Oxford Studies in Medieval Philosophy Volume 9 9780192844637, 0192844636

Table of contents :
Cover
Oxford Studies in Medieval Philosophy Volume 9
Copyright
Contents
Perception and Extramission in De quantitate animae
1. Augustine and Extramission
2. The Textual Evidence
3. Dialectical Context
4. The Definition
5. Augustine’s Counterargument
6. The Grades of Extramissive Involvement
7. Extramission in the Augustinian Corpus
a. De musica
b. De Gen. ad lit.
c. Sermon 277
d. De Trin.
8. Conclusion
Bibliography
Believing Miracles in 10th–11th Century Baghdad: Four Christian Philosophers on a Problem of Epistemic Justification
1. Introduction
2. Yaḥyā b. ʿAdī (d. 974): Miracles and Epistemic Agency
3. Ibn Zurʿa (d. 1008): Towards a Rational Assessment of Miracles
4. Ibn al-Samḥ(d. 1027): Granting Belief to Reports of Miracles
5. Ibn al-Ṭayyib(d. 1043): Doing without Miracles
6. Conclusion: Christian Philosophers in the Light of Islamic Theology
Bibliography
Now Is Not the Time: Revisiting Avicenna’s Account of the Now
1. Aristotle and His Commentators on the Now and Its Relation to Time
2. The Now in Avicenna’s Philosophy of Time
3. The Now as wāṣil and fāṣil
4. Is There a Flowing Now Producing Time?
a. The Outermost Sphere
b. Indivisibles Are Incapable of Moving
c. A Point Cannot Produce a Line
5. Is There a Flowing Now Producing Time in The Lecture on Nature II.12?
6. Conclusion
Bibliography
a. Primary Sources
b. Secondary Sources
Thirteenth-Century Aristotelian: LogicThe Study of Scientific Method
1. Preliminaries
a. Problem 1—‘Science’ across Historical Periods
b. Problem 2—The Polysemy of ‘Scientia’
c. Thirteenth-CenturyAristotelian Logic (AL-13) and Medieval Science
2. The 1230s, Nicholas of Paris
3. The 1250s, Albert the Great
4. 1290s, Radulphus Brito
a. Brito’s Logic of Intentions
5. Conclusion
Bibliography
Bonaventure, Aristotle, and the Being of Universal Forms
1. The Current Consensus: Bonaventure’s Forms and his Attitude towards Aristotle
2. Forms: Universal or Individual?
3. The Relationship between Seminal Reasons and Universal Forms
4. Conclusion
Bibliography
On Loving God Contrary to a Divine Command: Demystifying Ockham’s Quodlibet III.14
1. Introduction
2. Five Claims of Quodlibet III.14
3. On the Contingency of Moral Status and the Goodness in Loving God
4. On the Command to Not Love God
5. On Obeying the Command to Not Love God
6. Conclusion
Bibliography
Productive Thoughts: Suárez on Exemplar Causes
1. Introduction: Two Problems
2. The Critique of Objective Existence Theories
3. The Act Theory
4. Acts and Their Causal Function
5. Conclusion
Bibliography
Primary Sources
Secondary Sources
Briefly Noted
Notes for Contributors
Index of Names

Citation preview

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Oxford Studies in Medieval Philosophy Volume 9

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

ADVI SORY B OARD Peter Adamson, Ludwig-­Maximilians University, Munich Deborah Black, University of Toronto Peter King, University of Toronto Henrik Lagerlund, Stockholm University John Marenbon, Trinity College, Cambridge Calvin Normore, University of California, Los Angeles Dominik Perler, Humboldt University, Berlin Eleonore Stump, St. Louis University Editorial Assistant Dawn Jacob, University of Colorado

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Oxford Studies in Medieval Philosophy Volume 9 Edited by

R O B E RT PA SNAU

1

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

1 Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © the several contributors 2021 The moral rights of the authors have been asserted First Edition published in 2021 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2021940881 ISBN 978–0–19–284463–7 DOI: 10.1093/oso/9780192844637.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Contents Articles Perception and Extramission in De quantitate animae1 Mark Eli Kalderon Believing Miracles in 10th–11th Century Baghdad: Four Christian Philosophers on a Problem of Epistemic Justification40 Alexander Lamprakis Now Is Not the Time: Revisiting Avicenna’s Account of the Now Andreas Lammer

77

Thirteenth-­Century Aristotelian Logic: The Study of Scientific Method Ana María Mora-­Márquez

147

Bonaventure, Aristotle, and the Being of Universal Forms Franziska van Buren On Loving God Contrary to a Divine Command: Demystifying Ockham’s Quodlibet III.14 Eric W. Hagedorn

187

222

Productive Thoughts: Suárez on Exemplar Causes Dominik Perler

246

Briefly Noted Adamson—Schabel—Schwartz—Panaccio

290

Notes for Contributors Index of Names

293 295

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae Mark Eli Kalderon

1.  Augustine and Extramission Augustine is commonly interpreted as endorsing an extramission theory of perception. Extramissive elements can be found in a number of his works (De musica 6.8.21, De Gen. ad lit. 1.16, Sermon 277.10, De Trin. 9.3). However, at least in his early work, De quantitate animae (henceforth, DQA) far from endorsing an extramission theory of perception, Augustine explicitly argues for its rejection. And yet on a common interpretation, Augustine is understood to endorse the extramission theory in DQA. (See, for example, Colleran 1949, 205 n55, 208–10 n73; Gannon 1956, 169 n56; O’Daly 1987, 82–3; Brittain 2002, 276 n64; Silva and Toivanen  2010, 248 n8; Brown  2012, 223–4; Toivanen  2013, 136; Silva  2014; Smith  2015, 154, 266. Nor is this a modern trend, thus Roger Bacon, Perspectiva 1.7.2 commends Augustine’s commitment to extramission even as Peter John Olivi, Quaest. in sec. lib. Sent. criticizes it. In addition, Peter Sutton, Quod. 1.24 attributes the extramission theory to Augustine’s early works, to which DQA belongs, even though Sutton holds that Augustine later abandons this view.) It is perhaps worth considering Augustine’s anti-­extramission argument in detail. That argument is a response to a puzzle or aporia that Evodius raises concerning the inextended, and, hence, incorporeal, nature of the soul. According to Evodius, we enjoy tactile sensation when we are touched anywhere on the body. If the soul is not itself extended throughout the body, then how could this be? Mark Eli Kalderon, Perception and Extramission in De quantitate animae In: Oxford Studies in Medieval Philosophy Volume 9. Edited by: Robert Pasnau, Oxford University Press. © Mark Eli Kalderon 2021. DOI: 10.1093/oso/9780192844637.003.0001

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

2  Mark Eli Kalderon We shall begin by discussing the textual evidence for attributing to Augustine an extramission theory in DQA 23. While suggestive, it is, by itself, inconclusive. Context matters if we are to understand Evodius’ puzzle and Augustine’s reply to it. In the third section, we shall review the project of DQA, in the fourth, we shall discuss Augustine’s definition of perception, before turning to the fifth, where we shall discuss Augustine’s reply in DQA 30 and explain why this commits him to rejecting the extramission theory. If Augustine explicitly rejects the extramission theory in chapter 30, then he could not have endorsed it in chapter 23, on pain of incoherence. If Augustine in DQA is committed to the rejection of the extramission theory, why is he persistently misunderstood? To address this question, the present essay ends with two sets of reflections. First, I shall provide a diagnosis for the misattribution in terms of unclarity about the commitments of the extramission theory. Specifically, we shall consider what it means to describe an account of perception as extramissive by distinguishing five different grades of extramissive involvement; some, if not all, are at the core of Augustine’s thinking. Second, in light of the distinguishable grades of extramissive involvement, we shall briefly review the strength of the evidence for attributing an extramission theory to Augustine on the basis of his other works. We shall see that, at best, it is neither required by scripture nor by reason, but represents authoritative opinion, such as Plato’s, Ptolemy’s, or Galen’s, and so is, by Augustine’s lights, a defeasible commitment, albeit a commitment he explicitly rejected in DQA. More importantly, of the five grades of extramissive involvement, only the first three are central to Augustine’s thinking about perception. But only the fourth grade involves a genuine extramissive element, and a full commitment to a pure form of the extramission theory only comes with the fifth.

2.  The Textual Evidence The textual evidence for attributing to Augustine an extramission theory occurs in chapter 23 of DQA. The primary evidence consists in two passages from DQA 23.43, but there is also a back reference to the second passage at the beginning of DQA 23.44. In this section, I shall argue that

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  3 the textual evidence is less conclusive than is commonly supposed. But that the extramissionist reading could not be right is only really established in section 5 where we discuss Augustine’s reply to Evodius’ puzzle. The first passage is as follows: Sight extends itself outward and through the eyes dart forth in every possible direction to light up what we see. (DQA 23.43; Colleran 1949, 66)

As Colleran’s translation indicates, the Latin pronoun is takes as its antecedent visus from the previous line. Visus can be translated as vision or sight, but sight is the appropriate translation as the present passage seems to be describing its actualization. On the extramissionist reading, this passage is describing the emission of the visual ray. The second passage involves the stick analogy that Alexander of Aphrodisias attributes to the Stoics in De anima 130.14: I say that by means of sight, reaching out to that place where you are, I see you where you are. But that I am not there, I admit. Still let us suppose that I were to touch you with a stick: I certainly would be the one doing the touching and I would sense it; yet I would not be there where I touched you.  (DQA 23.43; Colleran 1949, 66)

On the extramissionist reading, seeing by means of the visual ray is ­likened to touching something with a stick. The visual ray, like the stick, is a rectilinear, continuous unity. Let us begin with the first passage. It can seem, and has seemed to some, that it is an unambiguous expression of the extramission theory. How else is one to understand visus from the eyes? (Here, I am deliberately echoing Turnbull 1978, 13–14.) That sight extends itself outwards through the eyes seems like a reasonable description of looking and seeing, at least as a Platonist conceives of it. But the outward activity of looking and seeing is not the exclusive provenance of the extramission theory. The Platonic element of Augustine’s description consists in two things, the first more explicit than the next. First, sight is a power of the soul that is exercised through the use of the eyes. It is an instrument of the soul (DQA 33.41, De Gen. ad lit.12.24.51). That the body is an

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

4  Mark Eli Kalderon instrument of the soul is a distinctively Platonic thought (compare Plato, First Alcibiades 129d, Plotinus, Ennead 4.7.8). Second, Augustine is here emphasizing how seeing is an activity of sight, a power of the soul. The superiority of the incorporeal soul is manifest in its ability to act upon the sensible and corporeal without the sensible and the corporeal being able, in turn, to act upon the soul (De musica 6.5.8–10; see Silva 2014 for discussion). So, it is the soul that acts in seeing and so places itself in the distal body seen (Colleran 1949, 205 n55). The emphasis here is on the activity of the soul as opposed to its being substantially located where the perceived object is. The outward activity of looking and seeing is not the exclusive prov­ en­ance of the extramission theory. One might object that this ignores the illuminationist imagery at the end of the passage—sight darts forth from the eyes and illuminates what it sees. While it is true that the visual ray of the extramission theory is often likened to light (Empedocles, DK 31b84, Plato, Timaeus 45b−c), the illuminationist imagery, considered by itself, is insufficient grounds for the attribution of the extramission theory to Augustine. Other thinkers, who have explicitly rejected thexextramission theory, have coherently embraced this imagery. The illuminationist imagery is undoubtedly of Neoplatonic origin, but ­neither Plotinus nor Porphyry are extramission theorists. Neither are Iamblichus, Proclus, Priscian, nor Pseudo-­Simplicius. In addition, later scholastic thinkers influenced by Augustine employ the Neoplatonic illuminationist imagery while disavowing the extramission theory. Thus, Peter John Olivi compares a perceiver’s gaze to light illuminating its object (Quest. in sec. lib. Sent. q. 72 35–6) even as he denies that seeing involves any real emission (Quest. in sec. lib. Sent. q. 58 ad 14.8). Merleau-­Ponty (1967, 185) will echo both aspects of Olivi’s position. For Merleau-­Ponty, the illuminationist imagery captures the active outward phenomenology of looking and seeing even if the extramission theory provides a false causal model of perception. How might the illuminationist imagery be understood if it is not, indeed, committed to the extramission theory (for discussion see Kalderon 2018, chapter 5)? The awareness afforded by visual experience is like a beam of light that manifests the latent presence of its object. Vision, like illumination, has direction. Light is emitted outward from

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  5 its source upon the scene that it illuminates. Vision too is outer-­directed. In seeing, the perceiver looks out upon the scene before them. Not only do vision and light have direction but also they are both rectilinear. Thus, we speak of lines of sight and do so innocent of any commitment to extramission. Moreover, just as illumination manifests the latent visibility of an object, seeing an illuminated object manifests its latent presence to the perceiver revealing it to be where it is. The explicit awareness of the natural environment afforded by visual experience is akin to light not only in its rectilinear directionality and its power to manifest latent presence, but in the manner in which it discloses distal aspects of that environment. Just as the illumination alights upon the object it il­lu­min­ ates at a distance from its source, the perceiver’s gaze alights upon the object of perception at a distance from their eyes. The imagery here not only emphasizes that vision is a kind of perception at a distance but invokes an active outward extension. The acceptance of the analogy, so explicated, is consistent with no part of the perceiver being substantially located where the perceived object is and so bears no commitment to the active outward extension being specifically spatial. In addition to the phenomenological aptness of the illuminationist imagery, Augustine may have had another motive in deploying it, one that is consistent with the rejection of the extramission theory. Corporeal light may be the means by which we see, but spiritual light is the means by which we understand. The doctrine of illumination was first stated in De magistro written during the same period as the present dialogue (on the doctrine of illumination see Allers  1952 and Matthews  2014). Augustine understands corporeal light as the image, in the Platonic sense, of the true spiritual light. Insofar as perception, like the understanding, involves a mode of awareness, the illuminationist imagery may be the joint result of Augustine’s doctrine of illumination—that spiritual light is that by which we understand—and the claim that perceptual activity is an image of intellectual activity. Finally, a defender of the extramissionist reading might object that we are overlooking the final line of the quoted passage where it is maintained that vision occurs where the object is seen and not at the place where sight goes out from. It is true that if vision occurs through the emission of a visual ray, then this is well explained. However, while the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

6  Mark Eli Kalderon extramission theory may explain this claim, that explanation is not indispensable. Augustine, here, is echoing a Neoplatonic doctrine, that perception places the perceiver in the object perceived, shared by Plotinus and Porphyry, neither of whom were extramissionists. Augustine seems to have in mind, here, not the extramission theory, but the phenomenology of attentive awareness. Perception places us in the perceived body. That is where the perceiver’s attentive awareness is. In selectively attending to an object of your experience, where is your attention? On the object selectively attended to, of course. The query “Where’s your head at?” gives expression to this. Your “head” is where the object of your attention is. In general, an act or episode of attention is where its object is. Attention is not the kind of thing that has location in itself. Attention is conceived by Augustine to be an incorporeal power of the soul to direct the mind to various objects. Insofar as attention can be said to have location, it must inherit this location from the location of its object. This, then, is the claim about the phenomenology of attentive awareness, that attention is located where its object is. One need not be an extramission theorist to endorse it (for further discussion of this point see Kalderon 2018, chapter 6.5). The first passage, while not, by itself, sufficient grounds for attributing to Augustine an extramission theory, may gain new force and significance, however, when considered in context with the second passage. Recall, there, Evodius likens looking and seeing a distal object to touching it with a stick. On the extramissionist reading, this is a description of visual rays extending from the perceiver to the distal object. Like the stick, the visual ray is a continuous unity that spatially extends from the perceiver to the object perceived. Should we accept the extramissionist reading? A number of observations are relevant here. First, Evodius is adapting the Stoic stick analogy (Alexander of Aphrodisias, De anima 130 14) (for recent discussion of the influence of Stoicism on Augustine see Brittain 2002, Byers 2013, and Sellars 2013). The Stoic analogy is criticized by Galen in De Plac. Hipp. et Plat. 7.7, and Tideus in De Speculis, and subsequently echoed by Descartes (1637) in La Dioptrique, though for his own, distinctively modern, purposes. That analogy is plausibly rooted in, and an interpretation of, Plato’s Timaeus 45b−c discussion of perception (see Lindberg 1977, chapter 1). A pri­meval

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  7 fire within the eye, that gives light but does not burn, extends out through the pupil. There it encounters what is like it, external light. And together they constitute a continuous unity, a compound of emitted light and external light, extending from the perceiver to the distant object of perception in a rectilinear path. The stick analogy captures the formal features of the Timaeus account, namely, that the compound of emitted light and external light constitutes a continuous, rectilinear unity—just like a straight stick. However, these formal features are preserved in accounts given by thinkers that explicitly reject the extramission theory. And if that is right, then the stick analogy is not sufficient for the attribution of an extramission theory. Consider, then, one prominent account. Aristotle rejects the extramission theory as providing a false causal model of perception (De sensu 2 438a26–b2), and yet his alternative causal model preserves these formal features. The illuminated media intervening between the perceiver and the distant object of perception already, according to Aristotle, constitutes a continuous unity (De anima 2.7 419a12–22). But the resulting account is broadly intromissionist in that it emphasizes the perceived object acting upon the perceiver, albeit mediately. Specifically, the perceived object immediately acts upon the medium, a continuous unity, which in turn immediately acts upon the perceiver. In this way, the perceived object mediately acts upon the perceiver. Acceptance of the Stoic stick analogy is not sufficient for commitment to the extramission theory. Indeed, Descartes, in La Dioptrique, accepts the Stoic analogy even though his mechanical explanation of vision is in no way extramissive. Nor, indeed, is it necessary. Galen, in De Plac. Hipp. et Plat. 7, rejects the Stoic analogy even as he embraces the extramission theory. Another worry concerns the specific use to which Evodius is putting the Stoic stick analogy. It is commonly accepted that the extramission theory is one way of modeling vision on the basis of touch. Touch may be a contact sense, but distal senses, such as vision, may be modeled on touch if the emitted visual ray extends to the distant object of perception and is in contact with it. A specific form of touch is here accepted as paradigmatic, namely, sensation by contact. But sensation by contact is  not the only form of touch. Thus, Broad (1952) distinguishes a

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

8  Mark Eli Kalderon dy­ nam­ ic­ al form of touch—haptic touch in modern psychological ­parlance—from sensation by contact. But the conception of touch involved in Evodius’ adaption of the Stoic stick analogy is not sensation by contact. Rather, it is a specific and atypical form of haptic touch. It is a form of haptic touch since touching something with a stick—“poking”—is an activity and so displays the dynamical character of haptic touch. What is distinctive about this form of haptic touch is that involves a form of distal touch, where the felt object is not in direct contact with the perceiver’s body. This form of haptic touch is atypical in that most forms of haptic touch involve contact and not all involve an instrument or intervening medium. So, consider feeling the wooden frame through the padding of a Victorian hobby horse. The perceiver feels the wooden frame even though they are not in direct contact with it. Instruments, such as a stick, can be exploited in distal touch. Thus, by means of a stick one may feel the texture of a distant surface, or its hardness and rigidity. One feels the tactile qualities of the distant surface through the stick, despite not being in direct contact with it. The advertised worry turns on two observations. First, Evodius’ appeal to distal touch is in tension, if not inconsistent, with the extramission theory. The extramission theory is motivated by the idea that the perceived object must be in contact with the sense. But, again, the paradigmatic conception of touch is sensation by contact and not ­distal touch. In a way, Augustine, at this point of the dialogue, has elicited a dialectical concession from Evodius (albeit one whose full significance Evodius will only realize in chapter  30). The extended discussion of perception is inaugurated by a puzzle Evodius raises about the incorporeal nature of the soul. If the soul is not extended throughout the body, then how can it feel wherever the body is touched? Evodius’ implicit thought, here—that the perceived object must be in contact with the sense, understood not merely as a sense organ, but that organ as animated by the sensitive soul, the principle of sensation—is now abandoned. One can feel what one is not in direct contact with. It is because Evodius is walking back from a previous commitment, shared with the extramission theory, that this is properly regarded as a dialectical concession.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  9 Second, there is an oddity in Evodius’ description of distal touch. Using an instrument, such as a stick, to feel a distant textured surface, it seems, from within, that we experience that texture at the end of the stick. “As if your nervous system had a sensor out at the tip of the wand,” as Dennett (1993, 47) observes. That is, in cases of distal touch, you feel tactile qualities where they are. But Evodius seems to disavow that: “I certainly would be the one doing the touching and I would sense it; yet I would not be there where I touched you” (DQA 23.43; Colleran 1949, 66). But such a disavowal seems to be in tension with, not only the phe­nom­en­ology of distal touch, but also with the extramission theory as usually understood. If the eye emits a visual ray that extends to the distant object of perception so that it is in contact with it, then at least a part of the perceiver is substantially located where the perceived object is, or is, at the very least, contiguous with it. And it is there, at the point of contact, that the object is sensed. Compare the view that Nemesius attributes to Hipparchus, “Hipparchus says that rays extend from the eyes and with their extremities lay hold on external bodies like the touch of hands” (De natura hominis 7, 104), not to mention Euclid’s third def­in­ition in his Optics, “Those things are seen upon which the visual rays fall; the things upon which the visual rays do not fall are not seen” (Burton 1943, 357). As we shall see in the next section, the dialectical situation is this: Augustine has just put forward in chapter 23 what is to be the main conclusion of the subsequent sequence of chapters culminating in chapter 30, that in seeing the perceiver is affected where they are not. Evodius in chapter 23 tries to resist this conclusion by means of the stick analogy. In chapter 30, after the many important clarifications of the intervening chapters, Evodius comes to accept that conclusion on the basis of Augustine’s argument, first bruited in chapter 23. So Evodius’ response, his use of the stick analogy in chapter 23, could not be Augustine’s considered view, and, hence, could not be a basis for attributing the extramission theory to Augustine or, indeed, anything else. What is more, if at least Dennett is to be believed, it is phenomenologically inept, and, more importantly, inconsistent with the principle driving the extramission theory, and so constitutes a dialectical concession, though one that Evodius himself will not recognize until chapter 30.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

10  Mark Eli Kalderon

3.  Dialectical Context The force of the textual evidence for attributing an extramission theory to Augustine, in DQA, crucially depends upon the dialectical context. Allow me to briefly review the place of Augustine’s account of perception in that dialogue. DQA is mainly charged with the task of arguing for the incorporeal nature of the living soul. In the dialogue, Evodius, like Augustine’s former self (Confessiones 7.1ff), has a hard time conceiving of something that is both real and incorporeal (compare the position of the Giants in Plato’s Sophistes 246a). Throughout DQA, Augustine will give accounts of the soul’s powers and activities that are meant to persuade us that these are not powers and activities of the body. The question of the soul’s incorporeal nature is linked with the question of its magnitude. Bodies are extended in three dimensions. If souls are inextended, if they lack extensive magnitude, then they are incorporeal. But, importantly, being incorporeal is consistent with the soul’s possession of superior virtual magnitude. That is to say that psychic powers, the powers and virtues of the soul, are superior to any corporeal power. Augustine thus distinguishes two senses of magnitude (DQA 3.4): 1.  Extensive magnitudes: magnitudes of extension, “How tall is Hercules?” 2. Virtual magnitudes: magnitudes of power, “How great is Hercules’ valor or prowess?” Evodius seeks an answer to the question in both senses. In the sense of extensive magnitude, Augustine denies that the soul has quantity at all. The soul is inextended, and, hence, incorporeal since corporeal bodies are necessarily extended in three dimensions. Augustine’s denial concerns continuous quantities like magnitude and not discrete quan­ tities like number. In denying that the soul has quantity, Augustine is not denying that there are numerically distinct souls. Augustine will maintain that the soul, while lacking extensive magnitude, nevertheless possesses virtual magnitude. Whereas the question how great is the soul in the sense of extensive magnitudes is answered in the negative in DQA

