being a consequent from known propositions. “Nothing prevents someone who knows both that A belongs to the whole of B and this to C from thinking that A does not belong to C … for he does not know that A belongs to C, if he does not simultaneously reflect on both premises” (67a33-37). Aristotle’s view of the error-free deductive behaviour of an epistemic agent can be characterized in terms of the closure of knowledge under syllogistic inference, which is a case of epistemic closure under deduction: (K1) Kφ & (φ →ψ) →Kψ. A related form is the principle of epistemic closure under a known implication (K2) Kφ & K(φ →ψ) →Kψ. Many contemporary authors have commented on these forms, most of them accepting K2 rather than K1 which represents Hintikka’s logical omniscience view, but there are also various arguments against K2.12 Aristotle does not distinguish between the syllogistic counterparts of these principles, but his discussion of the first and second examples in An. pr. II.21 is in agreement with a syllogistic version of K1. The examples consist of universal first figure syllogisms which he characterizes as perfect or complete in the sense that they need nothing else beside their 12

See the discussion in Hollyday (2015); Luper (2016).

92

S. KNUUTTILA

premises in order for the necessity of the conclusion relative to the premises to be evident (An. pr. 24b22-24). Aristotle’s perspective is on the acceptability of answers in a dialectical evaluation of consistency and the possibility of bad answers based on false beliefs. He says at the beginning of the chapter that we also fall into error in connection with the terms, possibly having in mind An. pr. I.33, but he does not address these errors here, concentrating on the acceptance of propositions in a syllogistic frame-work. The mistakes are not taken to derive from the ignorance of the formal syllogistic setting, the participants in the discussion being considered competent in this respect, but they do follow from their insufficient attention to the requirements of consistency. In the last paragraph of An. pr. II.21, Aristotle discusses the example of syllogisms in which the premises and the conclusion express identities and are organized syllogistically in accordance with the transitivity of identity (A=B, B=C, therefore A=C). He argues that one’s believing (hupolambanein) that the premises are true entails one’s believing the conclusion, thus applying the syllogistic closure to the notion of belief as well.13 I have translated Aristotle’s term “hupolepsis” as “belief” in An. pr. II.21; it could be also translated as “assumption”. Aristotle sometimes uses the term “doxa” in the same way, but a difference is that while knowing that pimplies the belief that p in the sense of hupolepsis, it does not necessarily imply the belief in the sense of doxa which could then be translated as “opinion”. In De anima III.3, 427b25 Aristotle says that “belief” (hupolepsis) is a genus term for the epistemic attitudes of knowledge and opinion (doxa in the narrow sense).14 Opinion is then analogous to the modal term of contingency (separated from necessity and impossibility). Belief is analogous to possibility proper (separated from impossibility). According to this terminology, a’s knowing that p implies that a believes that p and that p is true. Belief or opinion that p does not imply that p is true.15 Medieval commentators followed Aristotle in stressing the difference between knowledge and opinion as 13

See also Strobach (2015), p. 477. For the same use of terms, see An.post. I.33, 89a38. 15 In An. post. I.33 Aristotle argues that one cannot have epistēmē and doxa about something at the same time, using doxa in the sense of opinion. For Aristotle’s terminology, see Barnes (1993), pp. 198-202. For knowledge and belief in Aristotle’s epistemology, see also Miller (2013). 14

EPISTEMIC LOGIC IN ARISTOTLE

93

incompatible epistemic attitudes. In his influential commentary on PosteriorAnalytics, Robert Grosseteste writes that opinion, understood in a broad sense, is cognition with assent (cognitio cum assensu) and that opinion in this sense is a genus of knowledge and opinion in a more proper sense, that is, acceptance of one part of contradiction with a fear that the opposite might be true (cumtimorealterius). Knowledge implies opinion in the former sense but not the latter. 16 Thomas Aquinas draws a similar distinction.17 Later medieval authors often use the term credere for what Grosseteste calls opinion in the broad sense and opinari for what he calls opinion in a proper sense.18 Hintikka calls the knowledge in K1 the strong concept of knowledge, which also implies the KK -thesis that “if a knows that p, then aknows that a knows that p”. Aristotle often says that we are aware of our acts and also that we understand that we understand, but these remarks pertain to the psychology or awareness rather than the concept of knowledge in itself.19 However, he regarded the immediately evident Barbara syllogism as the main tool of scientific knowledge and his very demanding conditions for this kind of knowledge, developed in Post.An. I.2-6, suggest that they include the KK -thesis in the sense that one who has scientific knowledge also knows that he has this knowledge. While Aristotle did not explicitly say so, this is how Thomas Aquinas understood his position. Combining the arguments in Post. an. 71b9-16 and Post. an. 99b26-32, Aquinas writes: “The philosopher speaks about the habits of the intellective part of the soul which, if they are perfect, cannot be hidden from those having them, since their perfection implies certainty. Therefore anyonewhoknowsknowsthatheknows, since to know is to know the cause of something and that it is impossible for things to be otherwise.” Aristotle does not deal with perfection or certainty in the texts mentioned. These terms are part of Aquinas’s argument that the knower knows that what he knows cannot be otherwise, which he takes to mean that his knowledge is certain and he knows that he knows in this

16 Robert Grosseteste, CommentariusinPosteriorumAnalyticorumlibros, introduction and critical edition by P. Rossi (Florence: Olschki, 1981), p. 278. 17 Summatheologiae II-2.2.1. 18 See William Ockham,Summalogicae III-1.30 (436.20-21); Paul of Venice (1981), p. 152. 19 Martin (2007), pp. 94-96.

94

S. KNUUTTILA

way.20 Aristotle ends the second chapter of Posterioranalyticsby stating that “anyone who knows simpliciter must be incapable of being persuaded otherwise” (72b2-4). In Aquinas’s approach, this could be understood as regarding one’s knowledge as a perfect and certain attitude which one knows to be incompatible with doubt.21 William Ockham and other early fourteenth-century logicians who became interested in the logic of epistemic notions argued that knowing the antecedent of a good consequence does not imply knowing the consequent because the subject may either not know the consequence or not think about the consequent. They did not base their general consideration on K1, which they regarded as mistaken as a principle of epistemic propositions in the composite sense on the basis of psychological reasons and because they preferred a notion of knowledge weaker than Aristotle’s scientific knowledge in their logical treatises.22 Things changed later in the fourteenth century when logicians continued to criticize K1 but began to defend K2, which became the mainstream position in late medieval epistemic logic.23 Medieval authors were more aware of the distinction between K1 and K2 than Aristotle was, one reason for this being their much more developed theory of consequences.24 Ockham and others divided formal consequences into those with an intrinsic and extrinsic middle, the latter being logical principles independent of the meaning of terms and the former being various conceptual links between the terms.25 One might think that K1 would apply to former consequences which do not require special knowledge of the meaning of the terms and that K2 is required for other ones, but no such distinction was drawn in treatises on the logic of knowing. 20 Thomas Aquinas, Quaestiones disputatae I: De veritate 10.10, ad 5. See also Thomas Aquinas, InAristotelislibrosPosteriorumanalyticorumexpositio I.4.4; II.20.4; Martin (2007), pp. 103-104. 21 In Aquinas’s Latin translation “anapopeistos” (incapable of being persuaded otherwise) is rendered “non incredibilis” which changes the meaning of this phrase in his commentary. 22 See for example Ockham, Summalogicae II.29 (342); Pseudo-Scotus,Librumprimum Priorum analyticorum Aristotelis quaestiones, q. 36 (328-329); Boh (1993), pp. 46-61. In commenting on Aristotle’s view of demonstrative knowledge about necessary propositions, William Ockham wrote that the knowledge of the premises organized in a syllogistic way makes one know the conclusion (op. cit. III-2.1 (506). This was the common view of demonstrative syllogism (syllogismusfaciensscire), but it was regarded as a special case of knowledge and not the basis for discussing epistemic logic. 23 Boh (1993), pp. 96-100, 108-110. 24 See Read (2010). 25 Ockham, op. cit. III.3.1 (588).

EPISTEMIC LOGIC IN ARISTOTLE

95

The rule thirteen of Ralph Strode’s Tractatus de consequentiis reads: “If the antecedent is known, the consequent is known”, but the author explains that knowledge is closed under known consequence. According to Strode, a consequence is sound by form “if as facts are adequately signified by the antecedent is understood, then as they are adequately signified by the consequent is also understood.”26 This interpretation of a good consequence shows that if the antecedent is known and understood in the strong sense of knowledge which includes the understanding of the consequent, Strode’s original rule is sufficient. Adding the clause of known consequence is redundant in this case, but makes the rule applicable to weaker notions to knowledge such as “evident comprehension of truth”.27 There was another medieval branch of logic which made use of a principle analogous to K1 with respect to consistent acceptance of propositions. In a standard obligations disputation, the rules of which were influenced by Aristotle’s TopicsVIII, a false and contingent proposition was first put forward by the questioner and accepted by the respondent. The rules of consistency defined how other propositions proposed by the questioner were to be evaluated by the answerer who was supposed to say ‘I grant it’, ‘I deny it’ or ‘I don’t know’. While irrelevant propositions were evaluated as such, relevant propositions were treated in terms of their logical relations with respect to the acceptance of the original position and other previous answers so that the set of answers remained consistent. A principle analogous to K1 is used here with respect to the rules of propositional logic because the explicit norms for answering in a consistent way presupposed that the answerer follow the basic logic. Not knowing it is not an excuse in this exercise.28 In an early obligations treatise tentatively attributed to William Sherwood, the author refers to Aristotle’s An.pr. II.21 as an explanation for accepting seemingly inconsistent propositions. The shortcomings in suntheorein (pertractare) are not acceptances of contradictory propositions and may shed light on the impossibilities used in disputations with an impossible position.29 26

See Boh (1993), pp. 90-91, 97. For the types of knowledge in Ockham, see op. cit. III-2.1 (506). 28 For medieval obligations logic, see Yrjönsuuri (ed., 2001). The closure under consequence in obligations logic is particularly clear in the discussions of sophisms which are solved by applying principles of logic assumed to be known. See, e.g., Walter Burley (1988), pp. 399-404. 29 Deobligationibus, Anonymous (1963). 27

96

S. KNUUTTILA

2. Singular Propositions Aristotle’s general explanation of the errors with respect to syllogistic consistency refers to an inclination to think about the premises separately instead of combining them from the syllogistic point of view. While he later illustrates this habit with the example of the universal knowledge about mules and the incompatible mistaken belief about a singular mule (67a35-37), his main example of knowing and not-knowing something about the same object is different. It also includes a universal and particular premise, but these are not associated with any mistaken belief although Aristotle misleadingly introduces the example as an illustration of being deceived with respect to particular things. The question which opens this discussion is that if someone knows that all triangles have inner angles equal to two right angles, does he know this about a particular triangle which is not known to him: Therefore, if someone knows that A belongs to everything to which B belongs, then he also knows that it belongs to C. But nothing prevents him being ignorant that C exists. For example, if A is two right angles, B stands for triangle, and C stands for a perceptible triangle, someone could believe C not to exist, while knowing that every triangle has two right angles, and consequently, he will at the same time know and be ignorant of the same thing. For to know of every triangle that it has two angles equal to two right angles is not a simple matter, but rather one is in virtue of having universal knowledge, and another way is in virtue of having the particular knowledge. In this way, then, by means of the universal knowledge, he knows of C that it has two right answers, but he does not know it by means of the particular knowledge; consequently, he will not possess contrary states of knowledge. (PriorAnalyticsII.21, 67a11-21) Aristotle does not explain how this argument is related to the previous cases, but he apparently thinks that it is an illustration of knowing and not-knowing with respect to of the same thing without any mistake. While the previous examples involve universal knowledge, the present case includes universal and particular knowledge which are regarded as two different kinds of knowledge. The ‘not contemplating together’ argument is not applied as an explanation here; Aristotle later offers it as the main analysis of inconsistent

EPISTEMIC LOGIC IN ARISTOTLE

97

epistemic attitudes in a syllogistic context. It is not applicable to the present case because there is no error caused by an acratic non-attending to what one ought to know because of knowing other relevant propositions.30 There is nothing the agent ought to know about a particular actual triangle the existence of which is not known to him or her, and consequently the agent does not hold contrary beliefs about it. Knowing about all triangles in a universal way includes knowing about all particular triangles in a way which is compatible with not knowing about a definite triangle that it has two right angles or that it exists. Aristotle says that when a particular figure is first seen and immediately identified as a triangle, this is an instance of induction (67a23). Here the term “induction” (epagoge) stands for subsumption; it is not used in this way elsewhere.31 This remark is part of Aristotle’s comment on Plato’s Menodialogue to the effect that what is regarded as recognition by Plato could be understood as applying a geometrical truth to a figure as soon as it is understood to belong to its scope.32 According to An. post. I.24, 86a23-29, knowing about all triangles that they have two right angles includes knowing potentially that this holds of isosceles even when one does not know that they are triangles. Many later commentators applied this view to the argument in Post.an.I.1 and regarded universal knowledge as potential knowledge with respect to its particular instances.33 Knowing in a universal way that every A is B means that one immediately knows of any particular A that it is B on becoming acquainted with it. Knowing that a universal premise (Every A is B) is true implies knowing that the corresponding particular premise is true (Some A is B).The truth of these forms implies that there are A-beings. Knowing about particulars and their existence in this generic way is implied by knowing the truth of the corresponding universal

30 Many commentators refer to the discussion of acrasy in Nicomacheanethics VII.3 as showing similarities with the argument in An.pr. II.21. 31 In An.post. I.1, 71a21 Aristotle uses the verb “epagein” in an analogous example. See the discussion in Strobach (2015), pp. 468-469. 32 67a21-26; see also An.post. I.1, 71a29-30. According to Aristotle, when one knows that every triangle has two right angles and then sees that a figure is a triangle, he knows at the same time (euthus, 67a24-25) that the figure has two right angles. This is in agreement with the Aristotelian epistemic closure. 33 See, for example, Aquinas, InAristotelislibrosPosteriorumanalyticorumexpositio I.3.6; I.38.7.

98

S. KNUUTTILA

proposition, but none of these implies knowledge about a particular A by means of peculiar knowledge (oikeiaepistēmē, 67a27). In commenting on Aristotle’s discussion of the triangle example in An. post. I.1, Paul of Venice repeats that the major premise is known earlier than the conclusion, whereas the conclusion is known at the same time as the minor premise is first known together with the major premise. The knowledge of a universal premise can be said to involve the knowledge of its particular instances confusedly but not distinctly. To know an individual object distinctly is to know it through its proper concept. To know an individual object confusedly is to know it through a universal concept, for example, to know Socrates through the concept of man or animal. This was traditional terminology also found in Giles of Rome and others. According to Paul, this remark holds of universal knowledge propositions read in the compound sense but not in the divided sense. This was a common fourteenth-century division. In medieval logic, modalities were divided into compounded and divided ones, sometimes also called modalities de dicto and de re. In a compounded proposition the modal term usually precedes the quantification term; in a divided proposition the order is other way round. Many authors considered this an illuminating interpretation of Aristotle’s concern in An.post. I.1.34 Aristotle summarizes this discussion on An. pr. II.21 by separating various ways of knowing about a particular thing and simultaneously not knowing about it and how these are compatible: For “to know’ can be used in three ways: as knowing by means of universal knowledge, knowing by means of the peculiar knowledge of something, or by means of exercising knowledge, and consequently “to be in error” is also used in the same number of ways. Nothing then prevents someone both knowing and being in error about the same thing (although not contrarily), which is also what happens to the man who knows a premise according to each kind of knowledge and has not previously examined them: for in believing that the female mule is pregnant, he does not have knowledge in the sense of exercising it, nor indeed does he have the error contrary to the knowledge as a result of his belief (for the error contrary to universal knowledge is a deduction.) (PriorAnalyticsII.21, 67b3-11)

34

See Knuuttila (2015), pp. 188-192.

EPISTEMIC LOGIC IN ARISTOTLE

99

Knowing universally about a particular being does not exclude the possibility of not knowing the same by peculiar knowledge, whether in the case that the particular thing is not known at all to the person who knows about it in a universal way (this being not a mistake) or when it is known to be one of those objects to which the universal knowledge applies but this knowledge is not combined with the content of universal knowledge (this is the mistake of not considering the premises together). The third use of “to know” is said to refer to exercising knowledge. This division seems to imply that one can know in the first and second sense by exercising knowledge or by having knowledge without exercising it. Knowing about a particular thing in a universal manner is potential knowledge before it is exercised in an inference with a singular premise about that instantiation.35 According to Aristotle, to know that every human is an animal is to know that the species of humans is not empty and that all instantiations are animals. This knowledge is about individuals only accidentally because it does not include any kind of individual identification. It could be called identificatory in the sense that the species of humans is identified as a subspecies of animals. The peculiar knowledge about a singular object can be identificatory in different ways. The first is that the singular thing is identified as a member of a species. On the basis of Aristotle’s special use of the term “induction” in An.pr. II.21, this could be called identificatory induction. Aristotle exemplifies this by knowing that something seen is a mule. Knowing this with the universal knowledge that all mules are sterile makes one virtually know that the animal is sterile, but in Aristotle’s example the person does not identify this animal as sterile because he does not combine the premises. These are the forms of identificatory knowledge about singular objects included in the summary in An. pr. II.21. None of these represent perspectival identification which is included in ‘knowing who’ contexts, but Aristotle has an influential example of this knowledge in OnSophistical Refutations 24: You know Coriscus and Coriscus is approaching; therefore, you know who is approaching.36 This argument was often discussed by medieval logicians when dealing with the failure of the substitutivity 35 36

Cf.An.post. I.24, 86a23-29. See also the discussion in Charles (2000), pp. 95-100.

100

S. KNUUTTILA

of identicals which was one of the main questions in fourteenth-century epistemic logic. It was taught that the compounded and divided readings of epistemic propositions about singular things referred to by demonstrative pronouns were interchangeable, but corresponding propositions with proper names did not behave in the same way. Take a look at some typical examples for valid inferences from divided premises to divided conclusions with proper names and their non-valid counterparts with compounded conclusions are: “This I know to be running, and this is Socrates; therefore, Socrates I know to be running”; “Of this I am uncertain that he is running, and this is Socrates; therefore, of Socrates I am uncertain that he is running”; “This you know to be true, and this is A; therefore, of A you know that it is true”. To conclude from these premises a proposition in a compounded sense is a mistaken form of arguing.37 It was generally assumed that one cannot replace a proper name in a compounded singular knowledge construction about something referred to by ‘this’ because the subject may not know the name of that for which ‘this’ stands. Corresponding divided propositions could be regarded as non-identificatory and compatible with such replacement. This is a much more elaborated theory than what is found in Aristotle, particularly the analysis of the role of demonstrative pronoun in singular epistemic propositions, but it has its origin in Aristotle’s logic and is in a very Aristotelian vein. LITERATURE Anonymous (1963), De obligationibus, ed. by Romuald Green in The Logical Treatise“Deobligationibus”:AnIntroductionwithCriticalTextsofWilliam ofSherwood(?)andWalterBurley. Ph. D. diss., University of Louvain. ARISTOTLE (1991), Analyticaprioraetposteriora, ed. W.D. Ross, Oxford: Clarendon Press. ARISTOTLE (1958), TopicaetSophisticielenchi, ed. W.D. Ross, Oxford: Clarendon Press. ARISTOTLE (1960), Topics, transl. E.S. Forster, (The Loeb Classical Library), Cambridge, Mass.: Harvard university Press. BARNES, Jonathan (1993), Aristotle: PosteriorAnalytics, second edition, Oxford: Clarendon Press. Boh, Ivan (1993), EpistemicLogicintheLaterMiddleAges, London: Routledge. 37

Johannes Venator Anglicus (1999),Logica,p. 419.

EPISTEMIC LOGIC IN ARISTOTLE

101

BOH, Ivan (2000) “Four Phases of Medieval Epistemic Logic” in Theoria 66 (2000), pp. 129-49. CHARLES, David (2000), AristotleonMeaningandEssence, Oxford: Clarendon Press. CHISHOLM, Roderick M. (1963), “The Logic of Knowing” in The Journal of Philosophy 60, pp. 773-795. HINTIKKA, Jaakko (1962), KnowledgeandBelief:AnIntroductiontotheLogic oftheTwoNotions, Ithaca, NY: Cornell University Press. HINTIKKA, Jaakko (1969), ModelsforModalities, Dordrecht: Reidel. FORSTER, E. S. (1960), Aristotle, Topics, (The Loeb Classical Library), Cambridge, Mass.: Harvard university Press. HOLLYDAY, W.H. (2015), “Epistemic Closure and Epistemic Logic I: Relevant Alternatives and Subjunctivism” in Journal of Philosophical Logic 44, pp. 1-62. JOHANNES Venator Anglicus (1999), Logica, ed. L.M. de Rijk, Stuttgart-Bad Cannstatt: Frommann-Holzboog. KNUUTTILA, Simo (2015) “Epistemic Logic in Paul of Venice’s Commentary on the Beginning of the Posterior Analytics” in J. Biard (ed.) Raison et démonstration Le commentaires médiévaux sur les Seconds Analytiques, Turnhout: Brepols, pp. 185-198. KNUUTTILA, Simo (2018), “Questions of Epistemic Logic in Hintikka” in H. van Ditmarsch & G. Sandu (eds.) Jaakko Hintikka on Knowledge and GameTheoreticalSemantics, Dordrecht: Springer, pp. 413-431. LUPER, S. (2016), “Epistemic Closure” in StanfordEncyclopediaofPhilosophy MARTIN, Ch. (2007) “Self-Knowledge and Cognitive Ascent: Thomas Aquinas and Peter Olivi on the KK-thesis” in H. Lagerlund (ed.) FormingtheMind. EssaysontheInternalSensesandtheMind/BodyProblemfromAvicenna toMedicalEnlightenment, Dordrecht: Springer, pp. 93-108. MILLER, F. D. (2013), “Aristotle on Belief and Knowledge” in G. Anagnostopoulos and F. D. Miller (eds.) ReasonandAnalysisinAncientGreekPhilosophy:EssaysinHonorofDavidKeyt, Dordrecht: Springer, pp. 285-307. PANACCIO, Claude (2012), “Ockham and Buridan on Epistemic Sentences: Appellation of the Form and Appellation of Reason” in Vivarium50 (2012), pp. 139-160. PAUL OF VENICE (1981), LogicaMagna I, Tractatusdescireetdubitare (edited with an English translation by P. Clarke), Oxford: Oxford University Press for The British Academy. PSEUDO-SCOTUS (1649),LibrumprimumPriorumanalyticorumAristotelisquaestiones in Johannes Duns Scotus, Opera omnia, ed. L. Wadding, Lyons, I, pp. 273-341. READ, S. (2010), “Inferences” in The Cambridge History of Later Medieval PHILOSOPHY I (ed. R. Pasnau), Cambridge: Cambridge University Press, pp. 173-184. ROBERT Grosseteste (1981), CommentariusinPosteriorumAnalyticorumlibros, introduction and critical edition by P. Rossi, Florence: Olschki.

102

S. KNUUTTILA

SMITH, R. (1989), Aristotle:Prioranalytics, trans. with introduction, notes and commentary, Indianapolis: Hackett. STROBACH, Niko (2015) Analytica priora Buch II, (transl. by N. Strobach and M. Malink, comm. N. Strobach), Aristoteles Werke in deutscher Übersetzung, Berlin: de Gruyter. Thomas AQUINAS (1964), InAristotelislibrosPosteriorumanalyticorumexpositio, Turin: Marietti. Thomas AQUINAS (1964b), QuaestionesdisputataeI:Deveritate, ed. R. Spiazzi, Turin: Marietti. Thomas AQUINAS (1948-1950), Summa theologiae, ed. P. Caramello, Turin: Marietti. Walter BURLEY (1988), Obligations, translated in The Cambridge Translations of Medieval Philosophical Texts, Volume One: Logic and the Philosophy of Language (ed. N. Kretzmann & E. Stump), Cambridge: Cambridge University Press, 1988), pp. 369-412. William HEYTESBURY (1988), Descireetdubitare, ch. 2 of his Regulaesolvendi sophismata (Venice, Bonetus Locatellus, 1494); translated in The CambridgeTranslationsofMedievalPhilosophicalTexts,VolumeOne:Logic and the Philosophy of Language (ed. N. Kretzmann & E. Stump), Cambridge: Cambridge University Press, 1988, pp. 436-79. William OCKHAM (1974), Summa logicae, ed. Ph. Boehner, G. Gàl and S. Brown, Guillelmi de Ockham Opera philosophica, 1, St Bonaventure, NY: St Bonaventure University. YRJÖNSUURI, Mikko (ed., 2001), MedievalFormalLogic:Obligations,Insolubles andConsequences, (The New Synthese Historical Library 49), Dordrecht: Kluwer.

REASON, EXPERIENCE, AND THE KNOWLEDGE OF THE PRINCIPLES IN ARISTOTLE’S POSTERIORANALYTICS2.19 Miira TUOMINEN

1. Introduction 1.1 EmpiricismcontraRationalism? Aristotle’s puzzling, even frustrating, account in Posterior Analytics 2.19 of how principles (ἀρχαί) of scientific demonstrations become known has caused much debate and disagreement among scholars. One disputed question concerns the respective roles of reason and perceptual experience in the acquisition of knowledge of the principles. While according to Aristotle the principles are known by a hexis1 of reason called nous,2 he also makes clear that the knowledge of the principles comes about from experience (100a6-8). The introduction of nous as the state that knows the principles has led some scholars to suggest that 1 In 100b5-6, nous is characterized as a hexis of reason (περὶ… τὴν διάνοιαν ἕξις). Hexis is often translated as ‘state’ in the context; see, e.g., Bronstein (2016), p. 226, Perelmuter (2010), p. 229, Barnes (1984), while others use ‘disposition’; see Gregoric and Grgic (2006); Modrak (1987), p. 173. This might sound like intuitionism and the translation of hexis as ‘state’ is perhaps used to avoid such connotations. However, someability or cognitive disposition must be assumed. If nous is the name of the state of knowing the principles, we must also be able to know them. Recognizing this, however, does not introduce an element of intuitionism to Aristotle’s theory. 2 I leave nous untranslated since its Greek name is relatively well known. Aristotle also calls the knowledge of principles ‘non-demonstrative’ (ἐπιστήμη ἀναπόδεικτος).Nous is usually identified with non-demonstrative knowledge on basis of Aristotle’s account in Posterior Analytics 1.33 (88b35-37) but some scholars have suggested that we should distinguish between the two. For the alternative reading of the lines, see Lesher (1973), p. 54-55; Harari (2004), p. 18; and Perelmuter (2010). I shall return to the issue in the body text.

104

M. TUOMINEN

Aristotle posits a form of rational intuition into the principles of scientific demonstrations.3 Although recently scholars have for the most part opposed such a suggestion, it still to an extent haunts the discussion. This entails that ascribing a positive role to reason in knowledge acquisition might be taken to endanger the empirical spirit of Aristotle’s methodology and his own inquiries. According to Jonathan Barnes, the opposition of what he calls “easy rationalism” and “honest empiricism”4 is especially pressing for the last chapter of the PosteriorAnalytics because the chapter concerns perceptual experience on the one hand, and seems to posit a form of rational ‘intuition’ on the other. This, he claims, makes the chapter Janus-faced.5 An important contribution to the discussion has been made by Pavel Gregoric and Filip Grgic, who argue that, instead of seeing experience and reason as opposed in Aristotle’s account of human cognitive capacities, reasoning enhances what we can learn from experience.6 In this essay, I shall argue along similar lines that a straightforward opposition between rationalism and empiricism is misguided with respect to Aristotle’s view of how human cognitive capacities function in the acquisition of scientific knowledge (ἐπιστήμη), especially of its principles. Rather than being diametrically opposed to each other, perceptual experience and the methods of making observations work together with the rational capacities of human soul. However, the rational capacities should not be identified with intuitive insight into the natures of things, 3 E.g. Irwin (1988), p. 134. Frede has underlined Aristotle’s rationalism (1996) while arguing against intuition of the principles (see, e.g., p. 167). For nous as intuition, see “intellectual intuition” (Grote 1880, p. 260) and “intuitive reason” (Lee 1935; Ross 1949, p. 675) also discussed in Baltussen (2007). For arguments against ascribing such intuition to Aristotle; see, e.g., Bronstein (2016); Charles (2000, p. 148 n3), who says he is “seeking to make it non-mysterious how a concept can be acquired”. Hankinson is also committed to demystifying the Aristotelian nous yet allows for the translation intuition “both for the faculty of nous and of its characteristic activity, as long as we hold fast to the quasi-perceptual force of the literal background to the English notion”, Hankinson (2011), p. 52. For nous as a faculty of discovery, see Kahn (1981). I shall avoid using the term ‘intuition’ because there does not seem to be a well-defined meaning for it and it typically implies a form of cognition of which we do not quite know how we acquire it. I also think that we should not speak of a distinct faculty of nous in the context of the PosteriorAnalytics. 4 Barnes (1993), p. 259. Hankinson detects in the debate traces of a long-standing battle of “decent British empiricism” against “some suspect continental power of intuition” (2011), p. 42. 5 Barnes himself is not committed to translating nous as ‘intuition’ but mentions it as a customary translation; Barnes (1993), p. 259. 6 Gregoric and Grgic (2006).

REASON, EXPERIENCE …

105

or an intellectual vision that guarantees the veridicality of inductive generalizations. Rather, such capacities are ordinary capacities of reason to process information on a general level: forming or acquiring general concepts - that Aristotle calls ‘universals’ in Posterior Analytics 2.197 - from experience and an ability to operate with propositions and making inferences. This means that, in PosteriorAnalytics 2.19, perception and experience on the one hand and reason on the other are not, as it were, pulling in different directions. Rather, they work together. We can acquire generally valid knowledge from experience because we have reason. I shall argue that the chapter presents in a concise and somewhat cryptic manner an outline of the cognitive conditions for more advanced knowledge acquisition. Aristotle gives an account of how we acquire such cognitive states through the capacities of perception, retention, experience,8 and reason, capacities that are not pre-existent in the soul and yet allow us to inquire into more detail into the natures and explanations of phenomena. This means that his account is not one of how we attain the first principles of scientific demonstrations. For such inquiry, more specific methods are needed, and Aristotle does not talk about them in PosteriorAnalytics 2.19. Therefore, contrary to more traditional views and my own earlier view,9 I follow David Bronstein’s recent monograph AristotleonKnowledgeand 7 Although many scholars have argued for the view that universals should be understood as propositions in PosteriorAnalytics 2.19, the fact remains that Aristotle’s examples are universal concepts such as ‘human being’ and ‘animal’ rather than propositional ones (“all human beings are animals”, “all phlegmatic patients are cured by leeches”). However, to say that the universals are concepts does not mean that they would merely exist in the soul. Quite the contrary. Aristotle says that they are in the instances, one and the same in all of them (100a7-8). There are of course different readings of what these words mean, and I shall return to them in the body text and the notes to the relevant lines. Modrak talks about “rudimentary universals of experience” as opposed to “universals of art and science” and argues that the difference “in scope and character” of the two kinds of universals is parallel to the one of linguistic meaning and scientific definition (2001), p. 98. I need to leave questions about language and definition outside the scope of this essay. However, I do not think that we should distinguish between different kinds of universals. It is true that, at the beginning, we do not for instance know the definitions for the species of which we have a concept, but I do not think the concept changes into another one when the definition becomes known. 8 In PosteriorAnalytics 2.19 Aristotle does not use the term, but he probably assumes that the name of the capacity of having experience is phantasia. 9 See also Lesher (1973). Frede (1996, p. 169), identifies reason (λόγος, 100a2) with nous, while Gregoric and Grgic talk about “noetic rationality”, Gregoric and Grgic (2006), p. 23. I shall return to the details of how my view differs from the earlier accounts in the body text.

106

M. TUOMINEN

Learning: The Posterior Analytics (2016) on recognizing that Aristotle does not call our capacity to acquire universal notions nous in Posterior Analytics 2.19. In that chapter, Aristotle only uses the term for the hexis that knows the first principles of demonstrations. There is no uncontroversial evidence in the whole Posterior Analytics for nous of universal notions. In the context of the last chapter, when speaking about our cognitive development that starts from perceptual experience, Aristotle talks about reason (λόγος, 100a2) and does not explicitly ascribe the acquisition of universal notions to nous. I also agree with Bronstein that PosteriorAnalytics 2.19 is not a methodological account of how we acquire knowledge of the first principles of scientific demonstrations. Therefore, the claim that our reason is capable of acquiring or coming to have universal notions from experience is not meant to replace the empirical research or compromise the methods of division and the inquiry into causes in terms of articulating the essential attributes of the objects of study that Aristotle introduces earlier in book 2. Therefore, I agree with Bronstein that Aristotle is more focused on the question of how we can begin to acquire knowledge of the principles, although as I shall explain in sections 2 and 3 below, I disagree with him on some details of the first part of Aristotle’s positive contribution in the chapter (99b32-100b5). I shall focus on that section (99b32-100b5) of PosteriorAnalytics 2.19 and, first, outline Bronstein’s reading and then explain why I disagree on the mentioned details. While Bronstein explains the progression from experience to universals in terms of inductive inference, filling in the details from the parallel passage in Metaphysics 1.1, I argue that such a reading relies too much on the parallel passage. The account in Posterior Analytics 2.19 does not refer to propositional generalizations but rather to general notions that Aristotle calls ‘universals’. This is important because, as I argue, PosteriorAnalytics 2.19 focuses on explaining the cognitive conditions on which we can start more systematic inquiry into things and more rudimentary knowledge acquisition through inference. While the passage in Metaphysics 1.1 gives more weight to the differences between theoretical knowledge and experience, PosteriorAnalytics is focused on how we can come to have knowledge from experience at all — and concerning this I agree with Bronstein. Since inductive inference requires subsuming individuals under general notions, coming to have such notions is a condition for inductive inference but not viceversa. Therefore, while

REASON, EXPERIENCE …

107

especially in adult human beings, it is perhaps almost impossible to identify a state of having a universal notion without any propositional cognition about it, the formation of notions seems to be prior to propositional universals in the sense just explained. My reading is based on the examples Aristotle gives in Posterior Analytics 2.19, and I also argue that it fits better with the details of the rest of his account in the chapter. Since the relevant cognitive condition of us being able to acquire knowledge from experience is our reason’s capacity to operate with general notions, recognizing this does not introduce an element of mysterious intuition into the account.10 Rather, the specific rational capacity is one of forming universal notions from perceptual experience through the kind of process described in PosteriorAnalytics 2.19, section 99b32-100a3.11 Some scholars talk about concept acquisition, and I suppose this is apt. However, in the context Aristotle talks about universals (τὸ καθόλου) not about concepts. Moreover, to have general or universal notions in Aristotle’s sense is not mere a priori cognition about one’s conceptual structure. Rather, it is a cognitive relation with the world. This also strengthens the point that reason and experience are not two opposing poles in Posterior Analytics 2.19 but, rather, reason provides us with another way of processing the material acquired from experience. 1.2 BrieflyonPosterior Analytics2.19inItsContext Despite the many debates about PosteriorAnalytics 2.19, there is general agreement that Aristotle’s positive account therein is structured by the two questions he poses at the beginning of the chapter. The questions concern how the principles (ἀρχαί) become known and what the cognitive hexis that knows them is. After the initial discussion of Meno’s Paradox and related difficulties, which I shall not discuss here, the rest of the chapter is divided into two parts with respect to the two questions. While the details 10 Baltussen says about intuition that “we seem to know intuitively what the word ‘intuition’ means” (2007), p. 56. This would mean that if we say we have an intuition about something, we think we know it, but we cannot give further justification for how we do. This, I argue, is not how the specific function of reason works in Aristotle’s account. 11 Bronstein also claims that “the universal is encoded in the representations we receive when we perceive particulars” (2016, p. 245). I agree that we receive the universals in perceptual experience. However, I argue that the perceptual faculty even if aided by memory, imagination and experience could not “decode” and thus grasp the universal notion. This remains a function of reason.

108

M. TUOMINEN

of his answer to the first question are difficult to decipher, his answer to the second question is clear. It is nous that knows the principles. However, this raises the question of what exactly it is that the nous knows and how the second part of the chapter is related to the first one, i.e., what role, if any, nous plays in the acquisition of the knowledge of the principles. In his monograph, Bronstein offers a comprehensive view of how Aristotle solves Meno’s Paradox in the treatise and what implications this solution has for the account of the knowledge of the principles.12 I have followed Bronstein’s strategy of allowing a rather wide range of meanings for the term ‘knowledge’. The general term that Aristotle uses for various kinds of epistemic states that can, with some accuracy, be translated as ‘knowledge’ is gnōsis that can also be translated as ‘cognition’.13 Such knowledge, if we choose to call it so, can be acquired, for example, simply by perceiving things, but the term can also refer to the knowledge of the principles of scientific demonstrations (as in An.post. 2.19, 99b22). The term epistēmē is usually restricted to a type of knowledge that can be called ‘scientific’, i.e., knowledge that requires demonstrations.14 In PosteriorAnalytics 1.2, Aristotle sets out to describe knowledge without qualification15 that requires knowing: (i) the cause of the thing or fact (πρᾶγμα), (ii) knowing that it is its cause, and (iii) knowing that the thing or fact cannot be otherwise (71b9-12). This indicates that knowledge 12 For Aristotle on Meno’s Paradox, see also Fine (2014), pp. 179-225; Charles (2010b); Adamson (2011). 13 Bronstein argues (2016), p. 17 n26) for the translation of gnōsis as ‘knowledge’ by saying that while cognition can be false, gnōsis only refers to true cognition. He opposes Fine’s suggestion (2010a), pp. 148-152, that there are instances of false gnōsis in Aristotle. I agree that the passages to which Fine refers do not seem to provide us with evidence for false gnōsis, but it is also the case to which Aristotle never says that gnōsis would always be true. There are other questions related to taking gnōsis as being knowledge, however, especially since it is ascribed to animals as well. In An.post. 2.19 (99b38-39) Aristotle says that animals without memory do not have gnōsis outside perception, which implies that even those animals that merely have perception have perceptual gnōsis. Bronstein takes such gnōsis to be knowledge as well (2016), p. 235. However, animal gnōsis raises the question of whether such knowledge can be analysed as justified true belief. Concerning Aristotle’s insistence that animals do not have opinions (doxai) because they lack conviction (pistis); see De an. 3.3, 428a20-22; Butler (2003), pp. 333-337. The notion of justification at least would need to be causal because animals lack the capacity to argue or reason. In Butler’s terms, animals have classificatory states that are “not even weakly reasons-dependent” (ibid., p. 337). I wish to address this question in more detail in future study. 14 I shall use the common translation ‘demonstration’ for the Greek ἀπόδειξις here, while in my earlier work I have used ‘proof’ (Tuominen, 2007; 2010) as McKirahan also does (1992). Burnyeat, however, has suggested that epistēmē should rather be translated as ‘understanding’. For one recent discussion, see Salmieri (2013). 15 The Greek for which is ἐπιστήμη ἁπλῶς.

REASON, EXPERIENCE …

109

without qualification is the knowledge of things or facts that are expressed in the conclusions of scientific demonstrations, the premises of which express their causes and/or essences. The principles or premises of demonstrations, by contrast, must be known in another way to avoid infinite regress or circular demonstration (An.Post. 1.3). Aristotle argues that not all knowledge is demonstrative. There must be a starting point (ἀρχή) for demonstrative knowledge, called “non-demonstrative knowledge” or nous.16 For example, in order to know a lunar eclipse in the unqualified way, one needs to know its explanation expressed in a demonstration. Eclipse belongs to all screening [of the sun’s light] by the earth. Screening [of the sun’s light] by the earth belongs to the moon. Eclipse belongs to the moon.17

The conclusion is thus demonstrated and knowledge about it demonstrative, whereas the cause (the screening of the sun’s light) is indemonstrable and the knowledge about it non-demonstrative. 2. What is Aristotle’s First Question about in Posterior Analytics 2.19? 2.1 Aristotle’sTwoQuestions After indicating that earlier in the PosteriorAnalytics he has discussed scientific demonstrations and how they come about, Aristotle goes on to state what he is going to do in the final chapter: (T1)

About principles: how they become known and what the cognitive hexis is that knows them will become clear to one who has first considered some difficulties (99b17-19).18

16 However, as mentioned, the identification between non-demonstrative knowledge (ἐπιστήμη ἀναπόδεικτος) and nous has been challenged by some scholars. I shall return to the question in section 4.3 below. 17 See also Bronstein (2016:), p. 161; Tuominen (2007), p. 86; Goldin (1996), p. 119. I shall not be concerned with the question of how exactly one should articulate Aristotle’s condition that the premises of demonstrations are necessary. See Malink (2013) for the distinction between “A necessarily belongs to all B” and “necessarily (A belongs to all B)”. He argues for the latter; for comments on Malink, see e.g. Leunissen (2015). A rather different view has been suggested by Angioni (2014), who maintains that the principles are necessary for explaining the fact expressed in the conclusion, which seems to slightly understate the significance of the necessity condition. 18 περὶ δὲ τῶν ἀρχῶν, πῶς τε γίνονται γνώριμοι καὶ τίς ἡ γνωρίζουσα ἕξις, ἐντεῦθεν ἔσται δῆλον προαπορήσασι πρῶτον.

110

M. TUOMINEN

As indicated above, Aristotle poses two questions: (Q1) (Q2)

How do the principles become known?19 What is the hexis that knows them?20

A traditional interpretation of the first question has been that it concerns methodology: (Q1a): [What is the] process or method by which we gain knowledge of first principles?21

The traditional answer to this is: induction (A1a). By contrast, Bronstein argues that Aristotle’s project in 2.19 is more modest and that the chapter “aims to provide a much smaller part of the answer [to Q1] than commentators usually think”.22 In Bronstein’s view, instead of explaining in a methodological sense (Q1a) how the first principles of demonstration become known, Aristotle is targeting another question: (Q1b): [What is the] original prior knowledge from which first principles become known?

Therefore, when Aristotle asks how the principles become known, Bronstein argues that the question of how (πῶς, 99b18) means “from what origin?”.23 The answer is, of course, “perception” (A2b).24 A crucial element in Bronstein’s account of 2.19 is the claim that we should not conflate Aristotle’s answers to the two questions (Q1 and Q2).25 As regards Q1b, Bronstein claims that Aristotle explains in An. post. 2.19 how what he calls “preliminary accounts” on the one hand and non-noetic knowledge of the definitions of the genera on the other are acquired by induction from experience.26 Therefore, despite formulating Q1 as concerning the origin of our knowledge of the principles, Bronstein is in partial agreement with the traditional answer (induction). This implies that, according to Bronstein, Aristotle assumes the traditional 19 20 21 22 23 24 25 26

περὶ δὲ τῶν ἀρχῶν, πῶς τε γίνονται γνώριμοι. τίς ἡ γνωρίζουσα ἕξις. Barnes (1993), p. 268. Bronstein (2016), pp. 225; 227. Bronstein (2016), p. 230. Bronstein (2016), p. 227. See also Hankinson (2011), p. 34; Frede (1996), p. 169. Bronstein (2016), p. 227.

REASON, EXPERIENCE …

111

question (Q1a) about preliminary accounts and non-noetic knowledge of the definitions of the genera, but not about the principles of demonstrations. For coming to know the first principles of demonstrations, Bronstein argues, induction is necessary, but not sufficient.27 If we take induction in the sense that Bronstein talks about it, I agree with him on this. However, it is not clear whether we should understand induction in that way, and I shall return to that question in section 4 below. Bronstein also stresses that the inductive explanation of the acquisition of the preliminary accounts and non-noetic knowledge of the definitions of genera are distinct from Aristotle’s methodological account of how we acquire the principles of scientific demonstrations. That methodological account is to be found in the earlier chapters of book 2. Bronstein insists that although nous is the state that knows the principles, it does not in any way explain theacquisition of that knowledge. It merely knowsthe principles when we have discovered them through division28 and other methods. Although I do not entirely agree with the details of Bronstein’s account of how Aristotelian scientific inquiry proceeds,29 I agree that Aristotle’s first question (Q1) is not about methodology. The methodological tools that one needs to employ to acquire the knowledge of first principles are given elsewhere (mainly in book 2). Bronstein’s strategy for avoiding intuitionism thus is the detachment of nous from Aristotle’s account of the acquisition of the first principles. In scientific inquiry, one must gather essential attributes of the thing one studies, be it a species like horse or cow or a natural phenomenon like lunar eclipse. Since nous does not take part in these operations, there is no danger of ascribing a rational intuition into the first principles to Aristotle. When, through empirical inquiry, we come to know all the essential attributes of the object of study we, in Bronstein’s view, have non-noetic knowledge about them. For instance, we can know non-noetically that earth’s screening of the sun’s light is an essential attribute of the loss of light that belongs to the moon without knowing it as the cause of the eclipse. We have nous or noetic knowledge about it only when we know earth’s screening as the cause of the loss of light that belongs to the moon that constitutes a lunar eclipse (A–C “loss of light belongs to the 27

See especially Bronstein (2016), p. 244 n56. For division as a preparatory procedure for finding the definition that is not sufficient for producing them, see Pellegrin (2010), p. 141. 29 For the present, I shall not be concerned with an argument about methodology. 28

112

M. TUOMINEN

moon” because of B “screening of the sun by the earth”). In the case of what Bronstein calls simple essences, one only has noetic knowledge after having “defined by division the undivided species of a genus G and explained (by demonstration) the in itself accidents of G”.30 Therefore, in Bronstein’s view noetic knowledge of definitions is holistic to a considerable degree.31 2.2 TwoKindsofPrinciples To discuss the role of reason in the acquisition of the knowledge of principles, we need to distinguish between two kinds of principles — a distinction that Bronstein also accepts although he phrases it in slightly different terms. As I have argued in previous works as well,32 ‘principles’ for Aristotle can mean: (FP1) starting points for inquiry and (FP2) first principles of scientific demonstrations.

This distinction is relatively uncontroversial. Any account that takes seriously Aristotle’s distinction between what is better known to us and what is better known in nature (Posterior Analytics 1.2, 71b33-72a5)33 must grant it. With this distinction in mind, let us go back to Aristotle’s two questions at the beginning of 2.19. (Q1) (Q2)

How do the principles become known? What is the hexis that knows them?

Now we need to ask which principles (FP1 or FP2) Aristotle is talking about. Let us begin with Q2. Given the earlier brief discussions on the matter in the treatise (especially in 1.3 and 1.33), we have come to expect that the cognitive hexis we are looking for is the one that knows the 30

Bronstein (2016), p. 222. I tend to disagree that it would be quite as holistic as that, but I shall not discuss this question here. Some scholars have also connected Aristotle’s quest for definitions to the practices of the Topics and books 7 and 8 of the Metaphysics (Deslauriers 2007). I cannot address the question about definitions in detail in this essay. See Charles (2010a), pp. 203355. 32 Tuominen (2007) and (2010). 33 See also An.pr. 2.23, 68b35-37; Top. 6.4, 141b3-14; Phys. 1.1, 184a16-26; Met. 7.3, 1029b3-12; EN 1.4, 1095a30-b5. 31

REASON, EXPERIENCE …

113

principles of scientific demonstration (FP2).34 This reading is also confirmed by the description of the principles that nous knows as being better known as well as truer35 by nature than demonstrations (100b9-10). Since the only things that in Aristotle’s classification are better known by nature than demonstrations are their principles, there seems to be no room for doubt that the second question of Posterior Analytics 2.19 is about principles in the sense of FP2. An important implication of Bronstein’s argument is that Q1 only concerns FP2 in a rather remote sense. When Aristotle asks how the principles become known, he is not asking about the process through which we acquire knowledge of FP2 but about its origin. One might object that it is odd to read “how” (πῶς) in this way. If Aristotle had wanted to underline the origin of the knowledge, a more natural question would have been something like “from where (πόθεν) do we start the acquisition of that knowledge?”. Bronstein adds that in Aristotle’s question of “how do the principles become known?” another question is lurking: “Do the principles become known?”.36 I am not sure if there are parallels in which Aristotle asks “how” in this way. Bronstein’s argument,37 however, is not of linguistic nature. The point is that, Meno’s Paradox looming, Aristotle needs to answer the question of whether it is possible to know the principles at all. This is a sound point and I do not wish to deny it. What I want to stress is the following. Despite his insistence on understanding Q1 in the sense of a question of origin (Q1b), Bronstein concedes that the question about methodology (Q1a) is assumed about principles in the sense of FP1 (that he calls “preliminary accounts”) and the non-noetic knowledge of the definitions of the genera.38 He also grants that in those cases the question is answered in the traditional way (induction). Moreover, for FP2, the traditional question and answer 34 As mentioned, some scholars (Perelmuter 2010; Harari 2004; see also Lesher 1973, p. 55) have argued that we need to distinguish between nous that is only about concepts and non-demonstrative knowledge that is about principles. I shall return to the question in 4.3 below. 35 The expression ‘truer’ (ἀληθέστερον) supposedly means truths that are more primary in the order of nature. For this reading, see also Lesher (1973), pp. 63-64. 36 Bronstein (2016), p. 230. 37 See especially Bronstein (2016), p. 230 n20. 38 Bronstein (2016), p. 242. See also ibid., p. 231: Aristotle’s second task in the first part of the chapter is to “identify the means (induction) by which we reach the preliminary accounts that are the objects of the proximate prior knowledge required for learning principles properly speaking”.

114

M. TUOMINEN

are also partially relevant because Bronstein accepts that for acquiring knowledge about them, induction is necessary but not sufficient.39 This means that Bronstein’s account of Aristotle’s answer to Q1 in fact goes beyond the question about the origin, to which the answer is simply that the origin of our more advanced types of knowledge lies in perception. A possible difficulty with Bronstein’s view is that it renders Q1 and Aristotle’s answer to it rather complicated. In order to see why, let us return to Aristotle’s questions. (T1)

About principles (Q1) how they become known and (Q2) what the cognitive hexis is that knows them will become clear to one who has first considered some difficulties (99b17-19).

On Bronstein’s reading, the second question (Q2) concerns FP2 and the answer is nous. By contrast, in Q1, Aristotle asks several things. About principles in the sense of FP2, he asks (i) how they begin to become known and (ii) whether they can be known at all given Meno’s Paradox. The methodological question (Q1a) is also allowed, but mainly about principles in the sense of FP1 (Bronstein’s preliminary accounts), although the non-noetic knowledge of the definitions of the genera is also included. Finally, this answer is partially relevant to how we come to know FP2 because induction is necessary, but not sufficient for the acquisition of the knowledge about the principles. This entails that, while on Bronstein’s reading Aristotle sets himself the task of asking the question about origin (Q1b) of the knowledge of the principles (FP2), i.e., how do they begin to become known, the answer Aristotle gives transcends this task. 2.3 AThirdReadingofQ1 I shall argue that we can avoid these complications by taking Q1 as neither a methodological question (Q1a) nor merely a question about origin (Q1b) — or, more specifically, not quite in the way Bronstein does.40

39

See especially Bronstein (2016), pp. 244 n56. For induction, see section 4 below. Hankinson (2011), p. 34, argues for yet another reading of the questions according to which the first question is (1) How do we form concepts and grasp first principles? And the answer: by way of epagōgē. The second question, by contrast, is the familiar one: (2) What should we call that state which consists in that grasp [of the principles]? Nous. As mentioned, I agree with Bronstein that epagōgē alone is not sufficient for the acquisition of the first principles of demonstrations. 40

REASON, EXPERIENCE …

115

This can be done if we read Q1 as concerning the cognitive conditions that are necessary for us being able to acquire knowledge through inquiry: (Q1c): How, i.e. on what cognitive conditions do the principles [begin to] become known?

It must be stressed again that this question is not intended to replace the methodological question. This is a corollary of saying that Q1 (taken in the sense of Q1c) is not a question about methodology (Q1a), a point on which I agree with Bronstein. Therefore, Aristotle is not suggesting in PosteriorAnalytics 2.19 that we should renounce empirical methods for inquiry, division, and comparison of competing explanations and rather opt for a rational intuition that would do all the work for us. The description of the cognitive conditions for the acquisition of knowledge is not a substitute for an account of inquiry, but an account of the cognitive conditions for having the kind of knowledge that is needed to engage in scientific inquiry.41 Therefore, I agree with Bronstein that Aristotle is focusing, at least in the first half of his answer to question 1, on more rudimentary levels on which these cognitive capacities and processes operate. As Bronstein also stresses, the cognitive processes described must be such that they conform to Aristotle’s solution to Meno’s Paradox. As Aristotle sees it, a Platonist could not accept his claim that perception produces universal notions in us through memory and experience without any innate or previously possessed knowledge states (ἕξεις).42 Induction, division, and comparing competing explanations through a search for essential attributes are methods that are needed in the acquisition of non-demonstrative knowledge of the principles, also called nous. In PosteriorAnalytics 2.19, Aristotle is focused on explaining on what conditions the methods can do so. The crux of his explanation is that 41 This means that, as Bronstein also notes (2016, p. 229), we should not take Posterior Analytics 2.19 as an almost independent treatise on the knowledge of the principles detached from the other things Aristotle says in the treatise but rather see it as being an integral part of the whole. 42 Peter Adamson argues (2011) that although Aristotle has more in common with Plato than has usually been thought, the crucial difference is that, for Plato, we have once had those knowledge states — he also points to the connection between ‘state’ (ἕξις) and ‘having’ (ἔχειν) (p. 5) — whereas Aristotle insists that when we acquire knowledge, starting from perceptions, we acquire it for the first time (p. 9). I fully agree that the hexeis Aristotle is talking about are such that we did not have before but acquire from experience.

116

M. TUOMINEN

our cognitive development starts from perception requiring no innate knowledge and, secondly, that we obtain (for the first time and not by recollecting them) universal notions or concepts from perceptions through memory and experience. When we have acquired such a basic level of cognitive development, we have reason and can operate with general notions, propositions, and make inferences — inductive, deductive, or otherwise. I thus agree with Bronstein that Q1 is mainly about how we begin to have knowledge, although I differ on how exactly to articulate the question and some details in the answer. My general agreement on this with Bronstein entails that my position also differs from such traditional accounts that take Q1 as a methodological question (Q1a) and answer it in terms of induction, but add to this answer the claim that induction produces first principles (FP2) in us if it is aided by nous. This means that my disagreement with Bronstein is rather small.43 In addition to the articulation of Q1, it concerns the question of what exactly the cognitive conditions for acquiring knowledge from experience are taken to be. Bronstein seems to assume that our reason is able to make inductive generalizations, filling the details of what this means from the parallel passage in Metaphysics 1.1. However, in the parallel passage induction (ἐπαγωγή) is not mentioned, and the examples in Posterior Analytics 2.19 are not generalizations of the propositional kind, such as “all patients of type T are helped by leeches”, but rather universal notions (‘human being’, ‘animal’). As we shall see, Aristotle also calls the universal “one besides the many”, and this does not seem propositional. I shall also argue that what I call ‘Part 3’ of Aristotle’s answer to his first question (Q1) is better understood as a process of forming universal notions than either as a process of inquiry or even inductive inference. Finally, as mentioned, inductive inference requires the capacity to subsume individuals under general notions, and this is why having universal notions seems to be a cognitive condition for inductive inference. Since, as I argue, PosteriorAnalytics 2.19 is exactly about such cognitive conditions and not about the difference between experience and knowledge (or art) more generally (as the parallel passage in Metaphysics 1.1), we need to fill in the details of Aristotle’s answer on the basis 43 I am grateful to the participants of the conference on nous in Prague (June 2018) for discussions on this point, and especially to Matyáš Havrda for his comments on my paper.

REASON, EXPERIENCE …

117

of the examples given in 2.19 rather than Metaphysics 1.1. It must be emphasized that for adults engaged in scientific inquiry, it might be almost impossible to just have general notions without knowing some propositions about them.44 However, since Aristotle’s account is not about what capacities grown-up scientists have but about how we can begin to have knowledge at all from experience, the capacity to have general notions must be prior to making inferences. 3. Aristotle’s Answer to Q1 3.1 TheFourPartsoftheAnswer Let me now move to consider Aristotle’s answer to the first question (Q1) in more detail. In order to make his answer clearer, it is useful to divide it into four parts: Part Part Part Part

1: 2: 3: 4:

99b32-100a3 100a3-9 100a10-14 100a14-b5

The articulation of the parts allows for a variety of readings, and I shall argue for one that takes Aristotle to be focusing on the same process from slightly different perspectives. In Part 1, he outlines the cognitive capacities that are needed to answer question 1. The highest of them is reason (λόγος 100a2) that is only said to emerge in some animals, i.e. the human ones, although this is not explicitly mentioned in the context. In Part 2, he describes the states that those capacities produce in us: the capacity of perception produces memories, thanks to the capacity of retention, and memories gather into experiences.45 In human beings, experience also produces universal notions or concepts (τὰ καθόλου) by virtue of our rational capacity. The name for the capacity that is responsible for experience is presumably phantasia that occurs in the parallel passage in 44 For instance, thunder may well just be noise in the clouds. However, there is nothing in my reading that would exclude that. The point is that a cognitive condition for knowing thunder as noise in the clouds (and then looking for its explanation) is that general or universal notions are formed in the soul to begin with. I am grateful to Christian Pfeiffer for a discussion on this point. 45 I am grateful to Thomas Kjeller Johansen, Klaus Corcilius, and Pieter Sjoerd Hasper for a discussion on this point.

118

M. TUOMINEN

the Metaphysics (1.1, 980b26),46 although in Posterior Analytics 2.19 Aristotle simply talks about experience (ἐμπειρία).47 Part 3 is a simile in which what has been said in Part 2 is compared to an army or an organized group of soldiers withdrawing in retreat or rout. One soldier stops, then another one, and then another, until a starting point (ἀρχή)48 has been reached. As I argue, this can be understood as an illustration of the way in which perceptions produce memories through retention, memories produce experience, and experiences give rise to universal notions. Aristotle grants, however, that the simile is not very clear.49 Part 4 is meant to clarify it, but is somewhat difficult to understand as well. In the fourth part, Aristotle states that the coming to rest of one of the undifferentiated things amounts to there being the first universal in the soul, because although we perceive particular objects, perception concerns universals as well. “A stand is made” in these until “the partless ones” and the universals are made to stand. Aristotle concludes his answer to the first question by saying: (T2)

It is clear then that it is necessary for the first ones to become known to us through induction, for perception also instils the universal [in us] in this way (100b3-5).50

It is no wonder that there has been much debate about what exactly is going on in Aristotle’s answer to Q1, and many important issues remain outside the scope of this essay. 3.2 UniversalsCometoRest I shall argue that by talking about universals coming to rest in our soul in Part 2 Aristotle refers to our capacity of forming or obtaining universal notions or concepts from experience. Thanks to our reason’s capacity (the name for reason,λόγος, is given in 100a2) of operating with universals, 46 Admittedly, the reference in the parallel passage is not to the capacity but to appearances that other animals than human beings can have. However, if the appearances are there, the capacity of having them (phantasia as a capacity) must be there as well. 47 I am grateful to Klaus Corcilius and Pieter Sjoerd Hasper for a discussion on this point. 48 For the simile, see Lesher (2010b; 2011); Salmieri (2010). 49 Although not all scholars agree that this is the intended reference. Waitz, for instance, takes the reference to be to 2.13, 97b7-25; Barnes (1993, p. 265), I think correctly, argues against the suggestion. 50 δῆλον δὴ ὅτι ἡμῖν τὰ πρῶτα ἐπαγωγῇ γνωρίζειν ἀναγκαῖον· καὶ γὰρ ἡ αἴσθησις οὕτω τὸ καθόλου ἐμποιεῖ.

REASON, EXPERIENCE …

119

we can make inferences, process information about general facts and engage in scientific inquiry. As explained above, reason’s capacity to operate with universals is not introduced to displace experience, but to explain on what conditions we can acquire generally valid knowledge through inquiry into the causes and definitions of natural facts and phenomena. The key condition is that because we have reason, we can obtain universal notions from experience and thus start processing information on a general level. One way of understanding the sequence that Aristotle outlines in Part 1 is what I call ‘Alternative 1’. According to this reading, Aristotle first explains the cognitive progression from perception to reason, naming the capacities that take us to reason from perception (in 99b32100a3): Perception → memory → experience (phantasia) → reason (as a capacity to make inferences)

This as such is relatively uncontroversial. The disputed questions concern what reason (λόγος)51 means here and how Part 1 is related to Part 2. What I call ‘Alternative 1’ takes reason as a capacity to make inferences, especially inductive generalizations from experience. Alternative 1 is combined with the suggestion that the universals that Aristotle talks about in Part 2 are propositional statements as the ones in Metaphysics 1.1, e.g. “all patients of type T were helped by this cure”, based on experience of this cure helping individual patients like Socrates and Callias (981a5-12).52 In Part 2, Aristotle says that the principles of art and scientific knowledge come about from experience or53 the universal that has come to rest in the soul. Alternative 1 maintains that by universals coming to rest in our soul Aristotle refers to our making inductive inferences. Alternative 1, Part 2: Perception → memory → experience [→ reason and inductive inference, assumed] → universal statements → principles of art and scientific knowledge

51 I have explained above my reasons for agreeing with Gregoric and Grgic (2006, pp. 21-23) as well as Bronstein (2016, p. 235 n30, where more references are given) that something like basic rationality is meant here by logos. Adamson discusses the Platonic parallel of the sequence (perception – memory – knowledge) in the Phaedo; see Adamson (2011), p. 10. 52 See, e.g., Charles (2000), pp. 150-152; Bronstein (2016), p. 238. 53 I shall return to the question of how to read the connective ‘or’ ἤ in 100a6 in the body text below.

120

M. TUOMINEN

I take it that Bronstein’s view is a version of this alternative54 and that he would add the following steps to complete the progression to first principles of demonstrations: Alternative 1, progression to principles (FP2): Perception → memory → experience [→ reason through inductive inference] → universal statements = preliminary accounts (FP1) and definitions of genera → first principles of demonstrations (FP2) through division and demonstration

Some scholars have argued, however, that we should not distinguish between experience and having universals but that we should read ‘or’ in “from experience or from the whole universal” (100a6-7) as being epexegetic.55 Alternative 1 epexegetic ‘or’ in Part 2: Perception → memory → experience = universal → principle of art and science

However, powerful arguments have been made against the suggestion that a universal coming to a standstill could be identified with experience.56 Most importantly, the parallel passage in Metaphysics 1.1 distinguishes between experience that is of particulars and art (τέχνη) that is of universals. This seems to entail that we need to distinguish between experience and having universals (that are formed or acquired from experience) in PosteriorAnalytics 2.19 as well. At the same time, it needs to be stressed that this distinction cannot be made in a way that compromises Aristotle’s criticism of Plato and his solution to Meno’s Paradox. Therefore, although I do not think that the universal should be assimilated with experience in Part 2 (100a6-7), this does not mean that the 54 However, Bronstein later subscribes to “the concept view” and supposes that Aristotle is not preoccupied with the distinction between propositions and concepts because he assumes that grasping concepts can be cashed out in terms of propositions (2016), p. 246 n63. However, for the present purposes it is important that parts 1 and 2 of Aristotle’s answer to Q1 are, in Alternative 1, articulated in propositional terms that is obtained through inductive inference. For other proponents of “the concept view”, see, e.g., McKirahan (1992), pp. 246247; Barnes (1993), p. 271; Charles (2000), p. 264 n37; Modrak (1987), p. 162, 164. 55 For this reading, see e.g. Barnes (1993), p. 264; LaBarge (2006), p. 38; Ross (1949), p. 674; Hasper and Yurdin (2014), pp. 122-123. For a more detailed reading of the passages by Bronstein, see his article (2012) and (2016), p. 235 n29. 56 Bronstein (2016, p. 238) also rejects the epexegetic reading on the basis of the parallel passage in Metaphysics 1.1 that distinguishes between experience as knowledge of particulars from craft that is of universals (981a15-16); see also Charles (2000), pp. 150-151, McKirahan (1992), p. 243, and Hankinson (2011), p. 47.

REASON, EXPERIENCE …

121

principles of knowledge and art come to be known from universals, as opposedtofromexperience. By contrast, I read the relevant line as stating that principles of art and knowledge can come to be known (using methods of inquiry) from experience on the condition57 that we have a capacity of reason, enabling universal notions to come to rest in our soul. However, the reading that assimilates universals with experience is not relevant for Alternative 1, and I shall not dwell on it. As regards Alternative 1, the important point is that it takes Aristotle to be saying that perception, memory, and experience give rise to reason as a capacity to make inferences, and the inferences are explained as ones from particular to universal statements in the style of what can be called “Humean induction” (“this swan is white, that swan is white… swann is white; therefore all swans are white”). This reading is attractive, but I do not think the evidence fully supports it. One reason is that Aristotle’s phrase “one besides the many” (τὸ ἓν παρὰ τὰ πολλά, 100a7), used in the description of what the universal is, has a parallel that refers to non-propositional universals. It occurs in Aristotle’s arguments against Platonic ideas such as the beautiful itself (e.g., in Metaphysics 13.10, 1087a9 and 14.3, 1090a17).58 This seems to suggest that the universal as “one besides the many” in PosteriorAnalytics 2.19 refers to a universal notion such as ‘human being’, ‘two-footed’, ‘mule’ and ‘planet’.59 57 Pieter Sjoerd Hasper has informed me in conversation that the sequence “or from the whole” (ἢ ἐκ παντὸς) has no manuscript support and is probably derived from the texts in the late ancient commentaries. This means that the debate on the meaning of ‘or’ (ἤ) would become obsolete. There is still the question of how to read the sequence “from experience the universal having come to rest in the soul” (ἐκ δ’ ἐμπειρίας ἠρεμήσαντος τοῦ καθόλου ἐν τῇ ψυχῇ), which I take to mean “from experience when the universal has come to rest in the soul”, meaning that we can start inquiring into the principles of demonstrations and acquire knowledge about them (through scientific inquiry) from experience onthecondition that universal notions come to rest in our soul. Therefore, my difference from Hasper’s reading is that he takes the universal coming to rest in the soul referring to experience, while I read it as a further condition. 58 In PosteriorAnalytics 2.19 Aristotle seems to accept the description “one besides the many”, while he normally uses it critically against Plato. For instance, in Metaphysics 10.2 he argues that no universal can be a substance exactly because of being “one besides the many” (1053b19). Bronstein, however, takes “the one apart from the many” to refer to the universal being apart from the many memories (2016, p. 239), i.e. he does not make use of the parallel expression in Aristotle’s criticism of Plato. 59 For attributes such as two-footed as universals in the context of the PosteriorAnalytics, see Lesher (1973, p. 61) referring to 87b32-33; see also 73b26-27. Hasper and Yurdin claim that even statements of the form “all Fs are G” can be formed on the level of experience (2014), p. 120. That is because they understand ‘universal’ in the restricted sense of

122

M. TUOMINEN

Secondly, as mentioned, the parallel passage in Metaphysics 1.1 does not mention induction, although from a modern perspective it seems natural to analyse the passage in those terms. Thirdly, Alternative 1 leaves unexplained how we acquire reason. On that reading, Aristotle says first that human beings have the capacity of reason in addition to the capacities of perception, retention, and (presumably phantasia that explains) experience. Then, according to Alternative 1, Aristotle in Part 2 points out that we acquire universal statements through inductive inference. Reading Part 2 in this way leaves it open how we acquire reason. Let me now consider another reading that does not leave these questions open and, I argue, corresponds better to what Aristotle says in Part 3. I call this reading ‘Alternative 2’. Part 1, Alternative 2: Perception → memory → experience → reason Part 2, Alternative 2: Perception → memory → experience → universal notions60

To recapitulate, taking the sequence in this way, Aristotle first (in Part 1) names the cognitive capacities — perception, retention, experience (presumably acquired through phantasia), and reason — that are needed in order to come to have generally valid knowledge from experience. Secondly, in Part 2, he names the cognitive states produced by those capacities. The crucial point that distinguishes my view from Alternative 1 is that the relevant state we have thanks to the capacity of reason is one of a universal notion coming to rest in our soul. Such a state can be seen as a cognitive condition for making inductive inferences that presuppose having general notions. Especially for adult human beings having a universal notion practically always means knowing some propositions about the things falling under that notion as well. However, Alternative 2 seems to me to be at advantage since, first, as noted, it relies on examples given in the context and, secondly, inductive inference PosteriorAnalytics 1.4-5. To me, it rather seems that ‘universals’ refer to general or universal notions, such as the ones that occur in statements like “a human being is an animal”. 60 Gregoric and Grgic talk about “noetic rationality”, introduced in (2006), p. 21 n40. Frede (1996, p. 169) ascribes such grasp to nous but, as explained in the body text as well, Aristotle does not explicitly call the grasp of universal notions nous in PosteriorAnalytics 2.19, but talks about reason (λόγος).

REASON, EXPERIENCE …

123

can be taken to assume the capacity of subsuming individuals under universal notions. Thirdly, it is more informative than Alternative 1 in the sense that, on this reading, Aristotle briefly explains what it means to acquire reason. Fourthly, I shall argue next that Alternative 2 fits better with what Aristotle says in Part 3. Perhaps, as to the third point just mentioned, one could object that Alternative 1 does not leave the acquisition of reason unexplained, but reason is acquired through inductive generalization. However, this explanation seems circular: reason means the capacity to reason, and we acquire such a capacity by reasoning, i.e. generalizing over particulars.61 By contrast, on Alternative 2, the explanation is not circular in this way. 3.3 SoldiersWhoStop Part 3 of Aristotle’s answer to Q1 opens with the statement that the cognitive states (ἕξεις) do not belong to us as definite and already existent (100a10). Since he has just finished his brief description of progression from perceptions to experience and universals, this seems to indicate that Aristotle is in Part 2 talking about such hexeis, as I have explained above as well. Therefore, Aristotle seems to be affirming his claim that we come to have cognitive hexeis from experience without having to assume pre-existent hexeis. It must be stressed that while the hexis that figures in Q2 and is thus central for the outline of the chapter is about first principles of demonstrations, those listed in Parts 1 and 2 of Aristotle’s answer to Q1 are not: he talks about memory, experience, and having universals at rest in one’s soul. What Aristotle says in Part 3 offers further support for the claim that Aristotle is talking about cognitive processes through which universal notions come to rest in our soul, rather than active inquiry and research that can yield knowledge about the principles of demonstrations (FP2). Therefore, Part 3 starts with the claim that knowledge acquisition does not require cognitive hexeis that would belong to us as already existent 61 I also have some reservations about “Humean induction” as an explanation of Aristotelian induction, since the passage that is used as evidence for it is the one in Metaphysics 1.1 in which Aristotle does not talk about induction (ἐπαγωγή). For the different ways in which Aristotle uses epagōgē, see Engberg-Pedersen’s somewhat provocative account (1979), where he is challenging his reader to produce counter examples (ibid., p. 311) to his reading; see also McKirahan (1992), pp. 250-257.

124

M. TUOMINEN

and definite, as Aristotle suggests Plato would have it. Rather, they are acquired from perceptual experience in a way that Aristotle notoriously compares to a retreat or rout of soldiers in a battle.62 When the soldiers have turned around, one stops, then another, and yet another, and (supposedly) so on until a starting point (ἀρχή)63 has been reached (100a11-13). Although Aristotle does not explain the details of the process,64 it seems to me that the simile can rather naturally be taken to be understood as a description of the process of how memories gather into experience. Individual soldiers that stop can thus be understood as memories of what has been perceived. When they stop together as a group, that amounts to experience. Although Aristotle does not explain how exactly it happens, he points out in Part 2 that experience gives rise to universal notions in the human soul.65 If the simile were meant to describe inductive inference and scientific research, one would expect it to refer to something like a commanding officer who organizes the troops in a specific way. However, such an element is missing from it. Finally, Aristotle finishes Part 3 by noting that the soul is such (and perhaps was already such from the beginning)66 that it can undergo (πάσχειν) the described process (100a13-14). Again, this seems a relatively natural way of referring to the cognitive processes of acquiring memories from perceptions through retention, experience from memories supposedly through phantasia, and universal notions from experience.67 62 Some scholars have underlined the disorderly nature of the retreat, i.e. that it is a rout. See Lesher (2010a; 2010b). For my reading the question of how organized the retreat is, is not central. 63 Although this is a somewhat difficult claim, I do not think we should emend the text (i.e., change ἀρχήν into ἀλκήν, ‘strength’) as Barnes suggests (1993), p. 265. 64 McKirahan says: “like some other famous similes of Aristotle’s, this one does nothing to make ideas clearer” (1992), p. 244. Lesher (2010b; 2011) has made important clarifications, however. 65 Salmieri (2010) offers another account of how the simile can be taken to explain the acquisition of universal notions. 66 I am grateful to Francesco Ademollo for the suggestion that the temporal dimension is important in the expression ὑπάρχει τοιαύτη οὖσα in 100a13-14 (with a parallel to Metaphysics1041a15). From that perspective, the point is that the soul is such from the beginning, i.e., that one has the capacity to acquire reason from birth. 67 Frede claims that acquiring universals from experience is a natural process because it is in our nature to acquire them (1996), p. 171. As Frede also points out, we need to distinguish between the way in which obtaining simple universals is natural and the way in which it is also natural to acquire the knowledge of the first principles which is more like becoming virtuous. Frede argues that universal notions is natural also because their acquisition is a causal rather than an epistemic relation (ibid., p. 172). Although I agree that the acquisition

REASON, EXPERIENCE …

125

By contrast, to say that our soul is capable of undergoing inductive inference or active inquiry seems strained.68 One might object, however, that we form concepts from experience rather than undergo a process of acquiring them. In a way, I suppose we do, but perhaps Aristotle is stressing a slightly different point here. By this I mean that there is a difference between a process of forming concepts and, say, initiating inquiry or making inferences at will. While the latter seems to be an activity we initiate, the same is not necessarily the case of the former. If Aristotle is talking about a rather rudimentary concept formation, we probably “undergo” a process of learning them at a rather young age. However, I do not think that acquiring or forming general concepts is limited to young age. Rather, Aristotle seems to allow in Part 4 of his answer to Q1 that notions on all levels of generality are acquired in a similar way, and in zoology, for example, there can be species that one has not encountered before. Therefore, I think one can form concepts also in a process of scientific or philosophical inquiry. What Aristotle seems to suggest in Part 4 is that, when we do this, it happens through notions coming to rest in our soul. This might seem awkward, but there is also something that quite well corresponds to experience: a difference between the process of trying to learn something or trying to understand some new concept and understanding them. I can start the process of trying to learn or trying to understand at will, read about new things and try to figure out what something means, but I cannot necessarily make myself understand something at will. Rather, there is an uncontrollable element in actually learning or understanding something. The claim that our soul is capable of undergoing the cognitive processes described in Parts 1-3 of Aristotle’s answer to Q1 can be taken to refer to such an element, provided that we adopt Alternative 2 in Parts 1 and 2 and the reading I suggested for Part 3.

of at least some universal notions from everyday experience is not an inferential process, I do not think that this excludes all reasoning towards more general universals. Even the simple ones can probably be given further support through pointing to instances, common features, distinguishing features, and so on. 68 One objection to this could be that, according to Aristotle (An.post. 1.1, 71a1-2), all intellectual learning comes about from some pre-existent knowledge (γνῶσις). This might be taken to suggest that learning from perceptual gnōsis must be through inference. However, to claim that all intellectual learning comes about from pre-existent knowledge does not imply that all learning from pre-existent gnōsis is acquired through inference.

126

M. TUOMINEN

3.4 UniversalsintheSoul Until now, I have argued that, in Parts 1-3 of Aristotle’s answer to Q1, he explains that universal notions come to rest in our soul in a way that is analogous to the way in which perceptions are retained in the soul by virtue of retention, and memories gather into experience by virtue of phantasia. Secondly, the process is analogous to the way in which soldiers in rout group into a more organized whole stopping around or after individual soldiers. At the beginning of Part 4, Aristotle notes that what he has said so far is not very clear and needs to be stated again. He continues: (T3)

When one of the undifferentiated things have come to rest, there is the first universal in the soul (for although a particular is perceived, perception is of a universal, such as of human being, not of Callias human being) (100a15-b1).

This is a notoriously puzzling dictum. Why is Aristotle saying that perception is of a universal, while denying that the capacities of the perceptive soul alone could process information on the level of universal notions or concepts? However, when combined with Parts 1 and 2 above, the remark can be seen as strengthening the point that universal notions are acquired from experience. The examples given in Part 4 also strengthen the claim that the kinds of universals that Aristotle has in mind are mainly universal notions rather than propositional universals. The example mentioned in (T3) is ‘human being’, also called ‘undifferentiated’.69 The 69 Many different readings have been introduced for ‘undifferentiated’ (ἀδιάφορα). One traditional reading is that, since the example is ‘human being’, Aristotle talks about infimaspecies, “least general universals” as McKirahan puts it (1992), p. 245; see also Barnes (1993), p. 245; Ross (1949), p. 677. However, Aristotle does not seem to suppose that such universals always come to rest at first: sometimes one can only first identify something as an animal in scientific inquiry. Bolton (1991, p. 6-9) has connected ‘undifferentiated’ with Physics 1.1, where Aristotle talks about how children first call all men ‘father’ because they have not learned to make the relevant distinction. Bronstein (2016, p. 244) and Salmieri (2010, pp. 165-166) take the expression in the sense of “the same species” (i.e., Socrates and Callias are undifferentiated by their species). Hankinson (2011, p. 49) makes another plausible suggestion that the undifferentiated things are properly formed, but unarticulated concepts of natural kinds; they are unarticulated in the sense that we do not know the specific differences yet. I basically agree with Hankinson, although it seems to me that the universal notions that one acquires through experience can also be of genera and attributes. In those cases, the expression would refer to different kinds being undifferentiated in genus. One relevant point is that they are undifferentiated in the sense that they are similar (i.e. having no difference) in a relevant sense: in species or in genus, or with respect to a certain attribute.

REASON, EXPERIENCE …

127

claim that perception is of the universal although only particulars are perceived could then be taken as a reference to the central claim of Parts 1 and 2 that universal notions are acquired from perceptual experience but that the capacities of the perceptive soul alone cannot grasp or process them. Universal notions can only be processed by reason. However, ‘human being’ is not the only example of a universal notion Aristotle mentions in Part 4. He goes on to state that “a stand is made” in these until “the partless ones” and the universals are made to stand, from “such and such an animal” to ‘animal’, “and with animal the stand is made in the same way” (tr. Barnes 1993, 100b1-3). To me it seems that this is most naturally read as a claim that universal notions can come to rest in our soul on all levels of generality, perhaps up to the notion of substance.70 The third example, “such and such an animal”, can also be taken to refer to a single unnamed animal species, although it is also possible to read it as a shorthand for a propositional statement (“a human being is a two-footed animal”). However, this must be completed by a conjecture, and the example does not yield uncontroversial support for the propositional reading. I have argued that Aristotle is explaining how we come to have universal notions from experience and that having them is a cognitive condition for coming to know the principles, and it is important to stress that those notions alone are not principles of demonstrations (FP2). Whether they can be seen as starting points for inquiry (FP1) is a more difficult question. Since we grasp simple universals from experience we can also learn that all human beings are animals, all planets are non-twinkling, eclipse belongs to the moon,71 and so on. It is coming to know facts of that kind that Aristotle sees as crucial for starting scientific inquiry (An. Post. 2.1-2).72 70 Ross suggests (1949, p. 678) that the series goes until the categories; see also Bronstein (2016), p. 247 n64. The suggestion seems plausible, although the expression ‘partless’ is unusual. 71 Aristotle treats celestial bodies that are the single representatives of their species as equivalent to terms denoting natural kinds. 72 Aristotle points out (An.post. 2.1-2) that finding out whether things exist or whether facts obtain is often a matter of perception. However, this is not always the case, as one might need to show whether something exists or not through argument (god, centaur 2.1, 89b31-32). Further, for determining whether an attribute is essential or not, he recommends (in An.post. 1.5, 74a35-b4) testing out which attributes can be abstracted without changing the essence (such as isosceles or brazen from a triangle). If an attribute (such as being a figure) cannot be thus abstracted, it is an essential one. This does not seem to me to be a strictly empirical test and thus I would not classify all Aristotle’s methodological tools as exclusively empirical. However, this remark is not meant to undermine the impor-

128

M. TUOMINEN

In order to learn such things and to inquire into the causes and definitions of natural kinds and phenomena, we must explore the world — ask what kind of animals there are, what their essential attributes are, how they live, how the non-twinkling of planets can be explained, how lunar eclipse can be explained, and so on rather than engage in apriori studies of a conceptual structure — although sciences may differ with respect to the role of experience and theoretical analysis in them. In any case, having universal notions that “come to rest in our soul” is a crucial condition for the acquisition of knowledge in general, although the notions alone are not principles in the sense of FP1 or FP2. From this it follows that my reading of the universals in Part 2 (100a6-7) as universal notions that can be rather elementary differs from those versions of “the concept view” that take the acquisition of universal concepts to amount to having definitions. If the universals that we acquire are rather universal notions such as ‘human being’, ‘animal’, ‘two-footed’, ‘tame’, they are only elements in the full definition of the object of study, such as “a human being is a tame, twofooted animal” — or whatever the exact definition should be taken to be — and the definitions must always be found through inquiry. 4. Aristotle’s Epagōgē and Induction 4.1 TheFirstOnesAreKnownThroughInduction Aristotle concludes his treatment of the first question (Q1) by saying: (T2)

It is clear then that it is necessary for the first ones to become known to us through induction (ἐπαγωγή), for perception also instils the universal [in us] in this way (100b3-5).73

Primafacie this passage seems to offer evidence for the traditional reading of Q1 as a methodological question (Q1a) with the traditional answer: induction (A1a). As indicated above, I agree with Bronstein that induction, at least understood in the “Humean” sense of simple generalization proceeding from individuals under one general notion,74 is not sufficient tance of empirical methods and systematic observation for Aristotelian scientific inquiry — and certainly not the claim that all cognitive development starts from experience. 73 δῆλον δὴ ὅτι ἡμῖν τὰ πρῶτα ἐπαγωγῇ γνωρίζειν ἀναγκαῖον· καὶ γὰρ ἡ αἴσθησις οὕτω τὸ καθόλου ἐμποιεῖ. 74 Such as: Socrates was helped by applying leeches, Callias was helped by applying leeches, and so on; therefore, all phlegmatic patients are helped by applying leeches.

REASON, EXPERIENCE …

129

for producing knowledge of the first principles of demonstrations (FP2). However, I am not convinced that this passage supports Bronstein’s reading. Contrary to the traditional reading that takes “the first ones” as being primary in the order of nature and hence first principles of demonstrations (FP2), Bronstein argues75 that they should be identified with the universal propositions we come to know first, i.e., the preliminary accounts. The difficulty with this view is the following. If we suppose T2 to make the claim that induction takes us to the starting points for inquiry (i.e., Bronstein’s preliminary accounts and non-noetic knowledge of the definitions of the genera), it is no longer clear how we can separate this account from the way in which perceptual experience instils universals in us according to the reading I have called Alternative 1. If we suppose that (i) “in this way” in T2 refers back to Answer 2 and to 100a3-9, (ii) the details of that passage are filled in from Metaphysics 1.1, and (iii) the “first ones” are the preliminary accounts, T2 is in danger of becoming redundant. On these assumptions, it becomes a claim that starting points for inquiry are acquired through inductive generalization because universal statements are formed through inductive generalization. This is not a literal tautology, and perhaps it could be read in the sense that because all generalizations are made by induction, the ones concerning preliminary accounts and genera are as well. However, if the details of what kind of generalizations Aristotle is talking about are filled in from Metaphysics 1.1, they are exactly the kind of statements that could function as preliminary accounts (“all phlegmatic patients were cured by applying leeches”) for which an explanation (“why were they helped by the cure?”) is sought in inquiry. Therefore, both the process (induction) and the result are the same in both halves of the claim made in T2. One might perhaps argue along the following lines that the reading does not make T2 redundant. To know the first universal propositions in the way that they can function as preliminary accounts, we need to know something more than knowing the generalization alone. For example, we need to know that there is some common feature because of which the generalization holds and that this makes the proposition an appropriate object for scientific study. For example, knowing that there is some 75

Bronstein (2016), p. 242.

130

M. TUOMINEN

common feature the phlegmatic people share makes them receptive to a cure that can be inflicted on them with the application of leeches makes one able to look for that feature in inquiry. This is a fair objection, but I am not sure it solves the problem because induction as a simple passage from the particular to the universal does not seem to provide us with that kind of knowledge. By contrast, if we take the universals (also in T2) as universal notions, we can distinguish between the universal notions that perception instils in us and the principles that come about from them, be they starting points for inquiry (FP1) or first principles of demonstration (FP2). Universal notions are not principles, but having them is a condition for knowing principles in either sense. This gives us at least three possible readings of T2. Let me begin by considering the first two. Aristotle could be taken to say that (i) we arrive at starting points for inquiry (FP1) through epagōgē because the kind of cognitive process described in the four parts of his answer to Q1 instils universal notions in our soul through a progression from perceiving particulars to having universal notions. Another possibility would be that (ii) the primary universals (i.e., the most general ones that are ‘first’ in the order of nature) are obtained through epagōgē because we also come to grasp universal notions (the first ones in the order of knowledge) through a progression from perceiving particulars to operating with universals. Although there is a parallel in Aristotle’s use of “the first ones” (τὰ πρῶτα) and ‘principles’ (ἀρχαί) in the sense of the first principles of demonstrations, it seems that T2 cannot be taken in the sense that first principles of demonstration are obtained through induction, if induction is understood as simple generalization. As has been repeatedly stressed, systematic scientific inquiry into the essential attributes and causes of things is needed to yield principles in the sense of FP2. This seems to suggest that we need to take “the first ones” in T2 either as the first in the sense that they are the starting points of inquiry (FP1) or the universal notions (such as substance) that are first in the sense of being prior in the order of nature. As I see it, both readings are possible for T2. With the first reading, Aristotle is saying that we acquire starting points for inquiry by simple generalization from experience,76 since we acquire general notions from 76

This is broadly speaking in agreement with Bronstein’s view.

REASON, EXPERIENCE …

131

experience through the process described in Parts 1-3 of Aristotle’s answer to Q1.77 The second reading would be that universal notions on all levels are formed similarly, be they the ones we acquire from experience without engaging in active inquiry (‘human being’, ‘dog’, ‘horse’) or the ones that are acquired through inquiry (such as ‘substance’). Since the essential attributes that we are looking for in inquiry are typically universal or general (they apply to more individuals than one), the same process can be taken to apply not only to universals in the category of substance, but also to essential attributes (‘tame’, ‘two-footed’, and so on). By inquiring into the common features of different kinds of things (such as human being, horse and mule as long-lived), we can find some essential attribute (say, low in bile) that, perhaps with the help of some other explanatory principles, can be understood to figure in the demonstration explaining why these animals are long-lived. This process resembles the one that instils universal notions in our soul because both proceed from the more particular to the more universal.78 However, they differ in the sense that generalization in systematic inquiry is guided by questions about essence and explanation, while the cognitive condition for all inquiry and generalization is the formation of general notions from experience. 4.2 WhatKindofInduction? As mentioned, Alternative 1 explains what it means to arrive at universals by reference to a parallel passage in Metaphysics 1.1. Despite its unquestionable resemblance to Posterior Analytics 2.19, however, the passage does not talk about induction (ἐπαγωγή). There are some instances in which Aristotle mentions particulars in the sense of individuals in connection with epagōgē (e.g. in An. Post. 71a20-21), but typically his examples proceed from individuals of different types or

77 This would differ from Bronstein’s reading since the progression from particulars to universals is not inductive inference (Alternative 1 above) but the way in which general notions “come to rest in our soul” (Alternative 2 above). 78 It is possible that such epagōgē can be used to teach why all these animals are longlived to the student; for epagein as leading someone to know, see Engberg-Pedersen’s type (1) (1979), p. 301. However, to do so presupposes that the teacher has acquired the knowledge in systematic inquiry into the causes and essential attributes of natural kinds.

132

M. TUOMINEN

different types and are thus not identical with “Humean” inductions in which the relevant particulars fall under the same general notion. I do not claim that we should rule out the possibility of articulating Aristotelian epagōgē in such Humean terms. I rather wish to emphasize that the examples Aristotle gives when articulating arguments that he calls ‘inductive’ in Topics 1.12 (105a13-16) and in PriorAnalytics 2.23 (68b17-24) are not of that kind. In the passage in the Topics, the particulars are of different types. Induction in Topics 1.12 The knowledgeable captain is the best. The knowledgeable charioteer is the best. Therefore, in all cases, the knowledgeable [expert] is the best.

One might object that we should not even expect the articulation of inductive arguments in the Topics to be relevant for how we come to know the principles of demonstration, since the Topics is concerned with how to enhance the credibility of conclusions, and that is not sufficient for establishing truth. That is fair objection, and perhaps the induction of the Topics is not relevant for PosteriorAnalytics 2.19. Yet, it remains the case that Aristotle does not provide an example of induction in Humean terms. What, then, about epagōgē in Prior Analytics 2.23? There Aristotle articulates an induction that he also calls “syllogism from induction” (68b15-16) in which the particular cases are particular species (human being, horse, mule) rather than individuals (say, Socrates, Bukephalos, Mule M). In addition, the inference is not a simple generalization, but consists of three terms (A, B and C) of which B and C are said to be convertible. A is “long-lived”, B “low in bile”, and C comprises (all) the long-lived animal species (human being, horse, and mule). The syllogism proceeds from the following premises: A of all C (being long-lived belongs to all human beings, horses and mules) C of all B (being human being, horse or mule belongs to all animals low in bile)

The conclusion is “A of all B”, namely that being long-lived belongs to all animals that are low in bile. This, however, is something that transcends induction as simple generalization from particulars to universals. It does not rely on listing individual cases or make “an inductive leap” to generalization, but rather proceeds from general types and assumes the knowledge that they exhaust the species that fall under the relevant kind

REASON, EXPERIENCE …

133

(being low in bile). This seems to presuppose that, when engaged in such induction, one is not concerned with the acquisition of preliminary accounts or definitions of the genera, but looking for the kinds of explanatory connections between essential attributes that are central in demonstrations proper. If then we wanted to read “the first ones” in T2 as referring to the first principles of demonstrations (FP2), this would seem to require that induction (epagōgē) should be taken to refer to the kind of inductive syllogism Aristotle talks about in Prior Analytics 2.23. It concludes that human beings, horses, and mules are low in bile and thus identify the cause for their longevity, which could be expressed in a demonstrative syllogism. Being long-lived belongs to all animals that are low in bile. Being low in bile belongs to all human beings, horses and mules. Therefore, being long-lived belongs to all human beings, horses and mules.

However, this would make the connection to Aristotle’s answer to Q1 rather remote. He would simply say that since we acquire universal notions from experience, we can come to know relevant explanatory connections between essential attributes of natural species and organize them into an inductive syllogism and thus establish principles of demonstrations. 4.3 NousandNon-DemonstrativeKnowledge If we read Aristotle’s answer to Q1 either as my preferred Alternative 2 or Bronstein’s Alternative 1, nous as a hexis of reason does not figure in it. As I have noted, Bronstein is right that, in the context of Posterior Analytics 2.19, nous only enters the stage in Aristotle’s answer to Q2.79 This also entails that universal notions are not principles,80 but that the knowledge of the principles comes to be “from experience when the universal has come to a standstill in the soul”.81 When finally introduced 79 Therefore, I have, as indicated, revised my earlier view (Tuominen 2010) with respect to this point. 80 For arguments to this effect, see Bronstein (2016), p. 237 n36; McKirahan (1992), p. 303 n23; p. 306 n95; see also Charles (2000), pp. 150-151. 81 ἐκ δ’ ἐμπειρίας (ἢ ἐκ παντὸς) ἠρεμήσαντος τοῦ καθόλου ἐν τῇ ψυχῇ… τέχνης ἀρχὴ καὶ ἐπιστήμης (100a6-8). As noted, ἢ ἐκ παντὸς does not have manuscript support but is perhaps derived from the commentaries.

134

M. TUOMINEN

in 100b8, nous is said to concern objects that are by nature better known than demonstrations. As we saw above, this must refer to the first principles of demonstrations (FP2). One might object that although the last chapter of the Posterior Analytics does not name nous as the state that knows universal notions, the rest of the treatise might do so. Especially one passage in book 1 could be taken to show that nous is not only concerned with propositional principles of demonstrations, but also with universal notions. The passage is found in chapter 1.3 in which Aristotle argues that there must be a starting point for demonstrative knowledge that does not require demonstration. In that context, the objects of such knowledge are called horoi (72b24-25). Because nous is identified as the starting point for knowledge (2.19, 100b15), this entails that nous knows the horoi. If we take horoi to mean ‘terms’, the passage could be taken to indicate that we have nous of universal notions. However, because horoi can also mean propositional definitions, the evidence is not conclusive.82 For similar reasons, I do not think that one can make the case for a systematic distinction between nous that is about non-propositional universals or concepts and “non-demonstrative knowledge” that is about propositional principles. This distinction is based on a suggested revision of how to read the passage in which Aristotle calls knowing the first principles of demonstrations “non-demonstrative knowledge” (1.33, 88b35-37). While he has usually been taken to employ ‘nous’ and “non-demonstrative knowledge” synonymously in the sequence (“neither nous… nor non-demonstrative knowledge”), the alternative reading is contrasting these two.83 The alternative reading of this passage is possible, but such a distinction does not seem pertinent in the 82 This is one of the reasons why Perelmuter’s (2010) case is not convincing. However, although plausible, I do not think Bronstein’s case against Perelmuter made in a long footnote (2016, p. 52 n2) is conclusive either. Bronstein argues that PosteriorAnalytics 1.3 fixes nous as the name of the capacity to grasp definitions and, since definitions are propositional premises, this shows that nous cannot be distinguished from non-demonstrative knowledge on the grounds of the argument given for the alternative reading. However, in An.post. 1.3 Aristotle talks about horoi (72b24; 73a9) not horismoi and in Perelmuter’s reading they could be taken to refer to the terms occurring in the definitions, not the propositional definitions as such. Fine rejects the alternative reading as well (2010b), p. 326 n8. 83 Perelmuter (2010) and Harari (2004), p. 18; the reading is also mentioned by Lesher (1973), pp 54-55.

REASON, EXPERIENCE …

135

context of the whole of PosteriorAnalytics, let alone 2.19. As we have seen above, the principles that are there singled out as the objects of nous are more naturally taken to be propositional first principles of demonstrations (FP2) rather than universal notions. 5. Concluding Remarks I have argued that any straightforward opposition between rationalism and empiricism is misguided for analysing PosteriorAnalytics 2.19. On the one hand, Aristotle states clearly that all knowledge acquisition begins from perception, and the knowledge of the principles of demonstrations comes about from experience (100a6-8). On the other hand, the principles are known by nous, a hexis of reason. Therefore, while our cognitive development must start from perceptions, it leads to generally valid knowledge only on the condition that we become able to process information on a general level: i.e. that we acquire general or universal notions from experience which we can use in inference. However, this is not a Janus-faced betrayal of Aristotle’s general empirical spirit, but rather the recognition of the fact that more advanced knowledge states presuppose that we are capable of processing information acquired from experience by reason: structure it by general notions, chain it into propositions and inferences, and asking questions about explanation and essential features in systematic inquiry. It is important to stress that, in my view, scientific inquiry is necessary for the acquisition of the knowledge of the principles of demonstrations. Therefore, I agree with David Bronstein that PosteriorAnalytics 2.19 is not an account of how we acquire the knowledge of the principles of demonstrations. I also agree that Aristotle’s first question in the chapter is more limited in scope than has traditionally been thought. I have disagreed on some of the details of Bronstein’s account and argued that Aristotle’s four-fold answer to the question should be read as describing the cognitive conditions on which we can (even begin to) acquire generally valid knowledge from experience. While the key condition for Bronstein is inductive generalization of the “Humean” kind to preliminary accounts and non-noetic knowledge of the definitions of the genera, I have argued that forming or acquiring general notions from experience is required even by such induction and that Aristotle identifies that condition in his account.

136

M. TUOMINEN

LITERATURE ADAMSON, Peter (2011), “PosteriorAnalytics II.19: A Dialogue with Plato?” in BulletinoftheInstituteofClassicalStudies, suppl. vol. S107 (Aristotleand theStoicsReadingPlato), pp. 1-19. ANGIONI, Lucas (2014), “Aristotle on Necessary Principles and on Explaining X Through the Essence of X” in StudiaPhilosophicaEstonica 7/2, pp. 88-112. BALTUSSEN, Han (2007). “Did Aristotle Have a Concept of ‘Intuition’? Some Thoughts on Translating Nous” in Close, Tsianikas and Couvalis (eds.), pp. 53-62. BARNES, Jonathan (1993), Aristotle: Posterior Analytics. Oxford: Oxford University Press. BERTI, Enrico (ed.) (1981), Aristotle on Science: The Posterior Analytics. Padova: Antenore. BOLTON, Robert (1991), “Aristotle’s Method in Natural Science: Physics I” in Judson (ed.), pp. 1-29. BRONSTEIN, David (2012), “The Origin and Aim of Posterior Analytics II.19” in Phronesis 57 (2012): pp. 29-62. BRONSTEIN, David (2016), AristotleonKnowledgeandLearning:ThePosterior Analytics (Oxford Aristotle Studies), Oxford: Oxford University Press. BURNYEAT, Myles (1981), “Aristotle on Understanding Knowledge” in Berti (ed.), pp. 97-139. BUTLER, Travis (2003), “Empeiria in Aristotle”. SouthernJournalofPhilosophy 41/3, pp. 329-350. CHARLES, David (2000), Aristotle on Meaning and Essence. Oxford: Oxford University Press. CHARLES, David (ed.) (2010a), DefinitioninGreekPhilosophy. Oxford: Oxford University Press. CHARLES, David (2010b), “The Paradox in the Meno and Aristotle’s Attempts to Resolve it” in Charles (ed.), pp. 115-150. CLOSE, E., TSIANIKAS, M. and COUVALIS, G. (eds.) (2007), Greek Research in Australia: Proceedings of the Sixth Biennial International Conference of GreekStudies. Adelaide. DESLAURIERS, Marguerite (2007), AristotleonDefinition. Leiden: Brill. DETEL, Wolfgang (1993), Aristoteles, Analytica Posteriora: Übersetzung und Erläuterung, 2 vols. Berlin: Akademie Verlag. ENGBERG-PEDERSEN, Troels (1979), “More on Aristotelian Epagoge” in Phronesis 24: pp. 301-319. FEREJOHN, Michael (2013), Formal Causes: Definition, Explanation, and Primacy in Socratic and Aristotelian Thought. Oxford: Oxford University Press. FINE, Gail (2010a), “Aristotle on Knowledge” in Elenchos 14: pp. 121-156. FINE, Gail (2010b), “Aristotle’s Two Worlds: Knowledge and Belief in Posterior Analytics 1.33” in ProceedingsoftheAristotelianSociety, vol. CX part 3 (2010), pp. 323-346.

REASON, EXPERIENCE …

137

FINE, Gail (2014), ThePossibilityofInquiry: Meno’sParadoxfromSocratesto Sextus, Oxford: Oxford University Press. FREDE, Michael (1996), “Aristotle’s Rationalism” in Frede & Striker (eds.) (1996), pp. 157-173. FREDE, Michael & STRIKER, Gisela (eds.) (1996), RationalityinGreekThought. Oxford: Oxford University Press. GOLDIN, Owen (1996), ExplainingtheEclipse, Ann Arbor: University of Michigan Press. GREGORIC, Pavel and GRGIC, Filip (2006), “Aristotle’s Notion of Experience” in ArchivfürGeschichtederPhilosophie 88, pp. 1-30. GROTE, G. (1880), Aristotle (2nd edn.), London: Murray. HANKINSON, J. (2011), “Avant nous le déluge. Aristotle’s Notion of Intellectual Grasp” in Morison and Ierodiakonou (eds.), pp. 30-59. HARARI, Orna (2004), KnowledgeandDemonstration:Aristotle’sPosteriorAnalytics. Dordrecht: Kluwer. HASPER, Peter and YURDIN, Joel (2014), “Between Perception and Scientific Knowledge: Aristotle’s Account of Experience” in Oxford Studies in AncientPhilosophy 47, pp. 120-150. JUDSON, Lindsay (ed.) (1991), Aristotle’sPhysics:ACollectionofEssays (Clarendon Aristotle Series). Oxford: Oxford University Press. KAHN, Charles (1981), “The Role of Nous in the Cognition of the First Principles” in Berti (ed.), pp. 385-414. KOSMAN, L. A. (1973), “Understanding, Explanation, and Insight in Aristotle’s PosteriorAnalytics” in Lee, Mourelatos and Rorty (eds.), pp. 374-392. LABARGE, Scott (2006), “Aristotle on Empeiria’ in AncientPhilosophy 26, pp. 23-44. LEE, H.D.P. (1935), “Geometrical Method and Aristotle’s Account of First Principles” in ClassicalQuarterly 29, pp. 113-124. LEE, H. D. P, MOURELATOS, A. and RORTY, R. (eds.) (1973), ExegesisandArgument, Assen: Van Gorcum. LENNOX, James and BOLTON, Richard (eds.) (2010), Being, Nature and Life in Aristotle:EssaysinHonorofAllanGotthelf, Cambridge: Cambridge University Press. LESHER, J. (1973), “The Meaning of ΝΟΥΣ in the PosteriorAnalytics” in Phronesis 18, pp. 44-68. LESHER, J. (ed.) (2010a), From Inquiry to Demonstrative Knowledge: New Essays on Aristotle’s Posterior Analytics (= Apeiron 43/2-3), Kelowna: Academic Printing and Publishing. LESHER, J. (2010b), “Just as in a Battle: The Simile of the Rout in Aristotle’s PosteriorAnalytics ii 19” in AncientPhilosophy 30, pp. 95-105. LESHER, J. (2011), “A Note on the Simile of the Rout in the PosteriorAnalytics ii 19” in AncientPhilosophy 31, pp. 121-125. LEUNISSEN, Mariska (2015), “Comments on Malink’s Aristotle’s Modal Syllogistic” in PhilosophyandPhenomenologicalResearch 90, pp. 733-741. MALINK, Marko (2013), Aristotle’sModalSyllogistic, Cambridge HA: Harvard University Press.