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  11 3.4, a full answer to the question how great is the soul in the sense of virtual magnitude, in its powers and virtues, only emerges in the ­hier­arch­ic­al­ly organized enumeration of the soul’s powers that ends the dialogue (DQA 33–6; see Brittain 2003 for discussion). This hierarchically organized enumeration of the soul’s powers is also, at the same time, a soteriology, at least in part, in that it describes the soul’s ascent to God (and it is in this sense that it is a theoretical articulation of the vision in Ostia that Augustine shared with Monica as reported in the Confessiones 9.10.24–5). Evodius will resist Augustine’s denial that the soul possesses extensive magnitude. Extensive magnitudes cited by Augustine are length, width, and strength. A third spatial dimension, height, is latter added, DQA 6. Strength here translates robustam which means the resistance offered by solid objects that occupy space. Specifically, then, Evodius doubts whether something without length, width, height, and strength so much as could exist. Here, Evodius is echoing the position of the Giants in the Sophistes. In response, Augustine will offer a negative argument and a positive argument. According to the negative argument, just because the soul lacks extension does not mean that it is not real (DQA 3.4–4.5). And according to the positive argument, the soul must be incorporeal since it possesses powers that bodies lack (DQA 4.6–15.25). In the Sophistes, the Eleatic Visitor convinces the Giants to modify their corporealism in order to allow for justice, since denying the existence of this virtue would be impious. Justice lacks length, width, and height. It cannot be grasped and offers no resistance to touch and hence lacks strength, robustam. And it is by means of the Eleatic Visitor’s argument that Augustine convinces Evodius that lacking extensive magnitude does not entail nonexistence. (It is unclear whether the Platonic influence is direct, but as Augustine had read the books of certain Platonists, Confessiones 7.9.13, possibly the Latin translations of Plotinus and Porphyry by Marius Victorinus, the Platonic influence may well be indirect.) Specifically, a tree, a sensible and corporeal object with extensive magnitude, exists. But so too does justice despite lacking extensive magnitude. Moreover, and importantly, justice is greater in value than the tree. The adaption of the Eleatic Visitor’s argument is meant to establish not only that virtues like justice may exist despite

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

12  Mark Eli Kalderon being inextended but that they may also be more excellent than any extended thing. Justice may lack extensive magnitude and yet possess greater virtual magnitude than any sensible body. Augustine may have established that the soul’s nonexistence does not follow from its lack of extension, but he has yet to establish what the soul positively is. Moreover, just because justice is real despite being inextended, it does not follow that the soul itself is inextended. Evodius is persistently attracted to the idea that the soul extends throughout the body that it animates and hence must itself be extended. Augustine will argue, in contrast, that the soul is inextended because it possesses ­powers that corporeal extended things lack. In effect, Augustine is arguing that the soul possesses greater virtual magnitude than any extended corporeal thing. It is the greatness of the soul that establishes its inextended and hence incorporeal nature. Augustine’s argumentative strategy is to emphasize how an account of the powers and activities of the soul establish that they are not powers and activities of a body. Indeed, so construed, Augustine offers not one positive argument but two: 1. In visually imagining a remembered body, my mental image is not constrained in the way that corporeal images are. Corporeal likenesses are only as large as the body in which the image occurs, incorporeal likenesses are not so constrained. (DQA 5.7–9) 2.  In conceiving of geometrical figures abstracted from three-­ dimensional bodies—such as planes, lines, and points that figure as parts of those bodies—the soul must be incorporeal since only like can conceive of like. (DQA 6.10–15.26) At this point the central case of the dialogue has been made. That is to say, Augustine has argued that just because something is incorporeal does not mean that it is less real or less valuable than something cor­por­ eal. Moreover, Augustine has argued that the soul must be incorporeal because it possesses powers that corporeal extended things lack. And that the incorporeal powers of the soul are either more valuable than their corporeal counterparts or at least more valuable than the corporeal objects of their activity.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  13 Evodius accepts that Augustine has established this. What are raised, at this point, are less objections per se to Augustine’s conclusions than certain residual puzzles or aporiai. There are two: 1. If the soul is inextended, how is it that the soul grows over time as the body grows? (DQA 15.26–22) 2. If the soul is inextended, then it is not extended throughout the body. But if the soul is not extended throughout the body, then how can it feel wherever the body is touched? (DQA 23–30) It is the second residual puzzle that prompts the extended discussion of perception. A tacit assumption is at work in Evodius’ second puzzle. Attending only to sensation by contact, and perhaps by regarding it as an exemplary form of perception, can suggest that the principle governing sensation takes a certain form. Its slogan might be: to be perceptible is to be palpable (see Kalderon  2015 for discussion). The idea is that the perceived object must be in contact with the principle governing sensation. The sensitive soul is the principle governing sensation. Wherever Evodius is touched upon his body he feels it. That means that the prin­ ciple of his sensation, Evodius’ sensitive soul, must be in contact with what touches his body, if it is to be perceptible. But it is perceptible. And so the sensitive soul, at least, must be extended throughout the body. Augustine is sensitive to this tacit assumption even though he does not make it explicit. This sensitivity is manifest in a curious shift in example. Whereas Evodius’ objection turns on an observation about touch, understood as sensation by contact, Augustine develops his def­ in­ition of perception with reference to vision, a form of distal perception. The objects of vision must be at a distance from the perceiver’s eyes. Augustine’s intent is to undermine the tacit assumption driving Evodius to think that the soul must be extended throughout the body if sensation by contact is to be so much as possible. Augustine’s intent is to undermine the assumption that to be perceptible is to be palpable, that the perceived object must be in contact with the principle governing sensation. This assumption not only drives Evodius’ puzzle, but the extramission theory as well.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

14  Mark Eli Kalderon

4.  The Definition When asked by Augustine what sense perception is, Evodius, good Socratic stooge that he is, merely responds with a list, the Peripatetic five senses—vision, audition, olfaction, taste, and touch. Augustine explains that they are presently seeking a single definition that would encompass all five senses (compare, for example, Plato, Theaetetus 145e–147c) and proposes a candidate definition for Evodius to defend or reject. This definition will subsequently be refined, though it will retain its basic form. It is worth, however, discussing the initial formulation of the definition: Perception, I think, is what affects the body not being hidden from the soul.  (Augustine, DQA, 23.41)

Augustine’s provisional definition comes in two parts: quod patitur corpus describes the object of perception whereas non latere animam describes the soul’s relation to that object. Consider the soul’s relation to the object of perception first. And bracket, for the moment, the significance of Augustine’s indirect description of that relation. In describing the object of perception as being not hidden from the soul, Augustine is defining perception, in the first instance, as a mode of awareness (Brittain 2002, 275). What is the significance of Augustine’s indirect description of the soul’s relation to the object of perception? Why describe the object of sensory awareness as something that is not hidden from the soul? Some have found this worrying. Bourke (1947, 112) flatly pronounces Augustine’s definition “unsatisfactory” on just this basis: “The persistent use of the negative formula (non latere) indicates Augustine’s inability to say positively just what sensation is.” And Bourke’s judgment is reaffirmed by McMahon (1947, 104 n1; Gannon 1956, 167–8, raises a similar worry). However, I wonder to what extent Bourke’s judgment is fair. If we accept, as seems evident, that Augustine had in mind a mode of awareness, then perhaps to describe the object of perception as not hidden from the soul is, after all, to provide a positive characterization of the sensory awareness afforded by perceptual experience.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  15 Etienne Gilson thought so: His purpose is to make sensation an activity of the soul within the soul itself. This is really the reason why he defines it in such a roundabout way. The phrase “non latet” indicates precisely that the soul is a spiritual force, ever watchful and attentive. In order to sense, it does not have to receive anything from the organ that it vivifies; it is enough if the changes undergone by the organs do not escape its notice, and come within the range of its attention.  (Gilson 1961, 63)

Notice how Gilson understands something not being hidden from the soul as the effect of the soul’s attentive vigilance. The positive char­ac­ter­ iza­tion that Gilson sees in Augustine’s roundabout expression is metaphysically significant. That something is not hidden from the range of the soul’s attention is due to the activity of the soul within the soul itself. I believe that Gilson was right to see in Augustine’s roundabout expression a positive characterization of sensory awareness. Consistent with Gilson’s suggestion, one might also understand Augustine as claiming that sensory awareness is a mode of disclosure. To describe the object of perception as not hidden from the soul is to understand sensory awareness not only as the soul’s attentive vigilance but also as a mode of disclosure—an activity whereby what was previously hidden from the range of its attention is now revealed to the soul. Moreover, this indirect description of the awareness involved in perception, non latere, has Platonic roots, μή λανθνειν, and Augustine may have picked it up from his encounter with the books of the Platonists (Confessiones 7.9.13). It can be found in Philebus 33d and 43c and in Timaeus 64a–c. This usage is picked up by later Platonists. It can be found, for example, in Plotinus’ Enneads 4.4.18–19. So, the negative formulation is not due to Augustine’s inability to say positively what sensation is, as Bourke contends, rather he is picking up on a Platonic gloss on awareness (see O’Daly 1987, 86 n15 and Brittain 2002, 275 n61 for further discussion). Consider now Augustine’s description of what is not hidden from the soul, the object of sensory awareness. Brittain (2002, 274–8) claims that there is a crucial ambiguity in quod patitur corpus. The verb patior

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

16  Mark Eli Kalderon means to suffer or undergo, to be affected. So, a natural understanding of this phrase might be what the body undergoes. Here corpus is understood as the subject of the verb patior. What the body undergoes is a bodily affection, the way in which the body is affected. However, quod patitur might also be read as Latin rendering of the Greek ὅ τι πάσχει. So understood, quod is nominative, not accusative, and is the subject of the verb patior. On this reading, the object of sensory awareness is less a bodily affection than what affects the body. What affects the body is the causal agent of the bodily affection, whereas what the body undergoes is the bodily affection elicited by that cause. On the former understanding, the object of sensory awareness is internal—it is the perceiver’s body being affected in a certain way. On the latter understanding, the object of sensory awareness is external, at least if we rule out cases of auto-­ affection—it is what affects the body from without. Whether or not quod patitur corpus is ambiguous in the way that Brittain claims it to be, understanding that phrase as what affects the body better coheres with Augustine’s own examples of the objects of perception in DQA. The objects of vision, for example, are external bodies located at a distance from the perceiver. Thus, Augustine sees Evodius, and Evodius sees Augustine, and neither is at the place where the other is. The objects of perception are external to the perceiver’s body. In modern psychological parlance, perception is exteroceptive. And this would remain true even if the sensory disclosure of external bodies involves the perceiver’s sensory organs being affected, or even the formation of incorporeal images in the soul occasioned by such affections. Understanding quod patitur corpus as bodily affection yields a definition of what is, at best, bodily sensation. It at best characterizes a form of interoceptive awareness. Interoceptive awareness contrasts with exteroceptive awareness in that it is an awareness, not of what is external to the body, but of the body itself. It is only on the understanding of quod patitur corpus as what affects the body—understood as the external causal agent of the bodily affection—that it plausibly yields a def­in­ ition of perception. Only so understood does the definition characterize a form of exteroceptive awareness. After all, what affects the body, the objects of perception, are external to that body, and the objects of visual

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  17 perception are located at a distance from the perceiver’s body. Moreover, as we have emphasized (section  3), it is dialectically important that Augustine is discussing vision, a distal sense. If perception fundamentally involves the soul’s sensory awareness understood as a mode of disclosure, and if the objects not hidden from the soul are external bodies, then the emphasis of the expression quod patitur corpus is on the fact that the soul’s sensory awareness of external bodies is mediated by the body’s affection. In this way perception contrasts with understanding. The soul’s awareness of the intelligible is not mediated by the body’s affection the way that the soul’s sensory awareness is. Augustine explicitly distinguishes perception (sensus) and the awareness that it involves from knowledge (scientia) and the awareness that it involves (DQA 29). Summarizing that discussion Augustine opens DQA 30 with the claim that perception is one thing, knowledge is another (aliud sensus, aliud scientia), even though not being hidden from the soul is a common element of both. However, what is not hidden from the soul differs in these cases in whether what is disclosed is made apparent through the frame of the body or through pure intelligence (sive per temperationem corporis, sive per intelligentiae puritatem). This clearly shows that quod patitur corpus or passio corporis are used to specify the relevant kind of awareness involved in perception, a kind of awareness disclosed through the body’s affection, a kind of awareness that Augustine explicitly contrasts with the awareness involved in knowledge. More refined versions of the preliminary definition will be given in DQA 25.48, 26.49, and 30.59. There are two salient differences between the preliminary definition and the refined versions. First, quod patitur corpus has been replaced by passio corporis (or, equivalently, corporis passio—the variation in word order makes no difference to the Latin grammar). Passio corporis is most naturally read, not as what affects the body, but the body’s affection, what the body undergoes when being affected from without. However, as we observed before, so understood, the definition does not define perception but bodily sensation. But it is clear and important that Augustine is not only discussing perception but a form of distal perception, vision. So passio

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

18  Mark Eli Kalderon corporis is a kind of Augustinian shorthand for quod patitur corpus that is meant to inherit its sense. So, this first difference is superficial and does not represent a genuine refinement of the definition. Second, each of the refined versions include a new phrase, per seipsam, which can be translated as “by itself ” or “directly.” Unlike the substitution of passio corporis for quod patitur corpus, this is not a superficial difference but does represent a genuine refinement of the definition. How are we to understand this refinement and what is the nature of its grounds? The refinement of the preliminary definition was prompted by a counterexample to its sufficiency. Growth is an affection of the body not hidden from the soul and yet it is insensible. According to Augustine, such failures, where the definition encompasses more than it undertook to explain, can sometimes be fixed with an emendation (DQA 25.47). Thus, Augustine’s challenge to Evodius (DQA 25.48): What qualification can be added to the definition to render it valid? Evodius’ answer (picking up on an earlier claim of Augustine’s, DQA 24.46, whose significance he did not, at that time, understand) is per seipsam. What is the force of the qualification? Growth is an affection of the body not hidden from the soul and yet growth is insensible. That is why it represented a failure of sufficiency of the initial formulation of the definition. However, it is not true that growth by itself is not hidden from the soul. Growth only falls within the range of the soul’s attention by the operation of reason and intellect. In contrast, the perceived object affecting the eyes is by itself not hidden from the soul. The structure of Augustine’s definition is metaphysically telling. The distal object acts upon the animated sense organ occasioning the sensitive soul to direct its attentive awareness to what affects that organ. First, notice that it is only the sense organ, and not the sensitive soul itself, that is affected by the object of perception. Plausibly, this is an instance of the prohibition on ontological inferiors acting upon ontological su­per­iors. The sensible and the corporeal may act upon the sense organ, but not the sensitive soul that animates it (De musica 6.5.8–10; see Silva 2014 for discussion). Moreover, as we shall see in the next section, the sense organ is only affected in the way that it is because it is animated by the sensitive soul (compare Priscian, Metaphrasis 2.1–2). Though the object of perception does not act upon the sensitive soul, it is not hidden from the range of the soul’s attention. That something is

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  19 not hidden from the range of the soul’s attention is due to the activity of the soul within the soul itself. Not only will Augustine criticize the preliminary definition to mo­tiv­ ate its refinement, but he will also clarify a number of issues that might stand in the way of accepting the definition. These tasks will be pursued in chapters 24 to 29. It is only when we have the refined definition, and the obstacles to accepting it have been eliminated, that Augustine applies that definition, in chapter 30, to answer Evodius’ challenge to the in­cor­ por­eality of the soul—to explain how it is that the soul feels wherever the body is touched if it is not extended throughout the body.

5. Augustine’s Counterargument Augustine’s counterargument begins with a restatement of the refined definition whose validity has been accepted by Evodius and Augustine (DQA 25.48, 26.49, and 30.59): 1. Perception is what affects the body by itself not being hidden from the soul. From the accepted definition it immediately follows that: 2.  If one perceives, one’s body is acted upon. Moreover, it was earlier (DQA 23.42) agreed that: 3.  Seeing is a form of perception. But seeing is in some ways a distinctive form of perception. It is a mode of distal perception. Importantly, it was also agreed that: 4.  A perceiver sees an object where the perceiver is not. If sight were confined to where the perceiver is at, at best the eye alone could be seen. But the objects of sight are located at a distance from the perceiver. So, the perceiver sees an object where they are not.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

20  Mark Eli Kalderon From these premises a startling conclusion follows. Seeing is a form of perception. If one perceives, one is acted upon. So, in seeing one is acted upon. And since what acts upon one when one sees is located where one is not, in seeing one is acted upon where one is not. The conclusion involves attributing to the animate eye a passive power to be affected where it is not. In Dutton’s (2020) apt phrase, the eye is subject to passion-­at-­a-­distance (Dutton’s essay is an excellent discussion of the present argument). This is a passive power not shared with soulless natural bodies. Inanimate natural bodies can only be acted upon by what is in contact with them. The living eye, part of a whole and healthy animal, is animated by the sensitive soul, and it is a manifestation of the su­per­ ior­ity of the soul that it endows the eye that it animates with the passive power to be affected where it is not. The sensitive soul may lack extensive magnitude, but it has, in this way, great virtual magnitude. This nicely fits the pattern we observed in Augustine’s positive argument for the inextended nature of the soul. This involved attributing superior powers to the soul not possessed by extended corporeal things. Thus, while corporeal images are limited by the size of the body upon which they are inscribed, the soul’s power to recall an image of an object previously seen is not subject to this limitation. Similarly, the soul possesses the power to conceive of incorporeal geometrical abstractions, and since only like may conceive of like, the soul itself must itself be incorporeal. It thus possesses a power that no corporeal thing may have. The passive power to be affected where one is not conferred by the sensitive soul is similarly a power that no soulless body may possess. Further testimony to the greatness of soul, conceived, not as greatness of extent but greatness of power or virtue. A commitment to the animate eye possessing the passive power to be affected where it is not is one that Augustine shares with thinkers who explain the eye’s affection in terms of sympathy (Cleomedes Meteōra; Plotinus Ennead 4.4–6; Porphyry Ad Gaurum; Epicurus, Letter to Herodotus also invokes sympathy in his account of perception, but not to explain action at a distance, but rather to explain the unity of the streams of eidola, see Lee 1978). The sympathetic affection of the animate eye is the result of action at a distance, and so the animate eye

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  21 possesses a passive power to be affected where it is not. In contrast with the Stoics, Epicureans, and Neoplatonists, Augustine does not appeal to sympathy as an explanatory principle, nor indeed does he explain the passive power in terms of any other principle (for discussion see Gannon 1956, 168 n53). We are merely meant to marvel at this power that the sensitive soul endows the eye with, a power superior to any ­passive power of a soulless natural body. How does this attribution of a passive power possessed by no purely corporeal thing bear on Evodius’ puzzle? Recall the tacit assumption behind Evodius’ use of sensation by contact—that the object of sensation must be in contact with its principle, that by which one senses. It is just this assumption that drives Evodius conviction that the soul must be extended throughout the body. For only if it were could it be in contact with what touches the body. But if the sensitive soul confers the passive power to be affected where one is not, then there is no need for the sensitive soul to extend throughout the body. The soul need not be where the body is affected, at least substantially, for this affection to be not hidden from the range of its attention (DQA 30.59). Though the writing has been on the wall since chapter 23, Evodius is stunned: That conclusion upsets me very much, so much, in fact, that I am completely stunned. I do not know what to answer and I do not know where I am. What shall I say? Shall I say that a bodily [affection] of which the soul is aware directly is not sensation? What is it, then, if it is not that? Shall I say that the eyes [are affected by] nothing when we see? That is most absurd. Shall I say that the eyes [are affected] where they are? But they do not see themselves and nothing is where they are, except themselves. Shall I say that the soul is not more powerful than the eyes, when the soul is the very power of the eyes? Nothing is more unreasonable. Or must this be said, that it is a sign of greater power to [be affected] there where something is than to [be affected by] it where it is not? But, if that were true, sight would not be rated higher than the other senses. (DQA 30.60; Colleran 1949, 87, with modifications)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

22  Mark Eli Kalderon The reason that corporeal things lack the passive power to be affected where they are not is because, being corporeal and extended, they are confined to the place where they are. The soul, being inextended and not so confined, may confer this passive power on a body that it animates. Augustine thus explicitly rejects the principle that drove Evodius’ aporia—to be perceptible is to be palpable to sense, that the perceived object must be in contact with the principle governing sensation. But this is the principle that drives extramission theories as well. Thus, Augustine’s counterargument is an anti-­extramission argument. A failure to recognize this, and so misattribute to Augustine a commitment to extramission in DQA, is due, at least in part, to unclarity about the commitments of the extramission theory.

6.  The Grades of Extramissive Involvement It is worth getting clear about what, exactly, it means to describe perception as extramissive. In doing so, however, we should bear in mind that the distinction between extramission and intromission theory arises in the historiography of perception, and that theories classified by his­tor­ ians as extramissive may have competing theoretical goals. Thus Lindberg (1978) distinguishes two forms of extramission theories (conspicuously leaving open that there are others). The goal of the first form of extramission theory was to provide a mathematically coherent account of perspective, refraction, reflection, and so on. This is most clearly exhibited in the tradition of geometrical optics which included Euclid, Hero, Ptolemy, and Al-­Kindi. The goal of the second form of extramission theory was not to provide a mathematically coherent account of these matters, so much as to provide a physiological and psychological explanation of vision. Galen’s account of vision in De Plac. Hipp. et Plat. is a good example of this second form of extramission theory. Augustine’s ambitions in writing about perception are more in line with Galen’s than Euclid’s and so we will focus on the second form of extramission theory. (On the influence of Galen’s physiology on Augustine’s account of perception see O’Daly 1987, chapter 3.1.)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  23 On the basis of our discussion so far, we are in a position to usefully distinguish different grades of extramissive involvement (the allusion to Quine 1953 is deliberate; for a similar approach to the commitments of the extramission theory see Lička  2020). The grades are meant to be applied successively, so, for example, a commitment to the fifth and highest grade involves a commitment to the previous four grades. It is only the fourth grade that introduces a genuine extramissive element, but it is only the fifth and highest grade that incurs a full commitment to a pure form of extramission: 1.  Perception must at least centrally involve the activity of the perceiver; 2. This activity is outer directed—in the case of vision, this outer-­ directed activity is rectilinear; 3. This outer-­directed activity of the perceiver constitutes, at least in part, their perception; 4. This outer-­directed activity that constitutes, at least in part, the perception of an object involves something spatially extending to the distal object of perception—in the case of vision, along a rectilinear path—so that at least part of the perceiver is substantially located where the perceived object is or is, at the very least, contiguous with it; 5. To be perceptible is to be palpable, the perceived object must be in contact with the principle governing sensation, a principle spatially extended from the perceiver to the perceived object. As we shall see, only the first three grades are central to Augustine’s thinking about the nature of perception. The first grade is far too weak to incur genuine commitment to extramission. Perception is not even identified with activity, let alone extramissive activity, but is only claimed to centrally involve activity. Many thinkers accept that perception at least centrally involves activity without accepting, as well, the extramission theory. The first two grades, considered jointly, are insufficient for extramission. Taken together they are equivalent to the claim that vision centrally involves rectilinear, outer-­directed activity. So understood, they might

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

24  Mark Eli Kalderon reasonably be taken to jointly describe looking and seeing. It is not implausible to think that in order to visually perceive an external scene, the perceiver must look at that scene, where looking involves directing one’s visual awareness to that scene. Looking, so conceived, is an outer-­ directed activity of the perceiver that is rectilinear. It determines a line of sight. Plausible though it may be, the principle—to see, one must look—is a substantive claim that not all may endorse. And yet it falls short of the extramission theory. Notice that the first two grades jointly capture Augustine’s claim when he writes “Sight extends itself outward and through the eyes darts forth far in every possible direction to light up what we see” (DQA 23.43; Colleran 1949, 66). But as these fall short of the extramission theory, this passage is insufficient grounds to attribute an extramission theory to Augustine as I have argued in section  2. Notice, as well, that in determining lines of sight the first two grades form a sufficient basis for geometrical optics of the kind developed by Euclid, Hero, Ptolemy, and Al-­Kindi. The third grade introduces a further substantive commitment. One may accept the principle that to see one must look, and yet deny that looking constitutes, even in part, seeing. So, the third grade is a further commitment. But even the three grades taken together are insufficient for a commitment to extramission. Contemporary enactivists, for ex­ample, such as Noë (2004), accept something like the first three grades, but enactivism is not a species of extramission. The fourth grade of extramissive involvement incurs a genuine extramissive element but falls short of a full commitment to the extramission theory. Only when the outer-­directed activity of the perceiver that constitutes, at least in part, their perception of the object is conceived as something spatially extending to the distal object so that it is in contact with that object do we get a genuine extramissive element but one that arguably falls short of a full commitment to a pure form of the extramission theory. (It may be sufficient if the goal of the extramission theory is to provide a mathematically coherent account of vision in the tradition of geometrical optics, but, recall, like Galen and Augustine, we are focusing on physiological and psychological explanations of vision.) So, consider the following. Francisco Suarez (Comm. una cum quest. in lib. Arist. de anima 3.17.1) marks a distinction between accounts