138

M. TUOMINEN

MCKIRAHAN, Richard (1992), PrinciplesandProofs, Princeton: Princeton University Press. MENDELL, Henri (1998), “Making Sense of Aristotelian Demonstration” in OxfordStudiesinAncientPhilosophy 16, pp. 161-225. MODRAK, Deborah K.W. (1987), Aristotle: ThePowerofPerception, Chicago: University of Chicago Press. MODRAK, Deborah K.W. (2001), Aristotle’sTheoryofLanguageandMeaning, Cambridge: Cambridge University Press. MORISON, Benjamin and IERODIAKONOU, Katerina (eds.) (2011), Epistēmē,Etc.: EssaysinHonourofJonathanBarnes, Oxford: Oxford University Press. MORISON, Benjamin (2012), “An Aristotelian Distinction between Two Types of Epistēmē” in ProceedingsoftheBostonAreaColloquiuminAncientPhilosophy 37, pp. 29-63. PELLEGRIN, Pierre (2005), Aristote:SecondsAnalytiques, Paris: Flammarion. PELLEGRIN, Pierre (2010), “Definition in Aristotle’s PosteriorAnalytics” in Lennox and Bolton (eds.), pp. 122-146. PERELMUTER, Zeev (2010), “Nous and Two Kinds of Epistēmē in Aristotle’s PosteriorAnalytics” in Phronesis 55/3, pp. 228-254. ROSS, W. D. (1949), Aristotle’s Prior and Posterior Analytics. A Revised Text withIntroductionandCommentary. Oxford: Clarendon Press. SALMIERI, Gregory (2010), “Αἴσθησις, Ἐμπειρία, and the Advent of Universals in PosteriorAnalytics II 19” in Lesher (ed.), pp. 155-185. SALMIERI, Gregory (2013), “Aristotelian Epistēmē and the Relation between Knowledge and Understanding” in Metascience 23, pp. 1-9. TUOMINEN, Miira (2007), Apprehension and Argument: Ancient Theories of StartingPointsforKnowledge. Dordrecht: Springer. TUOMINEN, Miira (2010) “Back to Posterior Analytics II 19: Aristotle on the Knowledge of Principles” in Lesher (ed.), pp. 115-143.

ARISTOTLE ON ADHOMINEMARGUMENTS Christoph HORN

In Topics VIII as well as in the Sophistical Refutations, Aristotle discusses a type of argument that he occasionally calls prostonanthrōpon (see Soph.El. 22, 178b17), rendered by Boethius as solutioadhominem. In dialectical debates, adhominem arguments are originally and typically used to attack the position of an interlocutor, mostly the respondent, by reference to his former concessions. In this primal sense, an adhominem argument does not so much entail an attack on formal fallacies or logical insufficiencies committed by the respondent, expressed at a certain time t1. The attack is rather directed against an alleged inconsistency between his present and his former standpoint, i.e. regarding a certain inconsistency between the opinions uttered at t1 and t2. Ad hominem arguments are commitment-based, and the commitments must have been made across time or lie somewhere in the background. This elementary version of adhominem arguments is usually called argumentumexconcessis. In dialectical exercises, the concessions in question must have been admitted by the interlocutor at some earlier time in the discussion (or, of course, before the present discussion, maybe even at a former stage of his biography). In this sense, the argumentum ad hominem is closely connected to peirastikoilogoi which are dedicated to test the questionable knowledge of someone, based on elements of his own position.1 A second aspect goes along with this quite immediately. Adhominem arguments are typically attacking the person who is defending a certain position. This personal attack can be meant as a blame of someone’s inconsistency (then it serves, as it were, as an enlargement of the argumentumexconcessis). Or it can be directed against other defects that an 1 This point is convincingly emphasized by G. Nuchelmans (1993), p. 38. Peirastikoi logoi are arguments that are used to test the knowledge of the interlocutor.

140

C. HORN

interlocutor allegedly has, especially an intellectual or moral shortcoming which diminishes his credibility within the dialectical debate. I will discuss this second aspect of ad hominem arguments, the personal attack, later in this paper. The general historical root of this sort of argument lies, unsurprisingly, in the juridical rhetoric of the fifth century BC.2 Already there, it has an ambiguous reputation. Somewhere in the background we might assume the presence of the Socratic elenchos (as the famous reference to Socrates in Sophistical Refutations 34, 183b7-8 seems to indicate), but since Aristotle ascribes the achievement of the dialectical methodos to himself, this should include also his remarks on adhominem arguments.3 Nevertheless, the influence of a certain Socratic or Academic background is plausible. However this may be, the argument seems to have, also in Aristotle’s view, a somewhat oscillating character. The aim of my paper is to show that Aristotle is right in discerning appropriate and inappropriate uses of adhominem arguments. He leaves room for a certain benevolence concerning them, and this makes good sense. On my view, it will turn out in the end that the idea of adhominem arguments plays a highly relevant role within Aristotelian virtue ethics, even if Aristotle does not say much on this topic on an explicit level. But indirectly, this idea is strongly present in his account of ethical and dianoetic virtues. I. The adhominem argument is traditionally subsumed under the heading of ‘fallacies’. But this is not generally supported by Aristotle’s treatment of it. He sometimes criticizes it, but does not seem to completely reject it, as was correctly pointed out, e.g., by Jaakko Hintikka (2004: 207). I would like to go one step further: I think that Aristotle basically acknowledges and appreciates adhominem strategies; for him, adhominem arguments ex concessis can be adequately used, in spite of their

2 For the ancient history of the argument see Chichi (2002); for its modern history see van Eemeren and Grootendorst (1993). 3 L. A. Dorion 2013 shows against R. Bolton that the SophisticalRefutations are not specifically directed against Socrates. Aristotle regards himself as the pioneer of the dialectical method.

ARISTOTLE ON ADHOMINEMARGUMENTS

141

frequently abusive character. The clearest statement in which he affirms their value can be found in book VIII of the Topics.4 [t1] Adverse criticism of an argument on its own merits, and of it when presented in the form of questions, are two different things. For often the failure to carry through the argument correctly in discussion is due to the person questioned, because he will not grant the steps of which a correct argument might have been made against his position: for it is not in the power of the one side only to effect properly a result that depends on both alike. Accordingly it sometimes becomes necessary to attack the speaker (πρὸς τὸν λέγοντα) and not his position (μὴ πρὸς τὴν θέσιν), when the answerer lies in wait for the points that are contrary to the questioner and becomes abusive as well: when people lose their tempers in this way, their argument becomes a contest, not a discussion (δυσκολαίνοντες οὖν ἀγωνιστικὰς καὶ οὐ διαλεκτικὰς ποιοῦνται τὰς διατριβάς). (Topics VIII.11, 161a16-24)

As Aristotle claims in [t1], it seems sometimes unavoidable to attack, within a dialectical debate, the respondent instead of attacking his position or argument. For there are cases in which the interlocutor is not willing to concede what he should admit in order to properly contribute to the progress of the discussion. In this case, the argument is not so much exconcessis, but exnonconcessis. Adhominem arguments of that type are hence legitimate whenever the respondent is blindly opposed to the questioner in order to avoid his imminent argumentative defeat. Note that already in [t1] there is a certain flavour of moral or personal evaluation implied. The respondent is described as a mere denier, someone who detracts himself from the discussion, thus turning the dialectical exercise into a contest. He practices agōnistikaidiatribai instead of doing dialektikai. Insofar the adhominem strategy is offered as an appropriate antidote. This is clearly a quite positive assessment of an ad hominem argument. We find a somewhat more ambiguous evaluation of adhominem arguments in SophisticalRefutations 33 where Aristotle takes the adhominem as one of three strategies within a dialectical debate: [t2] Just as it is possible to bring a solution sometimes against the argument, at others against the questioner (ὁτὲ μὲν πρὸς τὸν λόγον ὁτὲ δὲ πρὸς τὸν ἐρωτῶντα) and his mode of questioning, and at others against neither 4 All translations are taken from J. Barnes, CompleteWorksofAristotle,TheRevised OxfordTranslation.

142

C. HORN

of these, likewise also it is possible to marshal one’s questions and reasoning both against the thesis (πρὸς τὴν θέσιν), and against the answerer (πρὸς τὸν ἀποκρινόμενον) and against the time (πρὸς τὸν χρόνον), whenever the solution requires a longer time to examine than the period available. (SophisticalRefutations 33, 183a21-26)

Passage [t2] seems to be quite ambivalent. Whereas it is clear that Aristotle accepts the two first solutions, i.e. that which is directed against the argument (πρὸς τὸν λόγον) and that against the questioner (πρὸς τὸν ἐρωτῶντα), a possible difficulty lies in the second half of the text: does Aristotle really permit that someone chooses a solution that takes less time that normally would be necessary? Aristotle’s remark that the adequate lusis would be a more time-consuming one (and that someone might therefore look for a shorter way) seems to amount to a critique. But from reasons of coherence I would rather take it in the sense that such a shortcut is sometimes admissible: seen in this way, the text allows for three types of solution, namely against the position (πρὸς τὴν θέσιν), against the answerer (πρὸς τὸν ἀποκρινόμενον and against the time (πρὸς τὸν χρόνον). If that is the right reading, Aristotle concedes that it can be justified to attack the person who, in a dialectical debate, asks the question or the individual who answers, instead of addressing the content of his questions or answers. And this would be in agreement with [t1]. Let us now consider two texts in which Aristotle apparently criticizes adhominem arguments. It will be instructive to see which his reasons for such a criticism are. A possible shortcoming of ad hominem arguments emerges, according to [t3], when the interlocutor concedes a point which turns out not to be the core mistake of the fallacy. In this case, the interlocutor is appropriately refuted by the adhominem strategy, but the crucial mistake remains undetected. Look at ch. 20 of the SophisticalRefutations: [t3] Some people solve this last refutation in another way as well. For, they say, if he has granted that he can do anything in the way he can, still it does not follow that he can harp when not harping: for it has not been granted that he will do anything in every way in which he can; and it is not the same thing ‘to do a thing in the way he can’ and ‘to do it in every way in which he can’. But evidently they do not solve it properly: for of arguments that depend upon the same point the solution is the same, whereas this will not fit all cases of the kind nor yet all ways of putting the questions: it is valid against the questioner, but not against his argument (ἀλλ’ ἔστι πρὸς τὸν ἐρωτῶντα, οὐ πρὸς τὸν λόγον). (SophisticalRefutations 20, 177b27-34)

ARISTOTLE ON ADHOMINEMARGUMENTS

143

In the context of ch. 20 of the Sophistical Refutations we see Aristotle attacking false conclusions such as the following: someone infers from “It is correct tosaynow that you are born” the fallacious conclusion “It is correct to say that you are bornnow” (177b21-2). The temporal adverb qualifies the act of speaking, not that of being born. Of course, we are confronted here with a very simple type of fallacy. The case under consideration in the quotation is that someone falsely concludes from ‘If you are able to play the kithara you can play it without actually doing so’ the sentence ‘You can play the kithara without doing so’ (as if someone maintained that he could play an instrument without being active at all). Sophistic fallacies of this type can easily be unmasked. But then Aristotle discusses a further strategy of dealing with them, namely the demonstratioadhominem. In this case, some people claim that the concession ‘One can play the kithara without playing it’ is included if someone admits that everybody will do what he is able to in every way. But that solution depends on a questionable concession made here by the interlocutor. It would not be valid for similar cases such as that of the erroneous inference from ‘I say now that you are born’ to ‘I say that you are born now’. What is mistaken about ad hominem strategies becomes clearer when Aristotle takes up the example a bit later in ch. 22: [t4] But, as was said also above, all these persons direct their solutions against the man, not against his argument (οὗτοι πάντες οὐ πρὸς τὸν λόγον ἀλλὰ πρὸς τὸν ἄνθρωπον λύουσιν). For if this were a genuine solution, then, suppose any one to grant the opposite, he could find no solution, just as happens in other cases; e.g. suppose the true solution to be ‘So-and-so is partly true and partly not’, then, if the answerer grants the expression without any qualification, the sophist’s conclusion follows. If, on the other hand, the conclusion does not follow, then that could not be the true solution: and what we say in regard to the foregoing examples is that, even if all the sophist’s premises be granted, still no proof is effected. (Sophistical Refutations 22, 178b16-23)

Adhominem arguments depend on factual concessions given by the interlocutor, more precisely in the defectiveness of these concessions. In the case under discussion, however, such a concession is not only unnecessary to reject the inference, it even impedes an adequate solution. A logical fallacy can immediately be identified as a fallacy, as Aristotle claims, whether or not the interlocutor concedes the erroneous point (and whether or not the admitted weakness is a right or a false one). Seen from this perspective, Aristotle again is not criticizing the adhominem strategy

144

C. HORN

in itself, but only its misapplication. This is the reason why Hintikka in his commentary correctly says that “What these Aristotelian passages foreshadow is not the so-called adhominem fallacy in any size, shape or form, but rather the idea of deductive inferential validity” (2004: 207). The problems of [t3] and [t4] are caused by false inferences. Neither in [t3] nor in [t4] Aristotle is criticizing ad hominem arguments in general; what he rejects is only their use in cases in which we should analyse the fallacy as such, not with reference to the speaker. But this observation might then lead to a real criticism. Ad hominem arguments seem to be, in many cases at least, either redundant or false: redundant if the fallacy is additionally admitted by the interlocutor, but can already be identified within the argument itself; false if the interlocutor concedes another point than the fallacious one and hence the fallacy is refuted for the wrong reason. In fact, Aristotle’s clearest rejection of adhominem arguments seems to be on this point; it is contained in [t5]: [t5] A sophistical refutation is a refutation not absolutely but relatively to someone: and so is a proof, in the same way. For unless that which depends upon ambiguity assumes that the ambiguous term has a single meaning, and that which depends on like verbal forms assumes that substance is the only category, and the rest in the same way, there will be neither refutations nor proofs, either absolutely or relatively to the answerer (οὔθ’ ἁπλῶς οὔτε πρὸς τὸν ἐρωτώμενον): whereas if they do assume these things, they will stand, relatively to the answerer; but absolutely they will not stand: for they have not secured a statement that does have a single meaning, but only one that appears to have, and that only from this particular man. (Sophistical Refutations 8, 170a12-19)

Primafacie one gets the impression that [t5] identifies sophistical inferences with adhominem arguments and hence seems to amount to a fundamental criticism of them. What Aristotle attacks is that in sophistical elenchoi or sullogismoi a person must (adequately or not) be motivated to concede e.g. the synonymy of a meaning which should correctly be analysed as homonymous. If one puts pressure on someone in a dialectical debate he might concede what he shouldn’t admit. Then the argument is valid only in this situation and only against this speaker. Seen from this perspective, even [t5] doesn’t contain a fundamental rejection of ad hominem-arguments. It is again only about their possible abuse. We might hence conclude: If we carefully avoid this abuse, nothing hinders an appropriate use of this strategy.

ARISTOTLE ON ADHOMINEMARGUMENTS

145

What can we hence learn from [t5]? Important conditions of a correct use of adhominems are that the respondent himself must have subscribed to the position that is seen to be inconsistent with the rest of his standpoint. And he should have done so in a voluntary, correct, and well-reflected affirmation. The assent must not be enforced, otherwise his opponent only assumes that the respondent should have to accept it. But the respondent need not do so; he might maintain a quite unusual or provocative position. In this case, we are rather faced with an argument from common sense, from common knowledge or from recognized knowledge (endoxa). Ad hominem arguments presuppose free and adequate self-commitments. The ad hominem argument ex concessis against someone’s position can typically be realized (a) by pointing to a formal contradiction within someone’s expressed set of beliefs, (b) showing the fatal consequences implied in it so that the overall standpoint can be shown to be selfdefeating, (c) by hinting at the inappropriateness or defectiveness of the concessions with regard to the topic or (d) by reference to a former moral or quasi-moral commitment that is now violated. To illustrate (c), take e.g. the case of someone who is commonly known as a liberal and tolerant personality but, in an ongoing political debate, shows some sympathies for a populist or nationalist party. In such a case, the relevant point is the moral tension that exists between t1 and t2. In antiquity, this version was well known from its application against philosophers who did not practice in their lives what they were preaching (it is usually called tu quoque argument). In its version (c), the adhominem-argument exconcessis comes very close to what is usually seen as the second basic type of adhominem argument, the attack on the speaker’s ethical or intellectual character. Maybe (c) must even be seen as the only genuine ad hominem-argument among the arguments exconcessis. Douglas N. Walton e.g. in one of his papers (2004) argues that we can speak of an ad hominem argument exconcessis only if an attack on someone’s ethical identity is at play. Otherwise we are only faced here with an argument from inconsistent commitments. I will discuss this in a moment. But before that, I want to briefly mention that there might also be a further candidate for the set of ad hominem arguments which addresses the specific behaviour of someone or the peculiar facts or circumstances of a given situation in which the speaker is. In this sense, Aristotle says, e.g., in his indirect argument in favour of the principle of non-contradiction that someone who refuses to speak or to utter anything “would be no

146

C. HORN

better than a plant” (Met. Gamma 4, 1006a14-5; cf. 1008b7-12). His overall strategy in this passage is to “prove” the principle of non-contradiction not directly (apodeixaihaplōs) which is impossible, but to prove it indirectly (apodeixaielenktikōs).5 Maybe this remark (ὅμοιος γὰρ φυτῷ ὁ τοιοῦτος ᾗ τοιοῦτος ἤδη) contains a certain moral blame, or maybe it reflects the intention to ridicule the person who doesn’t want to express himself. Then it would be a reproach or a joke, not an argument. But it seems more likely that Aristotle’s intention is to show him a drastic imminent consequence of his position, namely the threat of rational selfdestruction. Let us take the words ‘he would not be better than a plant’ in the sense of a thread of self-destructive consequences: whoever declines to speak in order to avoid his confutation risks, according to Aristotle, the danger of undermining his rational identity. Indeed, Aristotle is in need of such a threatening consequence because his elenctic strategy can only succeed if the interlocutor in fact utters something. Now, it might be true that, simply by his refusal to speak, this person implicitly admits that, if or only if he said something distinctive, the elenctic strategy of Aristotle wouldbe successful — since he then would express a distinctive meaning and thereby acknowledge the principle of non-contradiction. But at least he is not actually refuted as long as he says nothing, and hence Aristotle needs an additional strategy that disqualifies this refusal. I am not sure if one should characterize that sort of argumentative strategy, i.e. Aristotle’s use of a thread of self-destructive consequences, connected with his indirect proof of the basis of our conditions as speakers or rational subjects, as an adhominem argument at all. In a certain sense this is not the case since the line of argument adopted by Aristotle in Metaphysics Gamma 4 and Kappa 5 does not take into account what someone personally concedes or believes and is not about his specific personality. It rather regards fundamental implicit commitments of any speaker, and so it points, as it were, to a part of the conditionhumaine. Only if we take the refusal of the interlocutor as a personal moral or intellectual deficiency — i.e. in the sense that he takes the risk of actually becoming like a plant — it seems to amount to an adhominem argument. We would then blame him for his decision to give up his full rational identity simply in order not to be refuted. 5 Nuchelmans (1993), p. 39, indicates that this dichotomy is taken up again, in a slightly different form in Met.K.5, 1062a2-9.

ARISTOTLE ON ADHOMINEMARGUMENTS

147

II. Let us now consider argumentative strategies which are directed against situative contradictions, performative tensions, or existential discrepancies. Should we integrate them into the class of adhominem arguments? In my opinion, we should do so only conditionally. A good example for what I have in mind is Schopenhauer’s famous story, to be found in his literary remains, of someone who, while living in Berlin, says ‘Berlin is really a terrible place to live’ and his interlocutor then replies ‘Why then don’t you leave it by the next train?’6 To my mind, this example should not count as an adhominem argument at all if it simply reports a tension in someone’s set of beliefs or a discrepancy between his opinions and his life or actions. The wide field of performative or existential selfcontradictions seems not to fall completely into the class of adhominem arguments. They are ad hominem arguments only if they contain an element of evaluation of someone’s personal character or other fundamental preconditions of his identity. In this sense, what is normally called the second type of adhominem arguments is, I think, the core of what is meant in general (to that extent, I share the position of D.N. Walton). Adhominem arguments must basically be directed against the proponent’s personality, character, identity, origin, set of attitudes etc. This standard type is sometimes called “direct” adhominem argument since it has the simple form of rejecting someone’s position by appeal to a lack of his credibility or trustworthiness. If someone is presented as a liar or deceiver, as a hypocrite or pretender, as narrow-minded or as an ideologist (a racist or chauvinist or whatever else), his position might be disqualified even without any reference to its content. The direct ad hominem argument can easily be exploited in a dialectical situation, and so it is sometimes rightly called “abusive”. With regard to this second form the adhominem strategy is characterized, in most of the relevant handbooks, as a fallacy from the outset. Taken as a fallacy, one normally interprets the adhominem argument in the sense of an argument that illegitimately alludes to negative aspects of someone’s identity. Using this type of adhominem strategy someone tries to shed doubt on his opponent’s arguments by attacking his personality. There seem to be three versions of it (a-c). (a) Direct ad hominemfallacy: If someone 6

Schopenhauer (2011).

148

C. HORN

undermines the credibility of his interlocutor by pointing out a negative character trait of this person, he commits a fallacy insofar as the position of this person need not be affected by the shortcomings of his character (It might be e.g. irrelevant for the research of a theoretical physicist that he is far from being a good father of his three children.) (b) Circumstantialad hominemfallacy: Someone might disqualify the position of his interlocutor by mentioning certain contexts that are constitutive for this position. Take e.g. the case in which someone raises the suspicion that a doctor recommends the medicaments of a certain pharmaceutical company after having been invited by them for a weekend trip. It can however be the case that the recommendation is absolutely uncorrupted by the self-interest of the doctor. (c) Tuquoquefallacy: If someone rejects the health tips of someone else by indicating that this person herself does not stick to it, he commits a fallacy since the tips can be valuable in themselves even if they are not strictly observed by the person who is putting them forth. All of these criticisms directed against the ad hominem strategy are based on the common idea that the adrem standpoint should always be prior to the ad hominem perspective. This idea seems basically right. A person can be fully trustworthy in his or her reports, narratives, opinions, recommendations and so on, even if we have good reason to doubt about his or her moral behaviour, social attitudes, family relationships or political preferences. To use an illustrative example: Leo Strauss in his NaturalRightandHistory (1953/1965: 42-3) mentions an extreme form of direct ad hominem fallacy which he calls the reductio ad Hitlerum; what Strauss means is that a view is not immediately refuted by the fact that it happens to have been shared by Hitler. Nevertheless, this is not the complete story: To be sure, there are also highly adequate attacks on someone’s ethical character. Someone’s judgment can in fact be decreased by factors that are constitutive for his or her (problematic) personality. It should also be noted that it is not always about ethical aspects of a person, but this sort of adhominem can additionally be based on evaluations of his intelligence, origin, social affiliation, philosophy, religion, and so on. And furthermore, the fundament of this argument need not be in all cases a negative evaluation of someone’s character. It can also be the other way around: if someone basically appreciates the moral character of his friend in general, he might feel obliged to point to a tension or discrepancy in his opinions or actions. Taken together, I would therefore like to call the second type of adhominem the ‘argument from personal

ARISTOTLE ON ADHOMINEMARGUMENTS

149

character traits’. The position of an interlocutor can and sometimes should be attacked with regard to the insufficiency of someone’s moral or cognitive features, attitudes, and habits. Seen from this wider perspective, one can discuss the ad hominem argument from personal character traits on the basis of virtue ethics. And here, Aristotle comes into play again. The fact that Aristotle basically has a positive attitude towards such arguments can be seen from three passages in which he uses the idea that someone suffering from a doubtful character or a lack of reality-perception has a more or less reduced credibility. The first one is to be found in the third book of the Rhetorics: [t6] Another method consists in counter-attacking the accuser; for it would be absurd to believe the words of one who is himself unworthy of belief (ἄτοπον γὰρ εἰ ὃς αὐτὸς ἄπιστος, οἱ τούτου λόγοι ἔσονται πιστοί). (Rhetorics III.15, 1416a26-28)

For Aristotle, it is justified to attack someone (in this case: the accuser in a juridical process) for his lack of credibility. Without naming this strategy by the label of an ad hominem argument Aristotle apparently accepts the legitimacy of what we just called “direct” adhominem argument.7 The pertinence of a standpoint in fact depends on the credibility of the person who defends it. In Physics II.1, Aristotle attacks an imaginary adversary on similar grounds: he first explains the concept of phusis (in the sense of an entity having its principle of kinēsis and stasis within itself) and then declares that those who wish to get a proof for the existence of such a phusis are neglecting an immediately given fact. These ignoramuses are like blind people speaking about colours, lacking the ability to perceive a basic feature of reality: [t7] That nature exists, it would be absurd to try to prove; for it is obvious that there are many things of this kind, and to prove what is obvious by what is not is the mark of a man who is unable to distinguish what is selfevident from what is not. This state of mind is clearly possible. A man blind from birth might reason about colours. Presumably therefore such persons must be talking about words without any thought to correspond. (Physics II.1, 193a3-9)

What is self-evident (phaneron) and known by itself (di’ hauto gnōrimon) cannot, according to our text, be proven by something that 7 Nuchelmans (1993), p. 43, mentions Rhet. I.1 for a certain criticism advanced by Aristotle against argumentative strategies lying outside the actual case.

150

C. HORN

is only indirectly known. Someone who ignores this fact is not a serious discussion partner. In a well-known passage of the Topics, he goes one step further in establishing a connection between individual character and the use of arguments. Some individual do not even deserve arguments at all, but they should be punished or need sense-perception: [t8] Not every problem, nor every thesis, should be examined, but only one which might puzzle one of those who need argument, not punishment or perception. For people who are puzzled to know whether one ought to honour the gods and love one’s parents or not need punishment, while those who are puzzled to know whether snow is white or not need perception. (Topics I.11, 105a3-7)

People who are wondering whether or not one should worship the gods or love one’s parents need punishment, not arguments. This seems to be an extreme expression of the principle of adhominem relativity of argumentation. Now, if Aristotle basically acknowledges the direct adhominem strategy in the sense of attacking someone’s position on the basis of character traits or in the sense addressing arguments according to someone’s character traits, it seems plausible to discuss the adhominem argument in connection with virtue ethics. And since human character (as we described it above) it has a moral and a cognitive part, it is tempting to discuss it on the basis of Aristotle’s theory of ethical and dianoetic virtues. The basic intuition seems to be this: the credibility of a person who is blamed of having a doubtful character should neither be seen as completely untouched nor as completely destroyed; we should not simply reject a position defended by such a person, but be cautious if the defective character might have influenced his position. In other words, we should both avoid absolute credulousness and extreme distrust. Let us for a moment look at contemporary virtue epistemology, as it is discussed e.g. in some writings of Ernest Sosa, Jay Wood, Linda Zagzebski, or Douglas Walton.8 A fundamental partition into two camps is that between virtue reliabilism and virtue responsibilism. Heather Battaly in one of her contributions on the topic (2010) develops the relevant point on the basis of two examples (I-II): (I) Suppose that a doctor who is known for his dogmatism in diagnosing people says that the patient has a bacterial infection. (II) Suppose that a doctor who is known for being cruel diagnoses that the patient has a 8

See especially Zagzebski (1996).

ARISTOTLE ON ADHOMINEMARGUMENTS

151

bacterial infection. As these two examples illustrate, the character of someone can play a role in assessing the credibility of his position. Even if it is true that cruel and dogmatic doctors can possess appropriate medical knowledge, their primafacie credibility is limited; it makes sense to take their statements with reservation. It depends on how their expertise might be deteriorated by their bad character traits. The difference between (I) and (II), as Battaly puts it, is that between a virtue-reliabilism on the one hand and a virtue-responsibilism on the other. While a virtuereliabilists claim that intellectual virtues should be characterized as stable and reliable capabilities (based on sound perception, memory, inductions, or conclusions), virtue-responsibilists think that intellectual virtues are close to ethical virtues in their quasi-moral character (such as open-mindedness, intellectual courage, intellectual autonomy, patience, creativity, intellectual phantasy, searching for the best possible solution, taking other opinions into account etc.). We are facing here two different issues: the soundness of a judgment can be negatively influenced by a defective cognitive attitude on the one hand and by a defective moral character on the other. While the first issue is quite obvious, it seems not so easy to deal with the other. How can we understand that someone who has a suboptimal moral character (in the wide sense of a lack of desirable character traits) sometimes arrives at and sometimes does not arrive at adequate judgments? What exactly is the force implied in his or her character which leads this person astray? I would like to call this the ‘problem of misguided intelligence’. The philosophical difficulty under consideration is, as it were, a theoretical equivalent to the problem of akrasia in practical thought. How can character exploit reason or intelligence? As e.g. the cases of Polus and Callicles in the Gorgias and of Thrasymachus in the Republic show, Plato is highly sensitive to the problem of the right argumentative attitude towards intellectually brilliant people who defend bad positions — and in the end go astray in their moral behaviour. Look at the following passage from Republic VII: [t9] Then the other so-called virtues of the soul do seem akin to those of the body. For it is true that where they do not pre-exist, they are afterwards created by habit and practice. But the excellence of thought, it seems, is certainly of a more divine quality, a thing that never loses its potency, but, according to the direction of its conversion, becomes useful and beneficent, or, again, useless and harmful. Have you never observed in those who are

152

C. HORN

popularly spoken of as bad, but smart men, how keen is the vision of the little soul, how quick it is to discern the things that interest it, a proof that it is not a poor vision which it has, but one forcibly enlisted in the service of evil, so that the sharper its sight the more mischief it accomplishes? --I certainly have, he said. (Republic VII, 518d8-519a5)

Does Plato suppose that intellectual virtue exists by nature? What he seemingly says in [t9] is that intellectual virtue, in opposition to “the other virtues”, is immediately present in some beings and never loses its divine force — while the other virtues resemble corporeal capacities in having their origins in training. Intellectual virtue can be used for good or bad purposes. Now, of course, this view can hardly contain Plato’s full position on intellectual virtue since it would be at odds with what we know from all of his other statements on the topic. It would rather sound like a Humean instrumentalism concerning reason. I think we can take it for granted that Plato, in [t9], doesn’t speak of virtues in the philosophical sense, but in popular way of speaking of excellent talents. Virtues in the philosophical sense cannot be misdirected or abused (see, e.g., Euthydemus 279c-280a for the invariant goodness of sophia which also rectifies the other, ambivalent goods). Nevertheless, the passage is one of the few ancient philosophical texts in which the author concedes that also brilliant intellectual abilities can be misdirected. To deal with the problem of misguided intelligence, Plato postulates a capacity of the soul which is analogous to the eye of the body; only this capacity can bring about the conversion (periagōgē) of someone. Is there a psychic faculty that can use intelligence appropriately or inappropriately? Aristotle deals with that problem of misguided intelligence when he discusses deinotēs in Nicomachean Ethics VI.13. Intelligence can in fact be used in a morally adequate or inadequate way, for the kalon or the aischron; in the latter case, it deserves to be called “knavery” (πανουργία). But Aristotle would not call knavery a virtue — even in its most developed form. For him, people who are bearers of full intellectual virtues cannot, at the same time, lack of a full realization of virtues of character, and viceversa. Intellectually talented people can be morally bad, but precisely insofar they are not dianoetically virtuous. And so, according to Aristotle, the description given in Republic VII cannot occur: namely that someone who has a defective character possesses intellectual virtue (if Plato in fact meant this by what he says in the quotation).

ARISTOTLE ON ADHOMINEMARGUMENTS

153

Let us finally have a look at Aristotle’s ideal of a mutual connection between the intellectual virtues of phronēsis and sophia and moral goodness. First he acknowledges that there exist brilliant cognitive abilities in people who are not morally good: [t10] This is why it is thought that these qualities are a natural gift, and that a man is considerate, understanding and intelligent by nature, though no one is a wise man by nature (καὶ φύσει σοφὸς μὲν οὐδείς, γνώμην δ᾽ ἔχειν καὶ σύνεσιν καὶ νοῦν). That this is so is indicated by our thinking of them as going with certain ages: we say that at such and such an age a man must have got intelligence and considerateness, which implies that they come by nature. [...] Consequently the unproved assertions and opinions of experienced and elderly people, or of prudent men, are as much deserving of attention as those which they support by proof; for experience has given them an eye for things, and so they see correctly. (Nicomachean Ethics VI.12, 1143b4-10)

Aristotle in this text discusses the difference that exists between intelligence (more exactly the cognitive competences of gnōmē, sunesis, and nous) as natural gifts and sophia as an intellectual virtue that nobody possesses by birth. As we know from NE VI.7, he considers sophia as the level of masterful intellectual competence in a field, especially in theoretical knowledge. Additionally we learn from NE VI that phronēsis has to play the role of appropriately guiding our practical deliberation, while sophia in its highest form is directed towards theoretical (invariant) objects and principles and is hence not concerned with action at all. Therefore, sophia cannot bring about happiness; all it does is to identify happiness as the right ultimate end of human action. On the other hand, phronēsis (which by the way can also originate from experience) can be misled if it lacks a theoretical and moral fundament in being good. In its experience-based form phronēsis is insufficient and prone to immoral or not fully moral behaviour. It must be supplemented by moral insight: [t11] We have therefore also to reconsider the nature of virtue. The fact is that the case of virtue is closely analogous to that of prudence in relation to cleverness. Prudence and cleverness are not the same, but they are similar; and natural virtue is related in the same way to virtue in the true sense. All are agreed that the various moral qualities are in a sense bestowed by nature: we are just, and capable of temperance, and brave, and possessed of the other virtues from the moment of our birth. But nevertheless we expect to find that true goodness is something different, and that the virtues

154

C. HORN

in the true sense come to belong to us in another way. For even children and wild animals possess the natural dispositions, yet without Intelligence these may manifestly be harmful. This at all events appears to be a matter of observation, that just as a man of powerful frame who has lost his sight meets with heavy falls when he moves about, because he cannot see, so it also happens in the moral sphere; whereas if a man of good natural disposition acquires intelligence, then he excels in conduct, and the disposition which previously only resembled virtue will now be virtue in the true sense. (NicomacheanEthics VI.13, 1144b1-15)

For Aristotle it holds true, both for intellectual and for moral virtues, that we can possess them by birth or experience. But such excellences can also be owned by children or animals. This shows that full intellectual and moral virtues must be different from natural ones. Natural perfections are not the true good that we are looking for. We see here how Aristotle wants to solve the problem of misguided intelligence: namely by the distinction he draws between natural talents and acquired dispositions. Since the first ones are not insight-based, they are possible objects of abuse, while the second ones cannot be abused since they are part of a morally and intellectually stable habit. We are hence searching for virtues which are based on insight, and this in a twofold sense, i.e. for prudence and for wisdom: [t12] But nevertheless it is not really the case that prudence is in authority over wisdom, or over the higher part of the intellect, any more than medical science is in authority over health. Medical science does not control health, but studies how to procure it; hence it issues orders in the interests of health, but not to health. And again, one might as well say that political science governs the gods, because it gives orders about everything in the state. (NicomacheanEthics VI.13, 1145a6-11)

Phronēsis, according to this text, looks how to bring about sophia. Prudence is hence conceived as preparatory and constitutive for wisdom. On the one hand, phronēsis is seen as leading us to the right practical deliberation and choice. On the other hand, sophia is, following Aristotle, the faculty that enables someone to ultimately develop an adequate picture of reality by grasping its principles and elements. According to the first aspect, phronēsis is analogous to political knowledge; according to the second aspect, sophia is the ordering faculty resembling the role of the gods in the universe. Aristotle here subscribes to the thesis that fully adequate thinking goes hand in hand with a fully developed character.

ARISTOTLE ON ADHOMINEMARGUMENTS

155

LITERATURE BATTALY, H. (2010), “Attacking Character: Ad Hominem Argument and Virtue Epistemology” in InformalLogic 30/4, pp. 361-390. CHICHI, G. M. (2002), “The Greek Roots of the Ad Hominem Argument” in Argumentation 16, pp. 333-348. DANCY, R. M. (1975), SenseandContradiction.AStudyinAristotle. Dordrecht/ Boston/London. DORION, L. A. (2013), “Aristotle and the Socratic elenchos” in C. Rapp & P.S. Hasper (eds.), Fallacious Arguments in Ancient Philosophy, Logical Analysis and History of Philosophy 15, pp. 323-343. EEMEREN, F. H. VAN & GROOTENDORST, R. (1993), “The History of the argumentum ad hominem since the Seventeenth Century” in Krabbe, E. C. W. & Dalitz, R. J. & Smit, P. A. (eds.), Empirical Logic and Public Debate. Amsterdam: Rodopi, pp. 49-68. EEMEREN, F. H. VAN & GROOTENDORST, R. (1994), “Relevance Reviewed: The Case of ArgumentumadHominem” in van Eemeren, F. H. & Grootendorst, R. (eds.), StudiesinPragma-Dialectics. Amsterdam: Vale. EEMEREN, F. H. van & GROOTENDORST, R. (1995), “ArgumentumAdHominem: A pragma-dialectical case in point” in Hansen, H. W. & Pinto, R. C. (eds.), Fallacies.ClassicalandContemporaryReadings. Pennsylvania: Pennsylvania University Press. HINTIKKA, J. (2004), Analyses of Aristotle. Dordrecht/Boston/London: Kluwer Academic. NUCHELMANS, G. (1993), “On the Fourfold Root of the ArgumentumAdHominem” in Krabbe, E. C. W. & Dalitz, R. J. & Smit, P. A. (eds.), Empirical LogicandPublicDebate. Amsterdam: Rodopi, pp. 37-47. SCHOPENHAUER, A. (2011), DieKunst,Rechtzubehalten [1830], Hamburg. WALTON, D.N. (1998), AdHominemArguments(Studies in Rhetoric and Communication), Tuscaloosa/Alabama. WALTON, D.N. (2001), “Searching for the Roots of the Circumstantial AdHominem” in Argumentation 15, pp. 207-221. WALTON, D.N. (2004), “Argumentation Schemes and Historical Origins of the Circumstantial AdHominem Argument” in Argumentation 18, pp. 359-368. ZAGZEBSKI, L.T. (1996), VirtuesoftheMind, Cambridge: Cambridge University Press.

REASON AND NATURE IN ETHICS

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES Thomas TUOZZO

For Aristotle, happiness consists in the activity of reasoning well, in either of two ways: in understanding the world or in acting in it. Excellent rational activity in those two areas shares a certain structure: the reasoner grasps certain first principles, and thinks through their explanatory and causal relationship to certain derivative phenomena. The first principles of action are moral values, and excellent practical reasoning consists in taking one of these values as a goal to be achieved and deliberating down to the action that is best suited to achieving that goal.1 In so doing the agent in effect reasons through how that action derives its choiceworthiness in this situation from its relation to her end. This deliberation, together with the ensuing action, constitutes the excellent activity of practical reasoning. In this essay I shall be concerned with the origin and nature of the virtuous person’s grasp of her moral first principles, and with the grasp of those same principles by other sorts of person who, in various ways, fall short of full moral virtue. Now on one widely-held picture, the virtuous person’s grasp of practical first principles is very special indeed. It is the last step, the crowning achievement in her development of practical wisdom and full moral virtue, both of which are attained only when, and as soon as, the appropriate grasp of these first principles is gained. It is my aim in this paper to bring this widely-held view into question, and to argue that the grasp of first principles takes place fairly early in a person’s moral development, as a result of moral habituation, and that that 1 I, like everyone else, hold that much of the time such deliberation consists in specifying what exactly an act instantiating the relevant moral value will consist in. Russell (2014) oddly misreads my argument in Tuozzo (1991), that Aristotelian deliberation is not involved in specifying the major components of happiness (courageous action, etc.), as an argument that deliberation never involves rendering an end more specific.

160

T. TUOZZO

grasp does not undergo any significant change as a person proceeds towards full virtue and practical wisdom. What is required for progress towards full virtue, beyond continuing moral habituation, is learning how to put those principles into practice: that is, what needs developing is the skill at deliberating how to attain one’s moral goals. As she gets better at doing this an agent learns how to determine what acting generously (etc.) comes to in, for example, particularly complicated or sensitive situations. She thereby comes to understands more of what her principles entail; and in that sense, one might say that she has a deeper or richer grasp of them. But this richer understanding is precisely the developed deliberative capacity to move from first principles to what they entail in particular situations. I shall argue that Aristotle does not think that, at the universal level, the fully virtuous person grasps or endorses or sees her principles differently from how she did earlier in her development, nor from the way in which some other, non-fully virtuous agents do so (for example, the un-self-controlled person, or akratēs). Rather, what is different, and especially admirable about her, is her ability to figure out how to put those general principles into practice. What we admire about people like Pericles is their ability to discern what is good for themselves and others2 in the manifold and complicated situations with which life confronts us, not their general views of what is worthwhile in human life. These can be shared, and endorsed in the same way, by many who are not as clever or as virtuous. Before advancing the argument for my position, it will be useful to say a few words about two general considerations that may explain the widespread acceptance of the view that for Aristotle the fully virtuous agent is characterized by a distinctive grasp of moral first principles. One motivation for this view is the parallel between the practical case and Aristotle’s account of theoretical reasoning, in conjunction with a widely-accepted view of how Aristotle thinks of the grasp of first principles there. In Aristotelian science the inquirer attempts to find the explanations for observed regularities in a field of study by tracing them back to their causes in the essences of the substances studied by the science. On the view in question, an inquirer first collects all the general truths of a science, formulated at their proper level of generality. She then proceeds to sort out which are causally prior to which; when she determines which 2

EN1140b5-10.

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES

161

are the first causes of them all, she recognizes them as expressing the explanatory essences of the relevant substances and so as the first principles of the science. On this view, the grasp of theoretical first principles assuch is indeed the crowning achievement that completes the science. If the practical case is parallel, then the grasp of first principles there, too, should come as the development of one’s rational powers is being completed. But there is some evidence that this picture misrepresents Aristotle’s views about theoretical science.3 On an alternative account, we attain the proper definitions expressing ultimate explanatory essences at an early stage of scientific investigation, by induction of universals from experience and their division into genera and species. We have these definitions already in hand as we conduct our further scientific inquiry, which consists in determining what the intermediate causal terms are that link these highest causes to the lower level explananda. If this alternative view is correct, then the progress of theoretical reasoning will in fact have some similarity to the picture I shall sketch of practical reasoning. Another motivation for the common view that for Aristotle the grasp of moral first principles is the last stage of moral development is more strictly philosophical, and is connected with the debate, stretching back to Kant’s reaction to Hume, over whether moral first principles are in some sense derived from or justifiable through reason. Aristotle’s insistence on the importance of habituation in the development of moral virtue, and his repeated claim that moral virtue makes the goal of deliberation correct, have long given support to the Humean side of the debate. An elegant and very influential way of accommodating this material while opposing the Humean interpretation was proposed some years ago by Burnyeat.4 Moral habituation, he argued, does not give us moral first principles; it only gives us the starting-points for moral inquiry. Just as an early stage of theoretical inquiry provides us general truths for which we must seek the causal explanations, ultimately in the first principles of the science, so too moral habituation supplies us with, to use Aristotelian language, the “that” — the moral facts, more or less — for which we must seek the “because” — the reasoned explanation — in fundamental moral first principles. The intellectual work of uncovering as it were the 3 See Bronstein (2016), especially ch. 8, who contrasts his deflationary account with the “explanationist account” defended by, e.g., Charles (2000) and Lennox (2001). 4 Burnyeat (1980).

162

T. TUOZZO

rational justification for the moral facts requires that we be made aware of those facts in the first place, and this requires (unlike in the theoretical case) early moral habituation. But determining what the basic principles are that explain these moral facts is, as in the theoretical case, the work of reason and inquiry. Indeed, Burnyeat holds that studying the Nicomachean Ethics is supposed to bring the student to this systematic and reasoned understanding of her previously unreasoned, low-level moral commitments. This is an attractive story, but its textual support it is not strong. The most convincing evidence for it is a passage in NE I 4, where Aristotle associates having been well brought up in good habits with acquiring the starting points students need before studying (or more accurately, listening to) the NicomacheanEthics. But it is not at all clear that the aim of the NicomacheanEthics is to perfect the practical wisdom of its students, nor is it clear where that work could be thought to deepen one’s grasp of to kalon or the fine, which (as it tells us) is the goal that the virtuous agent seeks to realize. A case can be made that the proximate goal of Aristotle’s treatise is, rather, a kind of theoretical knowledge5 that will be useful to potential statesmen, 6 and that what supplies the starting points for an inquiry of that sort is in fact a grasp of the moral principles that serve as the starting points of the good person’s deliberation. On this view, when Aristotle says in NE I 4 that the person who has been well brought up “has or can easily get” the starting points (1095b7-8), he is talking about the principles that the virtuous person makes the starting points of her deliberation. This is not to say that his students must be perfectly virtuous; for, as I shall argue, possessing and endorsing these first principles in the way that the virtuous person does is not a sufficient condition for full virtue. Much of the point of the debate as to whether Aristotle should be grouped with Hume or Kant on the question of the origin of our moral principles derives from the presupposition that desire is basically noncognitive. Burnyeat’s notion that moral habituation makes us familiar with the “that” — understood, as above, as moral facts — already goes some way to undermining that presupposition. In recent years other 5

See, for example, Nielsen (2017). Note that he there says that “it is necessary for the one who is going to listen adequately to lectures about fine things and just things and ingeneralaboutpoliticalthings to have been brought up finely in his habits” (1095b4-6). 6

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES

163

writers have done so more explicitly, arguing that for Aristotle desire is a multi-sided phenomena comprising cognitive, motivational, and affective dimensions, and that the cognitive dimension is in some sense prior to the others. 7 Desire involves grasping something (whether by perception, phantasia, or thought) as having value for one, and grasping something as having value in this way is in itself both pleasurable and motivating. If, as this view suggests, desire is in its essence a cognitive grasp of value, then the prospect that desire should be involved in our grasp of moral principles no longer so clearly threatens the rational basis of our moral practices. It is clear that we cannot ascribe to Aristotle any discursive justification of moral values or principles. But Aristotle’s not allowing any such justification is consistent with his being a moral realist and his holding that we recognize the truth of moral principles. Moral first principles are grasped by induction from veridical perceptions of moral value, and understanding them as first principles does not require any further discursive reasoning. The place of discursive reasoning in the acquisition of full virtue lies elsewhere. 1. First Principles in the Three Domains of Reasoning Let me return first of all to the parallels in Aristotle’s treatment of the different ways in which human reason can be employed. 8 There are in fact three different ways human reason can be used: in addition to understanding the world and acting in it, we may also use our reason in making things. In each of these areas there are first principles that serve in some way to explain other, derivative phenomena. In the particular case of theoretical knowledge of the natural world, the first principles, namely, the essences of natural substances, serve as explanatory final causes, and the inquirer seeks to understand the parts and properties of animals, as well as their growth and other processes, as directed towards the achievement of these goals. When a natural science like biology is being developed, the first order of business, even before the various genera and species have been sorted out by the process of division, is to determine 7 Authors advancing variations of this view include Tuozzo (1994), McDowell (1998), Achtenberg (2002), Charles (2006), Moss (2012). For most of these authors De an. 431a814 is a crucial text. 8 For good discussions of parallels between theoretical and practical knowing, see Allen (2017) and Charles (2017).

164

T. TUOZZO

what these final causes are. In his own inquiries Aristotle determines these final causes in his identification and discussion of the various life functions in the Deanima. Productive knowledge also consists in knowledge of an essence and of the process which is required to bring something with such an essence into existence. The essence here is not that of a natural substance, but rather of an artefact; but here too, the essence is a certain function. When a craft is first being developed, this function is, to a degree, known in advance: it is because we wish to have a place to shelter ourselves and our property that we attempt to figure out how to build a house. Our experiments with materials, structures, and procedures are evaluated by reference to the way they contribute to producing a satisfactory dwelling. When the art of house-building is finally established, then we have an understanding of what materials must be used and the order in which they must be put together to construct a house.9 We certainly have a much richer sense of what the essence of a house entails than we had at the start; indeed, one might say that possessing the form (eidos) of a house, as the house-builder does,10 just is understanding how to construct it. But this enriched understanding was brought about precisely by figuring out what the essence of a house requires in order to be built: that is, by discovering what derives from the essence of a house. It is perhaps worth noting a difference between Aristotle’s treatment of craft and the other two types of rational activity. While he recognizes the difference between first principle and derived phenomena in all three cases, he does not in the case of craft distinguish between a state that grasps the first principle and a state that grasps the things that follow from it. Craft, science, and practical wisdom are all states of the soul “with a logos”, and in the case of science, at least, Aristotle associates this logos with the first principles that discursive reasoning requires.11 But while Aristotle names psychological states whereby we possess the first principles for science and practical wisdom, he does not do the same for craft. The reason he does not is no doubt because the first principle of an art, the function which its product is supposed to fulfil, is known, grossomodo, all long, and never a matter for controversy. It is in fact the 9 For the parallel between natural and artifactual production, see, among other places, Physics 199a9-19. 10 See Met. 1032b1. 11 See EN 1140b31-35.

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES

165

recognized need for something to serve this function that prompts the development of the art in the first place. The case of practical wisdom is, I suggest, somewhere in between that of craft and theoretical science. In the theoretical case, no one need ever raise the question of how to explain the regularities of nature, much less theorize about the essences that, as first principles, will ultimately serve to explain them. As Aristotle tells us in Metaphysics A 1, such speculation arises only after humans have developed the crafts needed to satisfy our immediate needs (as well as the crafts that help us relax from those labours). In a way, we have no choice but to employ our productive reason: every community needs to find a way to produce shelter, food and the like. There is a similar sort of necessity in the practical case: we need to figure out how to conduct ourselves with others, and what to aim at when we do so. But since these goals are not provided uncontroversially by our physical needs, Aristotle needs to give an account of how we come to grasp them. Aristotle broaches this issue in EN VI after remarking that phronēsis is a disposition “with a logos” (1140b5). He goes on to explain the origin of its first principles as follows: “And this is why we give sophrosune its name, since it preserves practical wisdom. And it preserves a conception (ὐπόληψιν) of this kind. For pleasure and pain do not destroy or distort every conception, such as that the triangle has or does not have angles equal to two right angles, but those concerned with what is done (τὸ πρακτόν). For the first principles of the things done are that for the sake of which they are done, and once someone is corrupted through pleasure or pain the first principle does not appear to him… For vice (κακία) is destructive of the first principle.” (1140b11-20)

This passage is one of several that suggest that moral virtue is what supplies the ends for practical reason. Sometimes the language of “preserving” here has been taken to suggest that something other than virtue — say, some act of intellect — provides the first principles of phronēsis, which moral virtue then safeguards. I would suggest rather that Aristotle uses this language because he believes that actually grasping moral first principles is not in fact so difficult or extraordinary an achievement. Rather, given the right background and upbringing, people will generally acquire these principles, and indeed endorse them as first principles. They need the right upbringing in order to guard against the corrupting potential of the experience of pleasure and pain, and one of the functions

166

T. TUOZZO

of moral virtue is to continue to safeguard our commitment to moral principles. The grasp of moral first principles does not, in itself, bring with it the possession of phronēsis. Rather, when one begins to develop one’s skill in deliberation, just as when one sets about to develop a way to build houses, one must bring with one a conception, grossomodo, of what one wants to achieve. 2. Moral Habituation and Deliberation Practical reasoning differs from the other two kinds of reasoning in that, as directed towards and finding its completion in action, it is intimately involved with desire. Aristotelian desire is a form of cognition, a cognition precisely of value, which, as I have previously mentioned, also has affective and motivational dimensions. But like theoretical cognition, this cognition of value is present in the early stages of human life only at the level of sense-perception and phantasia; it takes some time for our rational cognitive abilities to develop. The early use of non-desiderative sense-perception and phantasia does not affect one way or another our ability to develop a theoretical understanding of the world later on when our ability to reason develops. The same, however, is not true of desire. The desires we experience before our reason develops, and especially the ways we do or do not satisfy those desires, have an effect on the sorts of desires we will find ourselves disposed to have when we do begin to reason. To put it otherwise: if our desires are perceptions of value in the world, satisfying particular desires is a way of reinforcing our attentiveness to the value that appeared to us in those desires. Unlike the theoretical case, where, for example, an unbalanced emphasis on looking at colours will not make it harder to hear sounds, a too ready indulgence in the desires for the satisfaction of our bodily needs can blind us to the other sorts of value in the world to which we would otherwise be sensitive. Hence the importance of moral habituation. For when reason does come, it is on the basis of our accumulated experience of perceiving value in the world that we perform our inductions and form our concepts of what is valuable. And when, somewhat later, we become capable of setting up goals for action that reach beyond the horizon of our immediate situation and require deliberation, the universals that will be available for our doing so will be those we have acquired from this induction.

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES

167

On my view we should not think of the child as originally solely having interest in and perceptual sensitivity to bodily pleasures, so that the task of moral habituation is to divert her interest towards new kinds of value. Rather, moral habituation works by encouraging an already existing natural sensitivity to values other than that of pleasure — in particular, a natural sensitivity, in a rudimentary form, to the fine. This natural sensitivity is what enables moral habituation to get a grip on a child in the first place. Excellent character is fostered by encouraging the child to act on impulses that respond to such value, and to refrain from action on other impulses when they run contrary to these. That Aristotle indeed recognizes some such natural sensitivity to the fine is manifest in his theory of natural virtue; for natural virtue is just the condition of those in whom this natural sensitivity is particularly strong. Aristotle describes natural virtue, and how it differs from virtue in the strict sense, at the beginning of EN VI 13: “Everyone regards each type of character as given us in some sense by nature; for from the moment we are born, we are just and temperate (in a way) and brave; but we look for something else to count as true goodness, wanting such qualities to be present in a different manner. For these natural dispositions (ἕξεις) are present in children and in wild animals, but without nous they are obviously harmful. So much seems manifest, that just as a mighty body moving without sight takes a mighty fall because of being unable to see, the same is true here. But if it acquires nous, it makes a big difference in action. And the disposition, from being like it, will become virtue in the strict sense.” (1144b4-14)

Aristotle here envisions dispositions similar to the developed traits of justice, generosity, and the like, except existing in an individual from birth; indeed, even as existing in children and wild animals. I would suggest that he has in mind the apparent sensitivity that children and animals have to good-making features of the world other than those relative to satisfying their bodily appetites, indeed, their ability in a rudimentary way to recognize the fine.12 This claim is not as far-fetched as it may sound; present-day studies have shown that young children recognize the difference between fair and unfair treatment, and that so do monkeys and other animals.

12 On the possibility that some animals have a rudimentary perception of the fine, see also Moss (2012), p. 210 n. 19.

168

T. TUOZZO

Further evidence for Aristotle’s account of natural virtue can be found in his account of courage, where he distinguishes genuine courage from a kind of courage based on thumos. This latter he calls the “most natural” (φυσικωτάτη) kind of courage, which he says is possessed both by some wild animals and by those human beings he calls “pugnacious” (μάχιμοι). Those with this sort of virtue do not act “on account of the fine” (διὰ τὸ καλόν), but this does not mean that they are not responding at a perceptual level to that which the fully virtuous would conceptualize as the fine. Otherwise, the development of natural courage into full courage that Aristotle envisions would be hard to explain. Aristotle tells us that “the courage that is due to thumos seems to be the most natural sort, and to be courage if reasoned choice (προαίρεσις) and the for the sake of which (τὸ οὗ ἕνεκα) are added” (1117a4-5), and that while the truly courageous act for the sake of the fine, even in their case “thumos works together with them (συνεργεῖ αὐτοῖς)” (1116b31). Animals and pugnacious human beings react on the basis of feeling (δια πάθος) to threats to themselves or others they care about. This is a precursor to real virtue, and certainly could be called, with a slight stretch of language, a nobler sort of reaction than running away. Human beings may on the basis of perceptions of the fitness of acting in this way form a general conception of the value of these actions, and so eventually become to set it up as a goal (οὗ ἕνεκα) and deliberate from it to a reasoned choice (προαίρεσις). This would be grasping the value of the fine as a first principle, and is a necessary step (though not the whole journey) towards full virtue. The idea that natural virtue may play a role in the acquisition of the goals that are moral first principles also appears in NE VII 8, in the course of a discussion of the difference between the un-self-controlled person and the vicious, intemperate person: “Virtue and vice respectively preserve and corrupt the first principle (τὴν αρχήν), which in matters of conduct is the for-the-sake-of-which (τὸ οὗ ἕνεκα), just as in mathematics it is the postulates (ὑποθέσεις). Neither in that case is it argument (λόγος) that serves as teacher of the first principles, nor is it here, but rather virtue, whether natural or habituated, [serves as teacher] of having the correct opinion (τοῦ ὀρθὁδοξεῖν) about the first principle. The temperate person is this sort of person, the intemperate person, the opposite.” (EN 1151a15-20)

Here Aristotle moves from the language of virtue’s preserving first principles, which we saw before, to the language of its teaching them; there

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES

169

seems to be no doubt that virtue is here given sole credit for the origin of our moral principles. (The suggestion sometimes made that “having the correct opinion” may denote a way of grasping moral first principles that falls short of the grasp that the truly virtuous have is belied by the fact that Aristotle here ascribes this grasp precisely to the temperate person.) The fact that Aristotle includes natural virtue together in a disjunction with habituated virtue as a source of moral first principles should be read, I suggest, as shedding light on the nature of moral habituation. Human beings have a natural tendency to recognize moral value. In some fortunate few that tendency is so strong that it provides them with a fund of experience sufficient to allow them to derive by induction a conception of fine action in one or more of the fields of human action, a conception that can serve them as a first principle. For most people, the tendency is weaker, and, unless it is properly supported and reinforced by moral habituation, is liable to be overpowered by the perception of the physically pleasant that is generally strong in all animals. On the view being sketched here, the acquisition of first principles is not a rare or difficult achievement; neither is it sufficient for true virtue. In EN VI 13 (quoted above), Aristotle tells us that natural virtues, when “without nous”, are harmful. Clearly he does not mean that they are harmful to the beasts that have them. Rather, they are harmful to humans who have been able to form first principles of conduct on their basis but do not know how to deliberate in such a way as to achieve them. Nous here does not indicate a special grasp of first principles, but rather, as usual in practical contexts, the ability to recognize, on the basis of deliberation, that a particular action is the way to achieve one’s goal (see EN 1143a32-b5). 13 This ability to deliberate correctly to achieve one’s moral principles is treated by Aristotle as the defining accomplishment of the person with practical wisdom. This ability certainly requires a prior acquisition of those moral principles, but that acquisition is only a first step. Aristotle compares the person with merely natural virtue to a person lacking sight. The sight she is lacking is the sight needed to see her way from the conception of the end to the required action. Earlier in NEVI Aristotle used the expression “the eye of the soul” to characterize the natural ability to deliberate well, considered in abstraction from any 13 For similar interpretations of what the person with natural virtue lacks, see Natali (2001), p. 51-3, Dominick (2006) and Moss (2012), p. 228.

170

T. TUOZZO

particular end to which it might be directed (1144a29-30). Aristotle calls this natural power of practical reasoning cleverness (δεινότης), and developing it so that one may attain the goals one has acquired through natural or habituated virtue is acquiring practical wisdom (φρόνησις). Developing phronēsis is like figuring out how to build a house: it is the acquisition of a rational disposition for attaining a goal that one can grasp without having that disposition itself. But the role of desire — affective and motivating cognition — in practical reasoning is involved not only in the setting of one’s deliberative goals, but also in the acquisition of the deliberative disposition itself. Scholars sometimes suppose that Aristotle thinks of deinotēs as a skill in deliberating that an adult could possess simply as such, having it at her disposal to apply to any end she might happen to have in addition to it. But this is as misguided as thinking that an adult could have the ability to feel anger without having any developed disposition to feel it one way or another. Cleverness may be thought of in abstraction from any given end, but cannot exist in this uncommitted form. The ability to deliberate, like the ability to feel anger, is a capacity (δύναμις), and its development necessarily involves the acquisition of a disposition (ἕξις).14 Some people have a natural talent for deliberating, others find it more difficult. But it is only in using it in the attempt to achieve certain ends that one develops it, and thereby develops the ability to see how the specific value one takes as one’s end might be affected by various constellations of factors. It is a matter of learning to discern the paths to the realization of your values that can be charted through the circumstances of the world. The ability to trace the path of a value in this way is specific to the value pursued. There can be no general cleverness any more than there can be a general skill at producing that makes one equally skilled at producing any product that might be asked of one. Deliberation begins by positing an end, a value to be achieved. But each stage of deliberation involves settling on a more specific version of that end, one appropriate to the particular context. Each stage therefore, as it involves a fresh practical cognition, involves a fresh affective and motivational response. This is, I think, what Aristotle means by calling moral virtue a hexisprohairetikē, a disposition for rational choice. The disposition for choosing does not just set the goal of deliberation; its 14

EN 1144a29-30.

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES

171

sensitivity to value is the necessary background for every deliberative step that one’s practical wisdom takes. Every new determination by intellect must be accompanied by the proper motivational and affective responses. Furthermore, this ongoing background of sensitivity to moral value, and the role it plays in deliberation, helps explain why Aristotle maintains the unity of the virtues, and why he associates this unity with the fact that phronēsis is necessary for the possession of moral virtue. In all deliberation, the agent is trying to move from her general commitment to doing what is fine to what is fine in the here and now. Her starting point will generally be the fine in some particular area: for example, the fine in facing danger, that is, the courageous. But she will need to be sensitive, at every stage of her deliberation, to the ways her contemplated paths of action are related to the fine in the domains of the other virtues, as well: if a contemplated courageous course of action would involve injustice, she needs to have that weigh in her calculations. One may have all the right commitments, and yet still not be sensitive to all their implications in different circumstances and combinations. The development of phronēsis is the development of one’s ability to deliberate, which requires simultaneous cultivation of the sensitivity to moral value in all domains. 3. Possession of Moral First Principles by the Less Than Fully Virtuous This view of the importance of deliberation and of its necessary affective dimension will make it easier to understand how someone without full moral virtue may nonetheless have a non-deficient grasp of moral first principles. We may first take the cases of the self-controlled and of the un-self-controlled (the enkratēs and the akratēs). Aristotle first mentions these figures in EN I in order to show that there is a semirational part of the soul that can resist the part of the soul that reasons about action: “There seems to be another nature in the soul which, while non-rational, nonetheless shares in a way in rationality (λόγος). For we praise the rational prescription (λόγος),15 and the part of the soul that possesses it, of the selfcontrolled person and of the un-self-controlled person. For it urges them on correctly, and towards the best things. Yet there appears to be also some 15 I adopt this translation for logos from Rowe and Broadie (2002), who translate logos in this way in, for example, Aristotle’s definition of moral virtue at 1107a1. In the passage quoted here, however, they employ the traditional translation “reason”.

172

T. TUOZZO

other natural thing in them besides the rational prescription (λόγος), which fights and resists the prescription (λόγος).” (1102b13-18)

Here by logos Aristotle is most likely referring either to the deliberative reasoning the (un-)self-controlled agent produces on a given occasion, or more particularly to the conclusion of that reasoning. And while this may imply possession of the correct first principle, Aristotle does not here explicitly make this point. We find a more explicit statement in connection with the un-self-controlled person in NE VII 8, in the continuation of a passage already quoted above: “There is a kind of person who is carried away by feeling (δια πάθος) against the correct prescription (τὸν ὀρθὸν λόγον), but not to the extent of being the sort of person to be persuaded that one should straightforwardly pursue such pleasures: this is the un-self-controlled person, one who is better than the intemperate person, and not bad without qualification, since the best thing is preserved, the first principle (ἡ ἀρχή).” (1151a20-26).

In this passage, at least, Aristotle seems to say fairly clearly that the selfcontrolled person possesses the proper moral first principle. It has been thought, however, that a different passage, in which Aristotle explains the difference between the ability he calls cleverness and phronēsis, shows that, though the un-self-controlled person may grasp moral first principles, her grasp is in some way deficient in comparison with that of the fully virtuous person.16 In that passage Aristotle writes, “The ability (cleverness) is not phronēsis, but phronēsis does not exist without this ability. But this eye of the soul does not get its [proper] disposition without virtue, as has been said and as is clear: for reasonings out of things done (οἱ συλλογισμοὶ τῶν πρακτῶν) have a first principle…., and this does not appear, except to the good person. For vice distorts and makes one go wrong about practical first principles.” (EN 1144a28-30)

If we take Aristotle to be asserting here that the first principle does not appear to anyone but the fully virtuous person, then, in order to preserve consistency with the passage discussed above that says that in the unself-controlled person the first principle is preserved, we need to suppose that notion of “appearing” has special significance here: while the unself-controlled person has the first principle, it does not appear to her. But since (in the paradigm case) she does take it as her end, it must 16

See Charles (2017), p. 81-84.

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES

173

appear to her in that way. Aristotle does not here explain what other sense of appearing he may have in mind. It is more natural, then, to think that, as the contrast with vice in the last line suggests, Aristotle is not here explicitly considering the case of the un-self-controlled. And this is understandable, since he is discussing the question whether someone who is good at deliberating towards her goals is, as such, phronimos. As we shall see, the non-virtuous nature of the un-self-controlled person’s condition reveals itself precisely in a deliberative failure: her failure to carry her deliberation all the way through to action. It is sometimes suggested that the grasp of first principles possessed by the un-self-controlled person differs from that of the fully virtuous in that the former hold those principles, to use Kantian language, heteronomously: on the strength of her parents’ say-so, or her Sunday school lessons, or the like, without having in some way truly recognized for herself their truth and so in that sense made them her own.17 This idea might seem to be supported by a comparison Aristotle makes between the un-self-controlled and students just beginning to learn a subject: “Those who learn a subject for the first time string together the arguments (λόγους), but do not yet know them. For they must grow to be part of their nature, and that takes time.” (1147a21-22)

But Aristotle makes this comparison when trying to explain the cognitive relation that the un-self-controlled person has, not to her first principle, but to the conclusion of her deliberation, that is, to what she has deliberated her moral principles require here and now. As is notorious, in Aristotle’s analysis of how the un-self-controlled person acts against her deliberated judgment of what it is best to do, it is her grasp on the applicability of her general principle totheparticularcasethat is in some way weakened or undermined. Her grasp of the principle ‘I should not act intemperately’ remains unclouded; it is its application, ‘I really should not eat this pie’, that she loses her grip on (see 1147b15-17). It is her grasp of the particular application that Aristotle likens, not only to the shaky grasp of a learner on her lesson, but also to the relationship of an actor to her lines or of a drunk person to Empedocles’ scientific demonstrations which, because written in verse, she can mindlessly recite. 17 So, for example, Moss (2012), pp. 225-226, whose argument here is uncharacteristically strained: “Aristotle’s thought mustbe …” (my emphasis). See also Charles (2017), p. 81.

174

T. TUOZZO

The self-controlled and the un-self-controlled possess the correct first principles, and begin their deliberations from them;18 but their characters are such that those deliberations do not go smoothly and, in the case of the un-self-controlled, are ultimately disrupted. Human beings are sensitive to different sorts of value, and one’s sensitivity to those values, and the strength of one’s response to them, are in part natural inheritance, in part modified by one’s experience. Moral habituation has (at least) two purposes: the strengthening of our attachment to the fine and the moderation of the strength of our response to non-moral value, especially the value of bodily pleasure. The vicious have so blunted their sensitivity to the fine that it no longer occurs to them to make it an object of pursuit; they are (in the case of the intemperate, for example) sensitive only to the opportunities for bodily pleasure, and deliberate how to achieve them. Those whose nature and/or upbringing have led to their acquisition of moral principles and a commitment to acting upon them, still must learn to be sensitive to the various ways in which those moral goals can be realized. Furthermore, they may have, by nature or as a result of faulty upbringing, strong desires for the pleasant that can conflict with their commitment to the fine. These desires may derail their moral deliberation in a number of ways. In the precipitous, it can make them fail to see the opportunity for the fine in the first place, only to realize their mistake later. For those who do begin deliberating towards the fine, it may make them overlook a possibility of moral action that would involve too great a sacrifice of bodily pleasure. And lastly, even when it does not prevent the agent from seeing on the basis of deliberation that a certain action is the fine one, it may weaken her grasp of that conclusion such that she understands it only as well as a drunk understands the abstruse theories of Empedocles that she can recite by heart. The person who does have self-control is presumably similarly threatened by a weakening of her grasp of the action to be done, but is able to go through with the action in spite of that. I have argued that there is little evidence to suggest that Aristotle thought that the fully virtuous agent is marked by a special way of grasping moral first principles, and that in particular he does not locate the difference between the fully virtuous and the (un-)self-controlled in a 18 Those, at least that do deliberate. Aristotle also recognizes precipitous un-self-controlled, persons, who are carried away by passion to act without calling their correct first principles to mind (see EN 1150b19-22).

ARISTOTLE ON GRASPING MORAL FIRST PRINCIPLES

175

difference between the way they grasp their moral first principles. I shall end with a few words on a different sort of agent who falls short of full virtue, yet whose grasp of moral first principles, again, does not differ substantially from that of the fully virtuous. I have in mind the agent whose upbringing has left her with appropriately restrained appetites for the merely pleasant and a strong commitment to doing what is right (i.e., fine), but still sometimes has difficulty figuring out what the right thing to do is. This is the kind of person that Aristotle seems clearly to have in mind in certain parts of his ethical treatises, for example in NE II 9, where he acknowledges how difficult it is to hit the mean, gives some general advice on how to steer clear of major errors, and encouragingly remarks that small errors are hard to avoid and not too serious. Presumably Aristotle has his (ideal) students particularly in mind: they have attained habituated virtue, and with it a commitment to moral principles, whose true value they appreciate. But having those principles is one thing; acquiring the practical wisdom to see their implications in all the different circumstances of life is another. And though Aristotle’s pointers may be of some help, attending his lecture course on ethics will not suffice to provide you with this wisdom, nor does Aristotle think it will. You may be lucky if a lifetime of experience is enough. LITERATURE ACHTENBERG, Deborah (2002), CognitionofvalueinAristotle’sethics:promiseof enrichment,threatofdestruction, Albany: State University of New York Press. ALLEN, James V. (2017), “Practical and Theoretical Knowledge in Aristotle” in BridgingtheGapbetweenAristotle’sScienceandEthics, edited by Devin Henry and Karen Margrethe Nielsen, Cambridge: Cambridge University Press, pp. 49-70. BRONSTEIN, David (2016), AristotleonKnowledgeandLearning. Oxford: Oxford University Press. BURNYEAT, Myles F (1980), “Aristotle on learning to be good” in Essays on Aristotle’sEthics, edited by Amelie O. Rorty, Berkeley, CA: University of California Press, pp. 69-92. CHARLES, David (2000), Aristotleonmeaningandessence, OxfordAristotlestudies, Oxford: Oxford University Press. CHARLES, David (2006), “Aristotle’s Desire” in MindandModality:Studiesin the History of Philosophy in Honour of Simo Knuuttila, edited by Vesa Hirvonen, Toivo J. Holopainen and Miira Tuominen, Boston, MA, USA: Brill Academic Publishers, pp. 19-40.

176

T. TUOZZO

CHARLES, David (2017), “Aristotle on Practical and Theoretical Knowledge” in BridgingtheGapbetweenAristotle’sScienceandEthics, edited by Devin Henry and Karen Margrethe Nielsen, Cambridge: Cambridge University Press, pp. 71-93. DOMINICK, Yancy Hughes (2006), “Teaching Nature: Natural Virtue and Practical Wisdom in the Nicomachean Ethics” in SouthwestPhilosophyReview 22 (1), pp. 103-111. LENNOX, J. (2001),Aristotle:onthepartsofanimalsI-IV. (Translated with an introduction and commentary), Oxford, UK: Clarendon Press. MCDOWELL, John (1998), “Some issues in Aristotle’s moral psychology” in Companionstoancientthought:Ethics, edited by Stephern Everson, Cambridge University Press, pp. 107-128. MOSS, Jessica (2012), AristotleontheApparentGood, Oxford: Oxford University Press. NATALI, Carlo (2001), ThewisdomofAristotle, Albany: State University of New York Press. NIELSEN, Karen Margrethe (2017), “Aristotle on Principles in Ethics: Political Science as the Science of the Human Good” in BridgingtheGapbetween Aristotle’sScienceandEthics, edited by Devin Henry and Karen Margrethe Nielsen, Cambridge: Cambridge University Press, pp. 29-48. ROWE, Christopher J., and Broadie, Sarah (2002), Nicomacheanethics, Oxford: Oxford University Press. RUSSELL, Daniel C. (2014), “Phronesis and the Virtues (NE vi 12-13)” in The Cambridge Companion to Aristotle’s Nicomachean Ethics, edited by Ronald Polansky, Cambridge: Cambridge University Press, pp. 203-220. TUOZZO, Thomas M. (1991), “Aristotelian deliberation is not of ends” in Essays inAncientGreekPhilosophy, edited by John P. Anton and Anthony Preus, Albany, NY: SUNY Press, pp. 193-212. TUOZZO, Thomas M. (1994), “Conceptualized and unconceptualized desire in Aristotle” in JournaloftheHistoryofPhilosophy 32, pp. 525-549.

‘LEARNING BY DOING’? ARISTOTLE ON THE HABITUATION OF MORAL CHARACTER Jörn MÜLLER

According to Aristotle, ethical virtues like justice and courage are not part of our inborn natural endowment but have to be acquired. The mode of their acquisition seems to be epitomized by the following quotation: “Again, in all the cases where something arises in us by nature, we first acquire the capacities and later exhibit the activities. (…) Virtues, however, we acquire by first exercising them. The same is true with skills, since what we need to learn before doing, we learn by doing; for example, we become builders by building, and lyre-players by playing the lyre. So too we become just by doing just actions, temperate by temperate actions, and courageous by courageous actions.”1 The idea that acquiring the ethical virtues follows the well-known educational principle of ‘learning by doing’, as Aristotle coins it here, is prominent in the Nicomachean as well as in the EudemianEthics.2 As the quoted passage already suggests, this idea is firmly rooted in at least two fundamental Aristotelian convictions. 1) Moral character is acquired by a process of habituation, an ethismos. To put it in Greek terms: êthos is ultimately brought about by ethos. This idea is already present in Plato3, and Aristotle nails down this connection with an etymological pun: “Character (ἦθος) exists, as the name signifies, because it develops from habit (ἔθος), and a thing gets habituated as a result of a pattern of conduct that is not innate, by repeated movement of one sort or another, so that it is eventually 1 Aristotle, EN II 1, 1130a31-b2 (tr. R. Crisp). Here and in the following I have used the translations indicated in the final bibliography, sometimes slightly modified. 2 See, e.g., Aristotle, EE II 1, 1220a22-31. 3 See Plato, Nomoi VII, 792e (τὸ πᾶν ἦθος διὰ ἔθος).

178

J. MÜLLER

capable of being active in that way”.4 This passage provides two further clues concerning our overall topic: firstly, the actions which give rise to moral virtue are not instinctive but have to be learnt in some way; and secondly, moral virtues are not acquired by singular performances of these acts, but habituation requires their repeated practicing, which creates a corresponding character.5 This ties in neatly with Aristotle’s definition of virtue as an instance of the genus “hexis”: this is a well-entrenched disposition which cannot easily be changed or removed because it is the product of many instances of previous acts.6 Aristotle links this notion of stable disposition (as ἕξις may be rendered) very closely with the concept of nature (φύσις) in order to account for the particularly lasting stability which is also visible in natural phenomena: “Frequency creates nature”.7 Thus, moral character seems to be a kind of “second nature”8 — to use another well-known but certainly not unproblematic catch-phrase to which I will return later. 2) By repeatedly acting in a certain manner we acquire an ethical quality, in the same way as we become a flute-player by continually practicing on this instrument. This is obviously the kernel of the second idea which Aristotle cherishes, namely that to become virtuous in a certain area is very much akin to learning the skills which belong to a certain art or craft (τέχνη).9 Aristotle tends to stress the parallel between virtue and craft — well known from Plato’s dialogues — especially with regard to their acquisition, but he also emphasizes other resemblances.10 Quite frequently he uses examples taken from 4

Aristotle, EE II, 1, 1220a39-b3 (tr. B. Inwood / R. Woolf). See also Aristotle, EN III 5, 1114a4-10, where he links the responsibility for character to the repetition of actions. 6 For the notion of hexis in general see Aristotle, Categ. 8 and Met. V 20. For a discussion of the five main features of these accounts and their ethical relevance see Lockwood 2013, esp. pp. 24-30. 7 Aristotle, De mem. II, 452a30 (tr. J.I Beare). 8 Aristotle never states this explicitly or verbally but has been taken to mean something like it by many later commentators and scholars; the usual reference for this idea is EN VII 10, 1152a30-33, where Aristotle quotes approvingly from the poet Euenos. 9 For “learning by doing” as the principle of learning an art see, e.g., Aristotle, Met. IX 8, 1049b29-32. The parallel between virtue and craft acquisition is stressed by Aristotle in Pol. VIII 1, 1337a18-21. 10 A comprehensive analysis of the different passages in which Aristotle compares virtues and crafts is provided by Angier 2010, esp. ch. 5. 5

‘LEARNING BY DOING’?

179

crafts in order to underscore his ideas on moral virtue as well as on ethics as a practical science. One may certainly wonder whether these passages indicate strong structural similarities or only weaker analogies which Aristotle nevertheless deems worthwhile for illustrative comparisons between the two areas of moral virtues and technical skills. There is certainly no need for accusing him of simply confusing moral and technical practice, as we will see later.11 But there is also no easy way around the fact that what inextricably links moral education and the acquisition of crafts in Aristotle’s mind seems to be exactly the idea of learning by repeated doing or practice in the area where the developed moral or technical capacity is also active afterwards. Therefore, we have to take this parallel between moral virtue and craft seriously instead of dismissing it out of hand as a remote echo from the time when Aristotle still belonged to the Platonic Academy. To sum this up briefly: the concept of ‘learning by doing’ in Aristotle’s ethics seems (i) to amount to an inculcation of habits by repeated practising and (ii) to rest on a parallel between virtue and craft acquisition. Although this Aristotelian ‘learning by doing’ idea has fared quite well in pedagogical literature — and especially in circles where moral education is viewed as a kind of practical character training12 — these two core elements of it have been severely criticised in the scholarly literature on Aristotle’s ethics as inadequate or at least highly problematic. In this paper, I will first outline the two main lines of criticism hurled at the underlying idea of habituation as learning by doing (part 1). Afterwards, I will offer some remarks and observations on the two main ideas articulated above, first addressing the relationship between moral virtue and craft and subsequently discussing the idea of repeated practice as the basis of moral habituation (part 2). In the conclusion (part 3) I will draw 11

See below, section II.1. Carr (2008), p. 106, describes the attractiveness of the Aristotelian idea of habituation for modern advocates of character education as follows: “Aristotle’s emphasis on training in certain precisely specified moral dispositions seems to offer the way to a more practical no-nonsense conception of moral education. (…) moral education is first and foremost a matter of initiation — under the instruction and guidance of parents, teachers and the general community — in socially approved standards of conduct.” For the approach to moral education in terms of virtue ethics see the contributions to Carr / Steutel (1999). 12

180

J. MÜLLER

a balance sheet of my considerations with a view to some influential interpretations of moral habituation in Aristotle (especially concerning the view of Nancy Sherman). The issues involved in these matters are on the whole too far-reaching and in detail also too thorny to be addressed sufficiently in such a paper. Nevertheless, I intend to offer a defence of the two cornerstones of Aristotelian ‘learning by doing’ against some prominent criticism of it. This stance is based on two convictions on my behalf which I will try to substantiate in the course of my paper: firstly, that ‘learning by doing’ is ultimately a fitting description of what Aristotle has in mind when it comes to the habituation of character; and secondly, that this conception captures some important aspects of moral development which contemporary ethical theory as well as the present philosophy of education should still take seriously. 1. Is ‘Learning by Doing’ a Misguided Conception of Moral Habituation in Aristotle? The question raised in the title of this section is to a certain extent ambiguous. It might be disambiguated in one of the following ways: on the one hand, it may insinuate that the interpretation of Aristotle’s remarks on habituation in terms of ‘learning by doing’ is — notwithstanding the quotations above — actually a misnomer or a misleading understanding of what Aristotle wants to tell us. On the other hand, it might also be taken as follows: even if Aristotle truly adhered to such a conception, he was misguided in doing so, because it is inadequate as an explanation of moral habituation. The critical remarks in the scholarly literature which I will address in the following are mostly inspired by one of these two concerns. But I will not divide them along this line but rather according to the two cornerstones of Aristotelian ‘learning by doing’ as laid out above. 1.1 VirtueasourSecondNature?ThePitfallsofMoralHabits Aristotle leaves no doubt that acting virtuously has as one of its fundamental conditions that the agent acts “from a firm and unshakeable character”13, a disposition or habit (ἕξις), which is produced by repeated 13

Aristotle, EN II 4, 1105a33.

‘LEARNING BY DOING’?

181

previous actions. Now, psychological habits tend to reproduce patterns of fixed behaviour like getting up early or drinking a pint at the pub after office hours.14 Aristotle even explicitly seems to support this idea when he states with respect to the moral virtues “that the activities that flow from them will consist in the same things”15 as they originate from. There seems to be a causal relation at work here in the production of virtuous acting from habit.16 This idea is also strengthened by the comparison of the working of an acquired habit with natural processes: when Aristotle frames recollection as a habitual event, the point is that the underlying habit reproduces the chain of thought in an effortless and efficient way, established by the same psychological associations.17 Nature always (or at least in most cases) produces the same results,18 a claim to which natural capacities stand testimony: fire always heats when it meets something which has the capacity of being heated — it simply cannot do otherwise. Now when Aristotle stresses that moral habituation creates such habits that are “virtually natural”,19 he seems to encourage the following view: ethical virtues are fixed habits which in a rather mindless manner repeat over and over again what has been acquired by previous practice. As soon as the occasion arises, the habit will just naturally “kick in” and reproduce the required activity without further planning or thought. This picture of acting from habit, although it does not seem to be utterly repellent to me at first blush in application to the moral sphere — who would not want a person to come to the aid of a drowning child at once 14 Aristotle explicitly names the plucking out of hair and the gnawing of nails as examples of bad habits which result from ethos; see EN VII 5, 1148b27-28. 15 Aristotle, EN II 2, 1104a29. 16 See Aristotle, Rhet. I 10, 1369b6-7 (tr. W. Rhys Roberts): “Acts are done from habit which men do because they have often done them before”. 17 See Morel (1997), pp. 138-140, who offers an insightful analysis of the causal relations between nature and habit. 18 For regularity as a characteristic feature of Aristotelian phusis, see Müller (2006), p. 18f. 19 See Aristotle, Rhet. I 11, 1370a6-7 (tr. W. Rhys Roberts; slightly modified): “For as soon as a thing has become habitual, it is virtually natural. For habit is similar to nature; what happens often is akin to what happens always, natural events happening always, habitual events often.” Note that Aristotle still sees a difference in the frequency of habitual and natural events (see also De mem. 2, 452b10-15) — therefore ethos is only “similar to” or “like” nature — there is no identity but also certainly more than a mere analogy, as Morel (1997) proves. Habits are also changed more easily than nature; see EN VII 11, 1152a27-30.

182

J. MÜLLER

and without hesitation? —, is not easy to square with the description that Aristotle gives of a full-blown virtue. Several problems arise immediately: a) Aristotle defines moral virtue as a ἕξις προαιρετική, i.e. a disposition resulting in acts of rational choice (προαίρεσις).20 Now choice — and one is inclined to say: especially moral choice — usually involves a decision between different courses of action available in a certain case. This is one of the reasons why we need practical deliberation (βούλευσις) in such situations in order to find out what is the best option. But how could such a deliberating and deciding capacity be instilled by more or less mechanical repetitions which simply issue their own causal reproduction? The idea of ‘habituated decision’ seems as paradoxical as that of ‘habituated reason’.21 b) Apart from rational deliberation, choice also entails certain forms of desire (ὄρεξις) which belong to the irrational part of the human soul and are manifested in emotions (πάθη). In consequence, Aristotelian moral virtue is not only concerned with actions but with passions too; it is also a disposition to feel in a certain way.22 Being courageous means not only to stand one’s ground in battle (i.e. to act in the appropriate manner) but also to experience the right amount of the passions of fear and confidence in the situation. But it is not easy to see how ‘learning by doing’, understood as a habitual exercise of certain behavioural patterns, will simultaneously instil the passions required for truly virtuous action.23 These two arguments point in the same direction: looking at the way in which Aristotle describes the context-dependency of virtuous actions and the corresponding feelings, it is hard to fit this in with the rather confined set of patterns of conduct which a dispositional habit usually consists of. The virtuous person is quite flexible when it comes to the requirements of a moral situation, while habits are rather fixed and seem to be the product of a kind of behavioural — or even behaviourist — conditioning of the agent. So if we think that moral habituation helps us to acquire and to understand full-blown virtues, there seems to be a recognizable gap 20 See Aristotle, EN III 2, 1111b5-6 (tr. R. Crisp): Choice “is thought to be very closely tied to virtue, and a better guide to character than actions”. 21 For a nuanced discussion of the problem of habituated reason in Aristotle see Kristjánsson (2007), ch. 3. 22 See Kosman (1980). 23 See Steutel/Spiecker (2004), p. 537f.

‘LEARNING BY DOING’?

183

here. For this reason, the view that Aristotle understands character formation as developing a rather “static” moral habit has been criticised heavily in favour of a more “dynamic” picture of this process and its outcome.24 The most prominent counter-attack on the idea that Aristotelian habituation is based on non-cognitive behavioural conditioning was launched by Nancy Sherman, who argues instead for a much more sophisticated concept of “critical habituation” or repetition, to which I will briefly return at the end of my paper.25 1.2 VirtueasaSkill?TheDrawbacksoftheComparisonBetweenἀρετὴ ἠθικήandτέχνη The second criticism of ‘learning by doing’ focuses on the frequent parallels drawn by Aristotle himself between virtue acquisition and the learning of arts or crafts (i.e. τέχναι) like flute-playing or house-building, which rely on purely technical skills.26 Two standard objections run as follows here. a) Aristotle himself distinguishes two kinds of acquisition of virtues: while ethical virtues are acquired by habituation, the intellectual virtues are learnt by teaching.27 Now technê is labelled unequivocally as an intellectual virtue by Aristotle in EN VI 4, and thus it has to be acquired in a very different mode from the moral virtues, namely by being instructed in the technical rules which one has to follow to achieve the right result. Thus, the parallel between moral virtue and technical skills used by Aristotle seems misleading with respect to the rather different way in which they are brought about according to his own account.28

24

See e.g. Lockwood (2013). See Sherman (1989), esp. ch. 5. She summarizes her approach towards this topic as follows: “Contrary to the popular interpretation according to which ethical habituation is non-rational, I argue that it includes early on the engagement of cognitive capacities. Thus, habituation is not mindless drill, but a cognitive shaping of desires through perception, belief and intention. These capacities are involved in acting from character, and, to a different extent and degree, in acquiring character. Thus, moral education will itself cultivate the perceptual and deliberative capacities requisite for mature character” (p. 7). 26 For Aristotle’s understanding of virtue and skill acquisition see Carr (2008), pp. 106-108. 27 See Aristotle, EN II 1, 1103a14-18. 28 See Stichter (2007), who argues that Aristotle ends up with a conception of virtue which comes perilously close to that of a skill. 25

184

J. MÜLLER

b) Secondly, the parallel also seems misleading regarding the kind of activity which is taught. Although the ethical habits are distinguished with regard to several spheres of activity, their actual range seems to be much broader than in the case of specialised arts or crafts, which might be reduced to a certain set of rules and a repertoire of standard activities safeguarding the quality of the product.29 Aristotle himself marks a significant difference between two kinds of action structures under the headings of poiêsis and praxis: while skills enable their possessor to produce something, an ergon, which is different from the activities which bring it about,30 moral virtues embody another kind of human action, which is self-sufficient in the sense that it is not oriented towards any further achievement apart from the action itself.31 And Aristotle states unambiguously that production and action are quite different so that neither of them is included in the other;32 in consequence he argues himself that “the case of skills is anyway not the same as that of the virtues. For the products of the skills have their worth within themselves, so it is enough for them to be turned out with a certain quality. But actions done in accordance with virtue are done in a just or temperate way not merely because they have some quality of their own, but rather in virtue of the agent acting in a certain state, namely, first, with knowledge, secondly from rational choice of actions for their own sake, and thirdly, from a firm and unshakeable character.”33 While the evaluation of production is dependent on the technical quality of the achieved external results, moral praise or blame is not accorded to actions solely on the basis of what is done: the moral judgment of an action is very much due to the way how it is done. And this way of doing something is to be 29

For the general differences between virtues and skills see Halbig (2013), pp. 73-78. See Aristotle, EE II 1, 1219a13-17 (tr. Inwood / Woolf): “But ‘function’ has two senses. Some things have a function that is over and above their use. For example, the function of the builder’s art is not building but a house, and the function of the art of medicine is not healing or treating, but health. With other things, their use is their function. For example, the function of sight is seeing and the function of mathematical knowledge is studying”. 31 See Aristotle, EN VI 5, 1140b6-7 (tr. Crisp): “For while production has an end distinct from itself, this could not be so with action (praxis), since the end here is acting well itself”. 32 See Aristotle, EN VI 4, 1140a1-6. At 1104b4, he states that making and action belong to a different genos. 33 Aristotle, EN II 4, 1105a26-33. 30

‘LEARNING BY DOING’?

185

characterized by the internal state of mind of the agent, which can be described by several conditions as given above: knowledge, purity of choice and stable disposition.34 Especially the “pure choice condition” certainly appears to mark a difference to skilful actions which are done for the sake of an external product. I will come back to these conditions later in some more detail (in section II). The fundamental difference between the poietic structure of crafts and the practical character of moral virtues is thus mirrored in two kinds of actions which seem to differ in their ontology,35 in their means-end-orientation as well as in the motivational structure they require on behalf of the agent: in the case of skills it is not guaranteed that they will be used by the agent when an opportunity to do so arises — it is not even certain that they will not be misused. In Metaphysics IX, Aristotle portrays craftlike skills as “rational potencies” which can be used both ways. The trained doctor knows how to cure but he also knows how to make someone effectively ill.36 The way in which he uses his medical knowledge is not determined by the possession of the artful knowledge but rather by his decision (προαίρεσις) how to act.37 It is exactly here that Aristotle pinpoints an important difference between skills and ethical virtues: “In skill the person who misses the mark voluntarily is preferable, but with practical wisdom (φρόνησις), as with the virtues, the reverse is true”.38 Concerning moral character, the decision to go wrong deliberately is not a mark of expertise; it is rather the signpost of a motivational deficiency. Obviously, the second nature built up by ethical ‘learning by doing’ is thick enough to prevent its misuse and to safeguard correct 34 Another internal condition for virtuous action which Aristotle does not name here but discusses in other places is that the agent acts with pleasure; see, e.g. EN II 3, 1104b3-9. This condition can be used to make a difference between the virtuous and the strong-willed (ἐγκρατής) agent: the latter one acts according to his rational desire but has to keep his recalcitrant emotions at bay, thus “spoiling” his emotional experience of doing the right thing. 35 See also the fundamental differences between kinêseis and energeiai which Aristotle marks in Met. IX 6, 1048b18-36: while they are still underway, processes are incomplete, whereas activities are at every time of their being complete in themselves. Tellingly, Aristotle uses the different activities involved in the craft of house-building as an example of an incomplete process (κίνησις); see EN X 4, 1174a19-29. Virtues, on the other hand, issue in energeiai; see, e.g., EN I 8, 1098b30-31. 36 See Aristotle, Met. IX 2, 1046a36-b7. 37 The use of rational potencies is always dependent on a previous desire (ὄρεξις) or choice (προαίρεσις); see Met. IX 5, 1048a1-13. 38 Aristotle, EN VI 5, 1140b22-24 (tr. R. Crisp).

186

J. MÜLLER

actions: when a moral disposition is in operation, the result is only of one kind,39 while rational potencies can produce opposite effects. The contrast drawn in the last quotation between skill and practical wisdom appears to provide further confirmation of the unbridgeable gulf between arts and crafts on the one hand, and moral character on the other hand: while craft consists in a poietic habit bent on production, phronêsis — being the intellectual virtue which is closely tied to the successful operation of the moral virtues — is a practical habit issuing in self-sufficient action.40 Not surprisingly there are also significant differences between technical and moral deliberation,41 notwithstanding the fact that Aristotle has a recognisable penchant for treating them in the same context (e.g. when discussing certain forms of practical syllogisms42). In conclusion: the two kinds of habits engendered in skills and in ethical virtues seem so entirely different from one another in their workings that it is hardly conceivable that they are instilled by the same (or even a similar) mode of ‘learning by doing’. Sherman summarizes this as one of the basic limitations of Aristotle’s parallel between craft and virtue acquisition: “acquiring an art, such as grammar, will be primarily a matter of internalizing certain procedural principles and producing a product that embodies that procedure (or knowledge). Virtue, on the other hand, will be a matter not of learning implicit procedures, but of having reliable motives, expressed in chosen actions which come to have intrinsic value. The actions will not be chosen by procedure, nor will what is brought about be valued apart from the actions that realize it”.43 2. A Defence of Aristotelian ‘Learning by Doing’ In my opinion, most of the criticisms sketched above make valuable philosophical points on Aristotle as well as on the question of moral 39 See Hutchinson (1986), pp. 35-38, for hexeis as dispositional traits or properties which never cause opposite results, in contrast to technêand dynamis. See esp. EN V 1, 1129a11-16 (tr. R. Crisp): “What is true of sciences and capacities is not true of states (ἕξεις), since it seems that contraries can both be the concern of the same capacity or science, while a state does not produce results contrary to itself. For example, as a result of health, we do not do actions contrary to health, but only those that are healthy”. 40 See Aristotle, EN VI 5, 1140b4-7. 41 See Broadie (1991), pp. 202-212. 42 See the examples of “technical” deliberation given in Demotuan. 7, 701a12-21. 43 Sherman (1989), p. 189.

‘LEARNING BY DOING’?

187

habituation. Nevertheless, I think they tend to overstate the case against ‘learning by doing’ in general as well as regarding Aristotle’s depiction of it. Therefore, I would like to take the bull by the horns and address them in a defence of this conception which will also require some efforts in interpreting Aristotle’s core ideas in the two main disputed areas as sketched above. I will discuss them in reverse order. 2.1 Practice,ExperienceandKnowledge:theAcquisitionofCraftsand Virtue One of the basic questions one has to address in interpreting Aristotelian ‘learning by doing’ is the following: why does Aristotle rely so heavily on the comparison between skill and virtue acquisition despite the considerable differences between production and practice which he himself demarcates and which were delineated in section I? I think that he has got several reasons for sticking to this parallel, some of them explicit, some implicit, which I will try to elucidate subsequently. First of all, it has to be noted that the favourite context in which Aristotle highlights ‘learning by doing’ as the common core of skill and virtue acquisition is connected with a sophistic puzzle: “Someone might, however, wonder what we mean by saying that becoming just requires doing just actions first, and becoming temperate, temperate actions. For if we do just and temperate actions, we are already just and temperate; similarly, if we do what is literate or musical, we must be literate or musical”.44 This suspicion of circularity is a serious aporia for any ‘learning by doing’-conception: how can we become capable of performing certain actions if the acquisition of this capability hinges itself on our previous performance of these acts — which seems to presuppose that this capability is already in place? Aristotle hints at least at two possible strategies to address this apparent dilemma. (a) One might distinguish, in the case of apprentices as well as in the case of the moral learners, between ‘doing skilled things’ and ‘doing them skilfully’ or between ‘performing virtuous actions’ and ‘performing them from virtue’.45 For example: The student may act on the explicit verbal instruction of a teacher; it is also possible that the teacher delivers 44 45

Aristotle, EN II 4, 1105a17-21. See Aristotle, EN II 4, 1105a21-26.

188

J. MÜLLER

a model performance which the student has to imitate step by step.46 So the actions of the student will not spring from an already formed habit or from his own understanding. In both cases there is an eminent need for a qualified teacher because a simple “trial-and-error”-approach might just backfire if the student gets accustomed to doing something in the wrong way: he still lacks the required expertise and he has no one who can judge his performance competently from the outside.47 This kind of “doing something under instruction and supervision”, which does not rely on an already existing habit on the part of the learner (but certainly in the teacher), may solve the aporiaof circularity. With a view to the acquisition of skills and virtues as dispositions, this might also be a significant first step in the learning process, and it cannot be seriously doubted that the repeated performance of certain tasks enables a student to do them more proficiently over time: guided repetition surely enhances efficiency. But this model does not naturally explain the transition between ‘learning to do something’ and ‘learning how to do it in a certain manner’ because it is ultimately restricted to relatively limited exterior actions like ‘repairing a bike’ or ‘helping an old lady to cross the street’. And an exclusive focus on outward actions misses something important about what has to be learnt “on the inside” in order to acquire the capability to its full extent. In acquiring a skill or a craft, the student will also learn the general rules which govern the procedures in which he partakes in exercising it: what turns him into an expert is not simply the ability to reproduce certain patterns of behaviour with the right results but the productive knowledge accompanying this capability: “Someone will be literate, then, when he produces something literate and does so in a literate way, that is, in accordance with his own literacy”.48 The condition of literacy as an example of a skill is the possession of knowledge, and this ‘knowledge condition’ certainly pertains not only to the absence of errors about the particular circumstances of this action, but also to the knowledge of what he does in terms of general rules. The dependence of a truly skilful performance on an internal condition of the agent, which is crucial in bringing it about, 46

For the role of the teacher in moral habituation see Steutel/Spiecker (2004), pp. 544-

546. 47 On the influence of teachers on the acquisition of virtues and skills see EN II 1, 1103b6-13. 48 See Aristotle, EN II 4, 1105a 23-26

‘LEARNING BY DOING’?

189

highlights an important parallel between moral and technical performances. In one passage, Aristotle discusses certain uncourageous and unjust acts, like fleeing in a battle or having sex with a (married) woman. He comments on this: “But to act in a cowardly or unjust way is not to do things of this kind, except incidentally, but to do them on the basis of having a certain character. In the same way, being a doctor or curing a patient is a matter not merely of operating or not operating, of prescribing or not prescribing, but of doing them inaparticularway”.49 Thus the performance of truly skilful as well as of truly virtuous actions cannot be separated from the internal conditions of the agent; and this is true in the case of crafts as well as in the case of moral virtues. There is a certain tendency in the scholarly literature to neglect this structural parallel between productive and practical performances because Aristotle downplays outright the ‘knowledge condition’ in the performance of moral virtues, deeming it less important in comparison with the two other conditions he specifies for moral practice, namely the ‘pure choice condition’ (that the action is done for its own sake) and the ‘dispositional condition’ (that the agent acts from an unshakeable character).50 These two other conditions do not hold for productive performances, as Aristotle explicitly tells us,51 and this explicit denial has led some commentators to diagnose an unbridgeable gulf between productive and practical performances as well as between technê and moral virtues. I will come back later to this point but for the moment it suffices to state that the difference which Aristotle makes here only works because it is based on atertiumcomparationis in which both kinds of action share (albeit in different form): truly skilful as well as truly moral performances are dependent on internal conditions of the agent which are not reducible to the outward action. 49

Aristotle, EN V 9, 1137a21-26 (my emphasis). For the understanding of the three internal conditions (knowledge, pure choice, stable disposition) which Aristotle names for virtuous action in EN 1105a30-33 see Taylor (2006), pp. 84-93. 51 See Aristotle, EN II 4, 1105a33-b2. Aristotle does not elaborate on this statement, but the background for it seems to be provided by some of the considerations already offered above, in section I.2: the “pure choice condition” does not sit too well with any form of instrumental action involved in skills and crafts; and the “dispositional condition”, which safeguards that a certain type of action is always produced, cannot easily be squared with the understanding of skills and crafts as rational potencies (see Met. IX 2 and 5) which may be used to produce opposite results. 50

190

J. MÜLLER

(b) Which lesson has to be drawn from this? ‘Learning by doing’ cannot be understood as a pure repetition of outward acts but also has to involve the cultivation of interior conditions in the agent. This idea touches on a second strategy which Aristotle may use in order to overcome the sophistic aporia. According to him, learning is a gradual process, a kinêsis,52 which is headed toward a certain completion. While it is still on its way, it is incomplete, but every stage of this gradual development already presupposes the previous stage(s) — there is an incremental change at work here which does not happen out of the blue and which is analysed on a theoretical level by Aristotle in his Physics.53 Let us transfer this idea first to the sphere of productive knowledge: the student of an art or a craft is still in the process of acquiring it to the full extent, but the performances of which he is capable right now are already based on a certain — if still incomplete — amount of ‘knowing how’ which he has acquired before. Now, extending and refining this already existing ‘knowing how’ by further performances and by the learning processes which accompany them is always possible: one may learn from eventual failures as well as from cues and tips given by a teacher, and the effected increase in knowledge will carry the process further towards its completion. Expertise develops in degrees54 — and looking at current developmentalist views of moral competence (e.g. Lawrence Kohlberg’s stages of moral judgment55), this idea receives some confirmation in contemporary educational theory.56 But there is still a gap: in order to support the idea that ‘learning by doing’ is the central element in such a developmentalist account, one has to show that the change in the internal conditions of the agent is intricately linked to, and somehow even caused by, the repeated outward performances. But this does not sit too well with the Aristotelian position 52

See Aristotle, Met. IX 6, 1048b29-30. See Aristotle, Phys. VI 6, 236b32-237a17: everything which is in a process of change has already completed an infinite number of changes. Aristotle alludes to this idea in Met. IX 8, 1049b32-1050a3, in order to solve the Sophistic puzzle — mirrored in Plato’s discussion of the paradox of inquiry at Meno 80d-e — that in order to learn something you must already know it somehow. This problem is in Aristotle’s view linked to the problem of how “learning by doing” can be understood; see ibid., 1049b29-32, and EN II 4, 1105b17-21. 54 See Makin (2006), p. 190. 55 See Kohlberg (1981). 56 For “developmentalist” accounts of Aristotle’s philosophy of education see Tobin (1989), Reeve (1998), Kristjánsson (2007), and Lawrence (2011). 53

‘LEARNING BY DOING’?