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  25 where vision occurs at the point where the ray meets the perceived object and accounts where the object’s species must somehow be reflected back to the eye first. Suarez cites Galen, in De Plac. Hipp. et Plat. 7, as attributing the latter kind of account to Plato (see, as well, Theophrastus, De sensibus 1.5 for a similar reading of Plato). The latter accounts, while involving extramissive elements, seem more aptly deemed interactionist, as vision is the result of the interaction of the perceiver’s activity and the activity of the object perceived (Smith 1996, 22–3; Ierodiakonou 2005; Remes 2014; Squire 2016; Kalderon 2018, 13). Such accounts not only involve extramissive elements but intromissive elements as well. Moreover, if the species must somehow be reflected back to the eye before vision occurs, then the extramissive activity is merely a precondition for intromission. So, the latter accounts are not purely extramissive. Indeed, they are fundamentally intromissive. The principle governing sensation lies within the perceiver which is why the effect of the perceived object must be transmitted within along the extended ray. Only the former accounts count as purely extramissive. If that is right, then the fourth grade may be accepted without a full commitment to the extramission theory. Consider, then, the former, purely extramissive, accounts where vision occurs at the point where the ray meets the perceived object. Perception is understood to be at least modeled on, if not a form of, sensation by contact. Its principle is: to be perceptible is to be palpable. It is the fifth and highest grade of extramissive involvement that generates the requirement that at least a part of the perceiver spatially extend to the distal object of perception. The principle governing sensation must be extended along with the ray so that it may be in direct contact with the object perceived. Only in this way could that object be palpable to perception and so perceived. On the latter interactionist accounts, perhaps the effect of the perceived object must reach the principle governing sensation lying within. But the fifth grade requires contact with the perceived object—which, in the case of vision, is at a distance from the perceiver—and not merely contact with its effect. Again, if something extends from the perceiver so that the effect of the perceived object may be transmitted within, the extramissive element is merely a precondition for intromission. So, it is the fifth and highest grade of

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

26  Mark Eli Kalderon extramissive involvement that incurs a full commitment to a pure extramission theory. Unclarity about the commitments of the extramission theory is aided and abetted, in certain circumstances, by the application of a certain methodological stricture. Some modern commentators have marked a distinction between the psychological and the physiological claims that Augustine makes. Moreover, there is a subsequent tendency to interpret Augustine’s commitment to extramission as a physiological claim about vision. So understood, the extramission theory is simply a false causal model of distal perception and may be dismissed as a piece of antiquated physiology. Allow me to make two observations about this. First, in discussions of the soul in late antiquity, psychological and physiological issues are intertwined, which is not to say confused. Consideration of the psychological is not so easily separated from the physiological. Second, and more fundamentally, there is more to the extramission theory than a false causal model. The extramission theory, in its fifth and highest grade, essentially involves a psychological claim, that the perceived object must be in contact with the principle governing sensation. And the first three grades of extramissive involvement make important claims about the active, outer-­directed phenomenology of vision. Thus, the psychologists Winer and Cottrell (1996) hypothesize that the ­tendency for extramission beliefs to persist into adulthood and their resistance to experimental intervention are partly explained by a phenomenological truth enshrined in extramission models: We assume that core aspects of the phenomenology of vision underlie extramission interpretations. Consider one phenomenologically salient aspect of vision, namely, its orientational or outer-­directed quality. When people see, they are generally oriented toward an external visual referent, that is, they direct their eyes and attention to an object in order to see it. In fact, this quality of vision is reflected in language. People talk about “looking at” things, and English has expressions such as “looking out of a window” and “looking out of binoculars . . .” (Winer and Cottrell 1996, 140)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  27 Notice that the phenomenological grounds that Winer and Cottrell ­postulate for the persistence of extramission beliefs are enshrined in the first three grades of extramission. It would be a mistake to dismiss the extramission theory merely on the grounds of being an antiquated physi­ology. The extramission theory essentially involves a principle governing sensation and makes important claims about the phe­nom­en­ ology of vision.

7.  Extramission in the Augustinian Corpus In this final section we shall briefly review the strength of evidence for attributing an extramission theory to Augustine on the basis of his other works. At best, it is neither required by scripture nor reason, but is authoritative opinion, and so a defeasible commitment, by Augustine’s lights. At times he complains of its subtlety and obscurity, emphasizing its failure to be clearly demonstrated. Many passages involving extramissive elements are, in fact, making a point about something other than perception. And very often this point is independent of the truth of the extramission theory. And sometimes Augustine will pursue his point even when it proves to be in tension with the extramission theory. I conclude that the extramission theory is not a central element in Augustine’s thinking about perception. That he was nonetheless persistently drawn to it is explicable. And not just because it was the authoritative opinion of Plato, Ptolemy, and Galen, at least as traditionally understood (we have entertained doubts whether Plato’s account is purely extramissive). Rather, perception is the soul’s activity using the eye as an instrument. Augustine understands this in a way that only commits him to the first three grades of extramissive involvement. On occasion Augustine flirts with the fourth grade of extramission. Where he does so, he thus abandons or at least brackets his early anti-­extramission argument of DQA. If the eye emits a ray that extends to the object seen, then it need not have the passive power to be affected where it is not. When Augustine seemingly accepts the fourth grade, he does so defeasibly, in deference to authoritative opinion, and he never commits himself to the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

28  Mark Eli Kalderon fifth and highest grade of extramissive involvement. What is central to and persistent in Augustine’s thinking about perception is the first three grades of extramissive involvement, but, again, only the fourth involves a genuine extramissive element, and only the fifth incurs a full commitment to a pure form of the extramission theory.

a.  De musica In De musica, Augustine compares the way in which the activity of memory comprehends the temporally distant to the way in which the activity of sight, as the extramission theory conceives of it, comprehends the spatially distant: Then, as the diffusion of rays shining out into the open from tiny pupils of the eye, and belonging therefore to our body, in such a way that, although the things we see are placed at a distance, they are yet quickened by the soul, so, just as we are helped by their effusion in comprehending place-­spans, the memory too, because it is somehow the light of time-­spans, so far comprehends these time-­spans as in its own way it too can be projected.  (De musica 6.8.21; tr. Taliaferro, 346)

Augustine’s point here is not about perception but memory, that it is somehow the light of time-­spans that discloses the temporally distant. That point can be made, and made as Augustine does by analogy with the extramission theory of vision, without commitment to the truth of the extramission theory. Indeed, Augustine’s doctrine of illumination is more directly relevant to understanding the way in which memory is the light of time-­ spans that discloses the temporally distant than the extramission theory (on the doctrine of illumination see Allers  1952 and Matthews 2014).

b.  De Gen. ad lit. In De Gen. ad lit., Augustine provides a phenomenology of focal attention that makes use of extramissive elements:

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  29 The shaft of rays from our eyes, to be sure, is a shaft of light. It can be pulled in when we focus on what is near our eyes and sent forth when we fix on objects at a distance. But when it is pulled in, it does not altogether stop seeing distant objects, although, of course, it sees them more obscurely than when it focuses its gaze upon them. Nevertheless, the light which is in the eye, according to authoritative opinion, is so slight that without the help of light from outside we should be able to see nothing. Since, moreover, it cannot be distinguished from the outside light, it is difficult as I have said, to find an analogy by which we might demonstrate the diffusion of light to make the day and a contraction to make the night.  (De Gen. ad lit. 1.16; tr. Taylor, 37–8)

The present extramissionist account is derived neither from scripture nor reason, but is accepted as received authoritative opinion. It is potentially revisable in the way that the deliveries of scripture and reason are not. Especially since the authoritative opinion is no mere record of observation. The authoritative opinion includes Plato’s explanation, in the Timaeus 45b−c, of the necessity of light for seeing, that the light emitted from the eye must be supplemented by external light in order for the perceiver to see. This has the consequence that the emitted light cannot be distinguished from the external light and so is not directly observable. A tension in the account raises a potential difficulty. Extramission theories are motivated by an apparent need to be in contact with distant sense objects if the perceiver is to be aware of them. But Augustine explicitly denies this in this passage “when it is pulled in, it does not altogether stop seeing distant objects, although, of course, it sees them more obscurely than when it focuses its gaze upon them.” The visual ray need not be in contact with the distal object in order to perceive it, but it does so less clearly than if it were. The passage provides a phe­nom­en­ ology of focal attention that seems to conflict with the requirement that the principle of sensation be in contact with the objects of sensory awareness. The present account is usefully compared to the extramissionist account that Nemesius attributes to the “geometricians” in De natura hominis 7. We get a similar description of the phenomenology of focal attention, also set within an extramissionist account, but where the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

30  Mark Eli Kalderon binocular character of vision is made explicit and exploited in an e­ xplanation for how we can see many things all at once. For consider a ray sent from a single eye. It would be natural to expect that it sees only that with which it is in contact. But what we see is not restricted in this way. Reflection on binocular vision provides an explanation. The rays from both eyes form a cone. Where the rays intersect is the point of focal attention where things appear exactly in a way that is meant to be consistent with many other things appearing as well if not exactly. However, even objects outside of the cone are in contact with visual rays emitted from at least one eye. So an extramissionist account of binocular vision would resolve the difficulty raised by the account in De Gen. ad lit. However, beyond speaking of the eyes in the plural there is no explicit discussion of binocular vision in that work. And though Augustine’s claims about focal attention generated the tension with the pure form of the extramission theory, Augustine does not acknowledge this tension, let alone consider its resolution by means of an extramission theory elaborated to account for binocular vision.

c.  Sermon 277 In 411 CE, on the birthday of the martyr Vincent, Augustine delivered the following as part of a sermon: In this very body, which we carry around with us, I can find something whose inexpressible swiftness astonishes me; the ray from our eye, with which we touch whatever we behold. What you see, after all, is what you touch with the ray from your eye. (Sermon 277.10; Hill 1994, 38)

Augustine is using the familiar tactile metaphor associated with extramission theories. But what is it a metaphor for? Arguably, touch is a metaphor for the presentation in sight of the object of vision. That we touch whatever we behold may be too weak, by itself, to establish that, but it is combined with the claim that what you see is what you touch

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  31 with the visual ray. On this reading, visual presentation is either reduced to or is at the very least modeled on tactile presentation. If that is right, then Hill (1994, 46 n17) is wrong to speculate that “presumably on meeting a visible object send back the message to the subject . . . or perhaps they bounce straight back to the eye like radar.” Recall Suarez’s contrast (section  6) between accounts where vision occurs at the point where the ray meets the perceived object and accounts where the object’s species must somehow be reflected back to the eye first. The rays, as Hill conceives of them, are merely part of the causal medium through which information about the perceived object is conveyed. But if the visual rays touch the objects of perception, then they are perceived where they are, and thus there is no need for a signal to return to the subject. Augustine explains occlusion as the obstruction of visual rays. This could not be an argument for the extramission theory, as occlusion is equally well explained on the intromissionist hypothesis. Rather, the example of a man obscuring a distant column is setting up the real topic of Sermon 277.10, the “inexpressible swiftness” of the visual rays, which will lead Augustine to an interpretation of Paul’s phrase “in the twink­ ling of an eye” (1 Corinthians 15:52) in Sermon 277.11, understood as the speed at which the body will be resurrected. Like the light of time-­ spans in De musica, Augustine’s point about the interpretation of the Pauline expression could be made without commitment to the extramission theory. Augustine speaks of the “inexpressible swiftness” of the visual rays, but strictly speaking their action is instantaneous. If two objects, one near (a man) and one far (a column) are visible to the perceiver in the circumstances of perception (and hence the man no longer occludes the column), then it is not the case that the visual rays reach the near object sooner than the far: You don’t get to him sooner and to it later; and here he is, nearby, and it’s a long way off. If you wanted to walk, you would get to the man sooner than to the column; because you wanted to see, you have got to the column as soon as the man. And yet, as soon as you open your

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

32  Mark Eli Kalderon eyes, lo and behold, you yourself are here, your ray is there. As soon as you wanted to see it, you reached it by seeing it. . . . Just opening your eyes constitutes reaching it.  (Sermon 277.10; Hill 1994, 39)

Notice that the “inexpressible swiftness” of the visual ray is contrasted with the speed of corporeal activities such as walking. Perhaps, like Philoponus, Augustine maintains that instantaneous action at a distance is only possible for incorporeal activity (In Arist. De anima lib. 325.1–341.9). After all, instantaneous action at a distance would require a body to travel at infinite speed, but bodies, no matter how swift, only travel at finite speeds. But if the “inexpressible swiftness” of the visual ray is contrasted with the speed of corporeal activities, the activity of the ray is plausibly spiritual and hence incorporeal. If so, nothing corporeal extends to the perceived object. And if only the corporeal admits of extension, nothing is substantially located where the perceived object is. Galen’s version of the extramission theory addresses this issue in a different way by denying that what travels with “inexpressible swiftness” is a body (thus also resolving some of Aristotle’s objections in De sensu 2). But neither is it an incorporeal activity, as Philoponus and perhaps Augustine maintain. Rather it is an effect of the emitted pneuma, a quality immediately instantiated by the external light. Like Aristotle, Galen maintains that the external light is already a continuous unity. The “inexpressible swiftness” would consist in the quality being instantiated, all at once, by that continuous unity (De Plac. Hipp. et Plat. 7.4–5). There is no evidence that Augustine entertained, let alone endorsed, the Galenic alternative. There are, then, two claims to be made about Sermon 277. First, like the De musica passage, the point being made by analogy with the extramission theory is independent of the truth of the extramission theory. Second, since the “inexpressible swiftness” of the visual ray is really instantaneous, the visual ray could only be a spiritual, incorporeal activity, with the result that nothing is substantially located where the perceived object is which is, at the very least, inconsistent with the fifth and highest grade of extramissive involvement.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  33

d.  De Trin. In De Trin., Augustine writes: For the mind does not know other minds and not know itself, as the eye of the body sees other eyes and does not see itself; for we see bodies through the eyes of the body, because, unless we are looking into a mirror, we cannot refract and reflect the rays themselves which shine forth through the eyes, and touch whatever we discern—a subject, indeed, which is treated of most subtly and obscurely, until it be clearly demonstrated whether the fact be so, or whether it be not. But whatever is the nature of the power by which we discern through the eyes, certainly, whether it be rays or anything else, we cannot discern with the eyes that power itself; but we inquire into it with the mind, and if possible, understand this with the mind. (De Trin. 9.3; tr. Haddan, 226)

Again, the present extramissionist account is derived neither from scripture nor reason. Nor is it, in this instance, even accepted as received authoritative opinion. Notice that Augustine, after having introduced the extramissionist imagery of rays, immediately brackets that commitment, claiming that it is treated subtly and obscurely and claims that the explanation of perceptual discernment by rays has not yet been clearly demonstrated. (This reinforces the doubt raised in subsection 7.2 since the composition of De Trin. overlaps the composition of De Gen. ad lit.) The central point of this passage is independent of the truth of the extramission theory. Moreover, it echoes a Neoplatonic theme. We get a contrast between the activity of the mind and the activity of the animated eye. Whereas the mind is spiritual, the animated eye is a compound of the corporeal and the spiritual. The animated eye is part of the living being, the compound of the organ and the sensitive soul that animates that organ and uses it as an instrument for the soul’s activities. The mind may apprehend itself in thought in the way that the power of sight acting through the eye could not apprehend itself in vision. In Neoplatonic vocabulary, the mind’s activity, being purely spiritual, is

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

34  Mark Eli Kalderon capable of reverting upon itself the way that the visual activity of the animated eye, being a compound of the corporeal and the spiritual, could not (compare Proclus’ demonstration of proposition 15 of Element. Theol.: “All that is capable of reverting upon itself is in­cor­por­eal” and Aquinas’ Super Lib. de Causis exp. propositions 7 and 15). Augustine’s point about the failure of sight’s activity to revert upon itself does not depend upon the truth of extramission. Augustine’s point is, in that sense, independent of the truth of the extramission theory.

8. Conclusion The extramission theory is not a central element in Augustine’s thinking about perception. That he was nonetheless persistently drawn to it is explicable. And not just because it was the authoritative opinion of Plato, Ptolemy, and Galen, at least as traditionally understood. Rather, perception is the soul’s activity using the eye as an instrument. Perception is not something done to the perceiver; it is the soul, through the use of the body, that perceives. Sensory awareness is an activity of the soul within the soul itself. According to Augustine, vision involves outer-­directed, rectilinear activity that constitutes the perception of the object. Augustine is committed to the first three grades of extramissive involvement. But a genuine extramissive element only occurs in fourth grade of extramissive involvement, and a full commitment to a pure form of extramission is only incurred with the acceptance of the fifth and highest grade. And in DQA at least, Augustine is committed to the rejection of the fourth and fifth grades. University College London

Bibliography Alexander of Aphrodisias. De l’âme [De anima], ed. M.  Bergeron and R. Dufour (Paris: Vrin, 2008). Allers, Rudolph. “St Augustine’s Doctrine on Illumination,” Franciscan Studies 12 (1952), 27–46. Aristotle. De sensu, in P. Siweck (ed.), Aristotelis Parva naturalia: Graece et Latine (Rome: Desclée, 1963).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  35 Aristotle. Tractatus de anima, Graece et Latine, ed. P.  Siweck (Rome: Desclée, 1965). Augustine. Confessiones, in P.  Knöll (ed.), Corpus scriptorum ecclesiasticorum Latinorum, volume 33 (Vienna: F. Tempsky, 1896). Augustine. De Genesi ad litteram libri duodecim, in J. Zycha (ed.), Corpus scriptorum ecclesiasticorum Latinorum, volume 28 (Vienna: F.  Tempsky, 1894). Augustine. De magistro, in G. Weigel (ed.), Corpus scriptorum ecclesiasticorum Latinorum, volume 77 (Vienna: Hoelder-Pichler-Tempsky, 1961). Augustine. De musica, in M. Jacobsson (ed.), Corpus scriptorum ecclesiasticorum Latinorum, volume 102 (Berlin: De Gruyter, 2017). Augustine. De quantitate animae [DQA], in W.  Hörmann (ed.), Corpus scriptorum ecclesiasticorum Latinorum, volume 89 (Vienna: Verlag der österreiachen Akademie der Wissenschaften, 1986). Augustine. De quantitate animae: The Measure of the Soul, tr. F. E. Tourscher (London: Peter Reilly, 1933). Augustine. De Trinitate, in W.  J.  Mountain (ed.), Corpus Christianorum Series Latina, volume 50 (Turnhout: Brepols, 1968). Augustine. The Immortality of the Soul; The Magnitude of the Soul; On Music; The Advantage of Believing; On the Faith in Things Unseen, tr. R. C. Taliaferro (Washington, DC: Catholic University of America Press, 1947). Augustine. The Literal Meaning of Genesis, tr. J. H. Taylor (New York: Paulist Press, 1982). Augustine. On the Trinity, tr. A. W. Haddan (Edinburgh: T&T Clark, 1873). Augustine. Sermon 277, in J.  Migne (ed.), Patrologia Latina, volume 38 (Paris: Garnier Fratres, 1863). Bourke, Vernon  J. Augustine’s Quest of Wisdom (Milwaukee: Bruce Publishing Company, 1947). Brittain, Charles. “Colloquium 7: Attention Deficit in Plotinus and Augustine; Psychological Problems in Christian and Platonist Theories of the Grades of Virtue,” Proceedings of the Boston Area Colloquium of Ancient Philosophy 18 (2003), 223–75. Brittain, Charles. “Non-Rational Perception in the Stoics and Augustine,” Oxford Studies in Ancient Philosophy 22 (2002), 253–308. Broad, C.  D. “Some Elementary Reflections on Sense-Perception,” Philosophy 27 (1952), 3–17.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

36  Mark Eli Kalderon Brown, Deborah. “Agency and Attention in Malebranche’s Theory of Cognition,” in M. Pickavé and L. Shapiro (eds.), Emotion and Cognitive Life in Medieval and Early Modern Philosophy (Oxford: Oxford University Press, 2012), 217–33. Burton, Harry Edwin. “The Optics of Euclid,” Journal of the Optical Society of America 35 (1943), 357–72. Byers, Sarah Catherine. Perception, Sensibility, and Moral Motivation in Augustine: A Stoic-Platonic Synthesis (Cambridge: Cambridge University Press, 2013). Cleomedes. Caelestia (Meteora), ed. R. B. Todd (Berlin: De Gruyter, 1990). Colleran, Joseph M. (tr.). St Augustine: The Greatness of the Soul, The Teacher (New York: The Newman Press, 1949). Dennett, Daniel C. Consciousness Explained (London: Penguin, 1993). Descartes, René. Discours de la methode pour bien conduire sa raison, chercher la verité dans les sciences: plus la dioptrique, les meteores, et la geometrie, qui sont des essais de cete methode (Leiden: Jan Maire, 1637). Diels, H. and W. Kranz. Die Fragmente der Vorsokratiker (DK), 6th edition (Berlin: Weidmann, 1974). Dutton, Blake D. “The Objection from Touch: Sensation, Extension and the Soul in Augustine’s The Quantity of the Soul,” History of Philosophy and Logical Analysis (2020), 1–28. Empedocles. The Extant Fragments, ed. and tr. M. R. Wright (New Haven, CT: Yale University Press, 1981). Empedocles. The Poem of Empedocles, ed. and tr. B.  Inwood (Toronto: University of Toronto Press, 2001). Epicurus. Letter to Herodotus, in M. Marcovich (ed.), Diogenis Laertii: Vitae philosophorum (Stuttgart: Teubner, 1999). Euclid. Optics, in J.  L.  Heiberg (ed.), Euclidis Opera Omnia, volume 7 (Leipzig: Teubner, 1895). Galen. De placitis Hippocratis et Platonis, in P. De Lacy (ed. and tr.), Galen On the Doctrines of Hippocrates and Plato, Second Part: Books VI– IX, Corpus Medicorum Graecorum (Berlin: Akademie-Verlag, 1980). Gannon, Sister Mary Ann Ida. “The Active Theory of Sensation in St. Augustine,” New Scholasticism 30 (1956), 154–80. Gilson, Etienne. The Christian Philosophy of Saint Augustine, tr. L. E. M. Lynch (London: Victor Gollancz, 1961). Hill, Edmund (tr.). Augustine: Sermons (Hyde Park, NY: New City Press, 1994).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  37 Ierodiakonou, Katerina. “Plato’s Theory of Colours in the Timaeus,” Rhizai: A Journal for Ancient Philosophy of Science 2 (2005), 219–33. Kalderon, Mark Eli. Form without Matter: Empedocles and Aristotle on Color Perception (Oxford: Oxford University Press, 2015). Kalderon, Mark Eli. Sympathy in Perception (Cambridge: Cambridge University Press, 2018). Lee, Edward N. “The Sense of an Object: Epicurus on Seeing and Hearing,” in P.  K.  Machamer and R.  G.  Turnbull (eds.), Studies in Perception: Interrelations in the History of Philosophy and Science (Columbus, OH: Ohio State University Press, 1978), 27–59. Lička, Lukáš. “The Visual Process: Immediate or Successive? Approaches to the Extramission Postulate in 13th Century Theories of Vision,” in E. Băltuţă (ed.), Medieval Perceptual Puzzles: Theories of Sense Perception in the 13th and 14th Centuries (Leiden: Brill, 2020), 73–110. Lindberg, David C. “The Intromission–Extramission Controversy in Islamic Visual Theory: Alkindi versus Avicenna,” in P.  K.  Machamer and R. G. Turnbull (eds.), Studies in Perception: Interrelations in the History of Philosophy and Science (Columbus, OH: Ohio State University Press, 1978), 137–59. Lindberg, David  C. Theories of Vision from al-Kindī to Kepler (Chicago: University of Chicago Press, 1977). McMahon, John J. (tr.). The Magnitude of the Soul, in The Immortality of the Soul; The Magnitude of the Soul; On Music; The Advantage of Believing; On the Faith in Things Unseen (Washington, DC: Catholic University of America Press, 1947). Matthews, Gareth  B. “Knowledge and Illumination,” in D.  V.  Meconi and E.  Stump (eds.), The Cambridge Companion to Augustine (Cambridge: Cambridge University Press, 2014), 171–85. Merleau-Ponty, Maurice. The Structure of Behavior (Boston: Beacon Press, 1967). Nemesius. On the Nature of Man [De natura hominis], ed. R. W. Sharples and P. J. van der Eijk (Liverpool: Liverpool University Press, 2008). Noë, Alva. Action in Perception (Cambridge, MA: MIT Press, 2004). O’Daly, Gerard. Augustine’s Philosophy of Mind (Berkeley: University of California Press, 1987). Peter John Olivi. Quaestiones in secundum librum Sententiarum, volumes 1–3, ed. B. Jansen (Florence: Collegium S. Bonaventurae, 1922–6).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