191

that technê as an intellectual virtue is acquired by theoretical instruction (διδαχή). Now, I think that this statement has to be qualified to a certain extent, because in several passages Aristotle himself seems to be rather critical of a ‘theory-first’ approach towards craft: without practical experience no one will benefit from theoretical teaching on skills.57 Anyone who has acquired the skill of driving a car will agree on this. It does not really help to study the manual of a vehicle or to memorise the traffic regulations before one has actually sat in the driver’s seat and gathered some practical experience in steering, accelerating, braking, etc. This is at first glance a very commonsensical idea, but it lies at the heart of Aristotle’s understanding of ‘learning by doing’ in the area of crafts and skills: the process of acquiring expertise in an art or craft is rather a kind of boot-strapping between practical experiences and working out general rules which govern this practice. The knowledge which is to be gained in a technê may differ considerably, depending on the level of complexity involved in the activity (becoming a doctor will certainly be more demanding than learning how to play the flute); but without some practice it will be impossible to get started or to proceed in the incremental manner as described above. To put it in epistemological terms: the student has to figure out universal concepts (like health in medicine) in the particular instances as well as to make up the general rules out of singular performances. Now, even after one has learnt “in practice” and “by heart” the general rules of a technê, it still often proves a tricky business to apply them correctly in a particular situation. There is no mechanical or automatic process in play here, as K. Kristjánsson has rightly emphasized: “a considerable portion of his treatment of technêdefies the rigid stereotype of the unreflective artisan. Some of Aristotle’s most vivid examples of the practitioners of technê involve not potters and carpenters, but medical doctors, army generals, navigators and performing artists”.58 To take up 57

See, e.g., Aristotle, EN VI 12, 1143b25-28; X 9, 1181b2-6. Kristjánsson (2007), p. 164. It has to be mentioned that Aristotle makes a difference in Met. I 1, 981a24-b13, between master-arts (ἀρχιτεκτονικαί) and purely manual crafts (χειροτέχναι). The latter act like natural potencies, merely on the basis of ethos (981b45) and experience, while the former enable their possessor not only to know what to do but also to grasp the causes and therefore to teach their profession to others. The contemporary criticisms leveled against a mechanical understanding of “learning by doing” may apply to the manual crafts as they are depicted here, but they do not hit the more sophisticated technai, which Aristotle obviously has in mind in his parallels between virtue and craft acquisition. 58

192

J. MÜLLER

this cue: the medical practitioner has to deliberate about the question of how to treat the patient at hand, with his particular bodily state, in the best way. This process involves bouleusis not only in the sense of technical deliberation but also with respect to choosing between various options of treatment available in this case.59 These key elements of performing the medical profession are not only helped by previous experiences, they even presuppose them to a certain degree, e.g. when it comes to spotting the salient features of the bodily state; it may even be the case that the experienced GP will recognize an individual exception to an otherwise well-established diagnosis or therapy. Aristotle’s bottom line is this: productive knowledge aims at certain forms of practice which take place in varying circumstances. Therefore, the practitioner as well as the student of it has to pay sufficient attention to the contingent details involved in this performance. And there is truly no way around acquiring the required experiences in practice. This is the more sophisticated meaning of ‘learning by doing’ in the area of crafts, instead of thinking of it as a mindless repetition of manufactural acts under constant supervision by a teacher or an expert. From a certain point onwards, the student will ultimately be the independent judge of his own performances and will be able to effect the incremental increase of productive knowledge mostly by himself (but still based on continuing practice). Now Aristotle describes medicine explicitly as a στοχαστικὴ τέχνη, and he also uses the term stochastikê conspicuously in his ethics to designate the ability of hitting the mean in actions under particular circumstances, which is the hallmark of moral virtue.60 This is no easy task, since the “mean relative to us” is of a quite complex nature, dependent on the agent and his individual conditions as well as on contextual circumstances. In consequence, there are many ways to go wrong and only one way to get it right,61 and this marks a fundamental difficulty in the moral area. According to Aristotle, crafts face a similar situation, and their best products are also characterized by hitting the mean: you cannot add to them or detract from them without diminishing their quality.62 59 For the deliberation of medical professionals see Aristotle, EE II 9, 1226a33-b2 and 1227a15-22. For the importance of deliberation in different forms of technai see EN II 3, 111a34-b20. 60 Compare Jaeger (1957) and London (2001), pp. 571-575. 61 See Aristotle, EN II 6, 1106b28-35. 62 See Aristotle, EN II 6, 1106b5-14.

‘LEARNING BY DOING’?

193

The basic difficulty is that hitting the mean in the moral as well as in the technical sphere is not possible by simply following completely standardized rules which only have to be applied to the individual situation at hand, but rather requires a genuine perception of what is needed in this particular case.63 The mere possession of general knowledge is thus insufficient for hitting the mean, so that “in order to do this, we must acquire a knowledge of particulars gained from an involvement in the actions of life”.64 The acquisition of moral virtues as instances of hitting the mean presupposes, as do the skills, a fine-tuned experience with regard to the salient features of a situation. Aristotle does not tire of stressing that actions in the sense of ethical praxis have to do with particulars65 and thus the habituation of moral dispositions is essentially linked with gaining this kind of experience which simply cannot be replaced by any kind of class-room instruction. This is one of the reasons why Aristotle deems young people unsuited as an audience for political science — they are simply too inexperienced (ἄπειρος) in practical matters to profit from theoretical teaching in this area.66 Experience also plays a crucial role in shaping the practical wisdom of phronêsis insofar as its deliberation has to do with the ultimate instances of human action which reside in the particular.67 And ethical experience is only to be gathered from being confronted with morally relevant situations over and over again — in other words, by frequent and repeated practice in this area. This requires above all time, so that a habitual knowledge can arise from it and become ingrained in the agent.68 Thus Aristotle opts for an ‘experience first approach’ in moral education as well as in the learning of skills which is due to the particular structure of the spheres they have to deal with: “Furthermore, in the case of every capacity and craft there are certain things, having to do with the work each is to accomplish, that must be a matter of prior education and prior habits, and so obviously this applies also to the activities of 63

See Aristotle, EN II 9, 1109a24-30. London (2001), p. 583. 65 See e.g. EN II 2, 1104a3-10; II 7, 1107b28-32. 66 See Aristotle, EN I 3, 1095a2-4. For the importance of experience in politics see also EN X 9, 1181a9-12. 67 See Aristotle, EN VI 8, 1142a14-15. 68 See Aristotle, EN VII 3, 1147a21-22 (tr. R. Crisp): “[T]hose who have just begun to learn can string words together, but do not yet know; it must grow into them (συμφυῆναι), and this takes time”. 64

194

J. MÜLLER

virtue.”69 And he sees a significant parallel in their “stochastic” mode of operation conditioned by repeated practice.70 This may well be viewed as an (indirect) criticism of a Socratic conception of craft in its technical and moral sense which tends to dissociate it from the notion of experience (ἐμπειρία).71 In Aristotle, technê and moral virtues substantially arise from previous experience,72 and they cannot be dissociated from it in their later performances. Therefore, Aristotle is sceptical of the Socratic approach towards moral virtues, according to which the question of how the virtues are to be acquired can only be settled after a definitional account of the concept of “virtue” has been given. Aristotle reverses the procedure: in order to establish what virtue truly is, we first have to understand how it is acquired.73 After all, Aristotle explicitly contradicts the Socratic dictum that moral virtues are simply identical with forms of knowledge,74 thus indicating a certain reservation towards stronger forms of ethical intellectualism. In my opinion this distancing from — Socratic or Platonic — intellectualism is also a direct consequence of virtue acquisition by a moral habituation which is not focused on theoretical learning but on ‘learning by doing’ (in the sense of undergoing practical experiences, issuing in an ability correctly to discern particular situations and their requirements). Theoretical knowledge alone does not enable us to act, neither 69

Aristotle, Pol. VIII 1, 1337a18-21 (tr. R. Kraut). London (2001) argues that there is a change from EE to EN: While in the early EE Aristotle leaves open the door for a “theory first”-approach, the more mature EN opts decisively for an “experience-first” approach. I am not convinced by this idea; the tendency to criticize Socrates’ view of ethics as a kind theoretical science is already present in the EE See e.g. the quotation below, in footnote 73. 71 See especially the discussion of Socrates with Polus in Plato’s Gorgias 462b-c, where he argues that rhetoric is not at all a technê but amounts only to experience (ἐμπειρία). For technê in the dialogues of Plato, see Angier (2010), ch. 1. 72 For the role of experience in the epistemological hierarchy of activities see Aristotle, Met. I 1, 980a25ff., and Angier (2010), p. 118f. 73 See Aristotle, EE I 5, 1216b6-24: “[H]e [i.e. Socrates] thought that all the virtues were kinds of knowledge, so that one turns out simultaneously both to know justice and to be just. After all one only has to have learned geometry and building to be a geometer or a builder. Hence Socrates used to investigate what virtue is, but not how and by what means it comes about. (…) Now it is indeed a fine thing to know each fine thing. Nonetheless when it comes to virtue, knowing what it is is not the most valuable point, but understanding what brings it about. For we do not want to know what courage is, but to be courageous, not to know what justice is, but to be just, as we want to be healthy rather than understand what being healthy is”. See also EE II 1, 1220a13-15. 74 See Aristotle, EN VI 13, esp. 1144b28-30. 70

‘LEARNING BY DOING’?

195

in the case of crafts nor in that of the moral virtues, but successful action is dependent on certain dispositions which can only be fully acquired in practice.75 In conclusion: in the scholarly literature on moral habituation, the complexity of Aristotelian technê and its parallels with moral virtues, which Aristotle stresses repeatedly, have sometimes been unduly marginalized. Accordingly, the value of this model for explaining how ‘learning by doing’ might function in the technical as well as the moral area has been underestimated. Aristotle does not use the crafts only occasionally in order to illustrate some aspects of moral habituation; in fact — and notwithstanding the justified assumption that Aristotle deliberately turns his back to a Socratic notion of a virtue-technê76 — Aristotle’s account of the acquisition of virtues is substantially informed by the understanding of ‘learning by doing’ in the area of skills and crafts. Aristotle is much more keen on stressing the similarities of action and production than on highlighting their differences.77 So what he has to say about the relationship between aretê and technê as instances of ‘learning by experience’ amounts to much more than a sheer analogy or a weak comparison.78 And if the crafts are understood as sophisticated forms of practice and not as the products of mindless repetition, the picture which emerges is complex enough to parry the first criticism, viz. that Aristotle is misguided in establishing this close relationship between the acquisition of crafts and virtues via the link of ‘learning by doing’.

75 See Aristotle, EN VI 12, 1143b20-28; EN X 9, 1179a35-b4, as well as 1181b2-11 (tr. R. Crisp): “For neither do doctors appear to acquire their skill from books. And yet doctors try to describe not only the treatments, but also how particular groups could be cured and ought to be treated, distinguishing the various states. This seems beneficial to those with experience, but useless to the ignorant. (…) But those who go through them without being in this state will not be able to judge well, unless by instinct (…)”. 76 See Aristotle, EN VI 5, and the comprehensive discussion provided by Angier (2010), ch. 2. 77 For an unorthodox but highly original approach to reduce the gap between ethical actions and technical production see Kontos (2011), pp. 17-31, who relies on the idea of a visible moral prakton (analogous to the external product of a craft) instead of stressing — as it is usually done — the internal conditions of moral actions. The idea that virtues and vices can be legitimately ascribed to crafts and their use in Aristotle’s ethics is analysed in Kontos (2014), pp. 207-211. 78 This is also the overall thesis of Angier (2010), ch. 5. He also refutes convincingly the claim that the crafts-model only possesses a propaedeutic and ancillary role in Aristotle’s account of moral virtue; see pp. 128-130.

196

J. MÜLLER

2.2 TheConnectivityofVirtuousActionswithEmotionandMotivation I will now turn to the related second criticism of learning by doing as an essential element of moral habituation, namely that repetition of acts may inculcate certain behavioural habits but is insufficient for acquiring the more refined dispositions of actions as we find them in the moral virtues. One rationale behind this criticism is certainly that the sphere of moral actions, even if we conceive of it only in terms of a singular virtue, is much too broad to be encompassed fully by such a training. Just think of the numerous forms of action which the virtue of benevolence may produce and the various circumstances in which its activation is required: could all these manifestations of benevolence really be trained beforehand on a basis of their frequent repetition by the moral student? Certainly not. Moral virtues simply seem to lack the specialization of a repertoire of conduct required in order to be teachable in this manner, as Gilbert Ryle (and some others) suspected.79 But such a narrow understanding of ‘learning by doing’ does not do justice to the application of this concept in Aristotle’s ethics. In the moral sphere, ‘learning by doing’ certainly does not mean ‘repeatedly doing the same thing’ but rather ‘repeatedly doing the right thing’80 in order to become virtuous by these performances: we become just by frequently doing just actions, and by repeated courageous behaviour we acquire the virtue of courage. Now, this requires the particular action in question to be done under a certain description: in standing one’s ground on the battlefield one basically performs a courageous act, and an important part of moral habituation is certainly the recognition on the part of the student that particular acts possess certain moral qualities which allow us to categorize them as just, courageous or benevolent.81 Learning to do something as an instance of a virtue seems to be the essential element in the acquisition of the moral “that” (ὅτι)82 which Aristotle sees as the result 79 See Ryle (1972), p. 436, who thinks that this is one of the reasons why we do not find “professors of probity, charity and patience”. 80 See Curzer (2002), p. 142. 81 On the meta-ethical plane this would presuppose a moral realism and a certain form of cognitivism which cannot be defended here; for one recent reading of Aristotle’s ethics in this vein see Kontos (2011), chs. 1-3. 82 For ethismos as a source of principles see Aristotle, EN I 7, 1098b2-4. In EN VII 1151a17-19, Aristotle remarks that the principles of good action are not acquired by (theoretical) teaching but that “natural or habituated virtue” provides us with the right opinion in this area.

‘LEARNING BY DOING’?

197

of moral habituation, preparing the ground for a deeper understanding of the ethical “why” (διότι), e.g. by reading the NicomacheanEthics. But the acquisition of the “why” seems to be a further step which presupposes habituation and is not a part of it.83 A moral training cannot ultimately amount to habituating certain particular acts like ‘helping an elderly lady to cross the street’. The hexis which is to be formed by such acts has to be a disposition to benevolent performances in general which reacts adequately to the particular circumstances of the situation at hand under a certain description.84 But this need for further specification does certainly not preclude that there are certain paradigmatic instances of virtuous behaviour which enable the moral student (a) to grasp an instructive example of what it means to perform a just, courageous or benevolent act, and (b) to be prepared by repeated performances of it to show similar behaviour under the same description in altered circumstances. In defining the subject matter of the singular virtues Aristotle is often quite specific in describing the kind of model paradigm which captures the essence of this virtue, and this also pertains to a standard repertoire of actions which regularly fall under their description: to stand one’s ground in a life-endangering battle is a paradigmatic instance of a courageous act, to abstain from unhealthy candies is quite a typical performance of moderation, etc. In this vein, Aristotle does not shy away from telling us that children should be trained to get accustomed to the cold as part of their education towards military courage.85 Furthermore, he lists several types of action (like adultery, murder or theft) which can never be done in a virtuous way and should therefore be avoided altogether;86 and he advises moral teachers to keep their

83 For the relationship between grasping the “that” and grasping the “why” which marks the transition from ethismos to phronêsis see Vasiliou (1996) and Angier (2010), pp. 116-120. 84 Aristotelian habituation thus involves grasping certain concepts, as McDowell (1996), p. 30, rightly emphasizes: “Aristotle does not see the product of habituation into the excellences of character as a collection of mindless behavioral tendencies. The result of habituation is a motivational tendency, but one with a conceptual and hence rational aspect. People with a properly formed character have learned to see certain actions as worth undertaking on the ground that they are noble; they have acquired that reasongiving concept, in a way that is inextricably bound up with acquiring the propensity to be motivated by thoughts in which it is applied”. 85 See Aristotle, Pol. VII 17, 1336a12-21. 86 See Aristotle, EN II 6, 1107a8-15; EE II 3, 1221b18-26.

198

J. MÜLLER

students away from situations where shameful forms of behaviour might be instilled indirectly.87 These are rather concrete examples of how an Aristotelian ‘learning by doing’-approach might proceed in habituating moral dispositions on the basis of paradigmatic and prototypical instances of their performance. It does not seem unreasonable that performing (or abstaining from) particular actions disposes the moral student to act similarly in altered circumstances even if the required performance is not identical with the one in which the original habituation took place.88 Aristotle explicitly states that even the games of very young children should already be devised in a way that their activities anticipate the requirements of practice in their adult life.89 This is based on the same principle: moral habituation does not produce dispositions to perform only the same actions but also similar ones. Therefore, moral training will mainly be concerned with paradigmatic or prototypical instances of the different virtues. An essential product of this rather concrete training may well be that the virtuous agent is also quite ready to react without delay when the circumstances call for it. This is even the hallmark of a truly virtuous character, as Aristotle reveals in a telling passage: “It therefore seems to be characteristic of the more courageous person to be unafraid and unruffled in sudden alarms rather than be so in those that are foreseen: it comes more from his state of character, because less from preparation”.90 This quasiinstinctive behaviour on the part of the virtuous agent is certainly one of the results of a previous ‘learning by doing’ which produces moral habits prone to a mode of natural operation. Thus the Aristotelian formula of habits as a “second nature” does not seem to be inappropriate, even when it comes to morals. Such a focus on paradigmatic performances as the basis of virtue acquisition is not restricted to outward actions but also touches on the emotional and motivational dimensions of moral dispositions. As we saw in the previous section, Aristotelian ‘learning by doing’ basically consists 87

See Aristotle, Pol. VII 17, 1336a39-b8. Contra Curzer (2002), p. 147, who doubts that “habitually resisting pressure in committee meetings and declining hot fudge sundaes disposes us to want to stand fast when we find ourselves in our first battle and run fast from our first seduction”. This idea does not appear to me as implausible as Curzer supposes. 89 See Aristotle, Pol. VII 17, 1336a28-34. For a similar advice see also Plato, Nomoi I, 643b-c. 90 See Aristotle, EN III 8, 1117a17-20. 88

‘LEARNING BY DOING’?

199

of ‘learning by experience’ with regard to the incremental acquisition of knowledge involved in this process. Now Aristotle stresses that moral practice also pertains to the passions involved in it: to act courageously is not restricted to the action of standing one’s ground firmly; it also means to have the right, i.e. appropriate amount of fear and confidence in the situation. This seems to be part of the moral “that” (ὅτι) of a properly educated person, and thus moral habituation as ‘learning by doing’ must already contain an element of éducation sentimentale. At least two issues are at stake here: 1) Although part of the moral training may be focused on the question of how to handle certain emotions appropriately, their ‘containment’ or suppression cannot be the benchmark of habituation.91 Aristotle explicitly distinguishes the strong-willed from the virtuous agent: while the first stays true to his choices in spite of recalcitrant passions and desires, the second is harmonious in that his emotions are already on the right level and thus do not oppose but support his virtuous actions.92 Now the capacity of being properly affected can also be trained by practice of virtuous action: we certainly cannot practice feelings “on command” but we might “engage in a certain range of conduct deliberately designed to make one the kind of person who will characteristically feel in appropriate ways, at appropriate times, and so on”.93 Action and feeling are closely intertwined in such a process. In repeatedly standing our ground bravely we also develop the affective dispositions of fear and confidence which have the power to effect a long-lasting emotional change in us.94 Thus sentimental education also presupposes practical habituation.95 2) An important part of Aristotelian moral education is that the student learns to take pleasure in virtuous actions and to dislike vicious ones.96 This is certainly intertwined with the éducationsentimentale 91 Contra Ryle (1972), who thinks that habituation only works for skill-like “virtues of will-power” but not for virtues as affective disposition. 92 See Aristotle, EN VII 9, 1151b32-1152a3. 93 Kosman (1980), p. 113. 94 For the details, see Oele (2012), who discusses this process on the basis of Aristotle’s ontology of dispositions and gives courage as an example. 95 See Steutel/Spiecker (2004), pp. 533ff. 96 See Aristotle, EN X 1, 1172a21-23. See Hutchinson (1986), p. 78: “[V]irtues and vices are dispositions to find certain things pleasant and certain other things unpleasant. In other words, they are each a disposition to like some other courses of conduct and

200

J. MÜLLER

of the passions since these are always related to pleasure and pain;97 it also has a vital motivational dimension insofar as Aristotle sees striving for pleasure and avoiding pain as fundamental tendencies of human behaviour which may be transformed by education but cannot be obliterated. But how may a morally right orientation in this area be achieved within a ‘learning by doing’-approach? This is a hotly debated issue, especially in the aftermath of Myles Burnyeat’s seminal paper on this topic. I would like to take a cue from a rather neglected passage in Aristotle’s treatise on pleasure: “For the pleasure proper to an activity enhances it, because those who engage in activity with pleasure show better and more accurate judgment. It is people who enjoy geometry, for example, who become geometricians and understand each aspect of it better, and similarly lovers of music, building and so on improve in their proper sphere by finding enjoyment in it. And the pleasures enhance the activities, and what enhances an activity is proper to it”.98 The main point which Aristotle targets here is that pleasure is not an add-on, which can be viewed separately from the activity; rather, it completes the activity itself and thus essentially belongs to it. On the other hand, pleasures also enhance the activity in that they effect a continual improvement in repeated activities of this sort. And significantly Aristotle uses crafts like music and building as examples here.99 I take him to mean the following: (a) The better one gets at doing something, the more pleasure or satisfaction one gets from it; successful repetition makes the act more pleasant. This is especially important in the gradual acquisition of a skill. (b) With growing expertise and mastery, dislike other courses of conduct. What this amounts to is that a trait of character is a taste in an area of conduct; virtue is good taste in practical matters”. 97 See Aristotle, EN II 3, 1104b8-1105a16, and EE II 4, 1221b32-39 (tr. B. Inwood / R. Woolf): “Base and excellent characters must, then, consist in the pursuit and avoidance of certain pleasures and pains (…). Capacities and states are about affections and affections are distinguished in terms of pain and pleasure. So from these considerations and from our earlier proposals it follows that every virtue of character concerns pleasures and pains”. 98 Aristotle, EN X 5, 1175a30-36. For the pleasures intrinsic to practice see Sherman (1989), pp. 184-191. But in her account, pleasure appears to be a mere “by-product” of habituation which stimulates further growth. This somehow misses the primordial importance which Aristotle attaches to pleasure in his conception of ethical virtues. 99 For incremental pleasure-taking as a common phenomenon in the Aristotelian view of craft-learning see also Angier (2010), p. 122f.

‘LEARNING BY DOING’?

201

which develops out of repeated practice, one does not only enjoy practicing this skill ever more, but one is also inclined to perform further acts of it. This does not strike me as a watertight conceptual connection, but rather as an empirical observation which, while it certainly does not hold in all cases, holds at least in many of them: we take joy in and are further inclined towards activities in which we have already acquired a certain kind of expertise and therefore do them well or get them right. This principle, which also explains the eagerness to improve and to expand one’s cherished abilities, alleviates to a certain extent the “motivational” ambivalence which Aristotle seems to ascribe to skills in Metaphysics IX (see above, section I.2): as rational potencies, acquired skills are on a conceptual level open to the opposites and therefore presuppose a decision to be activated in one of these two available directions. But from a psychological perspective, their owner will certainly be strongly inclined to use them — and probably also use them in the right manner, although he can deliberately produce mistakes: a trained musician will take pleasure in hearing and producing harmonious sounds instead of a cacophony because he has acquired an according taste.100 And once more Aristotle seems ready to draw a parallel between the artistic and the moral sphere: “The good person, in so far as he is good, enjoys actions that are in accordance with virtue, but is appalled by those done from vice, as the musician finds pleasure in noble tunes, but is pained by bad ones”.101 It is the mark of the good man to enjoy virtuous actions for their own sake, and it is also indicative of a moral habit that the noble action is done with pleasure.102 Without doubt Aristotle portrays acting from a disposition, be it from a craft or from a virtue, as inherently pleasant. But may we simply transfer these pleasures intrinsic to successful practice to moral habituation 100 Pavlos Kontos has pointed out to me that the notion of craft (τέχνη) does not entail any form of striving (ὄρεξις) for its performance. This is true on the conceptual level, but my observation is rather a psychological one which refers to the craftsman, who usually develops a penchant for his skills and the activities which they enable because of his previous successful performances. 101 Aristotle, EN IX 9, 1170a8-11. See also EN X 3, 1173b28-31 (tr. R. Crisp): “[P]leasures differ in species; those from noble sources (καλόν) are different from bad ones (αἰσχρόν), and we cannot experience the pleasures of the just person without being just, nor that of the musical person without being musical, and similarly in the other cases”. 102 See Aristotle, EN I 8, 1099a11-20.

202

J. MÜLLER

where the habit still has to be acquired? Do we truly learn to take pleasure in performing virtuous acts so that our moral development and motivation is ultimately fuelled by this gradual increase in enjoying these activities? This idea has been contested on several levels, especially with a view to virtues like courage: is courageous action not rather concerned with withstanding pressure and pain?103 Obviously, Aristotle delivers some examples which might rather be described with the slogan: “Virtue acquisition hurts”.104 I do not intend to take a decisive stance concerning the question of whether the Aristotelian path to virtue is ultimately pleasant (as Myles Burnyeat thinks) or painful (as Howard Curzer supposes); this might even depend on the particular virtue in question, and so it does not come as a surprise that Aristotle is reluctant to deliver a general blueprint of moral habituation.105 Perhaps the blunt opposition between pleasure or pain as means of moral habituation over-simplifies this complex matter. But it is worthwhile at least to mark the following point: it is certainly implausible that the courageous person literally enjoys (and that the moral student should learn to enjoy) standing the pain inflicted on him. But the point of virtue-oriented ‘learning by doing’ is, as we saw above, that the activity is performed under a certain description, namely as a courageous act. I suggest that the moral learner has to be habituated to find pleasure in courageous performances as courageous — and not as painful events. This presupposes that the motivation of the person will be fixed towards the virtuous act itself — and not to any accompanying circumstances or subsequent effects of it. And this requirement is at the heart of the internal conditions which the agent has to fulfil in order to act virtuously in the full sense of the word: he has to decide in favour of the virtuous action for its own sake just because it is the fine (καλόν) thing to do.106 103

See Aristotle, EN III 9, 1117a33-35. See Curzer (2002), pp. 158-161, on moral progress through pain. 105 Aristotle does repeatedly point out that his Nicomachean Ethics only provides a broad outline (τύπος) which has to be spelled out more completely; for this model of ethics as an “outline science” see Höffe (1996), part II. 106 See Aristotle, EE III 1, 1230a26-32: “Now all of virtue is connected with decision. (…) virtue makes everyone choose for the sake of something, and this ‘something for the sake of which’ is what is fine. That being so, it is clear that courage too, being a virtue, will make us endure what is fearful for the sake of something, and that will be due neither to ignorance (…) nor to pleasure, but because doing so is fine”. 104

‘LEARNING BY DOING’?

203

The motivation to perform the virtuous act as something fine is to a certain extent different from other possible reasons of its enactment, like doing it for pleasure or because it is useful. Therefore, moral education has to be quite cautious not to start “on the wrong foot” in this area. Positive or negative reinforcement by inducing pleasure or pain for performing an action runs the risk of a purely behavioural conditioning in which the learner associates the action with its external reward or punishment on a psychological level. But if the aim of moral development is that the agent takes pleasure in the virtuous action itself (and not in its consequences), it is not easy to see how the transition between these two kinds of motivation will be managed.107 Thus moral habituation must already cultivate this motivational structure in the agent right from the start and not a later stage when the habit is already in place. But how is this possible? This is an intricate question which ultimately exceeds the scope of my paper but I do not wish to bypass it completely. Aristotle offers at least one clue to a possible answer in his account of musical education in the 8th book of his Politics: music is not to be learnt because it is useful (unlike the other subjects of the curriculum such as gymnastics, grammar and drawing) but because it is fine (καλόν) and worthy of a free man.108 In other words, it is practiced and enjoyed for its own sake. This forges a natural link with virtuous action which Aristotle develops as follows: “And since it so happens that music is one of the pleasures, and virtue has to do with enjoying, loving and hating in the right way, obviously one must learn and become accustomed to nothing so much as correctly judging and enjoying decent characters and noble actions. In rhythms and melodies there is the greatest likeness to the true natures of anger and gentleness, and also of courage and temperance, all of their opposites, and the other characters. This is obvious from the facts: we undergo a change in our souls when we listen to such things. Someone who is accustomed to feeling pain and pleasure in things that are likenesses is close to someone who reacts in the same manner to the true things”.109

107 This seems to be the fundamental problem facing Burnyeat’s account of moral progress through pleasure; see the criticisms by Curzer (2002), pp. 144-150, Fossheim (2006), pp. 106-109, and Hitz (2012), p. 280ff. 108 See Aristotle, Pol. VIII 3, 1338a30-37. 109 Aristotle, Pol. VIII 5, 1340a14-25 (tr. R. Kraut).

204

J. MÜLLER

This musicalethismos ultimately relies on our natural love for mimetic performances as representational activities, as Hallvard Fossheim has convincingly shown.110 Now, what is important for Aristotle’s ‘learning by doing’-conception is his emphasis that children should not only listen to musical performances by others but also be trained to perform themselves. The underlying idea is that the development of critical faculties in this area is ultimately dependent on internalizing the right standards of performance by way of practice. But in this way moral habituation by way of musical education is once more adequately described in terms of ‘learning by doing’, although in a different manner than before: by practicing the right kind of music the character of the child is cleansed from disturbing emotional influences111 and simultaneously formed in such a way that he is able to perceive the fine (καλόν) and to be motivated to pursue it for its own sake. According to Aristotle, it is one hallmark of young people that they are ready to choose what is fine over what is advantageous,112 and this tendency can be strengthened in this manner. Therefore, musical education as a special sort of ‘learning by doing’ might serve as a basis for perceiving certain actions as virtuous and being motivated to perform them for their own sake, which is a vital part of moral habituation in the sense of the ‘pure choice condition’.113 This idea bolsters my defense of ‘learning by doing’ as a key to the understanding of Aristotelian ethical habituation, but it certainly cannot tell the whole story of how moral agents learn to appreciate and perform virtuous actions for their own sake. In my opinion, the ‘pure choice condition’ ultimately still marks a difference between performances of moral virtue and of craft. Notwithstanding the fact that the craftsman also takes pleasure in enacting his expertise, it would be problematic — at least in Aristotelian terms — to say that he performs these acts only for their own sake or that he is motivated mainly by the desire for the fine (καλόν). 114 This is due to 110 See esp. Aristotle, Poet. 4, 1448b4-10, and the excellent discussion provided by Fossheim (2006), pp. 109-117, on the relationship between mimêsis and moral development. 111 For this musical catharsis see esp. Pol. VIII 7, 1341b36-41. 112 See Aristotle, Rhet. II 12, 1389a34-36. 113 Apart from Fossheim (2006) see also Hitz (2012), pp. 298-301 and Sherman (1989), pp. 181-183. 114 This fundamental requirement for virtuous action is stressed repeatedly by Aristotle; see ENIII 7, 1115b12-13; III 8, 1116b30-31; 1117a8; IV 1, 1120a23-25; IV 2, 1122b6-7 and 1123a24-25; VI 12, 1144a19-20; X 6, 1176b2-9. For a sustained discussion of this “pure choice condition” see Hutchinson (1986), pp. 93-107.

‘LEARNING BY DOING’?

205

the fundamentally poietic structure of technai: poietic acts aim at a certain product, and it is this intention which ultimately governs the performances required for fulfilling this target. Because of this basic orientation, productive acts are simply not viewed as worthwhile in themselves in the manner that moral praxis is to be sought and enacted for its own sake. And this divergence must also be accounted for in the special mode of their acquisition, although Aristotle stays silent on this issue. But in my view this difference does not detract too much from the substantial core of ‘learning by doing’, which is common to the acquisition of crafts and virtues. To conclude this section: because of the multiple connectivity between action and passion, between repeated doing and its inherent pleasure, Aristotle’s account of ‘learning by doing’ does not even come close to a merely mechanical drill of outward actions. In fact, he tries to show that moral practice is necessary for acquiring the right feelings and desires, finally issuing in choices of virtuous acts for their own sake. Thus the first line of criticism, targeted at ‘learning by doing’ as an insufficient means to develop appropriately complex moral habits, does not really concern Aristotle’s account of habituation. 3. Conclusion: a Plea for a Truism After having defended Aristotle’s account of ‘learning by doing’, I would finally like to offer some closing comments on the topic. Nancy Sherman once pinpointed the task in analysing Aristotelian moral habituation as follows: “Aristotle’s account of habituation extends well beyond the truism expressed in the phrase that ‘we learn by doing’. At issue is how we learn by doing (…)”.115 This is certainly true but also reflects a rather dismissive attitude towards the allegedly “truistic’ idea of ‘learning by doing’ as the central core of Aristotelian moral education. Therefore, Sherman prefers to speak of “critical habituation”, a complex and dynamic process involving the perceptive, emotional and deliberative capacities of the ethical learner right from the start. But attempts like these to bridge the supposed gap between moral habituation and ethical instruction as smoothly as possible have a tendency to neglect the fact that Aristotle clearly offers a developmental 115

Sherman (1989), p. 199.

206

J. MÜLLER

approach towards the acquisition of virtues in which ethos is explicitly distinguished from phusis as well as from logos.116 By the cognitive “overload” of habituation in Sherman’s account it is not easy to recognise these transitions any longer.117 But Aristotle is quite unambiguous in stating it to be clear “that one should educate by means of habit before educating by means of reason”.118 On such a sequential view of moral education this cognitive overload of habituation might even prove unnecessary: moral habituation as an earlier stage simply does not have to capture and mirror all the aspects which full-blown virtue as the final product of ethical development engenders. There are recognizable differences between ‘acting to form a habit’ and ‘acting from a habit’ in the moral sphere, most of them connected with the acquisition of prudence (φρόνησις) as an intellectual virtue which is closely intertwined with moral character. According to Aristotle, habituation is necessary but not sufficient for the acquisition of moral virtue. This reservation concerning Sherman’s idea of “critical habituation” and similar cognitivist approaches does not amount to a plea for (re)turning to a non-cognitivist or even behaviourist reading of Aristotelian ethismos and ‘learning by doing’ instead. Every attempt to cut off habituation completely from any rational development runs exactly the opposite risk of playing down the complex way in which moral virtues (as well as the crafts) are acquired via ‘learning by doing’ according to Aristotle.119 Aristotle coherently stresses the connectivity of virtuous practice with rational 116 See Aristotle, Pol. VII 13, 1332a35-b10, and EN X 9 (similarly: EE I 1, 1214a1421), where Aristotle distinguishes three ways to become good: by nature (φύσις), by habituation (ἔθος) and by teaching (διδαχή). Significantly, Aristotle restricts the range of argument and teaching as follows (1179b23-31; tr. R. Crisp): “Argument and teaching, presumably, are not powerful in every case, but the soul of the student must be prepared beforehand in its habits, with a view to its enjoying and hating in a noble way, like soil that is to nourish seed. (…) There must, therefore, somehow be a pre-existing character with some affinity for virtue through its fondness for what is noble and dislike of what is disgraceful”. 117 Aristotle is quite explicit about the fact that it is the acquisition of thought (νοῦς) which makes all the difference for the development of full-blown virtues: before that, moral habits do only resemble true virtue; see Aristotle, EN VI 13, 1144b12-17. 118 Pol. VIII 3, 1338b 4-5 (tr. R. Kraut). See also Pol. VII 15, 1334b5-25, on the priority of education by ethos over logos and nous (which is the ultimate goal of the whole process). 119 The other extreme is certainly Curzer who argues forcefully against Burnyeat and Sherman in order to keep habituation and instruction as tidily apart as possible. But this seems to overstate the non-cognitivist case.

‘LEARNING BY DOING’?

207

knowledge as well as with emotional dispositions and the moral motivation arising from their interplay. Therefore, his account of ‘learning by doing’ also needs to be defended against and rescued from over-simplified versions of it which turn Aristotle into a kind of proto-behaviourist. The elucidation of Aristotle’s philosophy of moral education still encounters two basic tasks. Firstly, it has to answer what Rosalind Hursthouse once called “the Question”: Why does moral virtue have to be acquired by habituation?120 This problem pertains especially to the contribution which habituation is supposed to deliver to the grasp of moral principles and of right ethical beliefs. Secondly, it has to deal with the paradox of “extrinsically habituated reason”:121 how can the ethismos in its dependency on the external supervision and guidance of suitable teachers prepare the ground for an autonomous performance of moral rationality, based on Aristotelian phronêsis? I would like to suggest that a sufficiently fine-grained understanding of ‘learning by doing’ as the core of Aristotle’s account of virtue acquisition might help to answer these questions in an illuminating manner. My defence of the popular ‘learning by doing’-idea against two forceful criticisms mainly aims at clearing the path for such a refined account of Aristotelian moral habituation and to indicate in broad strokes the general direction which it might take. Hence, my conclusion finally amounts to a plea for a truism — which may nevertheless still be fitted to inspire Aristotelian scholarship as well as our contemporary thinking about moral development.122 LITERATURE Translations ARISTOTLE, Nicomachean Ethics (2000), translated and edited by Roger Crisp, Cambridge: Cambridge University Press. ARISTOTLE, EudemianEthics(2013), translated and edited by Brad Inwood and Raphael Woolf, Cambridge: Cambridge University Press. TheCompleteWorksofAristotle.TheRevisedOxfordTranslation (1995), edited by Jonathan Barnes, 6th ed., Princeton: Princeton University Press. 120

See Hursthouse (1988), p. 202. For discussion see Kristjánsson (2007), ch. 3. 122 I would like to thank Jon Bornholdt for his attentive reading of this text and his helpful suggestions for its improvement. 121

208

J. MÜLLER

Studies ANGIER, Tom (2010), TechnêinAristotle’sEthics:CraftingtheMoralLife, New York: Continuum. BROADIE, Sarah (1991), EthicswithAristotle, New York / Oxford: Oxford University Press. BURNYEAT, Myles F. (1980), “Aristotle on Learning to Be Good” in Rorty, A. O. (ed.), EssaysonAristotle’sEthics, Berkeley: University of California Press, pp. 69-92. CARR, David (2008), “Character Education as the Cultivation of Virtue” in Nucci, L. P. & Narvaez, D. (eds.), HandbookofMoralandCharacterEducation, New York / London: Routledge, pp. 99-116. CURZER, Howard (2002), “Aristotle’s Painful Path to Virtue” in Journalofthe HistoryofPhilosophy 40 (2002), pp. 141-162. FOSSHEIM, Hallvard (2006), “Habituation as Mimesis” in Chappell, T. D. J. (ed.), Values and virtues. Aristotelianism in contemporary ethics, Oxford, New York: Clarendon Press, pp. 105-117. HALBIG, Christoph (2013), DerBegriffderTugendunddieGrenzenderTugendethik, Frankfurt a.M.: Suhrkamp. HITZ, Zena (2012), “Aristotle on Law and Moral Education” in OxfordStudies inAncientPhilosophy 41, pp. 263-306. HÖFFE, Otfried (1996), PraktischePhilosophie.DasModelldesAristoteles, 2nd ed., Berlin: Akademie-Verlag. HURSTHOUSE, Rosalind (1988), “Moral Habituation: A Review of Troels Engberg-Pedersen” in OxfordStudiesinAncientPhilosophy6, pp. 201-219. HUTCHINSON, D. S. (1986), TheVirtuesofAristotle, London / New York: Routledge. JAEGER, Werner (1957), “Aristotle’s Use of Medicine as a Model of Method in His Ethics” in JournalofHellenicStudies 77, pp. 54-61. KOHLBERG, Lawrence (1981), EssaysonMoralDevelopment,Vol.I:ThePhilosophyofMoralDevelopment, San Francisco: Harper & Row. KONTOS, Pavlos (2011), Aristotle’s Moral Realism Reconsidered. PhenomenologicalEthics, New York / London: Routledge. KONTOS, Pavlos (2014), “Non-Virtuous Intellectual States in Aristotle’s Ethics” in OxfordStudiesinAncientPhilosophy 47, pp. 205-243. KOSMAN, L. A. (1980), “Being Properly Affected: Virtues and Feelings in Aristotle’s Ethics” in Rorty, A. O. (ed.):EssaysonAristotle’sEthics. Berkeley: University of California Press, pp. 103-16. KRAUT, Richard (1998), “Aristotle on Method and Moral Education” in Gentzler, J. (ed.), MethodinAncientPhilosophy, Oxford: Clarendon, pp. 271-290. KRISTJÁNSSON, Kristján. (2007), Aristotle,Emotions,andEducation, Aldershot: Ashgate. LAWRENCE, Gavin (2011), “Acquiring Character: Becoming Grown Up” in Pakaluk, M. & Pearson, G. (eds.), MoralPsychologyandHumanActionin Aristotle, Oxford: Oxford University Press, pp. 233-283.

‘LEARNING BY DOING’?

209

LOCKWOOD, Thornton C. (2013), “Habituation, Habit, and Character in Aristotle’s Nicomachean Ethics” in Sparrow, T. & Hutchinson, A. (eds.), AHistoryof Habit.FromAristotletoBourdieu, Lanham: Lexington Books, pp. 19-36. LONDON, Alex J. (2001) “Moral Knowledge and the Acquisition of Virtue in Aristotle’s Nicomachean and EudemianEthics” in ReviewofMetaphysics 54, pp. 553-583. MAKIN, Stephen (2006), Aristotle. Metaphysics Theta (Clarendon Aristotle Series), Oxford: Clarendon. McDOWELL, John (1996), “Deliberation and Moral Development in Aristotle’s Ethics” in Engstrom, S. P. & Whiting, J. (eds.), Aristotle, Kant, and the Stoics.RethinkingHappinessandDuty. Cambridge / New York: Cambridge University Press, pp. 19-35. MOREL, Pierre-Marie (1997), «L’habitude: une seconde nature?», in Morel, P.-M. (ed.), Aristoteetlanotiondenature.Enjeuxépistemologiquesetpratiques, Bordeaux: Presses universitaires de Bordeaux, pp. 131-148. MÜLLER, Jörn (2006), Physis und Ethos. Der Naturbegriff bei Aristoteles und seineRelevanzfürdieEthik, Würzburg: Königshausen & Neumann. OELE, M. (2012), “Passive Dispositions: On the relationship between πάθος and ἕξις in Aristotle” in AncientPhilosophy 32, pp. 351-368. REEVE, C. D. C. (1998), “Aristotelian Education” in Rorty, A. O. (ed.), Philosophers on Education. Historical Perspectives, London / New York: Routledge, pp. 51-65. RYLE, Gilbert (1972), “Can virtue be taught?”, in Dearden, R. F. & Hirst, P. H. & Peters R. S. (eds.), Educationandthedevelopmentofreason, London: Routledge, pp. 434-447. SHERMAN, Nancy (1989), TheFabricofCharacter.Aristotle’sTheoryofVirtue, Oxford: Clarendon. STEUTEL, Jan & SPIECKER, Ben (2004), “Cultivating Sentimental Dispositions Through Aristotelian Habituation” in JournalofPhilosophyofEducation 38, pp. 531-549. STICHTER, Matthew (2007), “Ethical Expertise. The Skill Model of Virtue” in EthicalTheoryandMoralPractice 10, pp. 183-194. TAYLOR, C. C. W. (2006), Aristotle.NicomacheanEthics,BooksII-IV (Clarendon Aristotle Series), Oxford: Clarendon. TOBIN, Bernadette M. (1989), “An Aristotelian Theory of Moral Development” in JournalofPhilosophyofEducation23/2, pp. 195-211. VASILIOU, Iakovos (1996), “The Role of Good Upbringing in Aristotle’s Ethics” in PhilosophyandPhenomenologicalResearch 56, pp. 771-797.

THE FUNCTIONS OF PRACTICAL DELIBERATION Øyvind RABBÅS

Aristotle’s project in ethics and politics is to identify and elucidate the good life for human beings. Starting from the teleological structure of human activity, he claims that there is a single, ultimate end that all human activity aims at — or ought to aim at (I 2). He follows common sense in terming this ultimate end or good eudaimonia, and he equates it with “living well” (εὖ ζῆν) and “doing” or “acting well” (εὖ πράττειν) (I 4). So the ultimate end of all human endeavour is to live a good life or, more precisely, to succeed in living life well. The question is what makes human life well lived, however, and in the crucial and well-known passage generally referred to as the “function argument” (EN I 7), Aristotle gives his substantive bid on the nature of the ultimate end: the human function, or characteristic “work” or “task” (ἔργον), is “activity of soul in accordance with, or not apart from reason (ψυχῆς ἐνέργεια κατὰ λόγον ἢ μὴ ἄνευ λόγου)”; hence “the human good turns out to be activity of soul in accordance with virtue (ψυχῆς ἐνέργεια κατ᾽ ἀρετήν)” (1098a7-8, 16-17).1 So eudaimonia, the highest good for human beings, the good life, is the activity of living a human life in an excellently rational way, i.e. living virtuously. Human action springs from decision, which in turn results from desire and thought: “Now the origin of action (πράξεως ἀρχή) — in terms of the source of the movement, not its end — is decision (προαίρεσις), while that of decision is desire (ὄρεξις) and rational reference to an end (λόγος ὁ ἕνεκά τινος). Hence 1 Translations of the NicomacheanEthics are by C. Rowe, Broadie and Rowe (2002), sometimes slightly modified. Thus I use the traditional “virtue” rather than “excellence” for ἀρετῆ.

212

Ø. RABBÅS

intelligence and thought (νοῦς καὶ διάνοια), on the one hand, and characterdisposition (ἠθικὴ ἕξις) on the other are necessary for decision; for doing well (εὐπραξία) and its contrary, in the context of action, are conditional on thought and character (διάνοια καὶ ἤθος).” (VI 2, 1139a31-5)

Desire and thought, in turn, are the activities or actualizations of the two ethically relevant parts of the human soul: one of these is rational, the other is “non-rational, although it shares in reason in a way” (I 13, 1102b13-14). Thus both are rational in some way and opposed to the truly non-rational part of the soul, which is the plant-like or nutritive part. Qua rational they work together in deliberation (βουλεύσις), the process of reasoning that ends in a decision. This distinction between two parts of the human soul then gives rise to a corresponding distinction between two kinds of virtue: virtue of thought or intellectual virtue (ἀρετῆ διανοητική), on the one hand, and virtue of character or ethical virtue (ἀρετῆ ἠθική), on the other. Virtue of character is defined as “a disposition issuing in decisions (ἕξις προαιρετική)” (II 6, 1106b36), and actions count as virtuous “first, (A) if [the agent] does them knowingly, secondly, (B) if he (i) decides to do them, and (ii) decides to do them for themselves, and thirdly (C) if he does them from a firm and unchanging disposition.” (II 4, 1105a31-3) Decisions, moreover, are based in deliberation (βουλεύσις). So virtue of character is crucially related to deliberation and decision. But so is intellectual virtue, i.e. practical wisdom (φρόνησις): “it is thought characteristic of a practically wise person to be able to deliberate well about the things that are good and advantageous to himself, not in specific contexts, e.g. what sorts of things conduce to [are πρός] health, or to physical strength, but what sorts of things conduce to the good life in general (πρὸς τὸ εὖ ζῆν ὅλος).” (VI 5, 1140a25-8) So both virtue of character and virtue of thought are essentially involved in deliberation, which raises the question of their respective roles or functions in deliberation. In his official discussion of deliberation, in III 3, Aristotle says that “we deliberate, not about ends, but about what forwards those ends.” (1112b1112) So there is a difference between the end (τὸ τέλος) and that which “forwards” — is “towards” or “conduces to” — the end (τὸ πρὸς τὸ τέλος), and deliberation is said to be concerned only with the latter. In other passages he seems clearly to assign these two items to virtue of character and virtue of thought respectively. Perhaps the most straightforward of these passages comes towards the end of Book VI, on virtues of thought.

THE FUNCTIONS OF PRACTICAL DELIBERATION

213

Here he says that “virtue [i.e. virtue of character] makes the goal right (τὸν σκοπὸν ποιεῖ ὀρθόν), while practical wisdom (φρόνησις) makes what leads to it (τὰ πρὸς τοῦτον) correct.” (VI 12, 1144a7-9) These passages, along with some others (see VI 13, 1145a5-7; VII 8, 1151a15-19; EE II 11, 1227b23-5), suggest a division of labour between the two parts of the soul distinguished in I 13. Thus the agent’s non-rational part sets up an end qua end of his activity, in the sense of (a) something apprehended as worth pursuing, as well as (b) something he is actually motivated to pursue. However, the question is exactly how this end is to be pursued, for the problem is often that this is insufficiently determinate. There is thus a second task to be performed, and that is the function of the second part of the soul, the rational part: to determine the end as concretely pursuable, as well as actually pursued. Consequently, when the non-rational part is virtuous, i.e. when the agent has ethical virtue, the end set up will be right, whereas when the agent is vicious the end will not be correct. Similarly, when the rational part is virtuous, i.e. when the agent has practical wisdom, his deliberation will lead to the correct decision about the best way to go about realizing the end set up. There has been a lot of controversy in the scholarly literature about how exactly to understand this division of labour, or indeed whether there is one at all. The reason is clearly a worry that such a division seems to commit Aristotle to a Humean picture of human motivation.2 This is seen as a problem because it seems to place the setting of practical ends beyond the reach of reason and to assign to reason a merely instrumental function. In a recent paper Jessica Moss3 has argued (1) that Aristotle does indeed hold what he seems to hold in these passages, viz. that there is this division of labour between the non-rational and the rational parts of the soul; but (2) that this by itself does not commit Aristotle to a Humean conception of human motivation. That such an implication does not follow becomes apparent once we realize that Aristotle’s notion of reason (λόγος) as a capacity of the soul is not our notion of the cognitive. When Aristotle talks about possessing reason (λόγον ἔχειν) what he has in mind is the possession of a capacity for a certain kind of rational activity, viz. reasoning (sc. deliberation), issuing in a certain kind of intellectual grasp. We can thus allow the deliverances of the non-rational part of the soul — viz. the ends 2 3

See Moss (2011), p. 206 n. 4 for references. See Moss (2011).

214

Ø. RABBÅS

set up for deliberation and action — to be rational in the sense of cognitively constituted, somehow, and hence rationally criticisable as being right or wrong, even when they are not the result of a certain kind of reasoning process (deliberation). It is not the function of this non-rational part to actualize itself in this kind of reasoning (or any other kind of reasoning, for that matter — this is what makes it non-rational “in a sense”). Indeed, the status of the deliverances of non-rational parts as either correct or incorrect is assumed as obvious in the very passages that articulate the division of labour: “virtue makes the goal right (τὸν σκοπὸν ποιεῖ ὀρθόν), while φρόνησις makes what leads to it (τὰ πρὸς τοῦτον) correct.” Talk of rightness or correctness implies rational normative standards that the virtuous character somehow “tracks”.4 Given what is said so far, it is clear that deliberation, if successful, ends with a decision about what to do, so deliberation has a heuristic function: it aims tofindoutwhattodo. Aristotle explicitly says as much: “deliberation is a certain kind of inquiry (ζητεῖν τι)” (VI 9, 1142a31-2; cf. b1-2: “someone who is deliberating is inquiring (ζητεῖ)”.) But I want to argue that deliberation has two further functions. The first is demonstrative: deliberation should not only discover the right decision to act (find out what to do), but also showthatandwhy this decision is in fact the right one — as well as show that the agent really is virtuous. And, thirdly, deliberation also has a formative function: it will form and stabilize the character of the agent. Or perhaps, strictly speaking, these three should only be regarded as aspects of one single function, which is to actualize human nature as a ζῷον λόγον ἔχων by living the human life well. 1. The Heuristic Function of Deliberation The deliberative problem arises in a concrete, practical situation that the agent takes to call for a practical response from him.5 Sometimes the agent enters the situation already having an objective of pursuit, while at other times he does not, such that his problem is to figure out which 4 Moss also argues (3) that the way virtue is developed and sets up its ends should be accounted for in a certain way, according to what she calls “Practical Empiricism” (2011, p. 255). I do not follow her on this third point, which she sees as her own main contribution to the debate, but I do accept the first two. 5 Broadie (1991), p. 227. The following account of the first function of deliberation is greatly indebted to Broadie’s seminal discussion.

THE FUNCTIONS OF PRACTICAL DELIBERATION

215

objective would be the suitable one to pursue.6 There are differences here, but I shall leave them aside here as I don’t take them to be relevant for present concerns. An example may illustrate. In the PosteriorAnalytics Aristotle tells of a man — let us call him Al — who went to a certain place to get some money to repay a debt and not act unjustly (I 24, 85b30-2). Suppose this action was deliberated: how would his deliberation have proceeded? Presumably it started when, in a certain situation — say when the day agreed upon had arrived — Al thought he should repay his debt. He thought this because he is an honest and just man, and because he remembered the agreement. But there is a problem, we may imagine: he doesn’t have enough cash, and this sets the deliberative problem: “How can I repay my debt?” It then occurs to Al that he might go to the bank and take out some of his savings to cover the debt, and since there do not seem to be any reasons against this option, he decides to go to the bank. So off he goes to get the money to pay what he owes. Let us look more closely at this. The process starts when an agent finds himself in a situation where certain ethical features are salient — the situation has a certain ethical shape, we might say. The agent will respond to the situation in a certain way. We should note several things about this practical response. The first thing to note is that the response is complex. The agent will not only perceive the facts of the situation, but he will also do so in an evaluative light. Thus, Al might see the financial advantage of not paying his debt and keep his money for himself but, being just, this consideration will carry no weight with him and he will fulfil his obligation and repay his debt. Similarly, an agent may see that the situation he is in is one of danger, or one where his friend needs help, or one where either he himself or somebody else is exposed to some injustice. As a result he will react emotionally, where this reaction not only has some hedonic valence as well as certain characteristic physiological manifestations, but also often involves a motivation to act in certain ways. For instance, if he is exposed to a threat he may see reason to flee into security or, if some greater value is at stake, he may take himself to have reason to endure and proceed according to plan in spite of the danger that threatens and elicits his fear. Similarly, upon seeing a friend subjected to an insult from 6

Broadie (1991), p. 233.

216

Ø. RABBÅS

somebody he may feel an obligation to speak up in his defence and criticize the insulting person. The two practically most salient elements of the response, however, are the emotional reaction and the motive to pursue a certain end set up by the response. The second thing to note is that this complex practical response is not a one-off response (it does not come out of the blue, as it were) but it manifeststheagent’scharacter. If this character is good, the agent has virtue. Virtue “has to do with affections and actions (περὶ πὰθη καὶ πράξεις)” (II 6, 1106b16-17, 24-5; 1107a4-5, 8-9), and is defined in terms of how the agent relates to “what is required both in affections and in actions … and both finds and chooses the intermediate.” (II 6, 1107a4-6) Moreover, a person’s character is not merely a set of psychological traits and dispositions — a personality; rather, it is a psychological structure constituted by a certain evaluative outlook on life. It consists, to be sure, in a set of perceptual abilities, evaluative tendencies, attention patterns, and conative dispositions — the agent is wont to respond in thisway. But this responsive structure is organized around a certain more or less articulate and reflected conception of what life is about, a general sense of what is of real importance and value in life — we could say: what the person stands for as a person. It is this conception that manifests itself in the various ways he responds perceptually, evaluatively, emotionally, and motivationally to particular situations. Thus the courageous person will typically endure the danger in order to proceed in his endeavour in spite of his fear, the temperate person will not let the lure of sensual gratification tempt him away from the healthy course. And the person with the virtue that Aristotle calls particular justice will never let the temptation of material gain impel him to acquire more than he is entitled to. The virtues of these people make them respond the way they do, and Al is an example of this. This implies the third thing to note about the practical response of an agent, viz. that it is subjecttorationaljudgement. This is so at two levels. First, the individual elements of the response may all, individually or collectively, be, or fail to be, adequate under the circumstances. The agent may, for instance, misperceive the situation or miscalculate its value if he overestimates the danger or the gratification that might be had. Alternatively, he might overreact: there is danger, but not sufficient to justify flight. Secondly, however, the underlying evaluative orientation might also be mistaken. Aristotle is a realist or objectivist about ethics: he thinks that there is a right way and several wrong ways of thinking about the most

THE FUNCTIONS OF PRACTICAL DELIBERATION

217

valuable things in life. He takes this for granted in his ethics, and he presents his own argument to justify, or articulate the justification for, his own view of what the right way to conceive of the ultimate end in life is. This means that the response of an agent to a particular situation is always the legitimate target of criticism. The person in Al’s situation, for instance, will apprehend the distributive problem raised by the situation, and correctly so. He will also, secondly, notice the cost or disadvantage of this for himself, and therefore, perhaps, feel some discomfort at this. But because he is virtuous he will not be moved by this feeling of discomfort, since this is required by justice. Similarly, the brave person will face up to the danger he fears and endure it, since there are greater values at stake than the ones exposed to the current threat, that require endurance. In this case we could perhaps say that the motivational force of the fear is neutralized by the realization of what is at stake and the commitment to stand by what is really worth protecting or pursuing. Similarly, a friend will see that his friend is in some kind of trouble and therefore needs help, he will sympathize with this, and be motivated to offer his help. However, the response of the agent, even if virtuous and hence correct, may not be determinate enough to issue in a decision. This is the fourth point to note about the practical response of an agent to the ethical situation he finds himself in. There may be more than one way of responding that seems adequate, or there may be no apparent way of responding. It is in such cases that the need for deliberation arises: the task of deliberation is to find out, to determine what to do, or how to react to the situation as registered. This is accomplished through a process of reasoning — deliberation — that requires and manifests the agent’s reasoning capacity, the quality of the reasoning part of his soul. If he has intellectual virtue, this capacity will be excellent and constitute practical wisdom, phronēsis, and this wisdom will guide him through the deliberative process. As David Charles puts it: “what is distinctive about practical thinking [is that] it is a type of thinking essentially directed at finding out what is good to do so as to do it.”7

7 Charles (2015), 75; cf. Allen (2015), p. 69: “the whole point and purpose of […] practical wisdom is to deliberate about and decide about how to act as occasions for action arise one after another”.

218

Ø. RABBÅS

Two misunderstandings should be cleared away at this point. First, even if deliberation gets started with the end provisionally set up in the agent’s immediate response, and is concerned with what promotes this end rather than the end itself, this doesn’t make Aristotle an instrumentalist about practical reason, one who claims that deliberation is simply about finding the means to bring about a given end. For the relation between the end and what promotes it doesn’t haveto be instrumental, it may also be constitutive. While Al’s end of going to the bank to withdraw money is a means to repay his debt, I might have as an end to have a nice vacation, and then going to Rome for two weeks might “promote” this end by constituting it — it is what the end consists in.8 Secondly, the objective of deliberation is still confined to finding the means — even if these are construed more broadly than on the usual instrumentalist account — while leaving the ends to be set in other ways, seemingly irrationally. This is presumably the main worry of scholars who fear that the division of labour between character and thought implies Humeanism. However, this does not leave the ends completely beyond the reach of deliberation. One might think that one could formulate the deliberative question as “How am I to attain O in C?”, where O is the objective proposed by the agent’s character as it responds to the situation, and C is the circumstances of the situation. But a better way of formulating this question would be “What will be involved in attaining O in C?” For it is not immediately clear what attaining this objective will actually amount to, nor what the circumstances really are. Therefore the process of deliberation will take the form of a reformulation or redescription of the initially proposed objective through which the desire for that objective is “transmitted” to action decided upon.9 To see this, let us return to our man, Al. The outline of his deliberation given above was very simple, and his deliberative problem easily solved, but we can imagine that it would have proceeded differently. He might, for instance, have come to realize that he had already overdrawn his bank account, so he couldn’t go to the bank — at least not to withdraw money from his own account. He might of course instead have asked for a loan at the bank, but the question is if that would have been wise: it would just 8

A point argued forcefully by John Cooper (1975), pp. 19-22. Cf. Broadie (1991), pp. 228-9. It is Jonathan Lear who speaks of the function of deliberation as the “transmission of desire” from a given desire for an object taken to be good by the agent to a deliberated choice to pursue that object; Lear (1988), pp. 141-51. 9

THE FUNCTIONS OF PRACTICAL DELIBERATION

219

put him in debt again. Alternatively, Al might ask his creditor, let’s call him Ben, for an extension, but that might have incurred a penalty interest, which would only put him further into debt and make things worse. This process is not just a move from the initial proposed end to the appropriate means, but also a way of scrutinising this end itself. That is why the end (O) is properly described as “assumed” or “proposed” — Aristotle says it is “hypothesized”.10 And the way it is being scrutinised is by seeing what its realization will involve; perhaps it will require means that are unavailable to us, or perhaps it will involve doing something which we can’t accept doing on other, independent grounds (the end doesn’t always sanctify the means).11 However, we could imagine a different course that his deliberation might have taken. In this case Al still assumes that he should repay his debt, and again he starts to think about what this would entail. He knows his creditor, Ben, and he knows that Ben is a person who does not have very good control over his economy. Moreover, he has lately, since lending his money to Al, engaged in some rather dubious transactions with criminal elements — say, through gambling. Al therefore knows that if he repays his debt to Ben now, it is highly likely that Ben will only spend the money on his illegal activities, which would not only contribute to further illegal activity, but would also get Ben into even deeper trouble. So repaying the money to Ben would be strongly against Ben’s interests, which speaks against repaying the debt now. On the other hand, is this really Al’s responsibility? Ben is a free and independent adult responsible for his own life, and if he has gotten himself into trouble with gangsters, that is his problem. So not only is this none of Al’s business, but intervening in Ben’s affairs in this way would actually be a way of showing disrespect for him — it would be paternalism. But then again, if Al and Ben are friends, then they mutually care for each other for each other’s sake, and that imposes on Al some obligation to show concern for Ben’s interests or welfare. Although, once again, even friends — or perhaps especially friends — owe each other respect: even among them care must be balanced against respect, which is a very delicate matter. Al may find it very difficult to arrive at an acceptable decision. But however he concludes, we should again note several points. First, as we have seen, Al’s character initially manifests itself not only in his apprehension of the situation as calling for a response from him, but also 10 11

See ENIII 3, 1112b15; VII 8, 1151a16-17; EE II 10, 1227a7-10. Broadie (1991), p. 240; Price (2011b), pp. 221-3.

220

Ø. RABBÅS

in the initial response that sets up the deliberative inquiry. It is because Al is an honest and just man that he thinks he should repay his debt to Ben. Secondly, as Al’s deliberation proceeds, his character continues to manifest itself in the considerations that come up along the way and in the way he responds to these. It is his justice, again, that makes him see the relevance of Ben’s involvement in criminal activity, and it is his friendship (which may not be a virtue, but is closely related to it, on Aristotle’s view; cf. VIII 1, 1155a3-4) that makes him care for Ben’s well-being, but also alerts him to the tension between this and the obligations of respect. Thirdly, therefore, we should note that as Al’s deliberation proceeds and he comes to see that his initially assumed end of repaying his debt should be abandoned (let us assume that his is what he ends up deciding to do), what makes him realize this is his reflection on what exactly is involved in his deeper commitments to justice and friendship. These virtues (or virtue-like commitments) make him apprehend the situation in a certain way and respond to it accordingly, but these very same virtues also require further reflection on the basic values that are constitutive of them. But what in the end enables him to complete his deliberation is his clarification of his own basic value commitments. That is to say, while he is clear that he wants to respect the requirements of justice as well as the requirements of friendship, what these requirements entail, precisely, is not clear, especially not in a situation where they seem to come into conflict. What he needs to clarify in the course of his deliberation, therefore, is in part what the deeper point or rationale of these commitments is, and in part he has to ask what they entail. There is thus a need for Al to articulate and include these deeper value commitments in his deliberation. He has to do this if he is to arrive at a correct determinate conclusion: a decision. Here it may be useful to distinguish between the processofreasoning leading to a conclusion in the form of a decision, and the logicalstructure of an argument justifying this conclusion. This distinction is important because the application of the term “starting-point” or “principle” (ἀρχή) is different in the two cases. It is also important to see that this distinction may come out differently in the case of theoretical reasoning from the way it does in practical reasoning or deliberation. As Aristotle conceives it, in the case of theoretical reasoning what comes first in the process of reasoning is not the premises from which the conclusion will be deduced (or inferred), but rather the conclusion — which at the outset has the status of a hypothesis, something to be investigated and, if possible, justified or

THE FUNCTIONS OF PRACTICAL DELIBERATION

221

demonstrated. One develops such a justification, or proof, by finding the premises that are such that the conclusion follows — or can be logically derived (demonstrated) — from them. So these premises come last in the process of reasoning, but first in the proof demonstrating the (necessary) truth of the conclusion. In the practical case, however, this seems to be different.12 For here the starting-point or principle is the same both in the process of reasoning (deliberation) and in the argument justifying the conclusion (decision): it is the particular end that provides the focus and direction of the deliberative process. However, if the account of deliberation outlined above is correct, we can see how premises supplied by commitments at a higher level than the provisional end set up by the agent’s character as the starting-point for his deliberation, can enter as deliberation proceeds, and assume the role of basic premises in the argument that finally justifies the decision that concludes deliberation. So even if deliberation doesn’t start with the aim of acting for the sake of these more fundamental values or ends, reference must be made to them if an adequate decision is to be arrived at. They constitute the ultimate ground both for the discovery and for the justification of the conclusion of deliberation, i.e., the decision. So in the argument that justifies the decision, these more fundamental ends will figure as the first starting-point or principle, as the archē, even though they enter at a later stage in the thought-process whose immediate archē is the assumed end set up by the agent’s character. The reasoning movement thus goes in two directions from the initially hypothesized end: “downward” to the concrete decision and “upward” to the ultimate justifying grounds. (The significance of this point will become clearer in section 3.) 2. The Demonstrative Function of Deliberation Now, continuing the line of thought that I have just outlined will show, I think, that the question of the function or task of deliberation is a very deep one that goes right to the heart of how Aristotle thinks of human beings and their life. Moreover, I believe that one reason why interpreters struggle to understand his thought here, is that they tend to assimilate too quickly Aristotle’s thought to more familiar modern ways of thinking about human agency and ethics. Let me explain. 12 See Broadie (1991), pp. 226, 229-30; Broadie (2002a), pp. 296-300; Broadie (2002b), passim.

222

Ø. RABBÅS

As we have seen, on Aristotle’s view deliberation starts with a hypothesized end set up by the agent’s character, which then initiates the deliberative process aiming at a decision determining what to do. Here it is natural for modern thinkers to take the function of this process to be to bring about something, viz. a decision to perform a certain action in order to realize the assumed end, i.e. to cause the agent to be motivated to perform the action. This decision is then in turn somehow supposed to bring about something further, viz. the specified action itself. However, if the purpose of deliberation is to bring about the decision (i.e. to cause the agent to be motivated to act in a certain way), one might think that this decision could just as well have been brought about in other ways, say by tossing a coin, following another’s instructions, or simply by luck. The result, the decision (or motivation) to do x in order to realize y, would be the same. This decision is logically independent from the deliberative process and defined solely by reference to the action it is supposed somehow to cause. But Aristotle thinks otherwise. For him the decision, which is a rational motivation, would not be the same if it were arrived at by tossing a coin as if it were the result of deliberation. The decision is defined not merely in terms of what is supposed to follow it — what it is a decision to — but also in terms of what it follows — what grounds it. A decision is a decision not merely to do x-for-the-sake-of-y (or in-order-to-accomplish-y) but to do x-for-the-sake-of-y-on-the-ground-that-z, where ‘z’ denotes some rational principle. Thus, e.g., Al may decide to withdraw money (x) in order to repay his debt (y) because that is what he takes justice (promise-keeping) (z) to require. But he might have done it for another reason, say because he was afraid of being killed by a hit-man (z’), or of being sued (z’’) — or because he flipped a coin (z’’’) — but then his action would not have been a virtuous one. The function of deliberation is therefore not merely (1) to discover x by making the determinable (or indeterminate) end y sufficiently determinate to be realizable, but also (2) to exhibit or demonstrate that and why x is the right way or means to realize y, as well as why pursuing y in this situation is justified.13 Tossing a coin, for instance, might accomplish (1) but not (2). 13 As James V. Allen nicely puts it: “if the object of theoretical enquiry is the grasp of a reasonedfact, that of practical enquiry is the performance of what we may call the reasoneddeed.” Allen (2015), p. 54.

THE FUNCTIONS OF PRACTICAL DELIBERATION

223

Deliberation is thus not to be regarded as a causal process leading to a decision (motivation) but, rather, as a form of activity or actualization (ἐνέργεια), whereby an agent rationally determines himself to do something — the English expression “makes up his mind” is very apt here. Since man is a rational animal whose reason can manifest itself both in desire and in thought, both the setting of the end through the desiderative/ orectic response and the determination through deliberation of the ways or means of realizing the end are ways the agent as such, as a rational human agent, a ζῷον λόγον ἔχων, truly acts. And only in this way will his action really be his own action, and only then will he be acting well (εὖ πράττειν), and hence living well (εὖ ζῆν). It is only then that his activity will be rational by manifesting the agent’s rationality (reason), so the function of deliberation is not merely to determine or discover what is to be done — what I called its heuristic function — but also to make the decision and action rational — this is the demonstrative function of deliberation. But there is another aspect to this demonstrative function as well, which can be seen from the following. Aristotle distinguishes between full virtue and conditions that resemble or approach full virtue without being so. Suppose Al thinks that he should repay his debt, and willingly does so, simply because that’s how he has been brought up: “This is the thing to do!” “You just don’t fail to repay your debts!” But if that is so, we may wonder whether he is really fully virtuous, or whether he merely has some kind of proto-virtue that resembles, but fails to be, full virtue. Aristotle seems to hold that if this were the case, the man would not have full virtue but merely some approximation. Here we should note two important things about these approximations to full virtue, for they can be of two kinds, both defective but in different ways. First, a person may be just, but inanon-reflectiveway. This is the state of the person who repays his debt because that is “the thing to do”. He knows this, i.e. that this is the thing to do, in a certain sense of “know”, but he is unable to say much about why this is so. He has “the that” but not “the why”, as Aristotle puts it (I 4, 1095b2-13). However, it is important to see that when he decides on what to do, he does so because he thinks this is the right thing to do — he decides to do the thing for its own sake, not because he has some ulterior motive for doing so. Secondly, however, a person may be such that he tends to act in the same ways as a just person would, but for reasons that a just person

224

Ø. RABBÅS

would not have, i.e. forthewrongkindsofreason. For instance, he might repay his debt (i) because he is afraid of the disgrace that might befall him if he didn’t, or he might do it (ii) because he regards this as prudent in the long run, or he might do it (iii) for fear of losing the pleasure of his creditor’s company. In short, the ultimate ground for the conduct of the person who repays his debt might be the desire (i) for honour, (ii) for profit, or (iii) for pleasure. Whenever a man enters or responds to a situation, he will do this within a certain framework defined by his idea of what is at stake, what doing well is all about, at the end of the day: is it (i) honour? (ii) profit? (iii) pleasure? At this point we need to address a question that has raised controversy. Some interpreters have argued that no such general idea of “what is at stake, what doing well is all about, at the end of the day” could be regarded as either required by the nature of virtuous deliberation or as ethically advisable. This “Grand End View” of practical reason, they argue, is unwarranted for both textual and philosophical reasons. 14 Instead, they argue, all that it takes for deliberation to be complete is the presence of a “dual perspective”: “We need to ascribe to the deliberator a dual approach that involves two features: [1] a focus upon a limited and accessible target, and [2] an openness to whatever considerations come into play. This involves distinguishing [2] ‘the defining end’, which is acting well (eupraxia), from [1] ‘the end premised in deliberation’, which is whatever it may be (such as health or wealth).” (Price (2011b), 222; italics and numbering mine)

These latter, moreover, should be thought of as “goods such as wealth and pleasure that are often but not always properly pursuable: the vicious agent pursues them too often, the virtuous agent only on occasions when they can be pursued commendably.” (ib. 221-2)

The mistake of the proponents of the Grand End View, on this approach, is that they take the notion of eudaimonia as the ultimate end to be a highly general specification of a substantive end to aim at in life, one that could figure as a reason for choosing one course of action rather than another. Instead, these critics claim, eudaimonia is the “formal” end of acting well, much like truth is the formal end of belief.15 14 15

See, e.g., Broadie (1991), ch. IV 4; Price (2011b), ch. C2 III. See Price (2011b), p. 23, 40 n. 8.

THE FUNCTIONS OF PRACTICAL DELIBERATION

225

I think this is mistaken. First, I think that the notion of eudaimonia, or a reference to it, can figure as — or in — a justification for choosing one course of action rather than another. And, more importantly, perhaps, a reference to eudaimonia can figure as the ground for criticism of the way a person lives his life. (We saw this above, pp. 9, 11.) But, secondly, in order for it to function this way, it has to be represented in a certain, determinate way. That is why the notion of eudaimonia as the ultimate end of living well must be understood as more substantive than the critics of the Grand End View have it.16 Two passages are particularly relevant here. One occurs at the opening of chapter I 5 of the Nicomachean Ethics where Aristotle reviews the commonly acknowledged candidates for the highest good or happiness. These are the life of pleasure, the life of honour, and the life of contemplation (or however we should translate θεωρία). These ends are not just any ends that one might pursue, but ends that answer the question what eudaimonia or success in life consist in. (The same holds for the ends mentioned in the passage from the EudemianEthics quoted below.) These ends are not merely ends that provide some specific content in life, they ground a conception of what living successfullyis. The hedonist takes it that whatever one does is done as a way of achieving pleasure, which is what he takes eudaimonia to consist in; the honour-lover takes it that whatever one does is done as way of achieving honour, which is what he takes to constitute eudaimonia, and so on. Only a very limited range of candidates could possibly play the role of such a highest good — music, football, collecting stamps could not, even though — indeed, precisely because — they might be pursued “too often”. Indeed, Aristotle argues in “the function argument” (I 7) that there is really only one candidate that is eligible for the title of the highest end of living well, and that is “activity of soul in accordance with virtue”.

16 And also why it differs from the notion of truth in relation to belief. We don’t justify our believing something — e.g., that a is F — by referring to the truth of this proposition, but by showing (or trying to show) that a satisfies the criteria for being F; talking about truth merely summarizes this specific, substantive justification. But we choose to live one way rather than another because we take this way of living to better satisfy the criteria we accept for living well as such: our conception of what living well amounts to does provide us with a substantive “reason for choosing one course of action rather than another”, and that is why we can point to this conception to justify or criticize a person’s conduct or way of life.

226

Ø. RABBÅS

In the passage from I 5 Aristotle is juxtaposing three different ways of living, three different forms of life, distinguished by their different basic orientations. These orientations manifest themselves not merely in what the agents would reply to abstract questions in the reflective mode about what makes life worth living, or what is of the greatest importance in life,17 but more significantly in how they concretely approach particular situations in which they find themselves in the course of living their lives. Thus, the end of wealth or of pleasure is neither the purely formal end of living well, which may play the same role as the notion of truth does in belief-formation, nor the material but highly general end that Price takes them to be — ends that “the vicious agent pursues … too often, the virtuous agent only on occasions when they can be pursued commendably”. The person who is vicious on account of his love of pleasure, is the pleasure-seeker who approaches any situation looking for pleasure — that is, he will be seeking to satisfy his desires whatever they are since that is his conception of eudaimonia. The flaw of this vicious person is not that he pursues pleasure “too often”, but that he does so allthetime — whatever he does, he does as a way of pursuing pleasure on the ground that this is eudaimonia. Similarly, a person who is vicious on account of his love of honour will always be keenly sensitive to what will bring him honour — or, more realistically, I take it — what he needs to do to avoid disgrace and the contempt of others, and will do what he does as a way of pursuing honour, i.e. — he thinks — eudaimonia. He is the thoroughgoing, conventionalist honour-lover. We should note that this does not have to mean that people’s conception of the highest end or eudaimonia needs to be explicitly articulated, or that it is the result of ethical reflection. On the contrary, Aristotle makes clear that most people don’t have such an articulate and reflective conception of the highest good. Nevertheless, they do implicitly have such a conception, as can be seen from the way they live (I 5, 1095b14-17). An even clearer expression of this thought comes in a well-known passage from the Eudemian Ethics, which parallels the opening of the NicomacheanEthics. Aristotle says: “Focusing our attention on these matters, everyone capable of living by their own decision ought to lay down some aim for living finely, be it honour or reputation or wealth or education, which they will look to in the 17

Cf. Broadie, op.cit.

THE FUNCTIONS OF PRACTICAL DELIBERATION

227

performance of all their actions, since not organizing one’s life in relation to some goal is a mark of great foolishness. We must, then, first and above all, determine, without haste or sluggishness, in which of our goods living well consists, and which by their absence prevent its attainment.” (I 2, 1214b6-14)18

I don’t see how the end here can be the purely formal end that the opponents of the Grand End View take it to be; nor do I see how the four examples Aristotle gives can be taken merely as the “formal end” of doing what is best in this particular situation, all things considered. Aristotle’s point surely must be stronger and more “material” or “substantive” than that, namely that what one does in this particular situation, the striving to attain this particular objective, is a genuine expression of one’s fundamental ethical orientation in life (which is how talk of a “Grand End” should best be taken). Moreover, the claim is not that everybody in fact leads their lives this way, but rather that this is what they ought to do. The claim is normative — it expresses an ideal, but an ideal that is based on a realistic picture of how ethical deliberation often proceeds.19 We can now see that only if we fully spell out the reasons underlying the person and his deliberations, can we determine whether the person really is virtuous, or whether he merely has some approximation or simulacrum of it. Aristotle makes this point when he states, at the beginning of his discussion of decision (προαίρεσις) in III 2, that “decision seems to be something highly germane to virtue, and to indicate the differences between people’s characters more than actions do.” (1111b5-6) Moreover, he also explicitly discusses the distinction between full virtue and states that are not virtue but somehow approximate or resemble them in his investigation of courage in III 8. Here he says that there are five 18 Translation Inwood and Woolf; reading δεῖ θέσθαι at b7 with Gigon. Quoted, in Solomon’s translation, in Broadie (1991), on p. 4. This passage, and the fact that Broadie quotes it, is discussed by Richard Kraut in his review of Broadie’s book; Ethics 103 (1993), pp. 361-74, sect. II. 19 A further, more contextual, piece of evidence in favour of this interpretation is that the idea of the existence of limited set of life forms, or “lifestyles” as we might say, is a widespread topos in ancient Greek literature. Perhaps the best known instance of it is in Pericles’ Funeral Oration where Pericles distinguishes three kinds of life that are all to be found among the Athenians, unlike other peoples (see Thucydides II 40.1-2). But the division is also fundamental in Plato; see, e.g., Republic VIII 580d-583a where Socrates distinguishes not only three parts of the soul, but also three kinds of pleasure and three corresponding types of human being; see also the way Socrates and Callicles are contrasted as representatives of two distinct kinds of life in Gorgias 484c-486d.

228

Ø. RABBÅS

conditions that resemble courage in decreasing order, and the crucial thing for us is that what disqualifies them from being genuine or full courage is precisely their fundamental value commitment. The first kind, the one closest to full courage, is “civic” courage (πολιτικὴ ἀνδρεῖα). This condition is disqualified from being full courage because “citizens seem to withstand the dangers facing them because of the penalties inflicted on them by the laws’ and people’s reproaches, and because of the honours” (1116a18-19), unlike the fully courageous man who “makes [his] choice (αἱρεῖται) and stands firm because doing so is fine (καλόν), or because not doing so is shameful (αἰσχρόν).” (III 7, 1116a11-12) The next deficient state in line belongs to the person who performs courageous acts under constraint (ἀναγκαζομένους): he fails to be really courageous because he acts “through fear (διὰ φόβον), and in order to escape not what is shameful but what is painful” (III 8, 1116a30-2). The fourth type belongs to the person acting out of temper (θυμός), and here Aristotle gives the clearest statement of what is lacking in order to count as full virtue. He says: “the ‘courage’ that comes about through temper does seem to be the most natural form, and to be courage once the factors of decision (προαίρεσις) and the end for the sake of which (τὸ οὗ ἕνεκα) have been added.” (1117a4-5) What must be added, I take it, is what we have just seen: correct deliberation starting from an immediately and truly apprehended end and referring to a true apprehension of what is at stake in a life lived well. So, to conclude this section, deliberation has a second function besides the heuristic one: the demonstrative. This demonstrative function has two aspects: Through deliberation the agent demonstrates (a) that and why his decision is the right one (this makes his decision a reasoned one), and also (b) that he himself is the sort of person to deliberate well and act correctly. However, I want to submit that there is still a third function of deliberation, the formative or paideutic function, which can be seen if we continue the line of thought just outlined. 3. The Formative Function The basic idea is that through repeated deliberation an agent will go through a process of ethical learning or formation (παιδεῖα). In part the point is the simple one that we become better at something by doing it over and over again — this holds for all virtues, and phronēsis, which is

THE FUNCTIONS OF PRACTICAL DELIBERATION

229

the virtue by which we “deliberate well about … what sorts of things conduce to the good life in general” (VI 5, 1140a26-8), is no exception. Deliberation is an exercise of reason, this can be done more or less well, and through experience we learn to do it better. But the point runs deeper than that. As we saw earlier, deliberation starts from an end set up hypothetically as to-be-pursued, and the deliberative question is what it will take to actually pursue this end successfully. The hypothesized end is the starting-point or principle of this process of deliberation. But we also saw that this end manifests the agent’s character, i.e. his general evaluative and motivational outlook on life and what matters. This outlook can be expressed in a higher principle or set of principles, e.g. that promises (including contracts) are to be kept, that one should care for and be loyal to one’s friends, that dangers are to be endured if greater values are at stake than the ones that are threatened, and so on — in general: that one should live well and pursue eudaimonia. However, as we also have seen, these principles are often so general and indeterminate that it is unclear how they should be brought to bear on the particular situation at hand. It is one thing to be a committed friend and quite another to know exactly what would be the right thing for a friend to do in the particular situation he finds himself in. This is the task for deliberation. But as one gains experience from deliberation on several occasions where one’s commitment to friendship is relevant, one will gradually get better at bringing one’s general commitment to friendship to bear on the situation at hand — at transmitting one’s desire from one’s general ends to the particular, determinate decisions one ought to perform. This learning process has two aspects. In part the learner’s ability to apprehend the particular situations he is confronted with will develop and become richer and more fine-tuned and flexible, so he will no longer need to deliberate as much as he originally did. As a result of this his “moral sense”, to borrow that loaded term, will develop to the point where the need for deliberation will diminish, though hardly disappear totally — real life is too messy to allow for that. But there is a second aspect to this learning process as well. For as the agent develops his moral sense and becomes better at focusing or transmitting his motivation from the hypothesized end to the particular decision, he will also get a clearer and more precise grasp of this end itself. Recall our man, Al. He owed his friend, Ben, some money he had borrowed and thought he should repay his debt. But, being Ben’s friend,

230

Ø. RABBÅS

Al also thought he should care about Ben’s well-being and, since Ben had involved himself in some dubious and risky affairs and was likely to spend the returned money on these affairs, Al came to think that he had better withhold the money for the sake of Ben’s well-being. However, as he continued to deliberate, Al also realized that friendship involves respect, which in this case would speak in favour of repaying the money after all and let Ben take responsibility for his own life. This deliberative process illustrates that a particular deliberative process may necessitate reflection on the higher end to be realized through action, and it will also direct the focus for this reflection. When this happens repeatedly, in various circumstances, the agent will gradually gain a clearer and more precise grasp of his own higher end. This enhanced grasp will not only enable him to articulate what the highest end involves, but it will also enable him to separate the essential from the inessential on particular occasions. Al, for instance, comes to realize that even though a concern for one’s friend’s well-being is often what friendship requires, the essence of friendship is not so much a matter of this kind of altruism but, rather, a concern for his friend as a person, as “another self”: whatever else this enigmatic notion amounts to, it includes the idea that my friend is a person like myself, i.e. somebody engaged in living his life, which is to say leading and shaping his own life, and therefore bearing the responsibility for it. This is one reason why only adults can be friends, properly speaking; children cannot, nor can animals, let alone inanimate objects: they are not persons. Let me, in conclusion, make clear exactly what deliberation does and does not do on the view I am proposing. (1) It does not setup the higher end, for this end is given at the outset as what provides the focus and impetus for the entire deliberative problem in the first place. The deliberator assumes the end of friendship as he deliberates. (2) Nor is the clarification of this higher end the object of deliberation, or the content of the conclusion. One does not deliberate inorderto clarify one’s higher end of friendship; the object is to arrive at a determinate decision about action, and the content of the conclusion is the action to be performed. Rather, (3) in order for deliberation to achieve its objective: a rationally justified decision, reflective clarification of the highest end is necessary. This clarification, thus, is a matter of reflection, not deliberation, but it is necessitated by, and gets its focus and direction from, the concrete deliberative process. Reflection is not deliberation, therefore, but it is

THE FUNCTIONS OF PRACTICAL DELIBERATION

231

necessary for deliberation to come to completion; it is not something merely undertaken by those with a special inclination for the part of philosophy called “ethics”. But this is where the third function of deliberation becomes apparent. For when a deliberator has consistently and competently deliberated well on numerous and varied occasions, thereby gradually developing his moral sense, he will also have gained a clearer and more precise grasp of his ends or principles: friendship, generosity, justice, and so on — in general: eudaimonia or living well. * * * In conclusion, my claim is that deliberation determines the right thing to do, and thereby demonstrates the correctness of decision and action. But thereby it not only manifests the agent’s character, but it also contributes to the development or formation of this character. We could therefore, perhaps, conclude that deliberation has one single and fundamental function, with three aspects, and that is the function of actualizing human nature as a rational animal: in wise deliberation, as in virtuous desire, a human being exercises his human nature — he is living well.20 LITERATURE ALLEN, James V. (2015), “Practical and theoretical knowledge in Aristotle” in Henry, D. & Nielsen, K.M. (eds.), Bridging the Gap Between Aristotle’s ScienceandEthics, Cambridge: Cambridge University Press, pp. 49-70. BROADIE, Sarah (1991), EthicsWithAristotle, Oxford: Oxford University Press. BROADIE, Sarah (2002a), “Introduction” and “Commentary” in Broadie and Rowe (2002), pp. 9-91 and 261-452. BROADIE, Sarah (2002b), “Interpreting Aristotle’s Directions” in J. Gentzler (ed.), Method in Ancient Philosophy, Oxford: Oxford University Press, pp. 291-306. BROADIE, Sarah & ROWE, Christopher (2002), Aristotle. Nicomachean Ethics. Translation, Introduction and Commentary, Oxford: Oxford University Press. 20 I am grateful to the participants at the conference in Helsinki, especially to David Charles, Jörn Müller, and Pavlos Kontos for pressing me to clarify my view of the third function of deliberation — I am uncertain whether they will be satisfied by the present version. Thanks also to Thomas Kjeller Johansen for comments on the penultimate draft of the paper.

232

Ø. RABBÅS

CHARLES, David (2015), “Aristotle on practical and theoretical knowledge” in Henry, D. & Nielsen, K. M. (eds.), BridgingtheGapBetweenAristotle’s ScienceandEthics, Cambridge: Cambridge University Press, pp. 71-93. COOPER, John M. (1975), ReasonandHumanGoodinAristotle, Cambridge MA: Harvard University Press. LEAR, Jonathan (1988), Aristotle.TheDesiretoUnderstand. Cambridge: Cambridge University Press. MOSS, Jessica (2011), “’Virtue Makes the Goal Right’: Virtue and Phronesis in Aristotle’s Ethics” in Phronesis 56, pp. 204-61. PRICE, Anthony W. (2011a), “Aristotle on the Ends of Deliberation” in Pakaluk, M. & Pearson, G. (eds.), MoralPsychologyandHumanActioninAristotle, Oxford: Oxford University Press, pp. 135-58. PRICE, Anthony W. (2011b), VirtueandReasoninPlatoandAristotle, Oxford: Oxford University Press.

PRACTICAL TRUTH IN ARISTOTLE1 Sarah BROADIE

1. Introduction Aristotle’s notion of practical truth occurs only at Nicomachean Ethics VI, 1139a26-30 (cf. 18). The main topic of ENVI is the intellectual virtue phronēsis, practical wisdom. This is the practical analogue of sophia, theoretical wisdom. Aristotle elucidates phronēsis partly by contrasts with productive expertise (technē) and sophia, partly by exploring its relation to “ethical” or “moral” virtue. Our text opens by distinguishing two parts of the rational part of soul: one concerned with things whose principles cannot be otherwise, the other with contingent things. Each rational part has its own kind of virtue, i.e. disposition that is best for the part’s proper work or function (ἔργον); and the proper work of both parts is truth.2 [1139a15] “So we need to understand what the best disposition is of each of these two, since that is the virtue of each, and the virtue relates to the proper work. Now, there are in the soul three things responsible for (τὰ κύρια) practical activity (πρᾶξις) and truth (ἀλήθεια): perception, intelligence (νοῦς), and desire (ὄρεξις). (But of these perception is not the source of any practical activity; this is clear from the fact that [a20] brute animals have perception but do not share practical activity.) What assertion and denial are in thought, pursuit and avoidance are in desire. So: since ethical virtue is a disposition for decision (προαίρεσις), and decision is deliberative desire, it must therefore be the case that both the logos is true and the desire correct (ὀρθήν) if the decision is a sound (σπουδαία) one, and [a25] the one must assert (φάναι) and the other pursue the same things. This (αὕτη), 1 I am grateful to Ben Morison for his uncompromising criticism of my earlier paper on this topic, and to Marko Malink and Jessica Moss for improvements to the present one. 2 “Truth”, “true” and “asserting truly” occur many times in ENVI: 1139a18; 24; 26-31; b12-15; 1140a10; 21; 1140b5; 21; 1141a3; 1142b33; 1143a24. See also EE II, 1221b27-30.

234

S. BROADIE

then, is the thought (διανοια) and the truth that is practical (ἡ ἀλήθεια πρακτική).3 In the case of thought that is theoretical and not practical nor yet productive, the well and badly are the true and false, for this is the business (ἔργον) of every intellectual part. But in the case of the part that is practical and [a30] thinks, is truth in agreement with correct desire. Decision is a source (ἀρχή) of action — the source from-which, not the one for-the-sake-of-which — and the sources of decision are desire and the logosthat says what the end is. Consequently there cannot be decision without intelligence and thought, nor without an ethical nature (ἦθος). For in the sphere of action doing well (εὐπραξία) and its opposite are impossible without thought and ethical nature. Thought as such moves nothing; what moves, rather, is thought that is for the sake of something and practical. For this in fact controls productive activity ... Decision is desiderative intelligence or desire that has been thought out (διανοητική), and this kind of source is man .... To sum up (δὴ): truth is the business of both the intellectual parts; so their virtues are those states whereby each part will be truthful (ἀληθεύσει) to the highest degree” (1139a15-b13).4

A question to bear in mind is whether the phrase “truth that is practical” (or “practical truth”) expresses a unitary concept or an amalgam of independently intelligible units: truth and practical. The concept is unitary if the truth of practical truth is to be understood as “inherently” practical (this could be cashed in different ways), or if the two sides turn out to be united at source: in any given case of practical truth, whatever legitimates calling it “truth” also legitimates calling it “practical”.5

3 This translation takes αὕτη as subject, ἡ διάνοια καὶ ἡ ἀλήθεια πρακτική as complement; thus e.g. Irwin (1999). An alternative is: “This thought, and truth, is practical”, taking αὕτη ἡ διάνοια καὶ ἡ ἀλήθεια as subject and πρακτικήas complement. (On “kind of” here, see Pakaluk (2010), pp. 145-159.) This may indeed convey Aristotle’s thought, but argument is needed, and will be given below, that he recognizes a distinct kind oftruththatis practical; so I have chosen a translation that is neutral on this score. On this translation what is the referent of αὕτη? It would be stylistically natural (1) to understand it as the nearest preceding feminine item, i.e. the sound prohairesis; many commentators reject this out of hand for philosophical reasons. Alternatively (2), hautē is due to attraction to the gender of dianoia and alētheia, and refers either (a) to the prohairetic complex true-logos-and-right-desire-etc. or (b) to the true logospart of that (mentioned at 1139a25). 2b is stylistically somewhat strained. Readings 1 and 2a support proposals A and C below. Proposals B, B*, and C* require 2b, or 2a taken as a whole-for-part synecdoche (the complex is referred to in virtue of containing the true logos). 4 Cf. EE II, 1221b28-30. 5 Olfert is the first, as far as I know, to make conceptual unity an explicit desideratum for interpretations of “practical truth”, Olfert (2014).

PRACTICAL TRUTH IN ARISTOTLE

235

I shall consider two kinds of interpretative approach. One, by far the most common, seeks to make sense of “practical truth” in terms of a fairly straightforward and familiar notion of semantic truth. A proposition is semantically true or false depending on whether the facts or realities are or are not as it says. In Aristotle’s hands this semantic notion is stated in terms of assertion. I shall consider several variants of this view, including G. E. M. Anscombe’s seminal and distinctive interpretation.6 According to the second type of approach, Aristotle’s practical truth is different from assertoric truth although closely related. 2. Assertoric Truth and Theoretical Understanding A natural starting point is Aristotle’s famous definition in Metaphysics IV: “To say that what is, is not, or that what is not, is, is false; whereas to say that what is, is, or that what is not, is not, is true” (1011b25-27). “To say” here surely means “to assert”, mentally or in words. I shall not undertake the close discussion that the definition demands.7 The main point for now is that the definition attaches ‘true’ and ‘false’ (or ‘being true’, ‘being false’) to judgments or assertions rather than to the propositional contents of assertions. See also a few lines down: “For every object of thought and understanding, thought either affirms it or denies it (ἢ κατάφησιν ἢ ἀπόφησιν) — this is clear from the definition - whenever it expresses truth or falsehood (ἀληθεύῃ ἢ ψεύδηται) (1012a2-4). I shall use ‘assertion’ to cover both affirmation and denial. At the basic level what is asserted are simple sentences (these include negations), which may be existential or predicative, singular or quantified. The truth and falsehood that attach to assertions I shall call ‘assertoric’ truth and falsehood. Truth and falsehood on the semantic or linguistic level are the truth and falsehood of assertions.8 On this basis, how should we understand the ENVI passage about practical truth? Well, since the practical intellect, there, is compared and contrasted with the theoretical intellect, let us first see how the definition in Metaphysics IV lines up with the theoretical side of things. In the 6

Anscombe (1965), pp. 143-158; Anscombe (1993). See Crivelli (2004), pp. 132-6. 8 Crivelli, ibid., pp. 86-95; also 7 – 39 passim. “Assertion” here covers any laying down of premisses, including ones that are assumed for the purpose of reductioadabsurdum. My thanks to Marko Malink for this point. 7

236

S. BROADIE

ENVI passage above Aristotle says that the well and the badly are the true and false, this being the ergonof every intellectual part of the soul. (i.e. the ergon is achieving the true and avoiding the false.) So does Aristotle hold that theoretical thought’s going well is nothing other than its achieving assertoric truth, and that doing this is the proper business, task, or object of the theoretical intellect? In a simple or straightforward sense of this question, the answer is ‘No’. This is because the object, for Aristotle, of theoretical intellectual activity is epistēmē, scientific knowledge, which involves grasping things on the basis of their causes or principles.9 We haveepistēmē in relation to p only if we understand why p is the case — only if we see it as grounded on something more fundamental: its cause or principle. More precisely: epistēmē in a field is a disposition for understanding and explaining things in that field. A, who has discovered that pis grounded inq, can teach this to B, i.e. bring B to see p as grounded in the more fundamental q, and hence to understand p. This teaching is an exercise of epistēmēby A although we would not expect A necessarily to renew, on each teaching-occasion, her or his own understanding of p by a fresh act of understanding; but we surely do assume that A’s ability to get others to understand p springs from A’s own prior act or acts of understanding p in terms of q. Such acts of understanding are what epistēmē is “all about”: itis a disposition defined by such acts, and these acts give value to the disposition. Thus the ultimate object of theoretical inquiry and teaching is the act of understanding. So is the disposition for such acts nothing other than a disposition for making assertions that are true (and none that are false) in the sense explained in the MetaphysicsIV definition? It does Aristotle no favour to ascribe to him such a view. If ‘p because q’ expresses an act of understanding p, the subject has done more than assert truly that pand thatq,10and also more than assert truly that p,and that q, and that qis cause or ground of p.For it is possible to do this without understanding or ever having understood p in light of its cause q, or without seeingq as cause of p.One could have it on external authority that pand that qand that the latter is cause of the former, yet still fail, oneself, to understand p on this basis. Given a suitable authority 9 At 1141a1720 Aristotle distinguishes the ability to derive, which he calls epistēmē, from the grasp of the principles as such, which he calls nous; the combination of these he calls sophia. 10 The point is noted by Olfert (2014), p. 210.

PRACTICAL TRUTH IN ARISTOTLE

237

K, for any rif Kasserts that r then r. So anyone in the circle of adherents could be placed by K in a position to assert truly that p and that q and that qis cause of p. But Khas not thereby transmitted understanding of pon the basis of q.In factnothing in this picture guarantees that even K understands pon the basis of q. So (short of supposing that Aristotle is oblivious to all this): if we still want to explain the theoretical intellect’s objective in a way that gives a central role to assertoric truth, we must take a slightly more subtle line. Here is a suggestion. Take the general notion of an intelligent Aristotelian theoretical inquirer. The notion itself implies that the inquirer’s aim is to understand something, p, in terms of its cause or ground, so that achieving this will constitute success. Then exhibit the understanding as supervening on the moment of the inquirer’s coming to possess all relevant assertoric truths. Admittedly, this is idealized. We have to assume an inquirer who, as well as being intent on coming to understand some datum, is undistracted, clear-headed, and thorough; but granted this, it seems plausible that access to all the relevant assertoric truths will ipso facto result in the “click”, the falling into place, that is the act of understanding.11 3. Practical Truth: Some Proposals Practical truth is evinced in a sound prohairesis, combining a logosfactor and a desire-factor which are as they should be: the logostrue, the desire correct, and in concord.12 The prohairesis, according to Aristotle, is formed by deliberating on what it is good for me to do so as to implement an end E. The short answer is ‘It is good to adopt means M’. Aristotle tends to focus on the short answer: thus he says that deliberation is of the means - i.e. is an inquiry about the means - to E, and that the eventually reached prohairesis is of the means i.e. specifies a determinate answer to precisely that inquiry (1144a8-9; III, 1111b26-29; EE II, 11 The aim here is not to give a non-circular analysis of theoretical understanding but to characterize it for the purpose of clarifying its practical analogue. This sketch also oversimplifies in treating understanding as supervening on the accumulation of all relevant assertoric truths; in fact we often accept a crucial assertoric truth on the ground that the best explanation requires it. 12 Olfert’s (ibid.) helpful discussion of this shows in detail the practical nature of both factors and how they respectively derive truth and correctness from a single standard, i.e. the unqualified human good (pp. 219-30).

238

S. BROADIE

1226b9-13; 1227a5-18). Deliberation resulting in the specification of M starts from a premiss such as ‘It would be good to get E’, other premisses being supplied by the particular situation. The fact that deliberation gets its direction from an end-specifying premiss does not mean that the deliberator is stuck with aiming for the end Eas if it cannot be abandoned (or can be abandoned only if found to be physically unfeasible). Deliberation with a view to Emay arrive at specifying M, but what is thereby specified as good to do is M-as-leading-to E. This is the longer answer to the deliberator’s question: it says more fully what one would be adopting in adopting M.In some cases the only discoverable means to E, given the circumstances,turns out to be simplyunacceptable on moral or prudential grounds, regardless of the nature of E.Since M is rejected, Eas end is abandoned, anyway here and now. This is analogous to (or is a case of) modustollens:we reject a premiss because we reject the conclusion. But often the only available means is not completely unacceptable but is here and now unacceptable because under the circumstances the specified end is not worth the means: it would be disproportionate, hence wrong, to pursue that end by this means, given the circumstances. This situation may be compared to the one in theoretical inquiry where in attempting to explain pwe reject q as a possible explanation even though we grant the truth of q: we reject its claim to be cause of p because it does not fit (e.g. because instances of q are known also to occur in the absence of any instances of p). Similarly, we reject M’s claim to be the sanctioned means to E because of lack of practical fit under the circumstances. What is distinctive about the Aristotelian prohairesis, and why it is essentially practical, is that it is an endorsement of some action as good or appropriate simpliciter, not merely good from a limited and overrideable point of view. It is an all-things-considered rational decision, one that represents my sense of how under these circumstances I as a human being should simply act or behave, or what I should simply go for, or what counts as unqualified doing well (εὐπραξία), for me here and now. This is by contrast with what counts as my acting well or doing well as a builder, farmer, money-maker, or lover of some type of pleasure. There are two important consequences. First, the prohairesis necessarily involves desire to do what, through all-things-considered deliberation, the agent has determined would be good to do (1139a4-5). Secondly, if truth attaches to the Aristotelian prohairesis or to the logos-factor within it, this truth includes truth about what is genuinely valuable in human life.

PRACTICAL TRUTH IN ARISTOTLE

239

The deliberative conclusion, ‘It would be good to implement M’, is part of what Aristotle at 1139a24 calls the logos. But the logos surely comprises the whole deliberative argument: premises, conclusion, inferences, and underlying assumptions, including ones about value and good conduct. If this complex logos is as it should be, the logos is all true.13 From here, detailed interpretation can go in different directions. ProposalA The good prohairesisas a whole is an assertoric truth: a true assertion of what it is good to do. Think of the logos-factor as supplying the asserted content while the desiderative factor constitutes the assertion of it. (Without assertion there is neither truth nor falsehood.) This would explain very cleanly why Aristotle says that desire is one of the three things in the soul that are responsible for practical activity and truth (1139a18). A good prohairesisis true, the bearer of truth, because thanks to desire it is a true assertion. This solution neatly integrates truth with practicality: the very same element, the right desire, both ensures truth, by ensuring assertion of a logos that represents things as they are, and ensures corresponding action. However, there are two objections. First, Aristotle elsewhere distinguishes prohairesisfrom doxa (belief, judgment, opinion) on the ground that we differentiate the latter by “false” and “true”, the former by “bad” and “good” (EE II, 1226a4; cf. ENIII, 1111b32-34; VI, 1142b11). Secondly, Aristotle says that the sound prohairesis involves a logosthat is true anda desire that is correct (1139a24-25), and that well-doing for practical intellect is truth inagreementwith the correct desire’ (1139a29-31). These expressions strongly suggest that the presence of the right desire is somehow additional to the truth of the logos-factor. Thus if the latter’s truth is assertoric, the truth involved in the good prohairesis has already been asserted by the intellect independently of the presence of right desire. ProposalB Truth in the sound prohairesis is simply the assertoric truth of the logos component, which in itselfis a doxa: but this doxais practical only if the 13

See note 22 on truth of arguments.