38  Mark Eli Kalderon Peter Sutton. Quodlibeta, in Ferdinand Etzkorn (ed.), “Petrus Sutton(?) O. F. M., Quodlibeta,” Franciscan Studies 23 (1963), 68–139. Philoponus. In Aristotelis De anima libros, ed. M. Hayduck, Commentaria in Aristotelem Graeca, volume 15 (Paris: Berolini, 1897). Plato. First Alcibiades, in J. Burnet (ed.), Platonis Opera, volume 2 (Oxford: Oxford University Press, 1901). Plato. Sophistes, in E. A. Duke et al. (eds.), Platonis Opera, volume 1 (Oxford: Oxford University Press, 1995). Plato. Theaetetus, in E.  A.  Duke et al. (eds.), Platonis Opera, volume 1 (Oxford: Oxford University Press, 1995). Plato. Timaeus, in J. Burnet (ed.), Platonis Opera, volume 4 (Oxford: Oxford University Press, 1905). Plotinus. Enneads, in P. Henry and H. R. Schwyzer (eds.), Plotini Opera, volumes 1–3 (Oxford: Clarendon Press, 1964–84). Porphyry. Ad Gaurum, in K. Kalbfleisch (ed.), Die neuplatonische, fälschlich dem Galen zugeschriebene Schrift Pros Gauron peri tou pôs empsukhoutai ta embrua: Aus dem Anhang zu den Abhandlungen der Königl (Berlin: Preussischen Akadamie der Wissenschaft, 1895). Priscian of Lydia. Metaphrasis in Theophrastum, in I.  Bywater (ed.), Supplementum Aristotelicum Voluminis I Pars II Priciani Lydi Quae Extant (Paris: G. Reimer, 1886). Priscian of Lydia. On Theophrastus On Sense Perception [Metaphrasis], tr. R. Sorabji et al. (London: Bloomsbury, 1997). Proclus. The Elements of Theology, tr. E.  R.  Dodds, 2nd edition (Oxford: Clarendon Press, 1963). Quine, W. V. O. “Three Grades of Modal Involvement,” Proceedings of the XIth International Congress of Philosophy 14 (1953), 65–81. Remes, Paulina. “Plato: Interaction between the External Body and the Perceiver in the Timaeus,” in J. F. Silva and M. Yrjönsuuri (eds.), Active Perception in the History of Philosophy (Dordrecht: Springer, 2014), 9–30. Roger Bacon. Perspectiva, in D. C. Lindberg (ed. and tr.), Roger Bacon and the Origins of Perspectiva in the Middle Ages (Oxford: Clarendon Press, 1996). Sellars, John. “Augustine and the Stoic Tradition,” in K.  Pollmann and W. Otten (eds.), The Oxford Guide to the Historical Reception of Augustine, volume 3 (Oxford: Oxford University Press, 2013), 1775–9. Silva, José Filipe. “Augustine on Active Perception,” in J.  F.  Silva and M.  Yrjönsuuri (eds.), Active Perception in the History of Philosophy (Dordrecht: Springer, 2014), 79–98.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Perception and Extramission in De quantitate animae  39 Silva, José Filipe and Juhana Toivanen. “The Active Nature of the Soul in Sense Perception: Robert Kilwardby and Peter Olivi,” Vivarium 48 (2010), 245–78. Smith, A. Mark. Ptolemy’s Theory of Visual Perception: An English Translation of the Optics with Introduction and Commentary (Philadelphia: The American Philosophical Society, 1996). Smith, A. Mark. From Sight to Light: The Passage from Ancient to Modern Optics (Chicago: University of Chicago Press, 2015). Squire, Michael. “Introductory Reflections: Making Sense of Ancient Sight,” in M. Squire (ed.), Sight and the Ancient Senses: The Senses in Antiquity (London: Routledge, 2016), 1–35. Suarez, Francisco. Commentaria una cum questionibus in libros Aristotelis De anima, in S.  Castellote (ed.), Comentarios a los Libros de Aristóteles Sobre El Alma, volumes 1–3 (Madrid: Sociedad de Estudios y Publicaciones, 1978–92). Theophrastus. De sensibus, in G.  M.  Stratton (tr.), Theophrastus and the Greek Physiological Psychology before Aristotle (New York: Macmillan, 1917). Thomas Aquinas. Super Librum de Causis expositio, ed. H.  D.  Saffrey (Louvain: Nauwelaerts, 1954). Thomas Aquinas. Super Librum de Causis expositio, in V.  A.  Guagliardo, C. R. Hess, and R. C. Taylor (trs.), St. Thomas Aquinas: Commentary on the Book of Causes (Washington, DC: Catholic University of America Press, 1996). Tideus. De speculis, in A. A. Bjømbo and S. Vogel (eds.), Alkindi, Tideus und Pseudo-Euklid: drei optische Werke (Leipzig: Teubner, 1912). Toivanen, Juhana. Perception and the Internal Senses: Peter of John Olivi on the Cognitive Functions of the Soul (Leiden: Brill, 2013). Turnbull, Robert  G. “The Role of the ‘Special Sensibles’ in the Perception Theories of Plato and Aristotle,” in P. K. Machamer and R. G. Turnbull (eds.), Studies in Perception: Interrelations in the History of Philosophy and Science (Columbus, OH: Ohio State University Press, 1978), 3–26. Winer, Gerald A. and Jane E. Cottrell. “Does Anything Leave the Eye When We See? Extramission Beliefs of Children and Adults,” Current Directions in Psychological Science 5 (1996), 137–42.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad Four Christian Philosophers on a Problem of Epistemic Justification Alexander Lamprakis

1. Introduction The debate surrounding miracles is usually connected to two areas of philosophy. The first is natural philosophy, including psychology. Miracles raise the question of how to explain and evaluate events and phenomena that, because they break with the normal course of events, cannot be reduced to a natural cause, and must instead be ascribed to supernatural interference. On the other hand, miracles also raise epis­ temo­logic­al problems. Philosophers are called on to decide whether or not we have reasons to believe perceived or reported miracles, whether we can base arguments on alleged supernatural events, and whether or not it is reasonable to argue for the truth of religion on the grounds of miracles. This second set of questions, which is the primary concern of this article, is associated in contemporary debates most of all with the name of David Hume, who, in his work An Enquiry Concerning Human Understanding (1748), discusses the conditions under which it is rea­ sonable to grant belief to reports of miracles.1 As past research has 1  Hume’s treatise Of Miracles is found in section X of his An Enquiry Concerning Human Understanding. Hume’s argument is still the subject of lively contemporary debate in social epistemology and the philosophy of religion. Alexander Lamprakis, Believing Miracles in 10th–11th Century Baghdad: Four Christian Philosophers on a Problem of Epistemic Justification In: Oxford Studies in Medieval Philosophy Volume 9. Edited by: Robert Pasnau, Oxford University Press. © Alexander Lamprakis 2021. DOI: 10.1093/oso/9780192844637.003.0002

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  41 shown, Hume was not the first to discuss miracles in the context of logic and epistemology: earlier figures of the Enlightenment, such as John Locke (1632–1704), and Antoine Arnauld and Pierre Nicole in their Port Royal Logic (1662), also raised arguments for and against the trust­ worthiness of miracles.2 But before miracles had become a topic among early modern empiricists and rationalists, they were also broadly dis­ cussed by medieval scholastic authors, for whom arguments from mir­ acles were an essential feature of any theology based on revealed religion.3 Scholarship dealing with philosophy in the Islamic world has so far mostly focused on major thinkers, such as Ibn Sīnā, al-­Ghazālī, and Ibn Rushd.4 These central figures pose the question of whether or not mir­ acles must be logically possible, and if so, whether they might be explained through natural philosophy. They furthermore discuss whether miracles were created by God as signs (āyāt) to lend credence to prophecy, and whether miracles can be considered authentic when directly perceived or when known through testimony and reliable chains of transmission. Miracles were discussed not only by these major figures of Islamic philosophy, however; they were also intensely debated by the philo­ sophers of the Baghdad School, who were predominantly Christian. The first of this group to elaborate on the probative function of miracles is the Jacobite Christian Abū Zakariyyāʾ Yaḥyā b. ʿAdī (d. 974). His views are preserved in a short treatise entitled On Establishing the Veracity of the Gospel by Way of Reasoning from Demonstration and Proof, and also scattered throughout some of his shorter apologetic treatises.5 Yaḥyā’s 2  For John Locke see his Essay Concerning Human Understanding, book IV chs. 15 and 16. For the Port Royal Logic see pt. 4, ch. 12. 3  For Augustine and Thomas Aquinas’ accounts, see, for instance, Joseph Houston, Reported Miracles: A Critique of Hume (Cambridge: Cambridge University Press, 1994), 8–33. A general overview on miracles in medieval Latin philosophy is given by Benedicta Ward, Miracles and the Medieval Mind: Theory, Record and Event, 1000–1215 (Aldershot: Scolar Press 1982), 1–20. 4  Cf. Fazlur Rahman, Prophecy in Islam: Philosophy and Orthodoxy (London: Routledge, 2008), 45–52; Michael Marmura, “Avicenna’s Theory of Prophecy in the Light of Ashʿarite Theology,” in W.  S.  McCullough (ed.), The Seed of Wisdom: Essays in Honour of T.  J.  Meek (Toronto: University of Toronto Press, 1964), 159–78 and 161–4; and Frank Griffel, “Al-­Ġazālī’s Concept of Prophecy: The Introduction of Avicennian Psychology into Ašʿarite Theology,” Arabic Sciences and Philosophy 14 (2004), 101–44, esp. 101–21. 5 Yaḥyā’s treatise On Establishing the Veracity of the Gospel (Fī ithbāt ṣidq al-­injīl ʿalā t ̣arīq al-­qiyās bi-­l-­burhān wa-­l-­dalīl) is edited in Paul Sbath, Vingt traités philosophiques et apologétiques d’auteurs arabes chrétiens du IXe au XIVe siècle (Cairo: H. Friedrich, 1929), 168–71. An

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

42  Alexander Lamprakis views on miracles were picked up by two of his direct disciples, the Jacobite Abū ʿAlī ʿĪsā b. Isḥāq Ibn Zurʿa (d. 1008) and Abū ʿAlī Ḥ asan b. Sahl b. Ghālib Ibn al-­Samḥ (d. 1027), whose bookstore at Baghdad’s eastern Gate of the Arch (Bāb al-­Ṭāq) was, according to al-­Tawḥīdī’s report, a much-­frequented hotspot for scientists and philosophers of 10th–11th century Baghdad.6 As I will argue, both of Yaḥyā’s students engage critically with the view on the epistemic value of miracles ­outlined by their teacher. The main sources for Ibn Zurʿa’s stance on this problem are some of his apologetical treatises7 and a treatise entitled A Demonstration Exonerating Those Who Cultivate the Sciences of Logic and Philosophy from Being Considered Irreligious,8 which is also quoted in three large fragments in Ẓahīr al-­Dīn al-­Bayhaqī’s (d. 1169) The Depository of Wisdom.9 The treatise relevant for Ibn al-­Samḥ’s views is a English translation was provided by Sam Noble online in the framework of the Tertullian Project. Some of Yaḥyā’s shorter works are found in Yaḥyā ben ʿAdī, Petits traites apologétiques: texte arabe, ed. and tr. Augustin Perier (Paris: Paul Geuthner, 1920). French translations of key passages are also found in Augustin Périer, Yaḥyā ben ʿAdī: un philosophe arabe chrétien du Xe siècle (Paris: Paul Geuthner, 1920). For a general overview on Yaḥyā, see Gerhard Endress, “Yaḥyā Ibn ʿAdī,” in U.  Rudolph, R.  Hansberger, and P.  Adamson (eds.), Philosophy in the Islamic World, Vol. 1, 8th–10th Centuries (Leiden: Brill, 2017), 435–68. 6  Cf. Elvira Wakelnig, “Ibn al-­Samḥ, Abū ʿAlī”, in K. Fleet, D. Krämer, D. Matringe, J. Nawas, and E. Rowson (eds.), Encyclopaedia of Islam, THREE (Leiden: Brill, 2019), 64. Ibn al-­Samḥ’s confession cannot be determined with precision. Most sources only refer to him as Christian. 7  Most relevant for the purpose of this article is his letter to Bishr b. Pinḥās b. Shuʿayb al-­ Ḥ āsib, edited in Sbath, Vingt traités, 19–52, and, more recently, in Peter Jonathan Starr, The  Epistle to Bišr b. Finḥās (Maqālah ʿamilahā ilā Bišr b. Finḥās) of Ibn Zurʿah (m. A. H. 398/A. D. 1008) (Ph.D. dissertation, Cambridge University 2000), 88–279. On Ibn Zurʿa in general see Gerhard Endress, “ʿĪsā Ibn Zurʿa,” in Rudolph, Hansberger, and Adamson, Philosophy in the Islamic World, Vol. 1, 468–80. Nicholas Rescher falsely states that he is a Nestorian, in “A Tenth-­Century Arab-­Christian Apologia for Logic,” in Studies in the History of Arabic Logic (Pittsburgh, PA: University of Pittsburgh Press, 1963), 56. 8 Extant in a single MS, Paris BnF Arabe 132, fols. 166v–170r, with the title Maqāla yubayyin fīhā barāʾat al-­nāẓirīn fī l-­mant ̣iq wa-­l-­falsafa mimmā yuʿrafūna min fasād al-­dīn. An English translation is found in Rescher, “Apologia for Logic,” 55–63. The Arabic MS is prob­ lematic, however, due to its many scribal mistakes, as pointed out by Cyrille Haddad, ʿĪsā ibn Zurʿa: Philosophe arabe et apologiste chrétien (Beirut: Dār al-­Kalima, 1971), 51: “Dans l’état où il nous a été transmis, ce Traité abonde en fautes de transcription; les digressions font perdre le fil des idées et rendent sa lecture réellement pénible.” Rescher’s translation therefore has to be consulted with caution. An improved edition of the text has been recently provided by Pierre Masrī with an introduction by Nadine Abbas. For this edition see Pierre Masrī and Nadine Abbas, “La relation entre la philosophie et la religion,” al-­Machriq 92 (2018), 381–393 (in Arabic). 9  For the Arabic edition see Ẓahīr al-­Dīn al-­Bayhaqī, The Depository of Wisdom (Tattimat ṣiwān al-­ḥikma), ed. R. ʿAjam (Beirut: Dār al-­Fikr al-­Lubnānī, 1994). Al-­Bayhaqī refers to Ibn Zurʿa’s treatise as Epistle Demonstrating that Philosophical Science Constitutes the Strongest Incitement to Abiding by the Religious Law (Risāla fī anna ʿilm al-­ḥikma aqwā l-­dawāʿī ilā mutābaʿat al-­sharāʾīʿ). The identity of the two treatises has not yet been investigated sufficiently. Moreover, one has to note that the passages quoted by al-­Bayhaqī are not always found

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  43 short work that comes down to us under the title On Reports Held by a Large Number,10 in which he also discusses the conditions under which one should grant belief to reports of miracles. Finally, among the most eminent and, arguably, the last of the Baghdad philosophers is the Nestorian Christian Abū l-­Faraj ʿAbd Allāh Ibn al-­Ṭayyib (d. 1043), whose critical views on the epistemic status of miracles are known from his short treatise On Knowledge and Miracles and the introductory remarks to his On Unity and Trinity.11 Although the epistemology of miracles has received significant attention in past scholarship, the arguments of the representatives of the Baghdad School have not yet been systematically taken into ­con­sid­er­ation.12 This neglect is partly due to their alleged lack of influence among later schools and authors, in contrast to the aforementioned Ibn Sīnā, al-­Ghazālī, and Ibn Rushd. Moreover, the debate on miracles among more canonical figures has often been taken to be an argument between Ashʿarites, who advocated for the possibility and veracity of

verbatim in the manuscript of Ibn Zurʿa’s treatise. For further material on al-­Bayhaqī see Frank Griffel, “On the Character, Content, and Authorship of Itmām Tatimmat Ṣiwān al-­Ḥ ikma and the Identity of the Author of Muntakhab Ṣiwān al-­Ḥ ikma,” Journal of the American Oriental Society 133 (2013), 1–20. 10  A facsimile of the single MS, British Library Add 7473, fols. 33v–40v is appended to Marie Bernand-­Baladi’s French translation, “Des critères de la certitude: un opuscule de Ḥ asan Ibn Sahl sur la crédibilité du dire transmis par un grand nombre,” Journal asiatique 257 (1969), 125–38 (The Arabic title is Qawl fī l-­akhbār allatī yukhbir bihā kathīrūn). For general informa­ tion on Ibn al-­Samḥ, see the discussion in Samuel Miklós Stern, “Ibn al-­Samḥ,” The Journal of the Royal Asiatic Society of Great Britain and Ireland 88 (1956), 31–44. However, neither Bernand-­Baladi nor Stern are aware of the fact that the author of On Reports Held by a Large Number is identical with the Baghdad philosopher Ibn al-­Samḥ; for further information on this treatise see Joel L. Kraemer, Humanism in the Renaissance of Islam: The Cultural Revival During the Buyid Age (Leiden: Brill, 1992), 130–2; Gerhard Endress, “Ibn al-­Samḥ,” in Rudolph, Hansberger, and Adamson, Philosophy in the Islamic World, Vol. 1, 490–6; and Wakelnig, “Ibn al-­Samḥ,” 64–6. 11  On Knowledge and Miracles (Fī l-­ʿilm wa-­l-­muʿjiz) is edited in Sbath, Vingt traités, 179–80. On Unity and Trinity (Fī l-­tawḥīd wa-­l-­tathlīth) is edited in Gérard Troupeau, “Le traité sur l’unité et la trinité de ʿAbd Allāh Ibn al-­Ṭayyib,” Parole de l’Orient 11 (1971), 75–89. An English translation of the former is provided online by Sam Noble in the framework of the Tertullian Project; a French translation is found in Gérard Troupeau, “Traité sur la science et le miracle et fragments du Traité sur les fondements de la religion de ʿAbd Allāh ibn al-­Ṯayyib,” in Études de civilisation médiévale, IXe–XIIe siècles (Poitiers: Centre d’etudes supérieures de civilisation médiévale, 1974), 676–7. As for the latter, a French translation is given in Troupeau, Sur l’unité et la Trinité, 71–91. A general overview on Ibn al-­Ṭayyib is found in Cleophea Ferrari, “Abū l-­Faraǧ Ibn al-­Ṭayyib,” in Rudolph, Hansberger, and Adamson, Philosophy in the Islamic World, Vol. 1, 496–506. 12  One noteworthy exception is Cyrille Haddad’s monograph dedicated to the philosophy of Ibn Zurʿa, in which he refers to his theory of miracles (ʿĪsā ibn Zurʿa, 227–33).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

44  Alexander Lamprakis Qurʾanic miracles, and peripatetics, who approached religion through reason and demonstrative proof, and put miracles within the remit of natural philosophy. I aim to show that, despite their Christian theo­logic­al premises and although much of their work has been lost to us, the Baghdad School also participated in and contributed to this debate. That philosophy and theology (or, more specifically, apologetic literature) must not be treated as two entirely disconnected areas of thought will also become apparent in the course of this article. For Christian theologians, just as for their Islamic counterparts, argu­ ments from miracles necessarily involve two claims: that the wondrous event indeed qualifies as a true miracle, and that the person who works miracles has some sort of epistemic trustworthiness.13 While later debates tend to focus on the latter,14 the authors whose views are ­discussed in this article focus on the former; that is, whether a miracle can be said to have actually taken place, and how one might investigate such events and their reports. Central to this debate is the concept of ‘belief ’ (taṣdīq), and the question of whether and under which condi­ tions mir­acles (muʿjizāt) are necessary for granting belief to the veracity (ṣidq) of prophethood and revelation.15 Whereas al-­Ghazālī (d. 1111) 13  Here, the ‘working’ of the miracle has to be distinguished from causing or creating it. It must be kept in mind that miracles are, generally speaking, caused by divine intervention, but carried out or worked through human agency. The most common Arabic terms used for the ‘working’ of miracles are aẓhara, faʿala, and taqaddara ʿalā. All the authors under discussion seem to hold that a miracle is something beyond the natural capacities of humans (or, more generally, of all creatures) to cause, and which breaks with the natural course of events. These divinely caused departures from natural regularities (nothing, however, logically impossible) ought to serve as a source of belief (taṣdīq). Also note that authors often refer to miracles (muʿjizāt) and signs (āyāt) interchangeably, and these need not be generally distinguished for the purpose of this article. 14  For the debate in al-­Ghazālī see Griffel, “Prophecy,” 121–31; for Fakhr al-­Dīn al-­Rāzī see, for instance, Ayman Shihadeh, The Teleological Ethics of Fakhr al-­Dīn al-­Rāzī (Leiden: Brill, 2006), 104–7 and 129–53, and, more recently, Tariq Jaffer, Rāzī: Master of Qurʾanic Interpretation and Theological Reasoning (Oxford: Oxford University Press, 2015), 114–17. 15  The Arabic term taṣdīq is notoriously difficult to translate. In the framework of the pre­ sent article I chose the translation ‘granting belief,’ or simply ‘belief,’ rather than the more com­ mon ‘assent,’ as it seems advisable for the arguments discussed in this article to employ a less technical translation. For a general overview of the different translations and interpretations of this term see Joep Lameer, Conception and Belief in Ṣadr al-­Dīn Shīrāzī (ca. 1571–1635), al-­Risāla fī l-­taṣawwur wa-­l-­taṣdīq (Tehran: Iranian Institute of Philosophy, 2006), 3–18. Arguments in favor of understanding taṣdīq as ‘belief ’ rather than ‘assent’ are provided in Lameer, Conception and Belief, 19–35, and Lameer, “Ghayr al-­maʿlūm yantaniʿ al-­ḥukm ʿalayhi: An Exploratory Anthology of a False Paradox in Medieval Islamic Philosophy,” Oriens 41 (2014), 403, note 16.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  45 and Ibn Rushd (d. 1198) famously rejected the inerrancy of miracles as signs of prophecy, Yaḥyā b. ʿAdī and his students Ibn Zurʿa and Ibn al-­ Samḥ regard miracles as guaranteeing the truth of religion, although they add additional requirements for believing in a miracle’s authenticity.16 Ibn al-­Ṭayyib, on the other hand, takes a much more critical approach, arguing for the sufficiency of demonstration in matters of religious truth. Thus, I will argue, the position held by the Baghdad philosophers is not at all unanimous and static, but rather underwent significant development which can be traced back to the influence of rationalistic peripatetic philosophy, Christian–Muslim polemics, and the paradigm shift from a Muʿtazilite to an Ashʿarite understanding of miracles among Islamic theologians of the 4th/10th century.