240

S. BROADIE

relevant desire is also present, namely the desire to do what the reasoning has marked out, e.g. M-for-the-sake-of-E. Since the desire may not be present (in which case there is no prohairesis but just a hypothetical picture of what one would do if one were doing what the reasoning showed to be good or appropriate), the truth of the logos is only contingently practical: it is truth which is (also) practical, and this is a conceptual amalgam according to the distinction in section I. ProposalB* B can be improved by arguing that, given the assertorically true logos (articulated through deliberation and asserted as the logosthat it is), the absence from it of the relevant desire is not merely a negative fact but a privative one. The desire ought to be present: it belongs with the true logos, which was only ever formulated in order to be a reason for doing this rather than that.Without the desire, this true logos is like a perfectly formed and viable child which never lived because someone smothered it at birth. On this account, surely more Aristotelian than B, the logos meets the norm for every logos, namely the norm of assertoric truth, but without desiderative support it lacks the opportunity to function as it was meant to function, i.e. by structuring action. Its opportunity is denied if a contrary desire has sprung up, one whose existence and ground in no way undermines the truth of the deliberative premisses or the correctness of the reasoning, but which the agent fails to control. This, of course, is akrasia, the betrayal of the deliberative intellect through interference from the non-intellectual part of the soul.14 According to B* the phrase “practical truth” does not indicate a mere conjunction of truth with the practicality of desire. Nor does it indicate a special kind of truth, since there is only one kind of truth, namely the truth of assertions that accord with the facts or the nature of things. Rather, “practical truth” points us to the ordinary assertoric truth of a special kind of logos, one whose formulation and assertion are pointless — a wasted or not properly exploited work of reason — 14 On this account the prohairesis comprises an executive desire for implementation. It follows that the acratic has lost his prohairesis. Aristotle seems happy to accept the metaphor of the acratic “jumping out of”, “abandoning”, his prohairesis: ENVII, 1145b11; 1146a18; 1151a1; 20; 26. The acratic’s motivational conflict, if there is one, is between the errant appetite and a wish that is, as it were, trying to be prohairetic.

PRACTICAL TRUTH IN ARISTOTLE

241

except in the presence of a fully executive desire to do what the logos truly says is good or appropriate. On this account, “practical truth” tells us that the practicality ensured by the right desire is more than the truth of the logosif we consider this truth by itself; but the “extra” afforded by right desire is not an external addition, any more than sight to the eyes although a naturally sighted animal might be blind. There is a non-contingent but defeasible connection between the true logos and the right desire. The situation where a true deliberated logosis present and the corresponding desire absent might be compared to that of a theoretical inquirer who makes a true assertion of the fact that is the reason or cause of what he seeks to explain, but fails to achieve understanding perhaps because of a false or confused theoretical assumption which has not been cleared out of the way. The assertion of what is in fact the cause meets the norm of truth, but (at least for now) the inquirer is no further forward in the effort to understand than if the assertion had been false. Similarly, the acratically diverted agent who “has” (in some sense) a true deliberated logos is for now no better off in a practical way than if the logos had been false through empirical ignorance of some circumstance. The truth of the acratic’s deliberated logosisnot only of no practical consequence but of no theoretical consequence either. The information it carries is about contingent things and particulars, which are viewed not theoretically (e.g. as confirming or disconfirming a general physical hypothesis, or constituting a perceptible diagram for the mathematician), but as setting up a need or practical problem and demanding or suggesting one or another practical response. 15 Neither B nor B* allows for a straightforward reading of Aristotle’s statement that desire, along with perception and intelligence, is responsible for practical activity andtruth. On B and B* one must read this as highly condensed, if not confused. He immediately corrects himself about perception: as such it does not control practical activity, which can be ascribed only to beings of a rational kind. Should he not also have immediately corrected himself about desire here, saying that it as such is not 15 Thus it is historically rather misleading to speak, as is often done, of the factual information that features in deliberation as “theoretical truths” (by which is meant that the propositions are non-evaluative and non-normative). From Aristotle’s point of view the acratic who has got it right about the circumstances and what he should do in them is not to be congratulated on having at least done well “theoretically”.

242

S. BROADIE

responsible for truth? If he means this latter correction, why does he not state it, given that (a) he does state the one about perception, (b) the one about desire would surely be just as important, and (c) the two corrections would mirror each other, one making perception responsible for truth but not practicality, the other making desire responsible for practicality but not truth? 4. Non-semantic Truth in Aristotle The proposals so far for interpreting “practical truth” have assumed that the truth indicated by “practical truth” is assertoric, the truth of an assertion that represents things as they are. We have not yet, therefore, parsed “practical truth” as naming a distinctive kindoftruth.16 (What, according to proposal B*, was of a “special kind” was the logos in the good prohairesis, not its truth.) As far as assertoric truth is concerned, it is not clear what it could mean to speak of different kinds of truth. Difference between the subject-matters of true assertions does not make for different kinds of truth. However, we shall soon (section V) be considering two further proposals, C and C*, which conceive of truth otherwise than as assertoric. On this sort of approach there can be different kinds of truth, one of which would be practical truth. But before getting to practical truth in particular, I shall bring to bear some passages where Aristotle speaks of thetruth, using the noun alētheia as distinct from the adjective alēthēs, in a sense that is richer, older, and perhaps less neatly definable than the assertoric sense.17 “The truth” in this richer sense indicates, simultaneously, (a) actual or possible cognitive achievement in relation to some reality, and (b) the reality itself insofar as it is or might be successfully presented to rational cognition. The truth in this sense may have as its object a thing, a fact, an underlying principle, or a whole domain. Aristotle alludes to truth in this sense when, in the context of theoretical inquiry, he speaks of the “study of truth (τῆς ἀληθείας θεωρία)” at the beginning of Metaphysics II, 993a30 and at DeCaelo I, 271b5-6; when he speaks of people “philosophizing about the truth” at 16

See note 3. For assertoric truth elsewhere he uses the adjective alēthēs or to alēthes, and the verb alētheuein. Thus if alētheia in our EN VI passage means assertoric truth, this is exceptional. 17

PRACTICAL TRUTH IN ARISTOTLE

243

MetaphysicsI, 983b2-318 and DeCaeloIII, 298b12-13; when he says that philosophy is rightly called “knowledge (ἐπιστήμη) of the truth” at MetaphysicsII, 993b19-20; when he speaks of people “philosophically seeking the truth and the nature of what is” at Physics I, 191a24-5.19 One might say that this is not very different, if different at all, from assertoric truth: “truth” in this so called rich sense makes a reference to the cognitive subject (“the soul”; cf. 1139a16-17), and so does “assertoric truth” since this assumes a subject who asserts. But there is a difference. Rich-sense “truth” in the passages cited indicates the full achievement of scientific knowledge and understanding, or the corresponding reality as it really is, i.e. as it would present itself to the ideal knower, whereas assertoric truth can attach to true opinions that fall short of being scientific knowledge by any standard, e.g. ones that owe their truth to a lucky guess at the facts, and ones that represent incidental combinations like ‘The doctor is building a boat’. Consider this from MetaphysicsII: “It is right that philosophy is called knowledge of the truth (ἐπιστήμην τῆς ἀληθείας). For the end of the theoretical kind is truth(ἀλήθεια) and of the practical kind it is a deed (ἔργον).20 (For practical thinkers, too, investigate how things are: they study [θεωροῦσιν] - not the cause per se, but — the cause that is relative to something and is now.) Without the cause, we do not know a truth. For every P, the thing that compared to the others is most P is the one whereby the characteristic that has that name is present in other things; e.g. fire is hottest as it is the cause of heat in other things. So that is truestwhich is cause of the derivative things’ being true. 18 W. D. Ross comments: “Aristotle does not mean either simply that these thinkers tried to reach the truth, as do inquirers in anyfield, or that they studied the nature of truth, as an ‘epistemologist’ does, but that they studied the truth in general, the ultimate nature of things”, Ross (1958), p. 128. 19 See also Metaphysics I, 988a20; II, 993b17; De Anima I, 402a5; Physics VIII, 251a6; Generation and Corruption I, 325a17; Physics I, 188b30 and Metaphysics I, 984b10 on being forced or led by the truth (cf. Parts of Animals I, 642a19). See also Eudemian Ethics I, 1215b1-2, and possibly 1216b31 and II, 1227a1-2. Other examples may be PhysicsVIII, 263a18; MagnaMoraliaI, 1185b39; II, 1200b23-4. My purpose in making the case that Aristotle sometimes operates with this richer sense of “truth” is not to sell it philosophically (as though Aristotle’s patronage is an automatic recommendation), but only to elucidate “practical truth” in ENVI. One can concede that what he says in the theoretical passages in terms of rich-sense “truth” could probably have been said just as well in cleaner to us terms like “scientific knowledge” and “understanding”. 20 It is not clear whether the theoretical and practical kinds here are kinds of ēpistēmē or of philosophy. Note that here Aristotle restricts alētheia to theoretical ēpistēmē/ philosophy.

244

S. BROADIE

Consequently, the principles of eternal things must be maximally true (ἀληθεστάτας). For they are not true merely sometimes, nor does anything cause them to be: they, rather, cause other things to be (τοῦ εἶναι). So, for each thing, the degree to which it has being is also the degree to which it has truth” (993b19-31).21

However exactly we unpick the metaphysics and epistemology of this, the truth that belongs to things in degrees depending on how ontologically fundamental they are (and therefore how explanatory they are for the scientific inquirer) is clearly not assertoric truth. An assertion that p is true as an assertion if and only if the asserter thereby says what is the case, whether what is said is metaphysically superficial or deep. Moreover, the significance of the passage cannot be dismissed by claiming that its degrees-of-truth vocabulary is just an archaic formulation of Aristotle’s well known idea that the primary things are the most knowable (MetaphysicsI, 982a25-b4). This may well be so. But even if degreesof-truth vocabulary is inter-translatable with degrees-of-knowability vocabulary, what matters for now is that Aristotle conceives of truth in a way that sanctions such translation.22 5. Practical Truth: Three More Proposals ProposalC Proposal C builds on the passages concerning theoretical inquiry where alētheia connotes the full measure of cognitive success.23 (I say “cognitive 21 «True» here is not a just another word for “existent” or “real”, since things are said to vary in truth as they vary in existence or reality. 22 It is interesting that Aristotle sometimes says that arguments, sullogismoi, are true and false, and in some cases for reasons other than assertoric truth or falsity of a premiss or the conclusion (on which see PosteriorAnalyticsI, 88a20 ff. and ENVI, 1142b21-26). At Topics VIII, 162b3-22he says that an argument is said to be false in four ways: (a) when it is invalid, (b) when it is valid but the conclusion is not the one proposed, (c) when the proposed conclusion is validly reached “but not according to the mode of inquiry appropriate to the case”, and (d) when a true conclusion is validly drawn from false premises (see also SophisticalRefutations 176b29-33). An argument that avoids all these errors, hence is “true”, does not necessarily possess truth in the elevated sense of scientific understanding since it may not be explanatory or even aiming at explanation. But the truth involved seems to be a matter of argumentative success rather than mere assertoric truth. 23 Cf. the naturalness of “The truth about ...” in titles of exposés, e.g. “The truth about the JFK assassination”. They promise not merely to substitute assertoric truths for false impressions but to do so by a narrative that makes best sense of all the data.

PRACTICAL TRUTH IN ARISTOTLE

245

success” in general, although our concern is with the intellectual kinds, because at 1139a18 Aristotle also counts sense perception as responsible foralētheia.) Proposal C transposes this into the practical domain. It says that “practical alētheia” in our ENVI passage marks the culminating intellectual achievement of practicalinquiry. This position gives the only clear reason I am aware of for treating practical alētheia as a distinctive kind of truth. Since practical inquiry is a distinctive kind of inquiry, its success is a corresponding kind of intellectual success; and “practical alētheia” is a synonym for that success as such.24 Thus “practical truth” according to C denotes nothing more nor less than the good prohairesis itself, with its true logos and its concordant desire.25 It is not a point against C that a prohairesis is not an assertion capable of assertoric truth and falsehood, for practical truth, on this proposal, is not claimed to be assertoric truth. (As to that, we have seen that understanding, the culminating achievement of theoretical intellect, cannot be reduced to a set of assertoric truths. Moreover, some scholars hold that for Aristotle the truth of sense perception, anyway for the special

24 According to Olfert (ibid. 208-10; 217), Aristotle treats the different kinds of intellectual truth as definitive of different kinds of rational activities, ‘achievements’, and excellences. Therefore the distinctive character of practical reason should be explained in terms of its distinctive truth. On this basis she objects to interpreters who try to understand what Aristotle means by “practical truth” by (a) treating it as the goal of practical reason and then (b) elucidating this goal by reference to Aristotle’s well-known discussions of deliberation and prohairesis. The objection rests on a misunderstanding. Grant that inquiry of kind P aims at P-truth: it does not follow that we should elucidate the nature of P-inquiry by first laying out whatever is conveyed by “P-truth”. Calling X a goal does not entail that one must treat X as definitive of the activity that aims at it. Rich-sense “truth” in the cited theoretical passages is not directly the name of something that explains what theoretical inquiry is or ought to be. Rather, it is an accolade marking the success of such inquiry. WhatAristotelian theoretical inquiry essentially is, as we know from innumerable passages, is the search for explanations in the domain of unchangeable facts; and from the PosteriorAnalyticsand elsewhere we know how Aristotle fills out the notion of “explanation”. None of this substantial information is conveyed or meant to be conveyed by his use of rich-sense “truth”. The aim of scoring more goals defines (in part) the game of football, hence “victorious at football” applies to the team which scores more goals; but it does not follow that “victory at football” defines or is meant to define the game — the phrase, unsurprisingly, says nothing of the nature of football. “Practical truth” according to proposal C marks (but does not define the nature of) the success as such of practical reasoning. Certainly Aristotle defines or explains good practical reasoning in terms of something other than itself: this is the good prohairesis, which in turn is analyzed in terms of the respective truth (ordinary, assertoric) and correctness of the two factors. 25 This was my interpretation in Broadie & Rowe (2002), pp. 362-3..

246

S. BROADIE

senses, is a non-propositional or non-predicational grasp of a proper sensible.26 On that view, the truth of perception is also non-assertoric.) Proposal C avoids one important objection to proposal A, namely that Aristotle distinguishes a prohairesisfrom a doxaon the ground that “true” and “false” apply to the latter and not to the former. Presumably this distinction is drawn in terms of assertoric truth and falsehood. But C’s claim is that the truth of the good prohairesis is truth in a different sense. Furthermore, on C “practical truth” indicates a natural unity, given that truth, the success of intellect, is instanced here by the prohairesis, a work of practical intellect. However, C faces a serious difficulty which also afflicted A. When Aristotle apportions alētheia to the practical intellect at 1139a29-31, he says of this alētheiathat it agreeswith correct desire. This represents the correct desire as logically extraneous to the alētheia. Hence C is mistaken in identifying practical alētheia with the good prohairesis as a whole: the correct desire is one element ofthe good prohairesis, andalētheiais assigned to the other element, the logos (identified at 1139a29 as the product of practical intellect). So it is the logosand not the whole prohairesisthathas or is an instance of alētheia. Should we then revert to saying (as in B and B*) that the logos’s alētheia is simply good old assertoric truth? Not necessarily. We can acknowledge that the logos-factor alone is the locus ofalētheia, and also, of course, that the logos is assertorically true (αληθής), without being committed to equating its rich-sense truth with assertoric truth without remainder. We can retain the idea that what belongs to this assertorically true logos partnered by correct desire is rich-sense truth, i.e. the culminating achievement of practical intellect. Just as in the theoretical sphere rich-sense truth supervenes when all the assertoric truths are in place, so here rich-sense truth of the prohairetic logoscan be seen as supervening on its assertoric truth given the presence of concordant correct desire. On this basis we should modify C to C*, as follows. ProposalC* According to C* the logos-factor is assertorically true, but rich-sense truth (indicated by Aristotle’s use of the noun) is not excluded from the 26 See e.g. Johnstone (2015), n. 28. See also Crivelli (ibid.), pp. 108-109 for discussion and references.

PRACTICAL TRUTH IN ARISTOTLE

247

scene. Rich-sense truth, according to C*, crowns the assertorically true logos alone, butonlyonconditionthatithascorrectdesireatitsside. Apologies to feminist Aristotelians, but this a ceremony where the queenconsort must be compliantly present for the king to receive the crown although his is the only head on which it can descend. C* therefore neatly explains why at the beginning of our passage Aristotle says that desire is one of the things responsible for alētheia. If this is understood as assertoric truth, Aristotle has mis-spoken. The acratic has formed a good prohairesis, but lacks the correct desire (it is non-executive). Therefore the acratic’s logos is assertorically true anyway: the right desire, if present, could make no difference to that. But according to C* the latter’s presence is necessary if the assertoric truth of the logos is to count as an unalloyed success of practical reason, i.e. as an instance of rich-sense truth. So correct desire is responsible not only for praxisbut foralētheia. Given an assertoric truth as to what it is good to do, the concordant correct desire is the source of its practical implementation and thereby of its elevation from being a mere assertoric truth to being an instance of alētheia. G. E.M.Anscombe’sInterpretation This famous interpretation is based on the simple and powerful idea that “the description of what [someone] does is made true by his doing it”. By “description” here she meant the description under which what is done is done. She concluded her 1965 paper as follows: “The notion of truthorfalsehoodinaction would quite generally be countered by the objection that ‘true’ and ‘false’ are senseless predicates as applied to what is done. If I am right there is philosophy to the contrary in Aristotle. And if, as I should maintain, the idea of the description under which what is done is done is integral to the notion of action, then these predicates apply to actions strictly and properly, and not merely by an extension and in a way that ought to be explained away.”

This is different from proposals A – C* above. The bearer of practical truth, according to Anscombe, is not the logosside of the good prohairesis, nor is it the prohairesisas such. Practical truth is the fit between the agent’s taking himself to be doing so and so and his doing it. Thus the bearer of truth is S’s agential assertion ‘I’m doing F’; it is made true (or makes itself true) by S’s doing F. Here, the primary source of truth (so I

248

S. BROADIE

understand Anscombe) is not the event of S’s doing For the fact that S does F: the primary source is an instance of praxis, namely the doing of F (by S). The event or fact may be called “practical” in that it has an action as a logical constituent (just as a fact is “biological” if its subjectmatter is biological), but it is not an instance of praxis. The source of truth is the doing of F (by S), which is why the truth here is practical truth. This spelling out of Anscombe’s interpretation is only partial. The other part, which she lays out in detail, follows from the fact that the Aristotelian prohairesis embodies an assertion to the effect that doing what it prescribes is a case of doing well simpliciter. Thus it expresses the agent’s fundamental values. Accordingly, the prohairetic agent’s agential assertion ‘I’m doing so and so’ includes the claim ‘I’m doing what is good or appropriate’ sans phrase. Depending on the situation, this can take such specific forms as ‘I’m doing what a just, courageous, wise, generous, etc. person would do’. There is only practical truth if the agential assertion is true at every level, so to speak. Hence there is only practical truth if the values expressed, as well as the agent’s interpretation of them in the particular situation, are those that a genuinely virtuous person would bring to bear.27 Over the last fifty years a fertile philosophy of action has sprung up from and around the thought that action is a self-implementing conception of what one is doing. ‘What one is doing’ is usually cashed in narrow and empirical terms such as ‘writing a letter’, ‘poisoning the water-supply to that house’, ‘returning a book to the library’. Such are the examples in Anscombe’s Intention(1957). The study of action on these terms is free of contentious meta-ethical assumptions about the truth-aptness of value judgments and ethical ones in particular. Twentieth-century philosophy of action would never have got off the ground had it not detached itself from this millstone of a problem. Anscombe set out to rehabilitate Aristotle’s notion of practical truth in the eyes of anyone who thinks that the predicates ‘true’ and ‘false’ are senseless as applied to whatisdone. But her own exegesis makes it abundantly clear that a full rehabilitation, 27 Here and elsewhere I try to accommodate the fact that goodness of prohairesisand the corresponding action is not restricted to possessors of virtue, the latter being a stable disposition for such conduct in a full range of situations (ENII, 1105a28-33).

PRACTICAL TRUTH IN ARISTOTLE

249

in the eyes of both her contemporaries and ours, would also include an adequate rebuttal of the view that ‘true’ and ‘false’ are senseless as applied to ethicaljudgments. Anscombe’s interpretation has the great merit of making sense of Aristotle’s statement that desire is responsible for praxis and for truth. Right desire gives rise to Anscombian practical truth by ensuring that the action is done. But if Anscombe is right, it is surprising that our text contains no explicit emphasis on the matter of a good agent’s actually doing what he takes himself to be doing. Instead the focus is surely on the good prohairesis and its factors.28 Of course, Anscombian practical truth is a close relative to the actual topic, since a prohairesisisnaturallyenacted, and the goodness and reasonableness of the action depends on that of the prohairesis. But given that practical truth has to do with practical goodness and reasonableness, the direction of dependence just mentioned is an additional reason for restricting “practical truth” to the prohairesis itself or one of its constituents. 6. Assessment and a Further Question Of the interpretations discussed, B* and C* surely come out in front. One might prefer B* on grounds of philosophical economy and clarity. B* rests on a single, fairly straightforward, understanding of truth as assertoric, whereas C* invokes an additional truth-notion, one that contemporary analytic philosophers would, perhaps rightly, regard as obscurantist and gratuitous. But B* is unable to accommodate Aristotle’s difficult remark that desire (as well as intelligence and perception) is responsible for alētheia(1139a18). C* supplies a natural solution to this problem. B* also fails to explain why, if assertoric truth alone is in play in our passage, Aristotle uses the noun alētheia, exceptional for him in this connection. So it seems to me that C* is ahead. But the supporter of proposal C* must explain why rich-sense truth figures at all in the account of the good prohairesis. This question arises because this notion of truth seems quite unnecessary to Aristotle’s analysis. The good prohairesis is shown to consist in a concordant logosfactor and desire-factor which are, respectively, assertorically true and correct. This is all Aristotle needs for explaining the function of practical 28

Cf. Broadie (1991), p. 224.

250

S. BROADIE

intellect, and what its virtue is. It is also all he needs for showing how this virtue rests on (or might even be said to subsume) ethical virtue (ENVI, 1143b18-1145a6). And it is all he needs for discussing the good deliberation that issues in the good prohairesis (1141b8-1142b33). So what does rich-sense truth contribute to the doctrines of ENVI? It adds nothing to our analytic understanding of what the good prohairesisis, what phronēsisis, and what it is to deliberate well. Rather, it presents these things in a certain way. It emphasizes that they are authentically intellectual. Why emphasize this? Suppose that in explaining the achievement of practical reason Aristotle had kept quiet about rich-sense truth: this might well have suggested, to his original audience, a deliberate withholding of the accolade. The message would have been that richsense truth is available to theoreticians alone.29 Now, we can all easily agree that the human being is essentially practical, a prohairetic cause (1139b4-5). Whether the human being is also essentially theoretical cannot be so easily taken for granted, on which see more below. At any rate, by assigning the goal of rich-sense truth to the practical sphere as well, Aristotle makes it clear that the human being as such — not merely a small and peculiar tribe of theoreticians — is of a nature to seek and attain cognition that deserves praise in the highest alethic terms. It is important for Aristotle to make this clear because there are considerations pointing in the opposite direction. He wants to resist going in that direction but also to do justice to the considerations. In fact he has reason to be somewhat defensive about the echt intellectuality of practical wisdom. In ENVI he rejects two models prominent in Plato: practical wisdom as a technical productive expertise, and as expertise in eternal verities about ethical abstractions and the metaphysical foundations of mathematics. To many then, as to us today, these must have seemed absurd intellectualizations of the personal qualities needed to make good decisions in the everyday human context. One natural response would have been to adopt a somewhat crude anti-intellectualism about these very real qualities. Aristotle responds by showing how phronēsis in its own way is a genuinely intellectual quality; but doing this involves him in much creative and careful philosophizing. He makes 29 To anyone brought up on the poem of Parmenides the restriction would have seemed natural. The fact that the domain of alētheia is now allowed to include physics, cosmology, and pluralistic metaphysics leaves untouched any tendency to dismiss even the best practical cognition as mere “mortal opinions” (Parmenides, fr. 8, 50-51).

PRACTICAL TRUTH IN ARISTOTLE

251

things harder for himself by admitting that the phronimos need not have articulate reasons (1143b11-13); by mentioning sympathetically (since it suits his purpose in the immediate context) the belief that some lower animals have phronēsis(1141b26-28);30 and by implying that euboulia, excellence in deliberation, is easy to confuse with the non-rational gift of eustochia, which hits the nail on the head without knowing how or why (1142a32-b6). A lot is at stake, because scepticism about the rationality of practical reason, so called, sends shock-waves through the system. Consider Aristotle’s transition from ENI’sdefinition of eudaimonia in terms of virtuous rational activity (1097b33-1098a15) to EN II’s discussion of the ethical virtues. This transition assumes that these virtues are intimately connected to human reason. He defines them by reference to prohairesis and phronēsis (II, 1106b36-1107a2; cf. VI, 1144b14-28 and X, 1178a1622); but these links support the connection to reason only if phronēsis is firmly established as an intellectual virtue. If courage, temperance, justice, and phronēsisitself, are only dubiously virtues of the human being qua rational animal, the activity expressing them is only dubiously any sort of eudaimonia: it is not a clear case of eudaimonia at all, even of a second-class kind (cf. X, 1178a7-22). So the only undeniable eudaimonia is entirely grounded in the only undeniably intellectual part of the soul, theoretical reason, with sophiaas the only undeniably intellectual virtue. From the point of view of human life, sophia is now rather dangerously out on a limb. How to reply to an anti-intellectualist about phronēsiswho infers that since the obvious human virtues are not really intellectual, the undeniably intellectual sophia is not in fact a virtue of the human being as such? Perhaps it is a virtue of the gods alone, and for us just an occasional beautiful contingent ornament, not a flowering of fundamental human nature. Then human individuals and societies are not fundamentally backward or blind to the full range of human good if they make no provision for theoretical activity. This line of thought shows how, if we play down the dignity of practical intellect in human beings, we undermine our ground for holding that the splendours of theoretical intellect represent an ideal for human beings as such. In this dialectical situation 30 See e.g. History of Animals VIII, 588a24 ff. on animal sunesis (“sagacity”); also Metaphysics I, 98027-b26. See PartsofAnimals II, 647b30 ff. and 650b20 ff., for a physiological explanation.

252

S. BROADIE

Aristotle would have missed a trick in not inserting rich-sense truth into his philosophy of practical reason.31 In motivating the study of parts of animals, an ignoble subject-matter compared to the immortal objects of astronomy, Aristotle retails what Heraclitus supposedly said to the guest who found him in the kitchen: “There are gods here too” (Part.an. I, 644b21-645a23). In ENVI, I have argued, Aristotle laces his account of the good prohairesis with the pointed assurance that - as well as logoialētheis - there is alētheia here too. 32 LITERATURE ANSCOMBE, G. E. M. (1965), “Thought and Action in Aristotle: What is Practical Truth?” in New Essays on Plato and Aristotle, ed. R. Bamborough, London: Routledge, pp. 143-158. ANSCOMBE, G. E. M. (1993), “Practical Truth”, WorkingPapersinLaw,Medicine and Philosophy, ed. J. Dolan, Minneapolis, reprinted in Logos 2/3, 1999, pp. 68-76, and in HumanLife,ActionandEthics, eds. M. Geach and L. Gormaly, Exeter, pp. 149-160. BROADIE, S. (1991), EthicswithAristotle, New York/Oxford: Oxford University Press. BROADIE, S. and ROWE, C. (2002), Nicomachean Ethics, Aristotle, translation, introductionandcommentary, Oxford: Oxford University Press. CRIVELLI, P. (2004), AristotleonTruth, Cambridge: Cambridge University Press, pp. 132-6. IRWIN, T. H. (1999), NicomacheanEthics,Aristotle,translatedwithintroduction, notesandglossary, Indianapolis: Hackett. 31

This is one of his many intellectualist emphases in this connection. In case it seems far-fetched that the virtue-status of sophia should piggy-back on the rationality of phronēsis, note this from the MagnaMoralia: “Is sophia a virtue or not? From this it should become clear that it is a virtue — from phronēsisitself. For if phronēsis is, as we claim, the virtue of one of the parts that have reason, and phronēsisis inferior to sophia because its subject-matter is inferior (for the subject-matter of sophia is the eternal and the divine, as we claim, whereas that of phronēsis is what is useful to the human being) — well, if what is inferior is a virtue, certainly what is superior reasonably a virtue: so that it is clear that sophia is a virtue” (MM I, 1197b3-10). The topic is evidently human virtues. The argument depends on the assumption that phronēsisbelongs to a rational part of the soul. That this argument exists shows that there was some anxiety about the status of sophia as a genuine human virtue. Aristotle feels the need to argue the point at ENI, 1103a8-10, and MM I, 1197b28-31 shows that someone had suggested that it is not for statecraft to accommodate sophia. 32

PRACTICAL TRUTH IN ARISTOTLE

253

JOHNSTONE, M. (2015), “Aristotle and Alexander on Perceptual Error” in Phronēsis60 (2015), pp. 310-338. OLFERT, C. (2014), “Aristotle’s Conception of Practical Truth” in Journalofthe HistoryofPhilosophy52 (2014), pp. 205-231. PAKALUK, M. (2010), “The Great Question of Practical Truth” in Acta Philosophica19, pp. 145-159. ROSS, W. D. (1958), Aristotle’s Metaphysics, Oxford: Clarendon.

This article has also been published as: Sarah Broadie “Practical Truth in Aristotle” AmericanCatholicPhilosophicalQuarterly Volume 90, Issue 2, Spring 2016, pp. 281-298 DOI: 10.5840/acpq20163284

PERSPECTIVES

CONSCIOUSNESS AND TEMPORALITY: HOW SARTRE RENEWS ARISTOTLE Annick STEVENS

When reading Sartre’s pages dedicated to the study of temporality in Being and Nothingness, one often feels a familiarity with what Aristotle says about time in the Physics and about memory in the ParvaNaturalia, even though the contexts and goals of the two studies are quite different. In particular, it often seems that Sartre formulates with a precise and rich conceptual apparatus the same theses that Aristotle was expressing with huge difficulties because he only disposed of a rudimentary vocabulary1. A closer examination of these convergences could lead us to a better understanding of some difficult passages in Aristotle, along with a highlighting of what is totally new in contemporary phenomenology, and even make us discover some gap that Sartre could have filled if he had better known Aristotle’s theory. Furthermore, the complementarity of the two theories may help us to progress in our own investigation about this fascinating topic. The guiding question of my comparative study will be to search in which way the existence of time is objective (in the sense of belonging to physical beings or to what Sartre calls “Being-in-itself”) and in which way it is subjective (due to the soul or consciousness). 1. The Subjective Conditions of Temporality In the section of BeingandNothingness dedicated to the phenomenology of temporality2, Sartre’s thesis is that the three dimensions of time — 1 For example, there is no term at this time to express the philosophical concept of consciousness, so that he has to use the verb aisthanesthai or the negation of lanthanein, which leads him to some problems that would have been avoided with the adequate vocabulary, as when he asks in De anima with which sense “we perceive that we perceive” (De an. III 2, 425b12-25). See on this point Stevens (2009). 2 Sartre (1984). All translations of quoted passages from Sartre and from Aristotle are mine.

258

A. STEVENS

past, present and future — only exist for a consciousness and have no reality by themselves. This thesis relies on the fundamental ontological distinction between the Being-in-itself and the Being-for-itself. The In-itself is the mode of being shared by all the physical bodies, which are just what they are, one-piece made, with a fixed identity and without any internal temporal multiplicity. On the contrary, the consciousness as a cogito, the “I” that accompanies each conscious act, would it be sensation, emotion, imagination or thought, i.e. the For-itself, is never simply what it is nor the necessary result of what it has been. Its mode of being is negation: it constitutes itself by distinguishing itself from its object, by negating that it is In-itself. As it is each time conscious of something else, it never totally coincides with itself; it can be defined neither by any of its acts nor by their totality, since the latter is unachieved and unpredictable. Neither is the For-itself a succession of present consciousnesses, else could it never escape from this present, and memory as well as projection into the future would be inexplicable. Sartre leaves open the question whether the For-itself is only human consciousness or if it is also, to some extent, the mode of being of other animals; this might be the case — he says — if they are “something else than a physicochemical system” (p. 152). There is no doubt that all perceiving beings possess an intentional consciousness, which Sartre expresses by the famous formula: “every positional consciousness of an object is at the same time a non-positional consciousness of itself” (p. 19). This means that feeling is not only catching an object as a presence, it is also feeling that this object appears to oneself, without taking oneself for an object. Aristotle clearly claims that any act of sensation involves this kind of consciousness, so that any animal must possess it. But the Sartrian For-itself cannot be reduced to the intentional consciousness: it is the existence that escapes determinism, that freely choses among its possibilities, and creates its future by re-interpretating its past. This is how the For-itself creates the dimensions of time. The Being-in-itself has no past, even though it is transformed following an irreversible physicochemical determinism: the fact that each state results from the anterior state only shows that it is “a phenomenon that survives its cause, and not the subsistence of the cause as a cause at the past state” (p. 152). For an event to become past, it has to be caught in the present as a past — which no Being-in-itself can do, but which the consciousness does: in the memory, it points to the event as past, and

CONSCIOUSNESS AND TEMPORALITY: HOW SARTRE RENEWS ARISTOTLE

259

doesn’t confuse it with a sensation or with another kind of image. The past exists only as relative to the present: “it is originally past of this present” (p. 149). This is the case, for instance, of a dead man, who now is as a past, only as far as it is an object of remembrance for the livings; in contrast, the dead man leaving neither memories nor traces is nothing at all anymore. The past being, would it be that of a dead or living person, is a transformation of the For-itself into In-itself, because it is now what cannot be changed anymore, what has exhausted his possibilities and what I have from now on to accept as such, even though I can still change the meaning I give to it with respect to my future. In the same way as the In-itself has a past only via a For-itself, the same holds for the present: the beings of the world are only co-present as far as a same For-itself is co-present to all of them. They are not present by themselves because they do not have by themselves this relation of comparison, which consists in recognizing things as contemporaneous. Only a For-itself can make this comparison, because its presence to the In-itself is not only a spatial proximity; rather, it is its original relation to the In-itself in general, this relation that lets it arise as a For-itself. By this relation, at the same time it adheres to the In-itself and it distinguishes itself by negation. The present is this escape or this run away from the In-itself; that is why it is impossible to catch it, to fix it even within an instant (p. 160-162). Finally, the future necessarily comes to being with the human reality, and does not belong to the In-itself. The future is not “an homogeneous and chronologically sorted succession of instants to come”. The proposition ‘the Moon will be full next night’ is true given the astronomical regularities, but this future is not the one of the Moon, it is the one of the humans who foresee it. The next phases donotexistinanywayyetinthe Moon; they exist only in our prevision. But the future is not our representation of what will be or might be; indeed, every representation is present. The future is a possible of the For-being. It is because the For-being is not simply what it is, because it is caught as unachieved, that there is a future, not a predictable and certain future, but only a possible one. The future is the lack that the For-being has to be and that he could not be (p. 164). Hence, the future of the world is laid down secondarily by the consciousness, as a background in which it will realize its project; and it is because it takes into account what the world can become from what it is that the consciousness is project and not only imagination of another possible.

260

A. STEVENS

One might be tempted to oppose to this conception that of Aristotle, for whom the future phases of the Moon exist in some way, on the mode of the “being potentially”3. However, this “being potentially” is not an objective future. It denotes a presentproperty of a body, so that this one will necessarily become some thing by virtue of natural regularities, or will possibly become some thing if a given number of conditions are realized, irrespective of whether these conditions are or not predictable4. As long as this realization does not become present, it can exist only in an act of imagination, produced by a human being or another imaginative animal (since Aristotle attributes to some animals a phantasialogistikē, a faculty of reasoning by imagining the future consequences of their actions5). Thus the notion of “being potentially” does not involve an objective or in itself given existence of the future. We will soon detail which kind of existence Aristotle attributes to this being, without relying, as does Sartre, on the mode of being For-itself. 2. The Past in De memoria Regarding the past, Aristotle provides us with a much more explicit phenomenological approach than regarding the future, since he studies in De memoria the way memory is formed, i.e. the connection from the present consciousness to its own past. This approach appears since the first attempt to define memories: “Each time one is in act of remembering, one says in the soul that one has heard or perceived or thought this before. Thus memory is neither a sensation nor a conception, but a possession or affection of one of them when some time has passed.” (449b22-25).

3 This is how the Hermes is already in the non-yet carved stone, and the wheat in the seed that has not yet sprouted (Met. Δ, 7, 1017b7-8). 4 See chap. 9 of the DeInterpretatione about the absence of truth or falseness of future sentences due to indetermination of human actions. 5 Because it needs certain faculties, the representation of future seems to be restricted to humans, though the possibility is left open to other animals: see De an. III, 10, 433b513, where reason is opposed to desires in animals that can feel the time, because intelligence orders to avoid something because of the future, and desires pushes to it because of the present (toēdē), “because one does not see the future (tomellon)”; 433b29-30: the other animals have the phantasiaaisthētikē, not the logistikē; 11, 434a7-10: willingness is a moving cause in the animals possessing the reasoning (entoislogistikois) and capable to wonder whether they will do this or that action.

CONSCIOUSNESS AND TEMPORALITY: HOW SARTRE RENEWS ARISTOTLE

261

The alternative “possession or affection” (hexis ē pathos) illustrates the difficulty to name precisely what is not simply the re-actualization of a sensation (which would be a simple act of imagination) but the synthesis between this sensation imprinted somewhere within us (450a31: οἷον τύπον τινὰ τοῦ αἰσθήματος) and the consciousness that it has been lived at a moment of our past. One of the consequences of this definition is that “only the animals that feel time can remember, and this by the same organ by which they feel” (449b28-30), i.e. by the common sensory faculty. However, the image itself cannot indicate that the event is past but the consciousness of a temporal distance must add up to the actualization of the image, as Aristotle writes a bit further: “one additionally feels that it was before” (450a21: προσαισθάνεται ὅτι πρότερον). The notion of consciousness is expressed, in the definition, by the words “to say in one’s soul”. The verb “say” is not used in its common meaning of articulated language: it is not necessary for a logos to take place in this operation that happens entirely in the central sensory faculty. In the De anima already, when Aristotle attributes to this faculty the synthesis and comparison of different sensible things (III 2, 426b8-427a16), he says that it judges (κρίνει) and tells (λέγει) the differences, which is a way to say that it takes consciousness of these differences. We shall find the same verb “say” in the Physics to express the consciousness of the time coming from the distinction between two instants (219a27). For memory to exist, it is not necessary to precisely quantify the elapsed time: “The sensation6 of time is of two kinds: sometimes we don’t remember the thing7 with the measure , for example that we have done this thing three days ago, sometimes we remember it with the measure; but we remember even when there is no measure. We use to say that we remember but that we do not know when it happened, when we don’t know by a measure the quantity of time.” (Dememoria, 2, 452b29-453a4).

In the latter case, knowing how much time has passed requires a reasoning, as the one indicated by Aristotle for recollection, by the 6 In this whole chapter, as well as in some parts of the Deanima, Aristotle uses kinēsis for the modification of a soul’s faculty, in the sense of an information that reaches it and of which it becomes aware. In the same way as the term aisthēsis is used to denote consciousness in absence of another available term, the term kinèsis compensates here the absence of a word to mean the appearance of data to the consciousness. 7 Most manuscripts have “auto”: “the thing in question”; Ross, following the ms SU, edits “auton”, refering to chronos, but I think this is a mistake because it is not time that we remember.

262

A. STEVENS

reconstitution of proportional lapses between different events (452b8-22). Even more, we can be wrong on the very fact that we lived an event, and it can happen that we cannot distinguish whether an image is a memory or not; however once we have established that this is the case, we can fix this status by several times repeating the image as a memory in order to recall it from now on as such8. This voluntary operation shows again that it is not the image of an event that allows us to situate it in our own past, and, without doubt, for Aristotle as for Sartre, it is the present consciousness which situates things in the past, because only consciousness is able to catch now what is not now. To explain the duplicity of memory, it is not sufficient to invoke the retention or the impression that events left in us, since, when these events are imprinted, they are present, and when they are re-actualized they are also present. As highlighted by Sartre, registering images would be of no use unless consciousness had the power to refer to them knowing them as its own. The retrospective relation of the consciousness to itself implies that it can feel itself as both identical and different: identical to enable self-reporting of the event, different to be able to say that it is no more the exact one that lived the event9. The temporal multiplicity of consciousness is also clarified by the way Aristotle explains the feeling of time. The thought experiment which opens the inquiry about time in the Physics indicates that time is not felt directly but through the sensation of a change: “When nothing changes in our thought, or when this change escapes us, it doesn’t seem to us that time has passed, as for those who, following the legend, slept in Sardinia beside the heroes, when they woke up; for they link the preceding instant to the following one and make one of the two, deleting the intermediary because of the lack of sensation. Thus in the same way that, if the instant was not other but one and the same there would be no time, so when it escapes that it is different it seems that there is no intermediary time. If thus not believing that time passes occurs when we don’t define any change but our soul appears to remain in a unique and 8 Further he will precise that we may believe that we remember when we don’t remember but not the opposite: we cannot remember without believing that we remember (452b24-28) — but he is playing here with the ambiguity of the verb oiesthai which in the first case expresses the (true or false) judgment we have of an image, while in the second case it expresses the consciousness of the act, confirmed by oukesti lanthanein. 9 This duplicity is different from that of the image in general, which is at the same time a sensible thing and a sign or a reference to another sensible thing (cf. 450b20-451a2; about a remembered image: 451a5-8).

CONSCIOUSNESS AND TEMPORALITY: HOW SARTRE RENEWS ARISTOTLE

263

indivisible state, while, when we perceive and define a change, then we say that some time has passed, obviously time doesn’t exist without movement and change.” (Phys. IV, 11, 218b21-219a1).

From this passage, it is clear that a single sensation cannot be enough to feel the time nor the movement, since at least two disjoint sensations, separated from each other, are required to notice a change. The gap between the two sensations is not of the same kind as a spatial separation, even though both are expressed in terms of anterior and posterior. Indeed, in Aristotle, anterior and posterior are not primarily defined chronologically, as before and after — would it be the case, the definition of time would be circular, since it would use notions themselves defined by the term to define. This is the reason why Aristotle takes care to specify that the antero-posteriority relationship lies first in the space (by position), then in the movement and finally in the time (219a14-19). In Metaphysics book Δ, he defines this relationship as the more or less large distance from a reference point, which can be defined “either absolutely and by nature, or relatively to some thing or some place or some persons” (Δ 11, 1018b9-11). In time, anterior and posterior do not correspond to past and future, because two past events can be anterior and posterior to each other (for instance, the Trojan war in comparison with the Median wars), and the same holds for two future events. Aristotle remarks that, if one situates events relatively to the present, then in the past the anterior is the most distant from the present, and in the future it is the closest (1018b1519). However, the antero-posteriority relationship is not modified by the observer’s situation and does not require the distinction between past and future. On the contrary, in space, the relationship is modified depending on the point from which both objects are considered. The difference obviously comes from the fact (not specified here by Aristotle) that temporal events are bythemselves successive whereas spatial objects are successive only by reference to a point of view. In order to observe the antero-posteriority relationship, one needs to consider together the two events as well as the distance between them. However, in the temporal extension, it is impossible to feel simultaneously two events if one of them is anterior and the other posterior, since there is no sensation apart from present10. One way to catch them together 10 Cf. Demem., 449b28-31: “Of the present there is no memory in the present, as has been said, but of what is present there is sensation, of what is future there is hope, and of what is past there is memory”.

264

A. STEVENS

consists in figuring time as a line whose points all co-exist. This is what we do in any theoretical consideration of time, for instance to situate historical events, as we just saw about past wars. In this case, we need neither to have lived those events, nor to have the memory of the moment when we learned them. However we thereby create a co-existence that would not exist objectively, and which cannot be object of sensation. Another way to catch together the anterior and the posterior is what we do in the internal experience of time, when we take consciousness that the present instant is not the same as the last instant we were aware of. In this case, time is not represented linearly but the present sensation is compared to the imaginary reactivation of the past sensation, and the difference between the two images indicates that something has changed, wherefrom we deduce that some time has passed: “We say that time has passed when we have a sensation of the anterior and posterior in the movement. And we define those by grasping them as others, with a different intermediary between them, for when we consider the extremities as different to the middle and the soul says that there are two instants, one anterior and one posterior, then we call this the time, for what is defined by the instant seems to be time; let us consider this as established.” (Phys. IV, 11, 219a24-30)

To feel the time we thus need to feel differences that do not coexist, thanks to the faculty of imagination, which retains sensory data. The synthesis of different sensory data by the common faculty includes the synthesis of successive data, and this is what guarantees the continuity of sensation or imagination. In addition, the re-actualized data must appear as something that we didlive and not as a mere image. This brings us back to the fundamental question about how we recognize our own past experience. Aristotle’s answer lies in the continuity of self-consciousness, which we will find in a fascinating passage of the De sensu. To demonstrate that there could be no time lapse small enough to be imperceptible, Aristotle invokes, as if it were obvious, that the consciousness of our own existence depends on the consciousness of passing time: “If, indeed, someone feels himself or feels somebody else in a continuous time, it is not possible that at this moment it escapes to him that he exists11, 11 And not “that this exists”, because the argument is not “I feel, thus what I feel exists” but “I feel thus I exist”.

CONSCIOUSNESS AND TEMPORALITY: HOW SARTRE RENEWS ARISTOTLE

265

and if in that continuum there is a time so small that it is absolutely non perceptible, then obviously at that moment it would escape to him if he himself exists or if he sees and feels.” (Desensu 7, 448a24-30).

There is an ambiguity in this text. The reasoning relies on the claim that, to feel our own existence, we must feel something during a continuous time, as small would it be. The form of the argument is a reductio ad absurdum: since it is obvious that we constantly have the consciousness of being, it is proven that any time is perceptible. We have seen that the sensation of time cannot occur in a single instant, but requires at least two successive sensations revealing a change. On the other hand, we also know that feeling and thinking are instantaneous acts and not changes12, so that, by virtue of the intentional consciousness, it would not be necessary to extend them during a certain time to ensure self-consciousness, but a single act would be enough. Hence the reasoning would not be valid anymore, since one could feel even in a non-sensible time. The only other passage where Aristotle evokes the consciousness of our own existence, in the NicomacheanEthics, is not accurate about this point; it only confirms that “feeling what we feel or what we think is feeling that we are (because to be was to feel or to think)” (IX, 9, 1170a 32-b 1). I think that the solution is the same as the one we are searching in order to clarify the relation of the present to the past: the consciousness of our own existence, which is considered here as obvious, is not the consciousness (of) oneself13 that accompanies any sensation, but the continuous consciousness that we have of our total existence, including the personal history that made us become what we are. This is also very likely to be the case in the passage mentioned in the NicomacheanEthics, because the quality of the person and of its life is an important element of the argument, and it could not be challenged without the consciousness of our past life. So, the continuity is part of the mode of being of consciousness, as Sartre seeks to establish in a much more explicit way by showing the absurdity of atomistic conceptions of consciousness (such as Hume’s or Berkeley’s). We shall confirm this essential role of continuity by examining now the objective conditions of temporality.

See Met. Θ 6, 1048b23-35. Sartre writes the “of” in brackets to indicate the absence of duality as in the reflexive consciousness. 12 13

266

A. STEVENS

3. The Objective Conditions of Temporality When reading Being and Nothingness, we have the impression that, if Sartre refuses any objective condition of time, it is due to the fact that he does not pay any attention to the objectivity of changes — objectivity that he cannot deny, but from which he does not draw any consequence about time. It is well known that Aristotle on the contrary bases time upon change. In the famous definition of time as “the number of movement according to the anterior and posterior”, the word “number” must be taken in the sense of an objective property of movement14, and not as a product of human knowledge. One must first discard the idea that the numbered number means the unit of measure. Change can be measured by units of time or change, and reciprocally time can be measured by units of time or change (for example a travel time can be measured by a course taken as reference). But the definition of time as number of the movement is not reciprocal; it expresses the unilateral ontological dependency of time upon change. The numbered number does not either mean the quantity of change, for instance if an alteration from the white went up to the black or only to the grey, because this “more or less” of the change is independent from the time taken to accomplish it. The key is given by the next terms of the definition: change occurs according to a succession of anterior and posterior steps. It is an intrinsic property of any changing body that its different states cannot co-exist, but succeed to each other in a determined order. This succession, continuous and ordered, of noncoexistent states of the change, is the time. This mode of being is expressed more explicitly in a passage of Physics III, in the frame of the study about the infinite (III, 6, 206a21-b2): “As ‘being’ is said in many ways, in the same way as the day and the Games15 exist by always becoming others, so does the infinite (...). That’s why we should not consider the infinite as a ‘this’, such as a man or a house, but as it is spoken of the day or the Games, whose existence has not begun like a substance but is always in the course of coming to be and passing away, limited but always different — but in the magnitudes it happens in such a way that the removed part remains, while in the time and the succession of perishing men, in such a way that it doesn’t remain.” 14 15

“Time is thus not movement but exists as far as movement has a number” (219b2-3). i.e. the Olympics Games by way of which the years were counted.

CONSCIOUSNESS AND TEMPORALITY: HOW SARTRE RENEWS ARISTOTLE

267

As far as it is defined as something of the movement, time is independent of the existence of a soul able to number: this is indeed the answer given by Aristotle in Physics IV, 14 (223a25-29): If nothing else is by nature able to number except the soul and the intellect of the soul, it is impossible for time to exist if the soul doesn’t exist, except for what time is at a moment, namely if it is possible that movement exist without a soul. Yet the anterior and posterior are in the movement, and time is those ones as far as they are numerable. However, as the successive steps of change do not persist, there is no objective past and future assuch, i.e. as parts defined by their situation relative to the present, because such a comparison can only be the product of a human soul. This does not imply that no objective relation exists between those steps, that there would be nothing linking them and justifying their order. Aristotle avoids falling into a Humean atomism, by demonstrating that change and time are objectively continuous, i.e. that their parts are mutually related by an internal and necessary link. But this continuity is not ensured by the co-presence of past, present and future steps; it is ensured on the one hand by the double function of the limit, on the other hand by the unity of the changing body, as Aristotle explains here: “Time is continuous by the instant and divided at the instant, because this also follows the transport and the mobile: the movement and the transport are one by the mobile, as it is one (not as it is at a moment, for there would be an interruption, but by its account), and on the other hand it defines the anterior and posterior movement. And this follows also in some way the point: the point makes the length continuous and defines it, as it is beginning of one part and end of the other. But when so considered, using the unique point as double, a stop occurs necessarily if the same point must be beginning and end, while the instant, due to the movement of the mobile, is always different.” (Phys. IV, 11, 220a5-14)16.

The analogy between the three continuums (magnitude, movement and time) is based on the mathematical notion of continuity: is continuous something whose parts are joined by common extremities. As each point of a line is common to two segments, so each state of a modified body is common to two parts of the change and each instant is common to two parts of the time. In mathematics, one can separate the line segments by duplicating 16 Same statement in 13, 222a10-20, where it is added that, accordingly, the instant divides the time only potentially and never actually.