2.  Yaḥyā b. ʿAdī (d. 974): Miracles and Epistemic Agency In his short treatise On Establishing the Veracity of the Gospel by Way of Reasoning from Demonstration and Proof, Yaḥyā claims that the Gospel’s veracity (ṣidq) can be proven by the fact that it is accepted (maqbūl) by a large number of people “whose countries are far from each other and who have different religions and diverse ambitions.”17 As he argues, there are causes for accepting something false and causes for accepting something true. The first involve emotions of fear, shame, and greed, or result from being intellectually tricked into a wrong syllogism. The mere presence of these causes does not tell us whether one’s belief is true or false, since they can lead to either result. The causes for accepting some­ thing true, however, on whose basis it is impossible to accept something wrong, are, according to Yaḥyā, five:18 (i) that it is perceived by the 16  One must note that, predominantly for apologetic reasons, the Christian authors dis­ cussed in this article also employ the concept of prophecy (nubuwwa), which is more relevant to Islamic than Christian theology. The Islamic acceptance of legitimate monotheistic religions (most notably, Judaism and Christianity, referred to as ahl al-­kitāb) is based on the grounds that they have a law (i.e. the Torah and the Gospel) that was established by a prophet (i.e. Moses and Jesus) whose authenticity was confirmed by miracles. In general terms, the refer­ ence to prophecy can be seen as a sign for the interreligious context of the respective treatises. 17  Cf. Sbath, Vingt traités, 168 (trans. Noble). 18  The following passage is found in Sbath, Vingt traités, 170. The Arabic of the quoted pas­ sages reads as follows: (i) mudrakan bi-­l-­ḥiss mawjūdan ḥāḍiran, (ii) awwalan fī l-­ ʿaql,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

46  Alexander Lamprakis senses while it is effectively present, (ii) that it is primary to the intellect, (iii) that it is made clear through proof, or that (iv) it is believed as being something widespread and generally accepted. The last cause for accept­ ing and granting belief to something (qubūluhu wa-­taṣdīq bihi) is that (v) “those who preach it demonstrate it through signs and miracles that man’s inability to perform bears witness to and they are not possible without the help of the Creator.”19 The initial causes enumerated by Yaḥyā for granting belief to the truth of the Gospel strongly correspond to the four sources of knowledge enu­ merated by his teacher al-­Fārābī (d. 950/1). In his Five Chapters on Logic (as well as in other places) he introduces four immediate sources of knowledge, which are themselves not derived from deduction. These are received opinions (maqbūlāt), generally accepted opinions (mashhūrāt), objects of sense perception (maḥsūsāt), and first intelligibles (maʿqūlāt uwal).20 However, as opposed to his student, al-­Fārābī does not mention miracles as a possible source of belief (taṣdīq) in any of his extant writ­ ings.21 So why add miracles to the immediate sources of knowledge? According to Yaḥyā’s account, miracles can produce belief because only those who are good (al-­akhyār), pure (al-­aṭhār), excellent (­al-­fāḍilūn) and trustworthy (al-­sạ̄ diqūn) can work them, those who (iii) yatabayyin bi-­burhān, (iv) shāʾiʿ dhāʾiʿ, and (v) āyāt wa-­muʿjizāt (trans. Noble, modified). Note that Yaḥyā refers to belief (taṣdīq) only in the context of accepting the truth. Accepting falsehood, however, does not qualify as belief. 19  Cf. Sbath, Vingt traités, 170 (trans. Noble). 20  Cf. Douglas Morton Dunlop, “al-­Fārābī’s Introductory Sections on Logic,” The Islamic Quarterly 2 (1955), 267. Similar accounts are also found in his Small Book on Syllogism (Kitāb al-­qiyās al-­ṣaghīr), in R. al-­ʿAjam (ed.), al-­Mant ̣iq ʿinda l-­Fārābī, vol. 2 (Beirut: Dār al-­Mashriq, 1986), 75 and his Book of Debate (Kitāb al-­jadal), in al-­Mant ̣iq, vol. 3, 19. For a general over­ view on the topic of ‘belief ’/‘assent’ (taṣdīq) in al-­Fārābī see Deborah  L.  Black, Logic and Aristotle’s Rhetoric and Poetics in Medieval Arabic Philosophy (Leiden: Brill, 1990), 94–102. 21  In fact, as pointed out by Rahman, al-­Fārābī’s only mention of miracles goes back to his strongly contested Fuṣūṣ al-­ḥikam. Other than that, he “has never spoken of miracles anywhere else in his extant works” (Rahman, Prophecy in Islam, 85, note 75). An exception to this rule is found in his long commentary on Aristotle’s Rhetoric, in which he refers to the miracles’ power to cause (rhetorical) belief. On this, see Frédérique Woerther, “al-­Fārābī and the Didascalia,” in A.  Alwishah and J.  Hayes (eds.), Aristotle and the Arabic Tradition (Cambridge: Cambridge University Press, 2015), 102. More evidence for al-­Fārābī’s critical stance can be derived from the way he depicts arguments based on miracles in his Enumeration of the Sciences (Iḥṣāʾ al-­ ʿUlūm li-­Abī Naṣr al-­Fārābī), ed. ʿAlī Bū Milham (Beirut: Dār wa-­Maktabat al-­Hillāl, 1996), 89. On this, see also the analysis in Ulrich Rudolph, “al-­Fārābī und die Muʿtazila,” in S. Schmidtke, C.  Adang, and D.  Sklare (eds.), A Common Rationality: Muʿtazilism in Islam and Judaism (Würzburg: Ergon Verlag, 2007), 61–6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  47 speak only the pure truth untainted by falsehood, only that which is wholly beneficial unmixed with harm.22

The underlying assumption expressed here is that performing miracles is an unmistakable sign of a person’s veracity and trustworthiness. Someone performing miracles can be regarded as an ideal witness for any sort of information, be it sensible, intelligible, theoretical, or prac­ tical. According to this theory, miracles are the deeds of a fully virtuous character who possesses both excellence and truthfulness. Unfortunately, we do not learn from Yaḥyā’s account how exactly he defines a miracle and what criteria it must fulfill. The only criterion he mentions is that the events must be outside the range of human beings (ʿajz al-­bashar ʿan fiʿlihā).23 Nor do we learn whether a miracle is suffi­ cient for ascribing veracity to a human being, nor whether additional conditions have to be met to distinguish a prophetic miracle from cases of sorcery and ordinary magic. However, since no further qualification is added, it is reasonable to assume that miracles are not merely a neces­ sary, but a sufficient condition for the trustworthiness of someone per­ forming them. After having enumerated all the possible sources for granting belief (taṣdīq) to that which is contained in the Gospel, Yaḥyā subsequently shows that belief in it cannot be granted other than through miracles. According to him, what the Gospel contains (i.e. its doctrinal content) was neither believed by early Christians on account of sense perception and immediate acquaintance, nor first intellectual principles, nor by apodictic truth, nor was it something widespread and generally accepted. It necessarily follows, according to Yaḥyā, that “it was accepted and believed on account of the miraculous signs that those who preached it [i.e. the apostles] used to perform.”24 Consequently, those who were called to embrace the doctrinal truth of the Gospel only did

22  Cf. Sbath, Vingt traités, 170 (trans. Noble). 23  Cf. Sbath, Vingt traités, 170. Similar expressions are also found in Périer, Petits traités, 105, where Yaḥyā says of miracles that it is “impossible for the innate nature of created beings (ʿalā t ̣ibāʿ al-­makhlūqīna) to enact them.” 24  Cf. Sbath, Vingt traités, 170–1 (trans. Noble).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

48  Alexander Lamprakis so “because of signs and miracles that God worked by the hands of those who preached it.”25 Yaḥyā is not the first Christian philosopher in the Islamic world to stress the importance of miracles in the context of epistemic justifica­ tion, but he is the first to express it using the resources of al-­Fārābī’s ter­ minology (or at least the terminology used in the Baghdad School). Around a century before, the Nestorian physician, scholar, and transla­ tor Ḥ unayn b. Isḥāq (d. 873) claimed in his treatise How to Discern of the Truth of Religion that “no one should claim that we have accepted that to which we adhere on any other basis but the display of signs and miracles (ẓuhūr al-­āyāt wa-­l-­muʿjizāt).”26 Like Yaḥyā, Ḥ unayn lists various (pri­ marily psychological) reasons for accepting falsehood, and among the four reasons for accepting the truth contained in the Gospel mentions the working of miracles first and foremost. Unlike Yaḥyā, however, Ḥ unayn does not rely on a peripatetic framework, aside from a refer­ ence to decisive proof (al-­burhān al-­muḍt ̣arr).27 The apologetic writings of Abū Rāʾit ̣a al-­Takrītī and Theodore Abū Qurra (both 9th century) similarly assert the priority of miracles over rational proof, but do not use any typically peripatetic vocabulary or arguments.28 25  Cf. Sbath, Vingt traités, 171 (trans. Noble). 26  Cf. Khalil Samir, “Ḥ unayn b. Isḥāq’s ‘On the Discernment of Religion’” (in Arabic), al-­ Mashriq 71 (1997), 363. An older edition of How to Discern of the Truth of Religion (kayfiyyat idrāk ḥaqīqat al-­diyāna) is found in Sbath, Vingt traités, 181–200 (here 185). For a general overview of this treatise see Barbara Roggema, “Kayfiyyat idrāk ḥaqīqat al-­diyāna, ‘How to Discern the Truth of a Religion,’” in D.  Thomas and B.  Roggema (eds.), Christian-­Muslim Relations: A Bibliographical History, vol. 1 (Leiden: Brill 2009), 775–9; for an English transla­ tion, see Diego Sarrió Cucarella, “On How to Discern the Truth of Religion, by Ḥ unayn b. Isḥāq: The Impersonal Recension,” Islamochristiana 45 (2019), 159–62. The same argument is also mentioned in Yaḥyā’s apologetic treatise on the humanity of Christ directed towards the opponents of Christianity; see Périer, Petits traités, 104–6; see also Périer, Yaḥyā ben ʿAdī, 211–13. According to Marwan Rashed, “New Evidence on the Critique of the Qurʾānic Miracle at the End of the Third/Ninth Century: Qust ̣ā ibn Lūqā vs. the Banū al-­Munajjim,” in P.  Adamson (ed.), In the Age of al-­Fārābī: Arabic Philosophy in the Fourth/Tenth Century (London: Warburg Institute, 2008), 277 the Melkite Qust ̣ā b. Lūqā also held the view that “miracle is the only proof a religion can claim for itself.” 27  Cf. Samir, “Discernment,” 353. The other two reasons for accepting the truth of a religion are, however, less clear than the two mentioned; see Sarrió Cucarella, The Truth of Religion, 160, note 14. 28  For Abū Qurra (d. ca. 825) see Theodore Abū Qurra, Traité de l’existence du créateur et de la vraie religion: introd. et texte critique, ed. I. Dick (Jounieh: Libraire Saint Paul, 1982), 261–8; and for Abū Rāʾit ̣a (fl. 9th cent.) see, for instance, his On Establishing the Christian Religion and the Holy Trinity (Fī ithbāt dīn al-­naṣrāniyya wa-­ithbāt al-­thālūth al-­muqaddas) edited with a parallel English translation in Sandra Toenies Keating, Defending the ‘People of Truth’ in the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  49 The underlying rationale behind Yaḥyā’s argument from miracles is that the ability to perform them is acquired by human beings through a theoretical and practical perfection of their character, which is both the cause and effect of their veracity and trustworthiness. The epis­temo­ logic­al dimension of this claim consists in the reduction of belief (taṣdīq) in p or non-p to belief in person A who is claiming p or non-p. Hence, the credibility of a proposition is derived from the trustworthiness of the person claiming it. Furthermore, there is no way of granting belief to the doctrines of religion, according to Yaḥyā, other than through mir­ acles and the inherent credibility of the person who performs them. We notice something odd, however, when we review Yaḥyā’s argu­ ment carefully. He starts his argument by claiming that the clearest sign of the Gospel’s veracity lies in the fact that a large number of people granted their belief (taṣdīq) to its truth.29 By the end of his treatise, how­ ever, Yaḥyā arrives at the conclusion that What we wanted to prove about the veracity (ṣidq) of what the pure Gospel contains is established, which is that the peoples who were called to believe it (al-­umam al-­madʿūwa ilā l-­taṣdīq) did so only because of signs and miracles (bi-­āyāt wa-­muʿjizāt) that God worked by the hands of those who preached it.30

By excluding all other potential sources of knowledge, Yaḥyā shows that miracles are the only possible cause of granting belief (taṣdīq) to the truth contained in the Gospel. One might legitimately ask why the Gospel must have been accepted on the strength of miracles and only

Early Islamic Period: The Christian Apologies of Abū Rāʾit ̣ah (Leiden: Brill, 2006), 82–145. The relevant passage is found on pp. 90–5. See also the analysis in Starr, Epistle, 56: “Abū Rāʾit ̣a goes on to argue that it is right that faith be proven by miracles which transcend the intellect, for even the clearest arguments may be rejected.” Interestingly, Abū l-­Ḥasan ʿAbd al-­Jabbār, Critique of Christian Origins: A Parallel English-­Arabic Text, ed. and tr. G.  S.  Reynolds and S. K. Samir (Provo, UT: Brigham Young University Press, 2010), 177, reports that the alleged founder of the Baghdad School, Abū Bishr Mattā b. Yūnus, himself a Nestorian Christian, claims that reports of miracles performed by monks and nuns “are worthless luck (hādhā bakht al-­nughūl).” 29  Cf. Sbath, Vingt traités, 168 (quoted above). 30  Cf. Sbath, Vingt traités, 171 (trans. Noble).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

50  Alexander Lamprakis miracles (a claim which is introduced by Yaḥyā as a historical fact), rather than any other equally valid immediate source of knowledge. The systematic connection between a large number of nations believ­ ing the truth of Christianity as transmitted by the Gospel and the role of miracles is worked out more thoroughly in one of Yaḥyā’s apologetical treatises defending the human attributes of Christ.31 At the end of his treatise, Yaḥyā gives the following argument: If the messengers of the Gospel were believed, it is necessary that this belief (taṣdīq) was granted because signs (āyāt) were performed and miracles (muʿjizāt) became evident. For this reason, it is necessary to remove the doubt concerning the veracity of what is reported about it [i.e. the apostolic miracles] in the Gospel. If this is necessary, and the Gospel contains in many places sayings of Christ and of those who spoke under his inspiration, that he is the incarnated God, granting belief to it (taṣdīquhu) and having faith in what it contains is mandatory.32

Since miracles are the only way to know the veracity of Christianity, Yaḥyā argues that all nations who were called to believe demanded their own miracles to testify for the miracles contained in the Gospel. Given that they all accepted Christianity, it follows that such miracles were indeed performed for them, and their belief in these miracles consti­ tutes grounds for contemporary Christians to continue to accept the truth of Christianity.33 Though the systematic coherence of Yaḥyā’s argument is now clearer, its weak spot remains the question of the con­ ditions under which miracles can be accepted as true rather than merely apparent. As will be seen, this is exactly the problem that two of Yaḥyā’s students, whose accounts have come down to us, address.

31  Edition and French translation in Petits traités, 87–109. 32  Cf. Sbath, Petits traités, 106–7. 33 Although most Christian and Jewish apologists hold that divine miracles have never ceased throughout history, unlike their Islamic counterparts.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  51

3.  Ibn Zurʿa (d. 1008): Towards a Rational Assessment of Miracles In the well-­known prosopography The Depository of Wisdom, Ẓahīr al-­ Dīn al-­Bayhaqī (d. ca. 1169) introduces Ibn Zurʿa as the most eminent logician of his time and depicts him as an advocate of rational thought and follower of the certain sciences.34 The passages quoted by al-­Bayhaqī focus on the idea that demonstrative knowledge does not contradict revealed law but rather strengthens adherence to it. Regarding the status of miracles, al-­Bayhaqī quotes Ibn Zurʿa, saying that one only knows the truthfulness of the prophet through miracles (al-­ muʿjizāt). One who does not know the truth of the miracles has no path to believing the prophet (taṣdīq al-­nabiyy), peace be upon him. One who knows the truth of miracles—and that is the truth of proph­ ecy (ḥaqāʾiq al-­nubuwwa)—is closer to granting belief (taṣdīq) to the prophet, peace be upon him, than one who closely follows the appear­ ance of things.35

In this report, Ibn Zurʿa alludes to the theory we came across in the work of his teacher Yaḥyā, namely that the truth of prophecy rests solely on miracles, which an ideal epistemic agent acquires the ability to per­ form through the theoretical and practical perfection of his character. However, he continues, there is no way of granting belief (taṣdīq) to those miracles other than by knowing their truth and authenticity through logic and the sciences. What does he mean by this?

34  On al-­Bayhaqī’s report in general see also Masrī and Abbas, “La philosophie et la reli­ gion,” 378. 35  Cf. al-­Bayhaqī, Depository, 77. The fragment is not found verbatim in the MS Paris BnF Arabe 132. The quoted passage might be also understood as a polemic against the veracity of Muḥammad’s miracles and a rallying cry to the Christian faithful, as seen by Rescher, “Apologia for Logic,” 57, for whom the “emphasis on this point doubtless stems largely from the need of Christian groups within the Muslim orbit for anti-­Islamic polemic and propaganda to assure continuing faithfulness”. However, Ibn Zurʿa does not mention such arguments in his apolo­ getic works directed against Muslims. In his reply to the Muʿtazilite theologian Abū l-­Qāsim al-­Balkhī (d. 931), he rather argues that the prophethood of Muḥammad must be rejected on the grounds “that with Christianity, God has already revealed the most perfect religion and the perfect moral law.” Cf. Endress, “ʿĪsā Ibn Zurʿa,” 476. See also Haddad, ʿĪsā ibn Zurʿa, 301.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

52  Alexander Lamprakis What Ibn Zurʿa is pointing at is that we can never be sure whether an extraordinary event is really a miracle or just something very rare in nature, like, for instance, the power of a magnet. As seen in the previous section, this is indeed a weak spot of Yaḥyā’s discussion, as he does not mention any further criterion for qualifying as a miracle other than being outside the reach of human beings. Not knowing the difference between rare natural events and real miracles, however, leads to the problem of how one distinguishes between a true miracle worker (be they a prophet or an apostle) and a wizard or an impostor—a worry that was frequently raised against the reliability of prophetic miracles in other stages of the debate.36 Examples of miracles whose possibility is often mooted in later iterations of the debate are Moses’ turning his staff into a snake and Muḥammad’s splitting of the moon.37 While Ashʿarite thinkers until al-­Ghazālī tend to accept these mir­ acles on occasionalist grounds, as produced by God and acquired by the prophets (i.e. Moses and Muḥammad), philosophers such as Ibn Sīnā rejected them, advocating instead for their interpretation as metaphors.38 In the case of the splitting of the moon, the fact that heavenly bodies (following Aristotelian premises) cannot be split, not being composed of the four sublunary elements but rather of the fifth in­des­truct­ible heavenly element, is grounds to reject the miracle. Moses’ miracle would likewise be dismissed on the grounds that only an entity’s accidents can be changed, not its substance.39 36  Ibn Zurʿa himself raises the worry of impostors in Paris BnF Arabe 132, fols. 168v–169r; see Masrī and Abbas, “La philosophie et la religion,” 390f. Again, according to Ibn Zurʿa, this threat can be warded off only by the study of logic. 37  The former is, for instance, mentioned in Qurʾān 7:107, the latter in Qurʾān 54:1–2. 38  Marmura, “Avicenna’s Theory of Prophecy,” 164 argues that Ashʿarites indeed accept the possibility of those events based on an occasionalist physics: “The true miracle is that which is beyond such a realm, as, for example, the changing of the staff into a snake by Moses. [. . .] The uniform pattern of nature to which we are accustomed is not in itself necessary, and can be changed by God at will. Miracles are instances of such changes meant to verify the genuine prophet’s claim to prophecy.” For philosophers, on the other hand, “such miracles are im­pos­ sible because they contradict the essential nature of things and the necessary causal laws.” 39  As Griffel, “Prophecy,” 115 points out, according to Ibn Sīnā’s Kitāb al-­Nafs in his Kitāb al-­Shifāʾ, prophets have “the capacity to cause storms, let rain fall, cause earthquakes, or cause people sink into the ground, but they are not capable of changing a piece of wood into an ani­ mal or of splitting the moon.” That Islamic accounts of miracles have been criticized on the basis of Aristotelian natural philosophy seems likely given the note in ʿAbd al-­Jabbār’s polemic against Christians, where he states that “Aristotle and other infidels say that it is not possible for the sun, the moon, and other heavenly bodies to be cleaved or divided” (ʿAbd al-­Jabbār,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  53 Being committed to an Aristotelian worldview, Ibn Zurʿa does not regard miracles as logically impossible events, but approaches them from the perspective of human nature. As we learn from his treatise On Exonerating Those Who Cultivate the Sciences of Logic and Philosophy from Being Considered Irreligious, he defines miracles as “actions which are—concerning our souls and man’s nature—impossible regarding the natural course of events (mumtaniʿa ʿalā l-­majrā l-­t ̣abīʿī).”40 To confirm whether the miracle in question matches this definition, Ibn Zurʿa sug­ gests that we need logic and knowledge of the sciences.41 Ibn Zurʿa provides two examples of miracles: the awakening of a dead man and restoring sight to a blind person.42 Unlike, for instance, the drawing of a round square or the existence of a triangle the sum of whose interior angles is not 180°, neither of those first two cases is logic­ al­ly impossible. There is, however, a difference between a medical and a miraculous cure of blindness. As Ibn Zurʿa argues, a gradual loss of sight can be healed through the art of medicine. On the other hand, healing an eye that is already entirely deprived of its capacity to see is impossible “according to the nature of human beings (t ̣abīʿat al-­bashar)”43 and is enacted, as he states earlier in the treatise, “in a way that is higher and nobler than nature (bi-­t ̣arīq hiya aʿlā min al-­t ̣abīʿa wa-­ashraf).”44 As he concludes, we only come to know the difference between the two cases through the art of medicine, which applies the art of logic.

Critique, 178). The miracles mentioned by Yaḥyā are the healing of the sick, the awakening of the dead, and the binding of living, dead, and inorganic bodies to one’s will (Petits traités, 105). It is an intriguing question whether philosophers like Ibn Sīnā would have accepted the men­ tioned miracles as possible for prophets to carry out. 40  Cf. MS Paris BnF Arabe 132, fol. 168v11–12. In my translation, I follow the readings of Masrī and Abbas, “La philosophie et la religion,” 390: “wa-­kānat al-­muʿjizāt ʿalā l-­ḥaqīqa hiya l-­afʿāl allatī hiya, fī nufusinā wa-­ʿinda t ̣abīʿat al-­bashar, mumtaniʿa ʿalā l-­majrā l-­t ̣abīʿī.” 41 Haddad, ʿĪsā ibn Zurʿa, 52 argues that science can tell whether something is possible or not and that a miracle has to be, consequently, naturally impossible. It is, however, not entirely clear from Ibn Zurʿa’s argument whether he also addresses logical impossibility. 42  Ibn Zurʿa discusses these two examples in Paris BnF Arabe 132, fols. 169v7–170r3; see Masrī and Abbas, “La philosophie et la religion,” 392f. Both examples are miracles reported to have been performed by Christ. As for the awakening of the dead see John 11:38–44, Mark 5:35–43, and Luke 11:17. For the healing of a blind man see John 9:1–6 (where it is important that the man is said to be blind from birth) and Mark 10:46–52. 43  Cf. Paris BnF Arabe 132, fol. 169v8. In the same place, Ibn Zurʿa also mentions that we know this “from the perspective of nature.” 44  Cf. Paris BnF Arabe 132, fol. 169r12–13 (following the corrections of Masrī and Abbas).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

54  Alexander Lamprakis Ibn Zurʿa accepts Yaḥyā’s argument from miracles as it is construed above, that every person who enacts a miracle is fully trustworthy and that the only reason for granting belief to a religion is through miracles. However, he also identifies the argument’s weak point: the difficulty of deciding solely on the basis of sense perception (or testimonial know­ ledge) whether a given agent has actually performed a miracle. Ibn Zurʿa therefore follows his teacher Yaḥyā in basing the truth of prophecy on miracles, but he demands that they be verified through logic and know­ ledge of the certain sciences, and hence implies that real belief in the truth of prophethood (or apostlehood) requires the mastery of these disciplines. In the passage quoted by al-­Bayhaqī, he subdivides belief (taṣdīq) in miracles into two categories: some grant belief to them based on appear­ ance alone, i.e. the sensual perception of the miracle accompanied by the psychological effect of wonder, while others verify them through logic and science, aiming ultimately at knowing their true state (ḥaqāʾiq al-­muʿjizāt). The question of why one has to consider miracles to be the supreme source of belief in the truth of revealed religion, rather than the other four canonical sources (i.e. sense perception, rational proof, first principles, and generally accepted or received opinions), is addressed by Ibn Zurʿa in a fashion similar to his teacher Yaḥyā, but more straightfor­ wardly. In his apologetic Letter to the Jew Bishr b. Pinḥās we encounter the following argument: It is clear that the message that [the apostles] called for was accepted (qubila), and [it is clear that] what a messenger says (qawl al-­dāʿī) about the ability to perform miracles (al-­qudra ʿalā l-­muʿjizāt) is not accepted by an intelligent person before demanding the performance of the miracle. And God Almighty does not perform miracles to lead people into error.45

45  Cf. Sbath, Vingt traités, 45 and the edition in Starr, Epistle, 239. For a more literal transla­ tion see Starr, Epistle, 238. As Starr, Epistle, 345 points out, the same argument is also found in one of Ibn Zurʿa’s answers to questions asked by Yūsuf b. al-­Buḥayrī.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  55 Ibn Zurʿa accepts Yaḥyā’s premise, according to which whoever carries out a miracle must be considered fully trustworthy, even strengthening it by claiming that God cannot possibly create a miracle to make people believe in something wrong.46 In none of his accounts, however, does Ibn Zurʿa reference the theoretical and practical excellence through which the status of prophet- or apostlehood is acquired. This might be traced back to the increasingly Ashʿarite environment of the Baghdad philosophers, according to whose doctrine ethical norms are only ac­cess­ible through the revealed law, and according to whom a prophet’s knowledge of ethical norms cannot therefore predate the revealed law establishing them. This point should not be given too much weight in Ibn Zurʿa’s overall theory, however, given the paucity of accounts that have come down to us from him. For Ibn Zurʿa, as for his teacher Yaḥyā, the truth of Christianity con­ sists in a miracle, namely the miracle of Christ’s divinity. This is clearly stated throughout his treatises.47 However, if Christianity is itself a mir­ acle, how can someone who did not witness the miracle of Christ’s res­ urrection grant belief to it? Not through sense perception, since the event is already past; nor can it be proven rationally or derived from principles; nor is it generally accepted. It is consequently only possible, for Ibn Zurʿa, through further miracles testifying to the veracity of the primary miracle. This is exactly the function he ascribes to the miracles performed by the apostles. As he explains, those who are called to grant belief to a miracle performed at a different time, in a different place, and for different people will necessarily demand the performance of their own miracle testifying to it, given that they have no other reason to 46  Peter Antes, Prophetenwunder in der Ašʿarīya bis al-­Ġazālī (Freiburg: Schwarz Verlag, 1970), 80 has argued that for Ashʿarite thinkers God’s omnipotence is not bound by anything, not even by his goodness. In his article on the problem of whether God would perform mir­ acles for liars and impostors Khaled El-­Rouayheb has, however, shown which solutions emi­ nent Ashʿarite thinkers provided for avoiding this problem. See “Must God Tell us the Truth? A Problem in Ashʿarī Theology,” in B.  Sadeghi, A.  Q.  Ahmed, A.  Silverstein, and R.  Hoyland (eds.), Islamic Cultures, Islamic Contexts: Essays in Honor of Professor Patricia Crone (Leiden: Brill, 2016), 411–30. 47  Cf. Haddad, ʿĪsā ibn Zurʿa, 227–9. Yaḥyā also claims that Christ’s divinity is at the core of Christian belief, as seen in Petits traités, 107 and Périer, Yaḥyā ben ʿAdī, 212–3. For Muslim attacks on Christ’s divinity based on a critical assessment of his miracles see David Thomas, “Early Muslim Responses to Christianity,” in D. Thomas (ed.), Christians at the Heart of Islamic Rule (Leiden: Brill, 2003), 239–43.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

56  Alexander Lamprakis grant belief to it. Acknowledging the truth of past miracles depends on the truth of present miracles that are performed to testify for them, just as logicians produce chains of deductions whose conclusions are guar­ anteed by the truth of premises that themselves may depend on the truth of other premises. In the case of Yaḥyā, this may even comprise three levels of testimonial miracles: the miracle of Christ’s divinity, mir­ acles enacted by the apostles bearing witness to the miracle of Christ’s divinity, and, finally, miracles enacted by missionaries testifying to the miracles performed by the apostles. The truth of these miracles per­ formed by missionaries, monks, and nuns, one might continue, is testified to by Christianity’s (miraculous) geographic expansion over time. In this manner, both Yaḥyā and Ibn Zurʿa ensure that believers of their time still grant their belief on the basis of the very same grounds Christianity ultimately derives its trustworthiness from, namely miracles.48 This argument is also mentioned in ʿAbd al-­Jabbār’s (d. 1024) polemic against Christians, in which he frequently refers to their claim that “their great men perform signs and these do not cease for them in any era, and whoever accepted Christianity did so because of miracles.”49 ʿAbd al-­Jabbār thus provides evidence for the fact that all non-­prophetic mir­acles—like the ones worked by the apostles or nuns and monks— were considered unacceptable by the Muʿtazilites,50 and so one might assume that it was in reaction to the Muʿtazilites that Yaḥyā and Ibn Zurʿa argued for the necessity of the occurrence of miracles. The problem of how to preserve the miracle of Christianity through time also seems to have had ramifications for Christian–Muslim debates on the primary miracle of Islam, the Qurʾān. The Qurʾān, unlike the truth of Christ’s divinity, does not depend on chains of transmission on account of its miraculous character (iʿjāz), which Muslim theologians held to be manifest at all times and places and among all nations. This  character was grounded in the Qurʾān’s inimitability and the

48  See also the analysis of Ibn Zurʿa’s argument in Haddad, ʿĪsā ibn Zurʿa, 232. 49 Cf. ʿAbd al-­Jabbār, Critique, 128 (tr. Reynolds and Samir). 50 Cf. ʿAbd al-­Jabbār, Critique, 125: “How could this be, when only the prophets—peace be upon them—in their era, performed miracles?” (tr. Reynolds and Samir).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  57 psychological effect of wonder it provoked.51 We see that the same cri­ter­ion Yaḥyā and Ibn Zurʿa applied to miracles was applied to the mir­acle of the Qurʾān by contemporary Islamic theologians, namely, that they are outside the natural capacity of a human being. The fact that the performance of the Qurʾān lies outside the capacity of a human being gives it its miraculous character, just as in the case of restoring the sight of an eye that is entirely deprived of its capacity to see. Accordingly, the 3rd–4th/9th–10th century Muslim theologian Abū ʿĪsā b. al-Munajjim can claim, in agreement with his future Christian interlocutors, that it does not matter “whether the signs are magnificent or subtle (kānat al-­ āyāt jalīlatan am lat ̣īfatan) as long as, by nature, the inhabitants of the world are incapable of imitating it [i.e. the Qurʾān].”52 Comparing Yaḥyā and his student Ibn Zurʿa shows that they both agree that the truth of the Gospel is only accepted on the strength of miracles performed by Christ, the apostles, and further missionaries. Whereas Yaḥyā arrives at this conclusion by excluding all other sources of knowledge acknowledged in the peripatetic tradition, Ibn Zurʿa adds positive reasons to believe that miracles have been the driving force behind the spread of Christianity. Moreover, as seen in the passage quoted by al-­Bayhaqī, he demands the investigation of miracles through logic and the sciences in order to discern their true nature. Only if one comes to know the true status of such miracles can one grant full belief in what they imply.53 This furthermore allows Ibn Zurʿa to counter attacks by Muslim contemporaries such as ʿAbd al-­Jabbār, who claims

51  In a well-­known ḥadīth, listed by Bukhārī (d. 870) in his al-Jāmiʿ al-­Ṣaḥīḥ, Muḥammad says “There was no prophet among the prophets not given signs (āyāt) based on which people had security or had belief, but what I was given was the Divine Inspiration which God revealed to me” (ḥadīth 7274, narrated by Abū Hurayra). 52  For the edition of the Arabic and a French translation see Samir Khalil Samir and Paul Nwiya, Une correspondance islamo-­chrétienne entre Ibn al-­Munajjim, Ḥ unayn ibn Isḥāq et Qust ̣ā ibn Lūqā (Turnhout: Brepols, 1981), 588. 53  Ibn Zurʿa’s rationalism (especially in contrast to earlier apologists such as Abū Rāʾit ̣a) is also evident from the following passage taken from his epistle to Bishr b. Pinḥās, where he states “as for the rational faculty, the activity of which is knowledge, he [i.e. Christ] urged us to exert ourselves in seeking knowledge, and research, and the study of books, so he said that the kingdom of God, as we know it, is enclosed within us, by which he referred to our understand­ ing of the noble intelligibles of eternal existence, and our knowledge of them in a true way, is the expected kingdom . . .” (Starr, Epistle, 143).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

58  Alexander Lamprakis that Christians accept miracles “without any demonstration or argu­ ment (bi-­ghayr burhān wa-­lā ḥujja).”54 Despite these differences, it is by no means necessary to see the views of Yaḥyā and his student Ibn Zurʿa as opposed. One should instead ask whether the generation following Yaḥyā had the same argumentative goals and target audience as their predecessors. One may well assume that Yaḥyā’s addressees consisted of philosophers attacking the rationale behind believing in miracles, whereas Ibn Zurʿa was already involved to a greater extent in interreligious debates, as several of his extant treatises show.55 That Ibn Zurʿa’s treatise was likely written for a Muslim audience calling for a rational assessment of their miracles is suggested by the connection he draws between miracles and prophecy. Attacking the miracles held by other communities is a powerful weapon in such debates, since Jews, Christians, and Muslims tend in various degrees to base the truth of their doctrines (e.g. the trinity, God’s unity and sim­ plicity, resurrection, reward and punishment in the afterlife, etc.) on miracles performed by their prophets or apostles. To do this, one must find means to investigate alleged miracles, understood as events that are still logically possible within the limits of natural philosophy but inex­ plicable by means of natural causation.

4.  Ibn al-­Samḥ (d. 1027): Granting Belief to Reports of Miracles In his small treatise entitled On Reports Held by a Large Number— headed On Belief (Fī taṣdīq) by a second hand in the single preserved manuscript56—Ibn al-­Samḥ reports a conversation he had with one of

54 Cf. ʿAbd al-­Jabbār, Critique, 166 (tr. Reynolds and Samir, though I have changed the translation of ‘ḥujja’ from ‘evidence’ to ‘argument’). 55  Besides his aforementioned letter addressed to a Jew, see also his letter to a Muslim friend (Risāla ṣannafahā fī maʿānī saʾalahu ʿanhā baʿḍ ikhwānihi) edited in Sbath, Vingt traités, 6–19. Cf. Endress, “ʿĪsā Ibn Zurʿa,” 475. In his A Demonstration Exonerating Those Who Cultivate the Sciences of Logic and Philosophy from Being Considered Irreligious, Ibn Zurʿa also speaks about “the need of all true religious laws” (MS Paris BnF Arabe 132, 168v3) for philosophy, by which he seems not only to mean Christianity, but also Judaism and Islam. 56  Cf. MS British Library Add 7473, fol. 33v.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  59 his friends concerning the question of when to grant belief to reports transmitted by a large number of people (akhbār min kathīrīn). Although the whole of Ibn al-­Samḥ’s treatise is worth detailed study, I will restrict my analysis to passages relevant to the epistemology of miracles. Whereas the few preserved quotations of Ibn Zurʿa focus on the prob­ lem of how to discern actual miracles using logic and science, Ibn al-­ Samḥ approaches the problem from a different angle, namely on what grounds and under which circumstances we can grant belief to reports, about both natural and miraculous events and containing both sensible and non-­sensible data, which are transmitted by a large number of ­people. Because it relates directly to the science of ḥadīth, this question seems to be genuinely Islamic, but it is noteworthy that the treatise uses no Islamic examples. The work, Kraemer claims, “implicitly (and tact­ fully) undermines the Muslims’ contention that their beliefs (say, in Muḥammad’s prophetic powers) are based upon the consensus of a great number of people,”57 because Ibn al-­Samḥ argues that, as long as certain criteria are not fulfilled, a large number of people reporting an event or state of affairs is not sufficient for granting belief in it. Whatever reasons Ibn al-­Samḥ had to exclude Islamic examples from his discus­ sion, his analysis must be placed together with those of earlier and later Muslim scholars.58 The treatise begins with three possible positions one can assume regarding reports transmitted by a large number of people. One might grant belief (taṣdīq) to all such reports, or one might disbelieve all of them. The third option is to grant belief only to some, while rejecting others. In what follows, Ibn al-­Samḥ argues for this option and lays down the fundamental criteria according to which one must grant, reject, or suspend one’s belief in testimony. Among the relevant criteria is that reports involving anything “whose existence is impossible, in

57  Cf. Kraemer, Humanism, 132. See also Endress, “Ibn al-­Samḥ,” 494: “Here the author is clearly targeting the truth criteria of religious tradition, in particular that of Islamic ḥadīṯ.” 58  For a general overview on the Islamic debate on the problem of tawātur (i.e. transmitted reports) see Josef Van Ess, Theologie und Gesellschaft im 2. und 3. Jahrhundert Hidschra: Eine Geschichte des religiösen Denkens im frühen Islam, Band III (Berlin: De Gruyter, 1992), 382–4 and Band IV (Berlin: De Gruyter, 1997), 334–5.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

60  Alexander Lamprakis such a way that they cannot exist anyhow”59 must be immediately rejected. As an example, he takes the case of someone who, based on reports, advocates for the fact that the sum of the two acute angles of a rectilinear triangle is larger than the right angle (rather than being equal to it). In this case “reason will reject this information because demon­ stration has proven its opposite.”60 It is questionable whether, based on this criterion, Ibn al-­Samḥ would dismiss Muḥammad’s alleged splitting the moon, since it can be proven by reason that the material condition of celestial bodies (being composed of the indestructible fifth element, given an Aristotelian worldview) would not allow for such an event to happen. Likewise, the report of Moses’ transforming his staff into a snake might have to be rejected by reason, as it is impossible to change the substance of any entity. At the same time, Ibn al-­Samḥ clearly states that a prophet “cannot be suspected of forgery.”61 If, then, prophets cannot be suspected of forgery, how can the transmission of such implausible reports arise? Ibn al-­Samḥ would argue, it seems, that the fault lies with the human beings, whether singularly or as a group, who bore witness to these putative miracles and transmitted accounts of them to posterity. Ibn al-­Samḥ appears to argue that one must reject any reports of events whose impossibility can be shown by reason, regardless of the number or trustworthiness of the people bearing witness to it. However, 59  Cf. British Library Add 7473, fol. 38r16–17: “wa-­minhā mā huwa mumtaniʿ al-­wujūd ḥattā annahu lā yumkin an yūjad al-­battata.” For a French translation see Bernand-­Baladi, “Critères,” 115. 60  Cf. British Library Add 7473, fol. 38v3–4: “al-­ʿaql yaruddu hādhā l-­khabar li-­anna al-­ burhān qad awḍaḥa naqīḍahu.” A stronger version of this claim is found in al-­Ghazālī’s Deliverance from Error (al-Munqidh min al-ḍalāl) which Stephen Menn calls “the serpent-­test.” See “The Discourse on the Method and the Tradition of Intellectual Autobiography,” in J. Miller and B.  Inwood (eds.), Hellenistic and Early Modern Philosophy (Cambridge: Cambridge University Press, 2003), 162. Al-­Ghazālī’s thought is the following: “For if I know that ten is greater than three, then if someone said to me, ‘No, three is greater, by the sign that I shall transform this rod into a serpent,’ and if he did transform it and I witnessed him doing so, then I would not for this reason doubt my knowledge: nothing would happen in me except wonder­ ment about how he could do this, but not doubt about what I knew” (tr. Watt, as quoted in Menn, “Autobiography,” 162). Ibn al-­Samḥ seems to hold the same view, namely that nothing surpasses certain knowledge. While al-­Ghazālī’s argument involves logical impossibility in the numerical case, it involves natural impossibility in case of the snake. 61 Cf. British Library Add 7473, fol. 35v10: “lā yuttahamu bi-l-­ iftiʿāl . . . ka-­wāḥid min ­al-­anbiyāʾ.” Again, the reference to prophets implies a specifically Islamic, or, at least, interreli­ gious, context for his treatise.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  61 he appears to make use of this criterion less strictly than his geometrical example suggests. For instance, he claims earlier in his treatise that rea­ son must reject what Jews claim about the miracles of Jesus, namely that someone [i.e. Jesus] steals one of God’s names (ism min asmāʾ illāh)— Most High and Almighty—that he cuts his leg and hides it [and] that this divine name escapes from his leg when the individual [i.e. Jesus] is exposed to torture.62

This example refers to a widespread polemic against Christ which explains his miraculous activities by claiming that he stole one of God’s divine, miracle-­performing names from the Torah, placed it in his leg, and enacted miracles with its power.63 Ibn al-­Samḥ notes that “reason rejects all information of this kind intuitively.”64 It cannot be ascertained whether Ibn al-­Samḥ, confronted with a report about a piece of parch­ ment bearing one of God’s secret Hebrew names working miracles while hidden in someone’s leg, would consider it to be logically or naturally impossible, or whether he at all makes this distinction. Given that he considers both the report of a rectilinear triangle, whose two acute angles are larger than the right angle, and Jesus’ allegedly forged miracle-­ working power as rejectable by reason, it seems to be plausible that Ibn al-­Samḥ considers them as comparable cases, in which reason is able to find proof (burhān) for their being impossible. Ibn al-­Samḥ’s treatise, hence, provides further evidence for the fact that, unlike early Ashʿarites or even David Hume, for whom miracles consist in a “transgression of a law of nature by a particular volition of the Deity, or by the interposition of some invisible agent,”65 we have no reason to believe that the philosophers of the Baghdad School regarded miracles as something impossible per se. Since Ibn al-­Samḥ explicitly discusses the transmission of miracles performed by Moses and Jesus,

62  Cf. British Library Add 7473, fol. 34v20–21. 63  This story is most probably taken from the Sefer Toledot Yeshu, a polemical work against Christians, originally composed in Aramaic. The oldest known version dates to the 11th cen­ tury. On this see Bernand-­Baladi, “Critères,” 122, note 9 and Starr, Epistle, 390. 64  Cf. British Library Add 7473, fol. 34v24: “yarudduhā l-­ʿaql bi-­badāhatihi.” 65  Cf. David Hume, Of Miracles (La Salle, IL: Open Court Classic, 1985), 114.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

62  Alexander Lamprakis he must have considered them to be, in principle, rationally accessible. What makes miracles miracles is not the fact that they contradict some logical principle, but rather, because they are outside of regular human capacities and the natural course of events, only chosen individuals, guided by God, can perform them. To borrow Augustine’s parlance, miracles are not contra naturam but rather supra naturam.66 Further criteria stipulate that such reports must be diffused among numerous communities and—importantly—that these communities should not “intend the affirmation or the denial of a fact established or denied by that report.”67 This criterion of unintentional testimony already appears in Yaḥyā’s aforementioned discussion of the truth of Christianity; according to Ibn al-­Samḥ’s teacher, “clear evidence” for the truth of a religion is that the peoples whose countries are far from each other and who have different religions and diverse ambitions “cannot collude nor is it thinkable that they would agree with one another.”68 This last criterion can scarcely be fulfilled by any reported miracles. Miracles performed by Christ, as Ibn al-­Samḥ’s example above shows, are evidently denied by the Jews. Even the miracles performed by Moses, which are accepted by Jews and Christians (and, as Ibn al-­Samḥ avoids mentioning, also by Muslims), are denied by the Mazdeans and the Sabians, who “would answer you individually and unanimously (ʿalā l-­infirād wa-­ʿalā l-­ijmāʿ) that their chronicles do not mention any.”69 Nonetheless, one should grant belief (taṣdīq) to the occurrence of such events only if both criteria are fulfilled, i.e. that the reported miracle is logically possible and transmitted by several independent groups who have no direct interest in establishing its truth or falsehood. Otherwise, one must find proof (burhān) for the reported events at stake. Where this is not possible, Ibn al-­Samḥ seems to suggest, one has to simply sus­ pend final judgment, regardless of one’s creed or confession. 66 As seen above, Ibn Zurʿa even uses the same expression in Arabic, namely aʿlā min al-­t ̣abīʿa. 67 Cf. British Library Add 7473, fol. 39r1–2: “lā taqṣidu fī l-­akhbār bihi ithbāt shayʾin tarūmu ithbātahu bihi wa-­lā tubt ̣il shayʾan tarūmu ibt ̣ālahu bihi.” 68  Cf. Sbath, Vingt traités, 168 (trans. Noble). 69  Cf. British Library Add 7473, fol. 34v4–7: “If you ask the Mazdeans (al-­majūs) and the Sabians (al-­ṣābiʾa) ‘Do your chronicles make any mention of miracles attributed to Moses? Indeed, I mean the ones which are attributed to him during his journey in Egypt and others,’ they would answer you individually and unanimously that their chronicles do not men­ tion any.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  63 It cannot be ascertained whether Ibn al-­Samḥ’s short treatise was read by a Muslim audience or what influence its arguments had on the debate surrounding the transmission of prophetic miracles. However, reading al-­ Ghazālī’s defense of such traditions—often explicitly directed at Christians—one may conclude that such arguments were more influen­ tial than one might initially think. In his Moderation in Belief, we read that If it is said that none of these events is transmitted through widely transmitted reports (al-­tawātur), then we say that even if this is true, it does not refute our conclusions, as long as the totality (bi-­l-­majmūʿ) of these events was transmitted widely. . . . If a Christian says ‘these matters have not reached me through widely transmitted reports, neither collectively nor individually,’ then one would say to him ‘if a Jew who has never mixed with Christians moves to a country and claims that the miracles of Jesus—peace be upon him—were never widely reported to him, and even if they were, they were reported by the tongues of Christians (ʿalā lisān al-­naṣārā), who take an interest in the matter (muhtammūn bihi), then how are you different from this Jew?’70

Although al-­Ghazālī is not directly referring to any specific author, he addresses both criteria that, according to Ibn al-­Samḥ, have to be met to grant belief in a miracle. Such transmitted reports (here called by their Islamic designation tawātur) have to be transmitted by various peoples independently and without them taking any interest in establishing or denying their truth or falsehood. Tellingly, al-­Ghazālī does not bother to rule out these worries but rather refers to the fact that they hold equally of the miracles enacted by Christ and the apostles. Accepting how mir­acles are transmitted in Christianity implies accepting it in the case of Islam, if one does not want to deny the general trustworthiness of prophetic or apostolic miracles.

70  Cf. Abū Ḥ āmid al-­Ghazālī, Moderation in Belief (al-­Iqtiṣād fī l-­iʿtiqād), ed. H. Atay and I. Çubukçu (Ankara: Nur Matbaasi, 1962), 209. The English translation is a slightly modified version of Yaqub’s (Moderation in Belief (al-­iqtiṣād fī al-­iʿtiqād), tr. A.  M.  Yaqub (Chicago: University of Chicago Press, 2013)), 205f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

64  Alexander Lamprakis Ibn al-­Samḥ, like Ibn Zurʿa, does not deny the possibility of mir­acles— he even defends Christ’s childhood miracles—but calls for a rational investigation of each case just as Ibn Zurʿa called for an assessment of miracles through logic and the sciences. On the other hand, Ibn al-­Samḥ offers a much more pessimistic approach towards the verification of reported miracles and reinforces the need for proof in cases of doubt. Moreover, this skepticism about the reliability of reported mir­acles brings their rational accessibility into doubt. As I will argue in the last section of this paper, this finally leads to the view held by the last ­representative of the Baghdad School, Ibn al-­Ṭayyib, for whom any adherence to miracles must be considered, from a philosophical point of view, inherently problematic.