268

A. STEVENS

the unique point that join them. Similarly, one can interrupt a change at any state, by linking it either to a resting state or to another change. But since time is implied by rest as well as by change, it will never stop and no instant will ever be final, at least as far as a changing body exists17. This raises the question of how the objective time is unique and identical for all changes. First of all, this objective time cannot be the infinite linear time of the theoretical representation. Aristotle invokes it somewhere else to answer to the “exoteric” aporia that only the present is, since the past is not anymore and the future is not yet (Phys. IV, 10, 217b30-218a3). This difficulty is solved by considering that the things that no longer exist are inthetime, as well as the things that will exist later, because being in the time means having its existence measured by time, contained in the time in the same way as in the space. That’s why nothing prevents us to conjugate the verb “being” in the past and the future tense as well as in the present (221b31-222a3)18. It is obvious that only a theoretical consciousness can have this conception of an encompassing and infinite time, by which time is caught in its totality including past, present and future. But Aristotle does not fall into the trap of objectivating this infinite time. The appropriate answer to the unicity of time comes at the very end of the study, by distinguishing two cases: “As far as there is movement, it is number of each movement. That’s why it is number of the continuous movement in general and not of a particular one. But it is possible that other things are moved now, for which there would be a number of each movement. Is their time different, and would there be two equal times together ? Or rather not, because every time equal and simultaneous is the same, and those that are not simultaneous are the same specifically. For, if we have dogs and horses and in both cases there are seven, the number is the same . In the same way, for the 17 Cf. 13, 222a33-b7: “Since the present is end and beginning of time, not of the same time but end of the past and beginning of the future, as the circle is in some way at the same place convex and concave, so the time is always at the beginning and at the end; and therefore it seems always other because the present is not beginning and end of the same thing, otherwise the contraries would be together at the same point. Consequently, the time will never cease because it is always at the beginning.” 18 And that’s why only those things are not in the time, that always exist whithout any change, like the mathematical truths (221b3-5; 222a5), or those whose existence is impossible. 19 Cf. 14, 224a3-15: for ten horses and ten dogs, the number is the same: it is the decade, but it is not the same individual decade.

CONSCIOUSNESS AND TEMPORALITY: HOW SARTRE RENEWS ARISTOTLE

269

movements that were accomplished together, the time is the same, even if one of them is fast and not the other one, or if one is a transport and the other one an alteration; however the time is the same, if the number is equal and simultaneous, for the alteration and the transport. And for this reason the movements are different and separate but the time is the same everywhere, because the number is one and the same everywhere for equal and simultaneous movements.” (Phys. IV, 14, 223a32-b12).

Since most changes are not equal and simultaneous, we must understand that most frequently the times of the different changes are only specifically identical, and that there is no global time numerically one. Even the time of the eternal first astral revolution is not a global time from which the other ones would be parts; this time is not necessary for the other ones to exist or to be compared with each other — this is made by any unit of measure, i.e. by any regular time such as the time of the earth revolution20. So the objective time is divided into as many times as there are movements, and has only a specific unity. On the other hand, the recognition of past and future as such and the theoretical linear representation of an infinite time encompassing all the particular times are products of the human soul. This is why in the Physics the question whether time exists or not without the soul couldn’t receive a simple answer. 4. Conclusion At the end of this survey, we can conclude that Sartre helped us to reveal in Aristotle some distinctions that Aristotle really made but in a less explicit way, regarding the subjective creation of the three temporal dimensions (past, present and future) by the sensitive consciousness. On the other hand, Aristotle’s ontological conception, which anchors time in the perpetual movement inherent to natural beings, and which defines it 20 It is not in order to affirm such a global time that Aristotle, in book VIII, demonstrates the infinity of the first astral revolution; this is rather established as a guarantee for the endlessness of time. And yet it was not the only possible guarantee, because the only requirement is that at each moment at least one change exists, so that the time would always continue along with one or another change. Aristotle could have founded this guarantee on his own conception of the material bodies, whose perpetual existence and change are not questionable for him. If he prefered the guarantee of a unique perpetual movement, it is probably because this conception was in accordance with the contemporary astronomical theories he had adopted from the Academy and that he describes in Met. Λ8.

270

A. STEVENS

as the succession of non-coexistent states, could have led Sartre, if he had known it, to qualify the absolute absence of time in the Being-in-itself. LITERATURE Stevens, Annick, (2009), “L’apparition de la conscience dans le De Anima et d’autres œuvres d’Aristote” in Van Riel, Gerd & Destrée, Pierre (eds.), AncientperspectivesonAristotle’sDe Anima, Leuven: Leuven University Press, pp. 35-48. Sartre, Jean-Paul (1984), L’ÊtreetleNéant, Paris: Gallimard, Tel.

ARISTOTLE’S THEORY OF NATURE FROM THE POINT OF VIEW OF OUR HERMENEUTICAL SITUATION Erwin SONDEREGGER

1. Some Assumptions about Aristotle’s Philosophy and the Need to Examine Them If we look at handbooks on Aristotle, old or new, at respective entries in philosophical dictionaries, at histories of philosophy, monographs, articles of current research etc., we find some common convictions. To the most important of them belong the claims, that Aristotle has developed a metaphysics of substance and a rational theology based thereon, and third that he has invented and practised science. These convictions or habits to understand Aristotle arose well after Aristotle’s death, some in the Late Antiquity already, some in the Middle Ages, some in modern times and they follow the mentality of different periods in their reception of Aristotle. During the last hundred and fifty years the conditions under which we can understand Aristotle have changed significantly, so we have a motive to verify these prevailing opinions about Aristotle’s philosophy. First, the philological treatment of the texts has contributed much to that change even if it doesn’t primarily concern the content. In the 19th century different philologues established criteria to edit the texts, they prepared critical editions of many classical authors, including Aristotle; some editions are still in use even if it was possible to improve the text here and there using new means and methods of text-editing. Due to their work, we can read the CorpusAristotelicum in the accessible form, which we know. W. Jaeger initiated a further major change; he showed us that the text is not of the kind, which was presumed so far. The difference between the old and the new view is comparable to the difference that was detected

272

E. SONDEREGGER

in the poems of Homer. Until Friedrich August Wolf’s Prolegomenaad Homerum(1705), it was thought that Homer composed his poems in the same way as poets of the Middle Ages, the Renaissance, the Baroque and so on, i. e. wrote some text on his desk, then corrected it considering his other drafts etc. and finally made a last written version. After some detours, it became clear by the research of Milman Parry and A.B. Lord that the origin of the Homeric poems lies in the oral poetry, and that therefore many formal characteristics of his songs must be evaluated quite differently than had previously been assumed. Similarly, Aristotle’s texts are not of the supposed kind, a book written with the goal to be published, as are the texts say of Kant or other modern philosophers, but they are scripts of a teacher who prepared his lectures and seminars. He used them over the years, added or omitted words and phrases and sometimes, but not as a rule, he also revised a text for publication as in the case of the Nicomachean Ethics. I also want to add that Hegel, being himself a speculative thinker, has recognised the speculative centre in Aristotle’s philosophy. Meanwhile this insight has been lost. In what follows, I will try to counteract this where possible. The third element of change is due to the hermeneutical philosophy, which helped us to better understand customs, people, texts, cultures, which are alien or strange to us. We have learnt that it is indispensable to take into account the conditionality of our own understanding if we want to understand a foreign culture. Doing so we integrate the foreign (dasFremde) in our world and can understand it by a “fusion of horizons” (Gadamer). If we wanted to understand the foreign in its own world then we had to replace our Doxa by the foreign Doxa, and the thing or event in the foreign world would be something else than the integrated thing or event. No foreign cultural or natural thing can be understood without getting acquainted with the world, which it is embedded in. In the third paragraph we will come back to this theme. But first, we will examine the claim that Aristotle has practiced science. We will see why his occupation with natural beings is not comparable with modern science and we will complement this negative part with a positive answer to the question of what the goal of his thinking about nature is.1 1 The other two claims are examined in my commentaries on Metaphysics Z (2012) and Λ (2008). An English version is underway.

ARISTOTLE’S THEORY OF NATURE

273

2. Aristotle Practices Science 2.1 TheThesis That Aristotle is doing science is probably the least contested claim in Aristotelian scholarship, besides the view that he has developed a metaphysics of substance.2 That seems to be plain just for the reason that half of the CorpusAristotelicum deals with natural beings.3 Especially in biology Aristotle is credited with good and correct insights. But there is no consensus concerning the Aristotelian science as a whole. Some think that it is amazing what Aristotle has discovered with his poor means,4 while others blame him for having hampered science for more than two thousand years.5 But both, “friend and foe”, assume that Aristotle at least has intended to proceed by science and that of course there is no other way to know something sound about nature than by science. In front of 2 In the words of R. J. Hankinson in Barnes (1995), p. 136: “... Aristotle’s claims to being a scientific empiricist.” Wolfgang Kullmann in Rapp (2011), p. 106, adds: “Der Entwurf einer Biologie ist eine der erstaunlichsten Leistungen des Aristoteles. Offensichtlich ist er Teil eines umfassenden Konzepts, die gesamte Natur in ihrer Vielfalt und allenihrenDetailszuerforschenundihreStrukturenzuerklären.” Wolfgang Kullmann has dealt with Aristotle’s science since 1974, WissenschaftundMethode.Interpretationen zuraristotelischenTheoriederNaturwissenschaft; meanwhile many times, e.g. 2007 and in his latest book from 2014, AristotelesalsNaturwissenschafter. This book he describes as “Fazit seiner Aristotelesforschungen.” In the first footnote he clarifies his position: “DerTitel[Wissenschaft und Methode]klingtinkritischerAbsichtandendes1960in TübingenerschienenenBuchesvonHans-GeorgGadamerWahrheit und Methodeanund sollte zum Ausdruck bringen, daß die naturwissenschaftliche Methode, die Aristoteles anwendet, universal ist und sich im Prinzip von der modernen naturwissenschaftlichen Methodik nicht unterscheidet und daß insofern auch die philologisch-historische AuslegungdesAristoteleskeinbesondereshermeneutischesProblemdarstellt.” 3 In the Handbuch,edited by Rapp/Corcilius there is no entry on “Naturwissenschaft”, but only on “Biologie”, the Companionhas an entry on “Science” with the typical ambivalent evaluation of the question, whether Aristotle was a scientist or not. Flashar, 1983, dwells on NaturphilosophieundNaturwissenschaft in his contribution to Aristotle. 4 Wolfgang Kullmann in Rapp/Corcilius (2011), p. 106, writes: “Der Entwurf einer BiologieisteinedererstaunlichstenLeistungendesAristoteles.” 5 Arthur March (1957), p. 18: “EsbestehtfürNaturwissenschaftlerkeinGrundindie Verehrungeinzustimmen,dieAristotelessonstgeniesst.ErhatdurchseineAblehnungdes Atomismus,dessenAusbausicherbereitsimAltertumzubedeutendenErgebnissengeführt hätte,denFortschrittderWissenschaftaufzweiJahrtausendeaufgehalten.Und,wasvielleichtnochschlimmerist:erhatalsUrhebereinerGeistesrichtung,diealleGrundsätze desphysikalischenDenkensverkannteunddieermitdemganzenGewichtseinerungeheurenAutoritätvertrat,aufdiespätereEntwicklungnichtblossderPhysik,sondernauch derübrigenNaturwissenschaftendenverderblichstenEinflussgenommen...,” quoted from the introduction by G. A. Seeck, Seeck (1975), p. XII.

274

E. SONDEREGGER

the different shortcomings, which modern scholars detected in Aristotle’s science, we must ask where the roots of them are. To come to the point: my answer will be that there are no such shortcomings at all because Aristotle’s research in the realm of natural beings either is ἱστορία or θεωρία περὶ φύσεως, and neither of them is science in its modern sense. Science does not belong to his world. 2.2 WhyAristotle’sOccupationwithNatureisnotScience Let’s have a look at that part of Aristotle’s occupation with natural beings, which Aristotle names “Theory about Nature” (ἐπιστήμη, θεωρία περὶ φύσεως). Distinguishing between Aristotle’s theory of nature and modern natural sciences we remind of three fundamental conditions to which modern science is subjected. First there must be a philosophy of science as a theoretical and methodological fundament, second modern science is based on a metaphysical decision that only the measurable, the countable and the ponderable really is, and third it presupposes that knowledge has a goal outside of itself, namely the control over nature in order to exploit it. The philosophy of science has to complete many tasks. It must construe a path leading from empirical data to generally valid knowledge. It must provide methods how to get rules and laws of nature from experiments and particular knowledge. Many say that in the Analytics Aristotle aims at a philosophy of science of this kind. Let’s see. The FirstAnalytics give a theory of syllogism. That’s a part of logic, but logic is only a necessary constituent of the philosophy of science and does not bear on the content. In the second book of the Second Analytics Aristotle enumerates four questions, which are relevant for knowledge. Τὸ ὅτι stands for the question ‘Is it true that S is P?’, τὸ διότι stands for the question ‘Why S is P?’, εἰ ἔστιν stands for the question ‘Is there any S?’, and τί ἐστιν stands for the question ‘What is S?’ Aristotle then develops and checks methods to answer these questions. Doing this, he encounters the question what knowledge in the form of a proof is and what the difference between definition and proof is (ὁρισμός and ἀπόδειξις). Definition and proof are different kinds of sentence (conclusion: a→b; definition: a=b), and we must not confound these two. In this line of thought, Aristotle never speaks about the transition from experience and experiments to rules, laws and theories, which is indispensable in modern natural sciences. Both Analyticstreat knowledge in general but not science.

ARISTOTLE’S THEORY OF NATURE

275

We find the first efforts to establish a philosophy of science about 1200 AD. At this time, when philosophers began to think about natural science, we see the opening of a dispute about Aristotle’s “science”. Those interested in science tried to dissociate their undertakings from elements attributed to Aristotle. According to Crombie Robert Grosseteste had a particular position in this quarrel and was the first to introduce methodological rules.6 Francis Bacon, John Stuart Mill, Auguste Comte advanced in the same direction, which has found several other expressions in later philosophy of science. The general attitude has changed in our time in so far as science is widely attributed to Aristotle. Hugo Dingler however says that the Greeks aimed at consistency and clarity and that they had results especially in logic and mathematics, but in the realm of changing things they couldn’t arrive at “bringing them in a precise shape” (p. 62), because they lacked the quantitative experiment,7 an observation which is confirmed by Flashar, 1983, 391: “Aber auch von einer ‘theoretischen Physik’ nach unserem Sprachgebrauch unterscheidet sich die aristotelische Physik, denn ihr korrespondiert keine Experimentalphysik.” But other scholars claim that Aristotle could handle experiments very well. Wolfgang Kullmann and Alfred Stückelberger made many attempts to prove this, pointing out sentences which refer to concrete observations and pieces of information by Aristotle which often resemble sentences in modern science. But neither of these two authors take into account that the respective fundamentals differ considerably. Of course, Aristotle based many statements about natural beings on experience, also in the case of astronomical observations, even if he is very cautious in this area;8 of course, there is control and repetition in 6 See Crombie (1953), p. 135; further information in Crombie (1995), Vol. II, pp. 27–28. Bacon, NovumOrganum, I 95 (cf. 70, 82, 100); concerning the dispute: Birch (1756), p. 57. Already Francis Bacon, 1622, had described the principle of test series. – J. St. Mill formulated four methods of experimental compliance: 1. MethodofAgreement, 2. Method of Difference, 3. Joint Method of Agreement and Difference, 4. Method of Residues,Mill (1843). 7 Dingler (1928). 8 See the provisos in Metaphysics Λ 8 und in DeCaeloI. Aristotle acts in a quite different way than Simplicio, who refused to look through the telescope. The real coevals of Galilei are even worse than the fictive Simplicio. See Blumenberg (ed.) (1965), the introduction and “Das Fernrohr und die Ohnmacht der Wahrheit”, Blumenberg (1965), with all desiderable details.

276

E. SONDEREGGER

experience, but this is also true for a child, which has found out something, and can not be named “scientific experiment”. We must distinguish between experience and experiment. As Ernesto Grassi said in a concise formulation: an experiment “questions nature in respect of a theory which is drafted already, to verify whether the experiment confirms or refutes the theory.”9 In the last chapter of the Second Analytics, B 19, Aristotle explains his concept of experience, a text which shows how this concept of experience differs from the modern concept of experiment. Aristotle asks what the first and the fundamental is in proof and knowledge, including knowledge acquired by experience. For animals, knowledge begins with perception, but perception is not conceived as a mere receptive and mechanical process because it has the ability to distinguish. Many impressions received by the animal confirm each other and remain, thus letting memory emerge. And if a memory is repeated, that gives rise to experience, ἐμπειρία. Using Plato’s characterisation of the ideas, Aristotle tells us that experience is the “One above the Many”. This experience is the origin of τέχνη if it is about becoming and producing, and of ἐπιστήμη, if it is about being. Many take for given that Aristotle practices science, but hardly anyone contests that the content of his Physics methodically and thematically falls outside of what modern physics does. To make clear the difference, it might be sufficient to mention how differently Aristotle and Newton conceive space, time, movement and their respective functions. For modern physics it is essential that measurements are repeatable, exact, unambiguous; physics builds up a mathematical model, which can be interpreted physically. Corporeal objects and their characteristics, further, how they react and relate to each other, must be captured insofar as all this is measurable. Against this, many fundamental conceptions in Aristotle’s Physics are notions of being, articulating the being of natural entities. Modern physics defines movement or motion as the change of the respective position of two bodies over time; it is thus of fundamental importance to measure and calculate velocity and the acceleration of corporeal objects. Aristotle on the other side defines movement as “the actuality of a possible being as such.” Such and similar statements (e.g. selected from Physics B) are at least misleading in the context of modern physics and, strictly speaking, they rather become senseless. One can 9

Grassi (1955), p. 135.

ARISTOTLE’S THEORY OF NATURE

277

accuse Aristotle’s Physics of lack of scientific character only from a standpoint, which is not Aristotle’s. 2.3 What is the Aim of Aristotle’s Research on Nature and Natural Beings? 2.3.1 Aristotle’s Physics as “Theory of Nature” We have seen some reasons why the statements about nature and natural beings in the CorpusAristotelicumhave neither a scientific aim nor use nor value; now we are going to consider how Aristotle speaks about nature and natural beings and what he aims at in doing so. For this purpose, we use the text of Physics Γ, where Aristotle introduces the concept of movement, and additionally Physics Θ (especially its first sentence); in this book Aristotle treats the same topic as in Metaphysics Λ, namely the question of how and where movement begins, what the origin of movement is. Book Θ very often was considered to confirm the theological interpretation of Metaphysics Λ. We want to examine whether rightly or not.10 When asking what nature is and what natural beings are, Aristotle carries on an earlier question in two respects. On the one hand the Presocratics, but Plato too, had dealt with nature and, as Aristotle, partly in a speculative manner. So Aristotle finds points of reference in the opinions and statements of predecessors and contemporaries. On the other hand, his topic is a continuation of his systematic project, i.e. of the question about the meaning of being, because natural beings are one of the different types of being (γένος τι τῶν ὄντων); since he had sketched the programme in Metaphysics Λ, he pursues his inquiry in adapting this same question to the different types of being. In Metaphysics E 1 Aristotle subdivides the being in general through the criteria “movable” and “separable” and their negations into four areas of being. Natural beings are the group of beings, which are separable and movable; 11 the knowledge interested in these beings is a theoretical one 10 See Corcilius, in Rapp & Corcilius (2011), p. 83; he toο translates τὸ πρῶτον κινοῦν with “unmoved mover”. which is false because the greek term is neutre, never masculine; the masculine translation is designed to evoke the Christian God. 11 A short note on “separable”: in this context no absolute autonomy is required, nor an independent existence (the line of thought which leads to substance), but the fact only, that something — the separable — can be considered to some degree in itself, without producing heavy errors: we look at something “as if it were alone’.

278

E. SONDEREGGER

because it does not consider natural beings with the intention to do something with them, and second because this knowledge has nothing to do with the manner in which we act and behave in relation to one another.12 The theoretical method to get knowledge allows us to ask for the world as the frame of our being, without using this frame as an implicit presupposition for further claims. This question can’t be asked and answered with methods which are applicable in research about things in our world; it would then be an “external question” of the kind Rudolf Carnap rightly dismissed. Theory is neither inductive nor deductive, it doesn’t proceed empirically nor does it presuppose axioms: theory asks for the first opinions in an opinion-based world.13 The basis of theoretical arguments is given through opinions (ἔνδοξα) about natural beings which are capable of movement and variable and which always include ὕλη. The knowledge about natural beings which Aristotle searches should be ἐπιστήμη, that refers only to beings or characteristics, which are necessary and a priori, non-empirical and non-contingent. In short, theory bears on the being of beings. Theory about nature has to do with the necessary and unchangeable in natural beings, which have the origin of movement in themselves and which are relatively independent. In this sense, the Aristotelian Physics is not science but “Theory of Nature”, as he names it in Physics Θ 114 and its main topics are possible to resume in the following way: because natural beings are beings capable of movement we have to ask what movement is as the being of this type of beings, and second, we have to ask what φύσις is as the principle of movement. 2.3.2 What is φύσις? Physics B treats the question what φύσις is. I quote some thoughts from this book. Aristotle says that natural beings are those, which have the principle of movement and rest in themselves (ἀρχὴ κινήσεως καὶ στάσεως),15 and that φύσις is the origin of movement and rest (192b14 For further considerations on the distinction ποίησις – πρᾶξις see ENVI 3. Cf. Erwin Sonderegger (2008), p. XXVI. 14 Physics Θ1, 250b17. In the doxography which introduces the topic Aristotle speaks about the presocratic approach as περὶ γενέσεως καὶ φθορᾶς εἶναι τὴν θεωρίαν; in the concluding sentence, 251a6, he says that movement is of great interest for the περὶ φύσεως θεωρίαν and the question about the first origin. 15 Note that these are two of Plato’s highest genera in TheSophist. 12 13

ARISTOTLE’S THEORY OF NATURE

279

and 21). Anything that has φύσις as the source of its movement, is οὐσία, a natural being (192b33). In this text, the term οὐσία has two entirely different meanings, which by no means are indicated. Aristotle explains the difference in Metaphysics Δ 8. First, οὐσία means beings (Seiendes). Examples for this use are earth, fire etc., any type of body and any parts of them; on the other hand, οὐσία means the being which gives the beings a ground for their being (ὁ ἂν ᾖ αἴτιον τοῦ εἶναι). It becomes clear that this difference is at work when we read that φύσις as οὐσία is a natural being, but on the other hand, that φύσις is the ground of the being of beings, because it causes their movement. If movement is the being of natural beings and if φύσις is the source of this movement, then φύσις is the origin of the being of natural beings. It is the being, on which becoming must be founded according to MetaphysicsΛ 6–9. Metaphysics Δ 4, 1015a13–16, has the same theme; after some considerations about different uses of φύσις we read that “if φύσις is taken in its first and most proper sense it is the οὐσία of that which has the origin of movement in itself”, which means that nature in the first and most proper sense is the being of natural beings. Similarly Aristotle defines φύσις in Metaphysics H 3, 1043b22–23: “In the range of the perishable only φύσις can be οὐσία.” Here ousiadoesn’t mean a particular being but the ground of natural beings. And physis is cause in each of the four forms of causes; of course, physis is an efficient cause, but neither exclusively nor primarily.16 2.3.3 What is Movement? For the definition of movement the other premises, exposed in Physics A, still apply, namely that natural beings are moved and that φύσις is the principle of movement. I report some thoughts from the two chapters at the beginning of book Γ. First Aristotle denotes the central question, which is about movement and distinguishes it from further questions (τὰ ἐφεξῆς, follow up questions). Then he exposes the conditions under which the question about movement must be asked. One of them is the 16 And never in the form of the “First Mover” as is repeatedadnauseam.The latest examples are some contributions in the anthology edited by Christoph Horn (2016); for one reason why the term “First Mover” is a false translation in Aristotle’s text, see footnote 10; further reasons why Met. Λ does not contain any theology are given by E. Sonderegger (2008), 93–98.

280

E. SONDEREGGER

modal distinction between actuality and potentiality; another is the categorical distinction. Given the categorical distinctions, movement belongs to the πρός τι, the relation, because in movement the moving and the moved are related to each other.17 Finally he determines where the movement is: there is no movement besides the things (παρὰ τὰ πράγματα; Γ 1, 200b26–201a9). After these terminological preparations he can determine movement: “Movement is the actuality of a potential being as such.”

The thoughts of his predecessors confirm his definition even if they seem to contradict it. Some said that movement is a difference, others that it is inequality, others that it is non-being. Aristotle responds that it is comprehensible to share such convictions because the status of movement is somehow vague, it is something between being and non-being, it is neither the mere possibility nor definitive actuality (Γ 2, 201b16–27).18 After some transformations of his definition partly to avoid misunderstandings partly to make it more explicit, he integrates this aspect in the definition, saying: “Movement seems to be actuality, but one that is not finished.”

The actuality of movement being incomplete, movement implies both modalities, while otherwise actuality and possibility must be separated (Γ 2, 201b27–202a3). Now we combine the statement about movement in Physics Γ with the question in Metaphysics Λ. Any natural being is moved. Movement is the being of natural beings, and as such, it is the actuality of a possible being as such. — That means that movement, κίνησις, is a concept of being (Seinsbegriff). The being of natural beings is determined by movement. In Aristotle’s Physics movement does not have the function to measure something or to make something measurable; it is not about to find out how quickly something moves or speeds up etc. Φύσις as ἀρχὴ κινήσεως is a principle, an origin of the being of natural beings.

17 Also in modern physics movement is a relation, but the relation between a moved body and an absolute or a rigid reference system. 18 Cf. the problem in the TheSophist that being is not pure being and becoming is not simply nothing. – Concerning the same problem, see G. W. F. Hegel (1967/1812), Erstes Buch.

ARISTOTLE’S THEORY OF NATURE

281

Remember the speculative question in Metaphysics Λ: What is the origin of becoming? The first answer in Metaphysics Λ is: Becoming is founded on being. The second question asks: On which being is becoming founded? The answer to the second question is: The natural comingto-be is founded on φύσις, which is the cause of being in all four senses of “cause”. Becoming aims at attaining its nature in the sense of the form, φύσις as εἶδος, as Aristotle says: ἔτι δ’ ἡ φύσις λεγομένη ὡς γένεσις ὁδός ἐστιν εἰς φύσιν, “φύσις in the sense of becoming is underway to φύσις.” (Physics, B 1, 193b12-13) This aim, to attain its εἶδος and by this tobeis the origin and the cause of movement (Γ 2, 202a9– 12). Achieving its εἶδος a being finds its place in the νόησις and in this way itis and is actually. So the Physicsanswers the speculative question asked in Metaphysics Λ. 2.3.4 The First Moving in Physics Θ and Metaphysics Λ Aristotle’s Physics is no more a “book” than his Metaphysics. It is a compilation of different texts, probably put together by a redactor. Normally the books A–Δ (I–IV), and the books E, Z, Θ (V, VI and VIII) are considered as groups, H (VII) is a separate book. The last book, Θ, begins as follows: Πότερον γέγονέ ποτε κίνησις οὐκ οὖσα πρότερον, καὶ φθείρεται πάλιν οὕτως ὥστε κινεῖσθαι μηδέν, ἢ οὔτ’ ἐγένετο οὔτε φθείρεται, ἀλλ’ ἀεὶ ἦν καὶ ἀεὶ ἔσται, καὶ τοῦτ’ ἀθάνατον καὶ ἄπαυστον ὑπάρχει τοῖς οὖσιν, οἷον ζωή τις οὖσα τοῖς φύσει συνεστῶσι πᾶσιν;

Did movement once come to be, not being before that, and will it pass away some day, so that nothing moves anymore, or didn’t it neither arise nor will it pass away anytime, but it ever was and will ever be, and that is an immortal and perpetual feature of the beings, like life for all natural beings?

The book opens with a question, an unusual beginning. The question is quite general, and indeed Aristotle often begins with general statements; to choose a beginning, which many can agree on, is a rhetoric and didactic device. Aristotle formulates his question concerning movement positively and negatively and considers the respective consequences (in other cases

282

E. SONDEREGGER

he is somewhat elliptic) and the phrase then ends with a comparison. So the text gets somehow cumbersome and redundant. It creates a solemn mood and differs clearly from Aristotle’s famous sober style. The question contains an anti-Parmenidean programme. Aristotle proposes the same alternative concerning movement as Parmenides did concerning being. According to Parmenides, we can’t say that being has arisen nor that it will pass away. In the dichotomy ἔστιν ἢ οὐκ ἔστιν there is only one possible choice: ἔστιν, that it is. Now Aristotle asks just the same concerning movement: Has movement arisen or not? If the question is a rhetoric one, then he already hints at his result, i.e. that movement has no origin — exactly the same as Parmenides claims for being. To make plain what he means, Aristotle cites Parmenides with the word ἄπαυστον, whose first occurrence is in frg. 8.27: ἐὸν...ἔστιν ἄναρχον ἄπαυστον. Subsequently, we find the word in choral songs of all three tragedians, then in the Timaeus where Plato describes how the demiurge formed the soul of the world (36e), “the way its [the soul’s] perpetual and rational life began.” Aristotle’s formulation is even closer to the locus in Kratylus 417c, where both words, ἄπαυστος καὶ ἀθάνατος are used to describe φορά, movement in place. The choice of words is emphatic, poetic and allusive.19 There are further parallels between Parmenides’ statements about being and Aristotle’s questions about movement.20 Theophrastus too uses the same terms describing the circular movement of the heavens, Met. 5a4, where he denotes it as συνεχὴς καὶ ἄπαυστος. And we find the same comparison later in Met. 10a15, οἷον γὰρ ξωή τις ἡ περιφορὰ τοῦ παντός, “the circular movement of the universe is like life.” So the first sentence of Physics Θ is in the tradition of Parmenides and Plato, and especially connected with questions which were discussed in the Academy and in Theophrastus’ Metaphysics.21 19 Further occurrences in the CorpusAristotelicum, besides Physics Θ1, 250b14: De Caelo, A 9, 279b1, B1, 284a9, B 5, 288a11, DeGenerationeetCorruptione, A 3, 318a25, Metaphysics Λ 7, 1072a21, Physics, Θ 6, 259b25; mostly used as an attribute to κίνησις, μεταβολή, κυκλοφορία. 20 Parmenides on being Aristotle on movement 8.3 ἀγένητον – ἀνώλεθρον γέγονεν – φθείρεται 8.10 ὕστερον ἢ πρόσθεν πρότερον 8.5 οὐδέ ποτ’ ἦν οὐδ’ ἔσται, ἐπεὶ νῦν ἔστιν ἀεὶ ἤν καὶ ἀεὶ ἔσται ὁμοῦ πᾶν 21 Ross and Wagner don’t mention these references; nor is there a word about them in the commentaries of Tricot, van Raalte or Laks-Most.

ARISTOTLE’S THEORY OF NATURE

283

These references and the solemnity Aristotle has chosen underline the importance of the question about the origin of the movement and at the same time the position Aristotle takes in regard to his tradition. Instead of the visionary revelation about being, which Parmenides received, instead of the life of the soul of the world, which Plato put into the form of an εἰκὼς λόγος, and where Theophrastus pointed on the cosmological aspect of the circular movement of the heavens, Aristotle asks for the origin and beginning of movement in a speculative approach. Aristotle treats the circular movement as the first spatial movement in Physics Θ 9; in Θ 10 he asks for the primary moving cause. The determinations of the first moving and the first movement are coordinated. Both have neither a temporal beginning nor an end, they are uniform, unique, they have no parts and take no space, both are unmoved but move. From the fact, that the first moving cause as the origin of movement must be unmoved itself, it results that it is not a natural being. From the fact, that the origin of movement can’t have parts or extension, one could conclude, that it is a mathematical or a geometrical entity. But Aristotle many times rejects the idea, that the origin of being could be of this kind. The eternal and unmoved being we are looking for, is not that of geometrical forms (Θ 2, 252b2). I try to summarise the results concerning the first moving cause. That φύσις is the origin of movement because it is the being of natural beings remains part of the accepted presuppositions.22 But φύσις is the ground of natural beings also in the sense of ὕλη, εἶδος and τέλος. The starting point of natural becoming is φύσις as ὕλη, the process of becoming reaches its end in the φύσις in the sense of εἶδος. When this state is reached, what is becoming has arrived at actuality (ἐνέργεια). Aristotle metaphorically says elsewhere that this goal is “aimed at”23 by the becoming beings (if I may use this oxymoron), in fact, this goal is the first moving cause (τὸ πρῶτον κινοῦν ἀκίνητον). In this context actuality (ἐνέργεια) does not mean “to exist” in our concrete and actual world, but it means that the coming-to-be has reached its τέλος in its actual and real φύσις in the sense of εἶδος. The εἶδος is noetic (that means it is not something like a natural or a geometrical shape). In reaching its εἶδος the being fits in a noetic world wherein it can be. Once that has happened, a factual mundane being has 22 23

B 1, 192b21; see the first sentence of Γ 1; in Θ 3, 253b5 he names that a ὑπόθεσις. Met. Λ 7, 1072a26.

284

E. SONDEREGGER

realised a particular noetic node of the world. Because the world wherein the new being is, is a noetic structure, the thing which has come to be can not be just for itself. In Aristotle’s terms that structure is νόησις. (More about this speculative moment in Aristotle’s thought below; for the moment I just want to stress that this noēsis is not the awareness of a particular mundane subject). This noēsis is the apriori unity of which being and actuality, the being aware and the being, which it is aware of, are results. Without this unifying common ground, which precedes noetically, no mundane being or event could be what it is; everything would lack the world wherein it can be. A particular being comes to be in the speculative sense, when it takes its place in this form of noēsis. Within this process, anything systematically (but not as a particular thing) provided in the νόησις can become real or rather actual in the sense of Metaphysics Λ 6–9, it can reach its ἐνέργεια, now no more only as an actuality of a possible being (which is in movement), but as a stable actuality. 3. Opinion-based Worlds I take up the result from PhysicsΘ and combine it with the speculation of Metaphysics Λ. The goal of the becoming of a natural being is to attain its φύσις as εἶδος; if that happens, then what was a becoming finds its actual being, enters into the structure of a world by realising its εἶδος; so far the result of Physics Θ. The starting point in Metaphysics Λ is the question about being, περὶ οὐσίας ἡ θεωρία. Aristotle explains this issue more precisely; we have to ask which is the being which grounds becoming. The first five chapters make clear that no natural being, nothing in the world is able to fulfil this function, not even the sun and the sphere of fixed stars. In no way it is possible to give this question a cosmological or an empirical answer, it must be answered speculatively. In the lines 1071b19–20 Aristotle provides the first part of that speculative answer: “The origin must be such that its being is an actuality.” By “origin” (ἀρχή) he makes clear that he speaks about the first moving cause. This statement can be translated in the formula first moving causedef {being – actuality}.

Actuality doesn’t mean that the first moving “exists already”, an answer which would remain in mundane limits. It’s not about the existence of

ARISTOTLE’S THEORY OF NATURE

285

something in an already existing world, but it’s about the actuality in the νόησις, as set out in chapter 7, 1072b14–21. The dash between “being” and “actuality” does not indicate identity or sameness because being and actuality do not relate to one another as two separate things, which have the chance to relate to one another afterwards. Their relation is rather something like that of a theme and the modus of that theme. Being and actuality are results of a process, whose origin lies in the νόησις, awareness, which now moves into focus.24 Without awareness (νόησις) there is neither being nor actuality. Awareness is before both; therefore I mark that with arrows, which start from awareness: the first moving cause is {being ← awareness → actuality}. In our everyday life, we have to distinguish that which is aware of something and that of which it is aware. But if we reflect on awareness we see that awareness can take notice of something else only if that which it notices at a given moment has already been in it — not temporally but systematically — as a node in a noetic network or as an interception point of some fundamental distinctions. Without that network of distinctions things would remain unnoticed, there would be no place for them to be. The result of this is, that the everyday distinction between that which is aware of something and that which it is aware of, is suspended in the process of entering in the network named ‘world’. A natural being “seeks for” its place in that structure, the place being its εἶδος. To find this place, to become actual in the νόησις, is the goal of the natural process, but this is only a formal structure. The content of a respective world depends on the Doxa, i.e. on the set of fundamental opinions (such as conceptions, distinctions, values). These make up the real content of a certain world. Therefore the formula for tobe as that on which becoming rests and therefore being the first moving cause can be completed as follows: ⎧

Being ← awareness → actuality ⎫ ⎬ ⎩ ⎭ Doxa ⎨

24 Usually, νόησις is translated by thinking, but as Kant said in the Preface to the second edition of the Critique of Pure Reason: “...I can think whatever I will if only I don’t contradict myself...”, Kant (1998 [1781/1787]), B XXVI, Footnote. In contrast to this I can be aware only of something that is. The translation is thus misleading. Even if awareness is connotated with modern mentalism, thinking is not less subjective; awareness has the advantage to have the same root as the German “gewahren”, which is much closer to νόησις than “thinking”.

286

E. SONDEREGGER

This formula, which is developed from the chapters Λ 6, 7 and 9, can be explained by a sentence, which Aristotle uses in Λ 9, and in which he concentrates his speculation as a whole. It is the famous phrase καὶ ἔστιν νόησις3 νοήσεως1 νόησις2. My explanation of this phrase will here be short.25 All previous explanations of this phrase that I know of identify the first mention of noēsis with the third, so that it seems that Aristotle speaks about the “thinking of thinking”. But in fact, noēsis has three mentions with different meanings, and the translation by “thinking” must be replaced by “awareness” or “being aware” (bemerken,gewahren; see footnote 25): – νόησις3 means the factual everyday noticing; to be aware of this or that particular thing, event etc., awareness in our normal use; – νοήσις1 means the noetic world before the actual world; it means the structure of a certain world; its content is defined by the fundamental opinions, in short, the Doxa; – νόησις2 is our faculty or ability to render this structure actual in a given case. Νοήσις1 is totally different from νόησις3. While νόησις3 is personal, individual, the usual being aware of this or that, νοήσις1 in contrast is impersonal; it is not an act of a subject, even not of a god, it is the mere noetic structure which makes possible to happen νόησις3. Only using the fundamental distinctions (= Doxa) of this νοήσις1 our ability to be aware of something can change into being aware actually of something as something. From the worldly point of view νοήσις1 is a mere possibility, a noetic structure, no more than a grid, but from the speculative point of view, it is the ground of every being. Without its world-constituting distinctions, there is no experience, no existence of anything. The sentence may be paraphrased like this: ‘Any of my perceptions, sensations, experiences, thoughts (= νόησις3) realises (= νόησις2) a noetic structure (= νοήσις1) in my mundane world; so the being aware and that which the being aware is aware of are and are real and true.’ Or expressed in another way: ‘To be aware of a particular being, event etc. (= νόησις3) is the realisation of a faculty to do so (= νόησις2), namely

25 For an extended analysis I refer to my commentary on book Λ, Sonderegger (2008); a revised English translation is in progress.

ARISTOTLE’S THEORY OF NATURE

287

to become aware of something as something (etwasalsetwasbemerken) in a world (= νοήσις1).’ There are several possible sets of fundamental opinions. Each set forms a world. Often these different worlds are considered as different cultures. But we must distinguish between worlds and cultures. The world includes cultures and is more comprehensive than culture; culture means primarily that which is formed by man; in the world, there are many things, which are not made by man. Aristotle does not speak of a plurality of worlds, but he gives reasons for such a view in stressing the function of the ἔνδοξα, the opinions. He shows that it is never possible to leave the limits of opinions. But we can distinguish common opinions and beliefs from the fundamental opinions and beliefs, which they rely on. And the task of the philosopher is to find the opinions, which offer the ground of everyday convictions (fundamental opinions, sometimes named the self-evident, dasSelbstverständliche). If all we know is within the limits of fundamental opinions and if the respective sets of opinions are different throughout time and place, as we in fact can see, and if these sets of fundamental opinions are the basis of respective worlds, then we can speak of a plurality of worlds drawing on the speculative sketch Aristotle has presented in Metaphysics Λ. Different worlds can have many distinctions in common. But they are not just different views on the same reality (as is common belief). There is no such objective world “above” the opinion-based worlds. How could we find it? We ought to have a position outside of any world to do that, which is impossibe. The world determines what can be and what can not be. There can be only things, which fit in the network of its world. In any world, it is possible to reach truth, but there is no truth above the different worlds, there is no “super-world”. Comparision, relation and translation between different worlds are possible inasmuch as they may have some fundamental opinions in common, which is very probable if there is only a finite number of fundamental opinions. 4. Compensation for Damages What I have said to this point seems to be negative only. No theology (some would agree),26 no metaphysics of substance in Aristotle’s philosophy 26

E. Sonderegger (2008).

288

E. SONDEREGGER

(I don’t know anybody so far who would like to agree, but I hope that will change)27 and moreover no science (for most people a scandal). All this seems, if not totally wrong, quite severe, considering the whole tradition and the history of reception, which dealt primarily with just these themes. So I would like to apologise and give some compensation. But first, we must remember the fact that the claims of the standard interpretation in the three above mentioned fields are in themselves somewhat inconsistent. One of the main claims is that substance translates οὐσία correctly, further that we find an elaborated metaphysics of substance in Aristotle’s texts, leading to theology. And finally, that it is nearly self-evident that his occupation with natural beings has a scientific goal, which is in principle comparable with the goal of contemporary science. Unfortunately, all these assumptions have in common that they at the same time must contend that Aristotle’s theoretical and scientific attempts have significant defects, which must be rectified. Strictly speaking, the substance is unique (no absolutely autonomous subsistent being can have another independent being beside it). If we want to avoid this, we have to give a vague meaning to the notion, what, like in the first case, renders the concept useless. Concerning the metaphysics of substance, there is no consensus among its defenders whether the particular being or rather the εἶδος is the core of substance. Neither proposal is satisfying. Even in the eyes of his friends, Aristotle’s reasoning about theology is not very convincing because the alleged proof of the existence of God is not conclusive. Finally, concerning his science some say that it is excusable that a project like this in its very infancy has some imperfections or even flaws, but that nevertheless his research on nature can be lined up with modern science in principle. Others condemn it because Aristotle worked rather with speculative prejudices than with empirical research, and thus impeded scientific progress for two thousand years. What is the positive result if we were right to contend, that Aristotle had neither a metaphysics of substance nor a theology resulting thereof, nor a science? Aristotle’s work on nature has several main lines; one is collective and empirical like the Historia animalium. The aim of these books is ἱστορία, to know as much as possible about beings in nature, how they are, to get facts and material. The objective of this research is not to apply it for technical or economical purposes. The only aim is to 27

E. Sonderegger (2012).

ARISTOTLE’S THEORY OF NATURE

289

know and to find subject matter to reflect on. Ἱστορία is not science in the modern sense. Other works have methodical intentions, like Departibus animalium. A superficial view could here suggest a proximity to modern science, but if we take into account the three conditions of contemporary science mentioned above, we see the difference. Other works are examples of θεωρία περὶ φύσεως, theory of nature, the most important of them being the Physics. Aristotle’s theory of nature fits very well in the speculative sketch outlined in Metaphysics Λ, which is a Programmschrift for his whole life. Having shown that it is possible to ask the question about being and having taught how to do it, he can question different classes of beings: ‘What is the being of x?’ In the Nicomachean Ethics x is man, in the Physics x is nature. In the theoretical works on nature he thus asks the question about being in the specific frames of nature: What is the being of natural beings? In such works, Aristotle runs the programme which he has outlined in Metaphysics Λ. These are the least scientific texts; they are speculative in the true sense of the word. Concerning Aristotelian theology, there is a new consensus coming up. Ritschl and Natorp raised first doubts in the 19th century; Natorp rightly noticed that it was impossible for him to see any theology in the text at all, whereas Ritschl wanted to separate metaphysical and theological insight in God. Their approach had little impact on the Aristotelian research (even Gadamer as disciple of Natorp interpreted MetaphysicsΛ as a theology), but in the last decennia some scholars like R. Bodéüs (1992), H. Lang (1993), B. Botter (2005) and S. Fazzo (2012, 2014) have provided us with arguments against a theological interpretation of Metaphysics Λ. However, many of those who deny that the book has a theological content replace theology by a metaphysics of substance. However, if agreeing that οὐσία is the topic, one should give reasons for the translation “substance” (which in fact translates ὑπόστασις) and explain why this concept is adequate. In my commentary on Metaphysics Λ I tried to show that Aristotle has developed a speculative answer to the question about the meaning of being. Some results of that commentary were presented in the last chapter. We can leave behind us the obsolete concept of substance, whose origin is in the Stoa and whose primary use is in the theology of the early fathers and the subsequent Christian theology. Instead, Aristotle can learn us a lot about method and the kind of philosophy to choose. In a situation, where the so-called metaphysics of

290

E. SONDEREGGER

substance has lost its interest,28 he gives us a lesson on speculative thinking,29 on how to ask the question about being. It is possible to ask this question in the form of a reflection on our Doxa. The method to ask the question is speculation in the topical attitude, with its means προτάσεις λαβεῖν, to distinguish the πολλαχῶς λεγόμενον, to use quotations as termini, to distinguish the various distinctions in our speech, in short theoretical philosophy in the manner of thought without claims, unbehauptendesDenken.30 If we take suchathoughtwithoutclaims seriously, it seems to have a far reaching impact: we have to reflect on the world we live in, and if that world is shaped by fundamental opinions, if the sets of fundamental opinions can be different, if there is no criterion to decide about truth between different worlds, but only about truth in a definite world, then we have to respect other worlds (what is more than other cultures).31 LITERATURE Ancient authors I quote Parmenides from Diels-Kranz, Plato from Burnet, Aristotle from the OCT editions, Theophrastus from the edition of André Laks and Glenn W. Most, Paris 1993. Modern authors BACON, Francis (1622), NovumOrganum, London. BARNES, J. (ed.) (1995), The Cambridge Companion to Aristotle, Cambridge: Cambridge University Press. BEIER, K. & Heuer, P. (eds.) (2010), Ontologie,ZurAktualitäteinerumstrittenen Disziplin, Leipzig: Leipziger Universitätsverlag. BIRCH,Th. (1756),TheHistoryofRoyalSociety,London.

28 Cf. the anthology Ontologie, Beier & Heuer (2010), where the contributors try to prove the contrary. The book is the result of a colloquium under the auspices of the Communauté Saint Jean (2008). 29 See Sonderegger (2010) and Sonderegger (2017). 30 For more information about this unusual term, see Sonderegger (2010), pp. 56 and 72–76. – The first to expose and use this form of thinking is Plato in The Sophist, see Sonderegger (2012), I, 2. 31 For the far-reaching consequences if we take our world as the unique possible one or if we accept a possible multitude of worlds, see Sonderegger (2013).

ARISTOTLE’S THEORY OF NATURE

291

BLUMENBERG, H. (ed.) (1965), GalileoGalilei,SidereusNuncius,Nachrichtvon neuenSternen, Frankfurt am M.: Suhrkamp. BLUMENBERG, H. (1965), “Das Fernrohr und die Ohnmacht der Wahrheit” in H. Blumenberg (ed.) (1965). BUCHHEIM, Th., FLASHAR, H., KING, R .A. H. (eds.) (2003), Kannmanheutenoch etwasanfangenmitAristoteles?, Hamburg: Meiner. CROMBIE, A. C. (1953), Robert Grosseteste and the Origins of Experimental Science1100–1700, Oxford: Oxford University Press. CROMBIE, A. C. (1995), TheHistoryofSciencefromAugustinetoGalileo(reprint of the editions 1959 and 1970), New York: Dover Publications. DINGLER, H. (2014), Das Experiment. Sein Wesen und seine Geschichte (Neudruck der Ausgabe von 1928), Münster: Mentis Verlag. FAZZO, S. (2012), IllibroLambdadellaMetafisicadiAristotele, Napoli: Bibliopolis. FAZZO, S. (2014), Commento al libro Lambda della Metafisica di Aristotele, Napoli: Bibliopolis. FLASHAR, H. (1983), GrundrissderGeschichtederPhilosophie,DiePhilosophie derAntike3,ÄltereAkademie,Aristoteles,Peripatos, Basel: Schwabe. GRASSI, E. (1955), “Enzyklopädisches Stichwort ‘Natur’” in Werner Heisenberg, DasNaturbildderheutigenPhysik, Rowohlts deutsche Enzyklopädie, Hamburg: Rohwolt. HANKINSON, R. J., “Philosophy of Science” in Barnes (ed.) (1995), pp 109–139. HEGEL, G. W. F. (1967), WissenschaftderLogik, Erster Teil [1812] (ed. Georg Lasson), Hamburg: Meiner. HORN, Chr. (ed.) (2016), Aristotle’s Metaphysics Lambda - New Essays, De Gruyter, Boston - Berlin. KANT, I. (1998 [1781/1787]), Kritik der reinen Vernunft (ed. J. Timmermann), Meiner, Hamburg. KULLMANN, W. (1974), WissenschaftundMethode.InterpretationenzuraristotelischenTheoriederNaturwissenschaften,Berlin/New York. KULLMANN, W., “Biologie, Zoologie” in Rapp & Corcilius (eds.) (2011), pp. 106–115. KULLMANN, W. (2014), AristotelesalsNaturwissenschaftler,Boston – Berlin – München: De Gruyter. KRÄMER, H. (2007), Kritik der Hermeneutik. Interpretationsphilosophie und Realismus, München: Beck. MARCH, A. (1957), DasneueDenkendermodernenPhysik, Hamburg. MILL, J. St. (1843), SystemofLogic, London. PARRY, M. (1987), TheMakingofHomericVerse:TheCollectedPapersofMilmanParry, Oxford: Oxford University Press. RAPP, Chr., CORCILIUS, K., (eds.) (2011), Aristoteles Handbuch, Leben – Werk –Wirkung,Stuttgart: Metzler. SEECK, G. A. (ed.) (1975), Die Naturphilosophie des Aristoteles, Darmstadt: Wissenschaftliche Buchgesellschaft Darmstadt. SONDEREGGER, E. (2008), Aristoteles Metaphysik Λ: Ein spekulativer Entwurf, Einführung,Übersetzung,Kommentar, Bern: Lang.

292

E. SONDEREGGER

SONDEREGGER, E. (2010), DerspekulativeAristoteles, Untersuchungen zur Frage nach dem Sein in den mittleren Büchern der Metaphysik und zur Frage nach dem Sein des Menschen in der Nikomachischen Ethik, Würzburg: Königshausen & Neumann. SONDEREGGER, E. (2012), Aristoteles, Metaphysik Z, Einführung, Übersetzung, Kommentar, Würzburg: Königshausen & Neumann. SONDEREGGER, E. (2013), Welten,Globalisierung in: philpapers.org SONDEREGGER, E. (2017), “Nichtempirische Begründungen von Wissen und Verstehen” in Trópos. Journal of Hermeneutics and Philosophical Criticism, 1 (X)/2017, pp. 11–30. STÜCKELBERGER, A. (1988), Einführung in die antiken Naturwissenschaften, Darmstadt: Wissenschaftliche Buchgesellschaft. WOLF, Friedrich August (1795), ProlegomenaadHomerum,Halle 1795.