5.  Ibn al-­Ṭayyib (d. 1043): Doing without Miracles Ibn al-­Ṭayyib marks an interesting turning point in the debate on mir­ acles, since he advocates a purely scientific attitude towards the truth contained in religion, stripped of any reference to miracles. As shown above, Yaḥyā b. ʿAdī argued against the possibility that the Gospel’s truth can be derived only through reason (ʿaql) and proof (dalīl). Before him, Ḥ unayn b. Isḥāq, too, argued for the priority of mir­ acles in matters of belief, along with earlier Greek, Syriac, and Arabic Christian apologists.71 In clear opposition to this tradition, Ibn al-­ Ṭayyib begins his On Knowledge and Miracles with what must be con­ sidered a direct attack on his predecessors, stating that, according to the Christian creed, demonstration is nobler than a miracle (al-burhān ashraf . . . min al-­muʿjiz). The argument Ibn al-­ Ṭayyib gives is likely 71  For an overview of late Antique Christian accounts on miracles see, for instance, James Carleton Paget, “Miracles in Early Christianity,” in G.  H.  Twelftree (ed.), The Cambridge Companion to Miracles (Cambridge: Cambridge University Press, 2011), 131–43. On early Christian Arabic accounts on miracles see Sidney  H.  Griffith, “Comparative Religion in the Apologetics of the First Christian Arabic Theologians,” Proceedings of the PMR Conference: Annual Publication of the Patristic, Mediaeval and Renaissance Conference 4 (1979), 63–87, and, ibid., “Faith and Reason in Christian Kalām. Theodore Abū Qurrah on Discerning the True Religion,” in S. K. Samir and J. S. Nielsen (eds.), Christian Arabic Apologetics during the Abbasid Period (750–1258), 1–43.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  65 derived from a late ancient peripatetic framework, in which intellect (nous) surpasses sense perception (aisthesis). Accordingly, he claims: Demonstration is a proof (dalīl) through which the intellect comes to grasp the truth to which the miracle worker (ṣāḥib al-­muʿjiz) calls.72

This is the first instance in the debate over miracles in which demonstra­ tion is treated not as a mere tool for assessing miracles—whether for investigating their logical and scientific possibility or their transmis­ sion—but as fully capable of producing belief (taṣdīq) in the doctrinal content of religion. Ibn al-­Ṭayyib does not neglect the points discussed by Ibn Zurʿa and Ibn al-­Samḥ, but explicitly refers to their abovementioned discussions in addressing the role demonstration plays in the critical assessment of miracles, and adds that demonstration also investigates their “­circumstances and the circumstances of the ones who transmit it and the condition of the reporter and the formation in the moment of its utterance.”73 However, whereas Ibn Zurʿa distinguished between believing miracles based on sense perception and based on a profound know­ledge of the sciences, Ibn al-­Ṭayyib generally places miracles on the same level as sense perception and in opposition to demonstration. Ibn al-­Ṭayyib depicts demonstration and miracles as two distinct paths towards the truth of religion. Whereas demonstration is the only way by which experts (khawāṣṣ), philosophers (falāsifa), and scholars (ʿulamāʾ) grant belief to the propositional content of any religion, mir­ acles are produced for the masses (jumhūr) and ultimately depend on sense perception.74 While Yaḥyā and Ibn Zurʿa argued that the miracles of Christianity must have been true because such great philosophers and scholars as the Greeks accepted them, Ibn al-­Ṭayyib argues that it is more likely that these great philosophers, rather than believing merely on the basis of miracles, evidence of which is ultimately derived from sense perception, found a rational proof for the truth of Christianity. By basing his judgment solely on rational proof, Ibn al-­Ṭayyib sides with 72  Cf. Sbath, Vingt traités, 179 (trans. Noble). 73  Cf. Sbath, Vingt traités, 179. 74  Cf. Sbath, Vingt traités, 179.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

66  Alexander Lamprakis the Muʿtazilite theologian ʿAbd al-­Jabbār, for whom “the only proof, without exception, of the validity of religions is argument (ḥujja) and demonstration (burhān).”75 Ibn al-­Ṭayyib’s rationalism comes at the price of breaking with the apologetic tradition of his predecessors Yaḥyā b. ʿAdī and Ibn Zurʿa. Given that intellect is nobler than sense perception, according to Ibn ­al-­Ṭayyib, believing on the basis of demonstration is nobler than doing so by reference to miracles. Even if the authenticity of a miracle is con­ firmed by reason, as demanded by Ibn Zurʿa, and its reports match the criteria outlined by Ibn al-­Samḥ, it will never transcend its origin in sense perception, which is, without exception, a weaker source of knowledge than demonstration. Ibn al-­Ṭayyib finds further support for his view in 1 Corinthians 12:28, where the apostle Paul states that God has placed in the church first of all apostles (al-­rusul al-­salīḥūna), after them the prophets (al-­anbiyāʾ), and after them the teachers (al-­ ʿulamāʾ), then those who perform miracles (alladhīna yafʿalūn ­al-­muʿjiz), then those who heal the sick, and after them those who master languages.76

Because Paul places ‘teachers’ before ‘those who perform miracles,’ Ibn al-­Ṭayyib infers that demonstration (which is what is taught by teachers, according to the Aristotelian tradition) is nobler than believing in miracles. Ibn al-­Ṭayyib’s short treatise includes a further argument, which, unlike the one just mentioned, does not depend on scripture: a miracle must be performed in a certain place, at a certain time, and among cer­ tain people. Consequently, if the place, time, and people perish, the miracle perishes too. Demonstration, on the other hand, exists 75 Cf. ʿAbd al-­Jabbār, Critique, 126. I have modified the English translation by Reynolds and Samir by translating ‘ḥujja’ as ‘argument’ rather than ‘evidence’. On the usage of ‘dalīl’ and ‘burhān’ in the context of Islamic theology see Josef Van Ess, “The Logical Structure of Islamic Theology,” in H. Biesterfeldt (ed.), Kleine Schriften by Josef van Ess (Leiden: Brill, 2018), 243, who reminds us that ‘dalīl’ is not equivalent to the Aristotelian concept of apodeixis, which is usually rendered ‘burhān.’ 76  Cf. Sbath, Vingt traités, 179–80. A discussion of this argument is also found in Franz Rosenthal, Knowledge Triumphant: The Concept of Knowledge in Medieval Islam (Leiden: Brill, 2007), 98.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  67 in­de­pend­ent­ly from place, time, and people and must hence be regarded as nobler than a miracle.77 Ibn al-­Ṭayyib seems therefore to hold, unlike Yaḥyā and arguably also his disciples Ibn Zurʿa and Ibn al-­Samḥ, that justifying and granting belief (taṣdīq) to the propositional truth con­ tained in prophecy is in principle possible only through demonstration. Ibn al-­Ṭayyib concludes that the truth of Christianity not only can but should be derived from demonstrative arguments alone, and that Christ performed miracles only for the masses, while establishing dem­ onstration for eminent philosophers whose judgment is not guided by sense perception. Ibn al-­Ṭayyib’s argument also seems to imply that longstanding and reliable transmission of a witnessed event cannot serve as epistemic justification in the same way immediate sense percep­ tion can. This is intriguing, since we saw above that establishing justified belief in miracles that take place at a different time, in a different place, and for different people—as Ibn al-­Tạ yyib formulates it—is something into which both Yaḥyā and Ibn Zurʿa put a lot of argumentative effort. We find the same view expressed in Ibn al-­Ṭayyib’s introductory remarks to On Unity and Trinity, where he claims that the wise accepted the teaching of the Gospel from the apostles only through reason: There is no doubt that the intellects of the experts investigated it [i.e.  the scripture], examined it closely, and accepted it based on a certain acceptance.78

On the other hand, he claims that the masses (al-­ʿawāmm) are necessarily called to grant belief (ilā taṣdīq) in it [i.e. in the scripture] through that which forces them to accept it: through miracles.79 For Ibn al-­Ṭayyib, blind imitation is legitimate in the case of common people, while ­philosophers have to be led to the truth through demonstrative proof. It is evident from this that he must hold that doctrines contained in the Gospel are indeed demonstrable and, consequently, rationally defensible 77  The first premise appears in a similar form in al-­Maqdisī’s Book of Creation and History, ed. and tr. C.  Huart (Paris: Ernest Leroux, 1907), 175. On this point, see also Tarif Khalidi, “Muʿtazilite Historiography: Maqdisī’s Kitāb al-­Badʾ wa-­l-­Tārīkh,” Journal of Near Eastern Studies 35 (1976), 6. 78  Cf. Troupeau, “l’Unité et la Trinité,” 77. 79  Cf. Troupeau, “l’Unité et la Trinité,” 77.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

68  Alexander Lamprakis without resort to miracles (which were only performed for those who have no access to demonstrative thinking), and thereby departs from the views expressed by his predecessor Yaḥyā b. ʿAdī and his two students. It is interesting to compare this both with al-­Fārābī and Avicenna, who famously argue that the truth must be communicated to people according to their intellectual capacities, and with al-­Ghazālī’s skeptical argument that, because sense perception is fundamentally fallible, one cannot achieve certain knowledge even of a miracle immediately present to the senses and must suspend judgment.80 To sum up, Ibn al-­Ṭayyib, like Ibn Zurʿa, identifies the weak spot of the overall argument from miracles as establishing with certainty whether someone has actually performed a miracle. While Ibn Zurʿa and Ibn al-­Samḥ try to firm up the evidence of miracles by demanding that they be rationally investigated in the light of logic and science, Ibn al-­Ṭayyib argues that, no matter how well established and investigated, any miracle is ultimately dependent on fallible sense perception. This does not mean that one should not undertake a rational assessment of miracles, but only that the acquired evidence does not differ from that provided by sense perception of regular events. Even multiple witnesses and flawless transmission do not elevate sense perception into the rank of certain knowledge.81

6.  Conclusion: Christian Philosophers in the Light of Islamic Theology Arguments from miracles were not new to Christian theology by the time of Yaḥyā b. ʿAdī and his students. Early Christian apologists 80  For al-­Ghazālī’s methodological skepticism concerning miracles see the reconstruction in Menn, “Autobiography,” 162–4. 81  One might also consider the fact that Ibn al-­Ṭayyib could have been influenced by his direct teacher al-­Ḥ asan Ibn al-­Khammār (d. after 1017), who, although we have no text of his on prophetic or apostolic miracles, is reported to have written a treatise entitled On the Agreement between the View of Philosophers and that of the Christians (Kitāb al-­wifāq bayna raʾy al-­falāsifa wa-­l-­naṣārā) whose current whereabouts are, unfortunately, unknown. The title of the treatise implies that the truth of Christianity is confirmed by philosophy, which would harmonize with Ibn al-­Ṭayyib’s claim as reconstructed above. See Gerhard Endress, “Ibn ­al-­Ḫammār,” in Rudolph, Hansberger, and Adamson, Philosophy in the Islamic World, Vol. 1, 487.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  69 frequently emphasize the key role miracles played in the justification of theological doctrines and the conversion of early Christians.82 To lend credence to their arguments in an increasingly philosophical milieu, however, Yaḥyā as well as later members of the Baghdad School of phil­ oso­phy began to fit miracles into the framework of their epistemology. While applying al-­Fārābī’s philosophical toolset, both Yaḥyā and Ibn Zurʿa show that none of his divisions of the immediate sources of know­ ledge apply to the experiences of the early Christians who embraced the truth of the Gospel. When regarding the differences and developments of the debate on miracles, one must also consider changes of attitude in Islamic kalām and its influence on the debate among the predominantly Christian philo­sophers of the Baghdad School. As Stroumsa notes, “the superior value of miracles is generally admitted on the Muslim side.”83 However, as she adds, it is not yet clear from which point of the development of Islamic thought prophetic miracles were regarded as an integral and undisputed part of doctrine.84 While it is difficult to discern the system­ atic status miracles occupy in the theology of early Muʿtazilism,85 it is 82 On the other hand, Stroumsa is right to point out that the topic of prophetic signs (including miracles) is not among the most pressing concerns of pre-­Islamic Christian and Jewish philosophy. See “The Signs of Prophecy: The Emergence and Early Development of a Theme in Arabic Theological Literature,” Harvard Theological Review 78 (1985), 103. On this point, see also Sidney Griffith, “Answering the Call of the Minaret: Christian Apologetics in the World of Islam,” in J.  J.  Van Ginkel, H.  L.  Murre–Van den Berg, and T.  M.  Van Lint (eds.), Redefining Christian Identity. Cultural Interactions in the Middle East since the Rise of Islam (Leuven: Peters, 2005), 116. 83  See Sarah Stroumsa, Freethinkers of Medieval Islam: Ibn al-­Rāwandī, Abū Bakr al-­Rāzī, and Their Impact on Islamic Thought (Leiden: Brill, 1999), 31. 84 See also Stroumsa, Freethinkers, 110: “By the fourth and fifth Islamic centuries, the mi­racu­lous beauty of the Qurʾān had become an integral part of the Islamic creed. But we know that in the third century some theologians, whose adherence to Islam cannot be ser­ ious­ly doubted, did not accept this doctrine as binding”; Paul  E.  Walker, “The Political Implications of al-­Rāzī’s Philosophy,” in C.  Butterworth (ed.), The Political Aspect of Islamic Philosophy: Essays in Honor of Muhsin S. Mahdi (Cambridge, MA: Harvard University Press, 1992), 86 also notes that “the doctrine of prophetic miracles—and even the specifics of the miracle of the Qurʾān—was not yet fixed [. . .] until the 5th Islamic Century.” 85 Muʿtazilite theologians arguing for the necessity of miracles are al-­Jāḥiẓ (d. 868/9), who in his The Signs of Prophecy (Ḥ ujaj al-­nubuwwa) mentions miracles as a sign of prophethood (cf. Griffel, “Prophecy,” 101); ʿAbd al-­Jabbār, who in his Summa on the Matters of Divine Unity and Divine Justice (Mughnī fī abwāb al-­tawḥīd wa-­l-­ʿadl) claims that “nothing but miracles are proper proof of the sending of [divine] messengers” (cf. Stroumsa, Freethinkers, 31f.); and ­al-­Mut ̣ahhar ibn Ṭāhir al-­Maqdisī (4th/10th cent.) who frequently refers to prophetic miracles in his Book of Creation and History (Kitāb al-­badʾ wa-­l-­tārīkh), cf. Khalidi, “Historiography,” 6ff.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

70  Alexander Lamprakis undisputed that they played a prominent role in contemporary Ashʿarite theories of prophecy as a result of the doctrine that all norms of human action are ultimately established by revelation and cannot be derived from reason. The character traits of a prophet, therefore, cannot be taken as a sign of the veracity of his prophecy, since it is exactly through his revelation that what is (morally) good and bad, and what counts as virtuous and vicious is established. As Griffel argues, al-­Ashʿarī (d. ca. 935/6) himself held that “one should put one’s trust only in the occur­ rence of true prophetic miracles.”86 In this light, it is likely that the accounts of the four Baghdad philosophers discussed here also reflect and react to discussions taking place in Islamic theology. Yaḥyā’s treatise is still very much grounded in Christian apologetic literature, which argued for the veracity of Christian belief by referring to the fact that even the Greeks, who were known to be scholars and philosophers, embraced the truth of the Gospel. Yaḥyā’s philosophical defense of miracles may be intended to counter attacks put forth by rationalist Muʿtazilite theologians and, similarly, by philosophers like Abū Bakr al-­Rāzī (d. 925 or 935), who rejected the necessity of any argu­ ment derived from prophetic and non-­prophetic miracles.87 The demand for a rational assessment of religious doctrine grew in the generation following Yaḥyā b. ʿAdī. With the rise of Ashʿarism in 4th/10th–5th/11th century Baghdad, the prophetological and therefore the epistemological importance of miracles increased significantly. Ashʿarites until al-­Ghazālī generally agreed with the proposition that miracles are unmistakable signs for the veracity of prophecy, which may have motivated later Christian philosophers such as Ibn Zurʿa and Ibn Unlike prophetic miracles (muʿjizāt), the possibility of non-­prophetic miracles (karamāt) was seen more critically by Muʿtazilite authors. 86  Cf. Griffel, “Prophecy,” 102. He also speaks of the “classical Ashʿarite view that prophecy is verified only through miracles” (140). For a general overview on al-­Ashʿarī’s views on pro­ phetic miracles see Daniel Gimaret, La doctrine d’ al-­Ašʿarī. (Paris: Cerf, 1990), 459–67. 87  See for instance the reports in Abū Ḥ ātim al-­Rāzī, The Proofs of Prophecy: A Parallel English-­Arabic Text, tr. T. Khalidi (Provo, UT: Brigham Young University Press, 2012), 139–44. Cf. Stroumsa, Freethinkers, 111; Marwan Rashed, “Abū Bakr al-­Rāzī et la Prophétie,” MIDEO 27 (2008), 169–81; and Philippe Vallat, “Between Hellenism, Islam, and Christianity: Abū Bakr al-­Rāzī and His Controversies with Contemporary Muʿtazilite Theologians as Reported by the Ashʿarite Theologian and Philosopher Fakhr al-­Dīn al-­Rāzī,” in D. Janos (ed.), Ideas in Motion in Baghdad and Beyond: Philosophical and Theological Exchanges between Christians and Muslims in the Third/Ninth and Fourth/Tenth Centuries (Leiden: Brill, 2016), 178–220.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  71 al-­Samḥ to concentrate instead on the question of which miracles can be considered authentic and which not. This clear shift of focus may be connected to the rise of interreligious debates concerning the truth of prophetic and apostolic miracles of Judaism, Christianity, and Islam. It is noteworthy that both Ibn Zurʿa and Ibn al-­Samḥ apply the language of Islamic kalām, which is strong evidence for their contribution to a gen­ eral debate on the epistemic value of miracles, not just between Christians, but also with Muʿtazilite and Ashʿarite theologians. Ibn Zurʿa seems to have reacted to Muslim attacks denying the possibility of Christian miracles, while Ibn al-­Samḥ attacked the way prophetic mir­ acles are transmitted within the Islamic community. While both Ibn Zurʿa and Ibn al-­Samḥ call for a rational assessment of miracles and reports bearing witness to them, Ibn al-­Ṭayyib, their successor and the last philosopher of the Baghdad School, abandons the philosophical use of miracles for establishing the truth of religion in favor of purely rational arguments. In his complete adherence to the method of rational proof and philosophical analysis, Ibn al-­Ṭayyib marks the end of the century-­long debate concerning the argument from miracles in the Baghdad School.88 Ludwig-­Maximilians-­Universität München

Bibliography ʿAbd al-Jabbār, Abū l-Ḥ asan. Critique of Christian Origins: A Parallel English-Arabic Text, ed. and tr. G.  S.  Reynolds and S.  K.  Samir (Provo, UT: Brigham Young University Press, 2010). Abū Qurra, Theodore. Traité de l’existence du créateur et de la vraie religion, ed. I. Dick (Jounieh: Libraire Saint Paul, 1982). al-Bayhaqī, Ẓahīr al-Dīn. The Depository of Wisdom (Tattimat Ṣiwān al-Ḥ ikma), ed. R. ʿAjam (Beirut: Dār al-Fikr al-Lubnānī, 1994). al-Fārābī, Abū Naṣr. Book of Debate (Kitāb al-Jadal), in R.  al-ʿAjam (ed.), al-Mant ̣iq ʿinda l-Fārābī, vol. 3 (Beirut: Dār al-Mashriq, 1986).

88  I would like to express my gratitude to Peter Adamson and to two anonymous reviewers who helped to improve both the argument and exposition of the present article.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

72  Alexander Lamprakis al-Fārābī, Abū Naṣr. Enumeration of the Sciences (Iḥṣāʾ al-ʿUlūm li-Abī Naṣr al-Fārābī), ed. ʿAlī Bū Milham (Beirut: Dār wa-Maktabat al-Hillāl, 1996). al-Fārābī, Abū Naṣr. Small Book on Syllogism (Kitāb al-Qiyās al-Ṣaghīr), in R.  al-ʿAjam (ed.), al-Mant ̣iq ʿinda l-Fārābī, vol. 2 (Beirut: Dār alMashriq, 1986). al-Ghazālī, Abū Ḥ āmid. Moderation in Belief (al-Iqtiṣād fī l-Iʿtiqād),  ed. H. Atay and I. Çubukçu (Ankara: Nur Matbaasi, 1962). al-Ghazālī, Abū Ḥ āmid. Moderation in Belief (al-iqtiṣād fī al-iʿtiqād), tr. A. M. Yaqub (Chicago: University of Chicago Press, 2013).

al-Maqdisī, Mut ̣ahhar ibn Ṭāhir. Livre de la création et de l’histoire (Kitāb al-Badʾ wa-l-Tārīkh), ed. and tr. C.  Huart, Part IV (Paris: Ernest Leroux, 1907). al-Rāzī, Abū Ḥ ātim. The Proofs of Prophecy: A Parallel English-Arabic Text, tr. T. Khalidi (Provo, UT: Brigham Young University Press, 2012). Antes, Peter. Prophetenwunder in der Ašʿarīya bis al-Ġazālī (Freiburg: Schwarz Verlag, 1970). Arnauld, Antoine and Pierre Nicole. Logic or the Art of Thinking, ed. and tr. J. V. Buroker (Cambridge: Cambridge University Press, 1996). Bernand-Baladi, Marie. “Des critères de la certitude: un opuscule de Ḥ asan Ibn Sahl sur la crédibilité du dire transmis par un grand nombre,” Journal asiatique 257 (1969), 95–138. Black, Deborah  L. Logic and Aristotle’s Rhetoric and Poetics in Medieval Arabic Philosophy (Leiden: Brill, 1990). Dunlop, Douglas Morton. “Al-Fārābī’s Introductory Sections on Logic,” The Islamic Quarterly 2 (1955), 264–82. El-Rouayheb, Khaled. “Must God Tell us the Truth? A Problem in Ashʿarī Theology,” in B.  Sadeghi, A.  Q.  Ahmed, A.  Silverstein, and R.  Hoyland (eds.), Islamic Cultures, Islamic Contexts: Essays in Honor of Professor Patricia Crone (Leiden: Brill, 2016), 411–30. Endress, Gerhard. “Ibn al-Ḫammār,” in U.  Rudolph, R.  Hansberger, and P.  Adamson (eds.), Philosophy in the Islamic World, Vol. 1, 8th–10th Centuries (Leiden: Brill, 2017), 480–90. Endress, Gerhard. “Ibn al-Samḥ,” in U.  Rudolph, R.  Hansberger, and P.  Adamson (eds.), Philosophy in the Islamic World, Vol. 1, 8th–10th Centuries (Leiden: Brill, 2017), 490–6. Endress, Gerhard. “ʿĪsā Ibn Zurʿa,” in U.  Rudolph, R.  Hansberger, and P.  Adamson (eds.), Philosophy in the Islamic World, Vol. 1, 8th–10th Centuries (Leiden: Brill, 2017), 468–80.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  73 Endress, Gerhard. “Yaḥyā Ibn ʿAdī,” in U.  Rudolph, R.  Hansberger, and P.  Adamson (eds.), Philosophy in the Islamic World, Vol. 1, 8th–10th Centuries (Leiden: Brill, 2017), 435–68. Ferrari, Cleophea. “Abū l-Faraǧ Ibn al-Ṭayyib,” in U.  Rudolph, R.  Hansberger, and P.  Adamson (eds.), Philosophy in the Islamic World, Vol. 1, 8th–10th Centuries (Leiden: Brill, 2017), 496–506. Gimaret, Daniel. La doctrine d’ al-Ašʿarī. (Paris: Cerf, 1990). Griffel, Frank. “Al-Ġazālī’s Concept of Prophecy: The Introduction of Avicennian Psychology into Ašʿarite Theology,” Arabic Sciences and Philosophy 14 (2004), 101–44. Griffel, Frank. Al-Ghazālī’s Philosophical Theology (Oxford: Oxford University Press, 2009). Griffel, Frank. “On the Character, Content, and Authorship of Itmām Tatimmat Ṣiwān al-ḥikma and the Identity of the Author of Muntakhab Ṣiwān al-ḥikma,” Journal of the American Oriental Society 133 (2013), 1–20. Griffith, Sidney  H. “Comparative Religion in the Apologetics of the First Christian Arabic Theologians,” Proceedings of the PMR Conference: Annual Publication of the Patristic, Mediaeval and Renaissance Conference 4 (Villanova, PA: Augustinian Historical Institute, 1979), 63–87. Griffith, Sidney  H. “Faith and Reason in Christian Kalām. Theodore Abū Qurrah on Discerning the True Religion,” in S. K. Samir and J. S. Nielsen (eds.), Christian Arabic Apologetics during the Abbasid Period (750–1258) (Leiden: Brill, 1994), 1–43. Griffith, Sidney  H. “Answering the Call of the Minaret: Christian Apologetics in the World of Islam,” in J. J. Van Ginkel, H. L. Murre - Van den Berg, and T. M. Van Lint (eds.), Redefining Christian Identity: Cultural Interactions in the Middle East since the Rise of Islam (Leuven: Peters, 2005), 91–126. Haddad, Cyrille. ʿĪsā ibn Zurʿa: Philosophe arabe et apologiste chrétien (Beirut: Dār al-Kalima, 1971). Houston, Joseph. Reported Miracles: A Critique of Hume (Cambridge: Cambridge University Press, 1994). Hume, David. Of Miracles (La Salle, IL: Open Court Classic, 1985). Jaffer, Tariq. Rāzī: Master of Qurʾanic Interpretation and Theological Reasoning (Oxford: Oxford University Press, 2015). Keating, Sandra Toenies. Defending the ‘People of Truth’ in the Early Islamic Period: The Christian Apologies of Abū Rāʾiṭah (Leiden: Brill, 2006).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

74  Alexander Lamprakis Khalidi, Tarif. “Muʿtazilite Historiography: Maqdisī’s Kitāb al-Badʾ wa-lTārīkh,” Journal of Near Eastern Studies 35 (1976), 1–12. Kraemer, Joel L. Humanism in the Renaissance of Islam: The Cultural Revival During the Buyid Age (Leiden: Brill, 1992). Lameer, Joep. Conception and Belief in Ṣadr al-Dīn Shīrāzī (ca. 1571–1635), al-Risāla fī l-taṣawwur wa-l-taṣdīq (Tehran: Iranian Institute of Philosophy, 2006). Lameer, Joep. “Ghayr al-maʿlūm yantaniʿ al-ḥukm ʿalayhi: An Exploratory Anthology of a False Paradox in Medieval Islamic Philosophy,” Oriens 41 (2014), 397–453. Locke, John. An Essay Concerning Human Understanding, ed. P. H. Nidditch (Oxford: Oxford University Press, 1999). Marmura, Michael. “Avicenna’s Theory of Prophecy in the Light of Ashʿarite Theology,” in W.  S.  McCullough (ed.), The Seed of Wisdom: Essays in Honour of T. J. Meek (Toronto: University of Toronto Press, 1964), 159–78. Masrī, Pierre and Nadine Abbas. “La relation entre la philosophie et la reli­ gion: traité écrit par Abū ʿAlī ʿĪsā Ibn Isḥāq Ibn Zurʿa, à certains de ses frères, dans lequel il démontre l’innocence de ceux qui s’adonnent à la logique et à la philosophie, de l’accusation de corruption de la religion” (in Arabic), al-Machriq 92 (2018), 365–99. Menn, Stephen. “The Discourse on the Method and the Tradition of Intellectual Autobiography,” in J. Miller and B. Inwood (eds.), Hellenistic and Early Modern Philosophy (Cambridge: Cambridge University Press, 2003), 141–91. Paget, James Carleton. “Miracles in Early Christianity,” in G. H. Twelftree (ed.), The Cambridge Companion to Miracles (Cambridge: Cambridge University Press, 2011), 131–43. Périer, Augustin. Yaḥyā ben ʿAdī: un philosophe arabe chrétien du Xe siècle (Paris: Paul Geuthner 1920). Rahman, Fazlur. Prophecy in Islam: Philosophy and Orthodoxy (London: Routledge 2008). Rashed, Marwan. “Abū Bakr al-Rāzī et la Prophétie,” MIDEO 27 (2008), 169–81. Rashed, Marwan. “New Evidence on the Critique of the Qurʾānic Miracle at the End of the Third/Ninth Century: Qust ̣ā ibn Lūqā vs. the Banū alMunajjim,” in P. Adamson (ed.), In the Age of al-Fārābī: Arabic Philosophy in the Fourth/Tenth Century (London: Warburg Institute, 2008), 277–95.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Believing Miracles in 10th–11th Century Baghdad  75 Rescher, Nicholas. “A Tenth-Century Arab-Christian Apologia for Logic,” in Studies in the History of Arabic Logic (Pittsburgh, PA: University of Pittsburgh Press, 1963), 55–63. Roggema, Barbara. “Kayfiyyat idrāk ḥaqīqat al-diyāna, ‘How to Discern the Truth of a Religion,’” in D.  Thomas and B.  Roggema (eds.), ChristianMuslim Relations: A Bibliographical History, vol. 1 (Leiden: Brill 2009), 775–9. Rosenthal, Franz. Knowledge Triumphant: The Concept of Knowledge in Medieval Islam (Leiden: Brill, 2007). Rudolph, Ulrich. “Al-Fārābī und die Muʿtazila,” in S. Schmidtke, C. Adang, and D.  Sklare (eds.), A Common Rationality: Muʿtazilism in Islam and Judaism (Würzburg: Ergon Verlag, 2007), 59–80. Samir, Khalil. “Ḥ unayn b. Isḥāq’s ‘On the Discernment of Religion’ (Maqāla Ḥ unayn b. Isḥāq ‘fī Kayfīyat idrāk ḥaqīqat al-diyāna)” (in Arabic), alMashriq 71 (1997), 345–63. Samir, Khalil (ed.) and Paul Nwiya (tr.). Une correspondance islamo-chrétienne entre Ibn al-Munajjim, Ḥ unayn ibn Isḥāq et Qust ̣ā ibn Lūqā (Turnhout: Brepols, 1981). Sarrió Cucarella, Diego. “On How to Discern the Truth of Religion, by Ḥ unayn b. Isḥāq: The Impersonal Recension,” Islamochristiana 45 (2019), 153–63. Sbath, Paul. Vingt traités philosophiques et apologétiques d’auteurs arabes chrétiens du IXe au XIVe siècle (Cairo: H. Friedrich, 1929). Shihadeh, Ayman. The Teleological Ethics of Fakhr al-Dīn al-Rāzī (Leiden: Brill, 2006). Starr, Peter Jonathan. The Epistle to Bišr b. Finḥās (Maqālah ʿamilahā ilā Bišr b. Finḥās) of Ibn Zurʿah (m. A.  H.  398/A.  D.  1008) (Ph.D.  dissertation, Cambridge University, 2000). Stern, Samuel Miklós. “Ibn al-Samḥ,” The Journal of the Royal Asiatic Society of Great Britain and Ireland 88 (1956), 31–44. Stroumsa, Sarah. Freethinkers of Medieval Islam: Ibn al-Rāwandī, Abū Bakr al-Rāzī, and Their Impact on Islamic Thought (Leiden: Brill, 1999). Stroumsa, Sarah. “The Signs of Prophecy: The Emergence and Early Development of a Theme in Arabic Theological Literature,” Harvard Theological Review 78 (1985), 101–14. Thomas, David. “Early Muslim Responses to Christianity,” in D.  Thomas (ed.), Christians at the Heart of Islamic Rule (Leiden: Brill, 2003), 231–55.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

76  Alexander Lamprakis Troupeau, Gérard. “Traité sur la science et le miracle et fragments du Traité sur les fondements de la religion de ʿAbd Allāh ibn al-Ṯayyib,” in Études de civilisation médiévale, IXe–XIIe siècles (Poitiers: Centre d’etudes supé­ rieures de civilisation médiévale, 1974), 675–9. Troupeau, Gérard. “Le traité sur l’unité et la Trinité de ʿAbd Allāh Ibn al-Ṭayyib,” Parole de l’Orient 11 (1971), 71–91. Vallat, Philippe. “Between Hellenism, Islam, and Christianity: Abū Bakr al-Rāzī and His Controversies with Contemporary Muʿtazilite Theologians as Reported by the Ashʿarite Theologian and Philosopher Fakhr al-Dīn al-Rāzī,” in D. Janos (ed.), Ideas in Motion in Baghdad and Beyond: Philosophical and Theological Exchanges between Christians and Muslims in the Third/Ninth and Fourth/Tenth Centuries (Leiden: Brill, 2016), 178–220. Van Ess, Josef. “The Logical Structure of Islamic Theology,” in H. Biesterfeldt (ed.), Kleine Schriften by Josef van Ess (Leiden: Brill, 2018), 238–71. Van Ess, Josef. Theologie und Gesellschaft im 2. und 3. Jahrhundert Hidschra: Eine Geschichte des religiösen Denkens im frühen Islam, Band III (Berlin: De Gruyter, 1992). Van Ess, Josef. Theologie und Gesellschaft im 2. und 3. Jahrhundert Hidschra. Eine Geschichte des religiösen Denkens im frühen Islam, Band IV (Berlin: De Gruyter, 1997). Wakelnig, Elvira. “Ibn al-Samḥ, Abū ʿAlī,” in K.  Fleet, D.  Krämer, D.  Matringe, J.  Nawas, and E.  Rowson (eds.), Encyclopaedia of Islam, THREE (Leiden: Brill, 2019), 64–6. Walker, Paul  E. “The Political Implications of al-Rāzī’s Philosophy,” in C. Butterworth (ed.), The Political Aspect of Islamic Philosophy: Essays in Honor of Muhsin S. Mahdi (Cambridge, MA: Harvard University Press, 1992), 61–94. Ward, Benedicta. Miracles and the Medieval Mind: Theory, Record and Event, 1000–1215 (Aldershot: Scolar Press, 1982). Woerther, Frédérique. “al-Fārābī and the Didascalia,” in A.  Alwishah and J. Hayes (eds.), Aristotle and the Arabic Tradition (Cambridge: Cambridge University Press, 2015), 92–104. Yaḥyā ben ʿAdī. Petits traités apologetiques: texte arabe, ed. and tr. Augustin Périer (Paris: Paul Geuthner, 1920).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time Revisiting Avicenna’s Account of the Now Andreas Lammer

»Völlig gleich wie oft du auf die Uhr schaust, es bleibt jetzt.« from “Ab heute immer jetzt” by Spaceman Spiff

The philosophy ofAvicenna (d. 428/1037) has long sparked the interest of modern interpreters in the West. It was in particular his major works on metaphysics and psychology, available also in Latin translation, that gradually but unanimously came to be regarded as “milestones” in the development of Western philosophy.1 More recently, the appreciation of Avicenna turned into a fascination with Ibn Sīnā as a Muslim author whose Arabic and Persian oeuvre, containing also treatises that were not—or only partially—translated into Latin, was considered valuable and appealing even beyond its direct or mediated influence upon Latin Scholasticism, either because of its intrinsic philosophical quality or because of its history of both sustained and intense reception in the Islamic world up to our modern times. As a result, the last three years have seen the publication of no less than fourteen monographs in 1  Classical studies in this regard are Amos Bertolacci, The Reception of Aristotle’s Metaphysics in Avicenna’s Kitāb al-­Šifāʾ: A Milestone of Western Metaphysical Thought (Leiden: Brill, 2006); Dag Nikolaus Hasse, Avicenna’s De anima in the Latin West: The Formation of a Peripatetic Philosophy of the Soul, 1160–1300 (London: Warburg Institute, 2000). Andreas Lammer, Now Is Not the Time: Revisiting Avicenna’s Account of the Now In: Oxford Studies in Medieval Philosophy Volume 9. Edited by: Robert Pasnau, Oxford University Press. © Andreas Lammer 2021. DOI: 10.1093/oso/9780192844637.003.0003

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

78  Andreas Lammer Western languages explicitly dedicated to various aspects of Avicenna’s thought and influence.2 One of the philosophical disciplines that in recent years has received a larger share of global attention in the investigation of Avicenna is physics—and within physics, it was the theory of time that proved to be a major incentive for interpreters.3 Now that there has been an increased 2  As far as I am aware, these are Tommaso Alpina, Subject, Definition, Activity: Framing Avicenna’s Science of the Soul (Berlin: De Gruyter, 2021); Mohammad Azadpur, Analytic Philosophy and Avicenna: Knowing the Unknown (London: Routledge, 2020); Fedor Benevich, Essentialität und Notwendigkeit: Avicenna und die aristotelische Tradition (Leiden: Brill, 2018); Silvia Di Vincenzo, Avicenna, The Healing, Logic: Isagoge: A New Edition, English Translation and Commentary of the Kitāb al-­Madḫal of Avicenna’s Kitāb al-­Šifāʾ (Berlin: De Gruyter, 2021); Daniel De Haan, Necessary Existence and the Doctrine of Being in Avicenna’s Metaphysics of the Healing (Leiden: Brill, 2020); Laura Hassan, Ashʿarism Encounters Avicennism: Sayf al-­Dīn al-­ Āmidī on Creation (Piscataway, NJ: Gorgias Press 2020); Damien Janos, Avicenna on the Ontology of Pure Quiddity (Berlin: De Gruyter, 2020); Alexander Kalbarczyk, Predication and Ontology: Studies and Texts on Avicennian and Post-­ Avicennian Readings of Aristotle’s Categories (Berlin: De Gruyter, 2018); Nora Kalbarczyk, Sprachphilosophie in der islamischen Rechtstheorie. Zur avicennischen Klassifikation der Bezeichnung bei Faḫr ad-­dīn ar-­Rāzī (gest. 1210) (Leiden: Brill, 2018); Andreas Lammer, The Elements of Avicenna’s Physics: Greek Sources and Arabic Innovations (Berlin: De Gruyter, 2018); Michael Noble, Philosophising the Occult: Avicennan Psychology and “The Hidden Secret” of Fakhr al-­Dīn al-­Rāzī (Berlin: De Gruyter, 2021); Sari Nusseibeh, Avicenna’s al-­Shifāʾ: Oriental Philosophy (New York: Routledge, 2018); Ahmed H. al-­Rahim, The Creation of Philosophical Tradition: Biography and the Reception of Avicenna’s Philosophy from the Eleventh to the Fourteenth Century  A.D. (Wiesbaden: Harrassowitz, 2018); Riccardo Strobino, Avicenna’s Theory of Science: Logic, Metaphysics, Epistemology (Berkeley: University of California Press, 2021). 3  The growing interest in Avicenna’s physics is borne out by the recent publication of Dag Nikolaus Hasse and Amos Bertolacci (eds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Physics and Cosmology (Berlin: De Gruyter, 2018), which—among other things— features two contributions on the Arabic reception of Avicenna’s temporal theory (viz., Adamson, “The Existence of Time in Faḫr al-­Dīn al-­Rāzī’s al-­Mat ̣ālib al-­ʿāliya,” 65–99 and Lammer, “Time and Mind-­Dependence in Sayf al-­Dīn al-­Āmidī’s Abkār al-­afkār,” 101–61). Arabic, German, and English studies of (various aspects of) Avicenna’s account of time from the last thirty years (or so) further include ʿAlāʾ al-­Dīn Muḥammad ʿAbd al-­Mutaʿāl, Taṣawwur Ibn Sīnā li-­l-­zamān wa-­uṣūluhū l-­yūnāniyya (Alexandria: Dār al-­Wafāʾ li-­Dunyā l-­Ṭibāʿa wa-­l-­ Nashr, 2003); Muḥammad ʿĀt ̣if al-­ʿIrāqī, al-­Falsafa al-­t ̣abīʿiyya ʿinda Ibn Sīnā (Cairo: Dār al-­ Maʿārif, 1983), 232–59; Udo Reinhold Jeck, Aristoteles contra Augustinum: Zur Frage nach dem Verhältnis von Zeit und Seele bei den antiken Aristoteleskommentatoren, im arabischen Aristotelismus und im 13. Jahrhundert (Amsterdam: B. R. Grüner, 1994), 103–13; Toby Mayer, “Avicenna against Time Beginning: The Debate between the Commentators on the Ishārāt,” in P.  Adamson (ed.), Classical Arabic Philosophy: Sources and Reception (London: Warburg Institute, 2007), 125–49; Jon McGinnis, “Ibn Sīnā on the Now,” American Catholic Philosophical Quarterly 73 (1999), 73–106; “The Topology of Time: An Analysis of Medieval Islamic Accounts of Discrete and Continuous Time,” The Modern Schoolman 81 (2003), 5–25; Time and Time Again: A Study of Aristotle and Ibn Sīnā’s Temporal Theories (Ph.D. dissertation, University of Pennsylvania, 1999); “Time to Change: Time, Motion and Possibility in Ibn Sīnā” in N. Bayhan, M.  Mazak, and A.  Süleyman (eds.), Uluslararası İbn Sīnā Sempozyumu Bildirileri, vol. 1 (Istanbul: İstanbul Büyükşehir Belediyesi Kültür A.Ş. Yayınları, 2009), 251–7; Andreas Lammer, arabisch-­ lateinische Rezeption der aristotelischen “Anzahl und Ausmaß: Die griechisch-­

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  79 interest in this specific topic, and a certain number of publications dedicated to it, it is advisable to step back a little, in order to assess the results and identify major lines of interpretation and agreement. One such universal agreement in the interpretation of Avicenna’s temporal theory is that time is to be regarded as the product of a ‘flowing now.’ This, at least, is what we are told by Muḥammad ʿĀt ̣if al-­ʿIrāqī, Yegane Shayegan, Jon McGinnis, ʿAlāʾ al-­Dīn Muḥammad ʿAbd al-­Mutaʿāl, and Toby Mayer.4 It is clear that this is no minor point in the interpretation of Avicenna’s theory of time and the understanding of his philosophical system. If time is said to be the product of a flowing now, then this means that time acquires its existence from another concept, viz., the ‘now.’ In consequence, such a claim concerns the very existence of time, i.e., the very existence of one of the most fundamental concepts in both philosophy and the human experience of reality. Of course, that the existence of one thing (in this case time) depends on the existence of another (in this case the ‘now’) or on one of its specific features (in this case the now’s ‘flowing’) is not problematic in itself. Yet, claiming such a dependence with regard to a concept as fundamental as time may well have repercussions for other theories as well—to mention just one prominent (or devastating) example: the eternity of the world. If, as Abū Ḥ āmid al-­Ghazālī (d. 505/1111) prominently described in his masterpiece The Incoherence of the Philosophers, the existence of the world depended on the existence of time, then literally everything (besides God) would appear to depend on the existence of time—and if time itself depended on the flow of a now, then ultimately everything would appear to depend on a now and its flow. Once more, this may itself perhaps not be problematic, were it not for the intrinsically unclear ontological status of such a concept as the now, let alone its at least equally unclear engagement in the activity of ‘flowing.’ We may well ask: what is ‘the now’ in the first place? Where did it

Zeitdefinition,” Das Mittelalter: Perspektiven mediävistischer Forschung 23 (2018), 109–27; Yegane Shayegan, Avicenna on Time (Cambridge, MA: Harvard University Press, 1986). 4  Cf. al-­ʿIrāqī, al-­Falsafa al-­t ̣abīʿiyya ʿinda Ibn Sīnā, 249–54; Shayegan, Avicenna on Time, 188–212; McGinnis, “Ibn Sīnā on the Now,” 97–106 (a reworked version of materials from chapter eight of Time and Time Again); ʿAbd al-­Mutaʿāl, Taṣawwur Ibn Sīnā li-­l-­zamān, 232–5 (following al-­ʿIrāqī); Mayer, “Avicenna against Time Beginning,” 142f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

80  Andreas Lammer come from? (Did God create it?) How did it end up being engaged in such a motion? (Did God put it into motion?) Where exactly does it flow? (And at what speed?) Such questions would have to be answered if one wanted to argue that time is the product of a flowing now with all its entailments and consequences for the world at large. Recently, I provided what I have called a “full” interpretation of Avicenna’s temporal theory and explained both time’s essence and existence, yet without having taken recourse to the now.5 Thus, my in­ter­pret­ ation stands in stark contrast to what could well be described as the current consensus among experts in Avicennian natural philosophy (viz., that the now is crucial, because its flow creates time). My aim in this paper is to address this contrast and to argue that the current consensus is mistaken—a consensus, it should be added, that has been reached on the basis of a historical investigation of Avicenna’s Greek predecessors and a philosophical investigation of Avicenna’s own writings. Perhaps the most thorough account that shaped the current consensus is chapter eight of Jon McGinnis’ impressive doctoral dissertation from 1999, published in the same year in a slightly modified form as an article called “Ibn Sīnā on the Now.” McGinnis’ overall thesis, which has gained universal acceptance, is that there are certain passages in Aristotle’s Physics that have influenced later commentators, especially Alexander of Aphrodisias (fl. ~ 200) and John Philoponus (d. ~ 575), in such a way that they came to argue that time is generated by a flowing now and that Avicenna, in turn influenced by reading the Arabic translations of Alexander’s and Philoponus’ commentaries, likewise came to argue that the now “produces time . . . through its flow.”6 In brief, the idea of a flowing now is that the now is a moment or instant that is always present. Since each moment that once was ‘now’ has been a different moment, while still having been once a present now, the now could be seen as constantly advancing and developing, so that it is both always the same (because it is always present) and always different (because it is ever marking a different moment). The moving now,

5 Lammer, The Elements of Avicenna’s Physics, 429–524. 6  McGinnis, “Ibn Sīnā on the Now,” 98f., 101; cf. generally McGinnis, Time and Time Again, esp. ch. 6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  81 then, is on the whole not unlike the seconds hand on a watch, which likewise is always the same, even though it constantly moves on, always pointing at another—further—position. Time, then, is regarded as the result of the now’s flow: it is what comes to be when the now, being always the same, advances in its motion, thus being always different. It is immediately apparent, even against the above listed questions that may (or may not) be irritating, that the conception of time as the result of such a flowing now—or as being itself nothing but that flow— answers to two powerful features of our common experience of time. On the one hand, time seems to be something that incessantly progresses forth. The now, which is always the same and also always different, captures this impression adequately. On the other hand, people often visualize or illustrate the extent of time by means of an extended time line. Such a line both symbolizes the continuous nature of time and helps us understand what we mean when we say that we had to wait ‘for five minutes’ (represented by a limited segment on the time line) and subsequently walked from A to B ‘in one hour’ (represented by a considerably longer segment on that same time line). On the whole, then, it is quite plausible to say that the incessantly progressing nature of time together with our strong sense of the present as well as the belief that time can be expressed—or expresses itself—in quantities is aptly portrayed by the idea that time is what is generated through the constant motion of an ever-­present temporal moment in analogy to a drawn line, which analogously can be said to be the product of a moving point (such as the moving tip of a ballpoint pen)—a prominent image also mentioned by Aristotle in his De anima.7 The contemporary discussion about Aristotle appears to have shifted from asking whether or not the now is moving (and flowing) to the related question whether time should be conceived of as dynamic (and flowing) or as static (and stable).8 Accordingly, the notion of a flowing 7 Cf. De an. I.4, 409a3–6. 8  Dynamic interpretations are offered for example by Fred D. Miller, Jr., “Aristotle on the Reality of Time,” Archiv für Geschichte der Philosophie 56 (1974), 132–55; Richard Sorabji, Time, Creation and the Continuum: Theories in Antiquity and the Early Middle Ages (Chicago: University of Chicago Press, 1983), ch. 4; Sarah Broadie, “Aristotle’s Now,” The Philosophical Quarterly 34 (1984), 104–28; “A Contemporary Look at Aristotle’s Changing Now,” in R. Salles (ed.), Metaphysics, Soul, and Ethics in Ancient Thought: Themes from the Work of Richard

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

82  Andreas Lammer now as such has by and large vanished from the available secondary literature on Aristotle. That is not the case for Avicenna; here the flowing now is very much en vogue and, indeed, initially it seems that the image of a moving point producing a line—on which the notion of a flowing now relies—could be more central to Avicenna’s concerns than to Aristotle’s. The reason for this is that Avicenna identified a different feature as the core of what time is than Aristotle had done. While Aristotle investigated in the fourth book of the Physics how time is a number or measure of motion, Avicenna developed his view of time against the background of the sixth book of the Physics (which investigates the essential features of continuous magnitudes) alongside late ancient Neoplatonic developments (which emphasized time as a—perhaps independent—quantity or “duration” measured by motion).9 As a result, Avicenna argued that time is a magnitude in the category of quantity which stretches from the beginning of a motion to its end.10 Whether or not this magnitude is already measured or numbered is for his conception secondary—and this is precisely one of the main differences from Aristotle, for whom time is primarily that measure or Sorabji (Oxford: Oxford University Press, 2005), 81–93; critical are Wolfgang Wieland, Die aristotelische Physik: Untersuchungen über die Grundlegung der Naturwissenschaft und die sprachlichen Bedingungen der Prinzipienforschung bei Aristoteles (Göttingen: Vandenhoeck & Ruprecht, 1992), 326f.; Norman Kretzmann, “Aristotle on the Instant of Change (II),” Proceedings of the Aristotelian Society: Supplementary Volumes 50 (1976), 91–114; Jon McGinnis, “Making Time Aristotle’s Way,” Apeiron 36 (2003), 143–69; cf. also the monographic analyses by Ursula Coope, Time for Aristotle: Physics IV.10–14 (New York: Oxford University Press, 2005), esp. ch. 8 (critically discussing Hussey’s adherence to the “moving-­now in­ter­pret­ ation” advanced in his commentary on Physics: Books III and IV) and Tony Roark, Aristotle on Time: A Study of the Physics (Cambridge: Cambridge University Press, 2011), esp. ch. 12, both leaning towards the former camp and discussing the literature. The contemporary discussion is frequently pursued as a response, in one way or another, to McTaggart’s famous investigation of time. In the following, I shall avoid addressing, or even approaching, the question whether Aristotle, Avicenna, or other mentioned philosophers adhered to an A-­series or a B-­series of time and how their overall conception relates to these. What is true especially for Avicenna, and perhaps similarly for Aristotle, is that motion serves as a basic feature of reality and as the ultimate explanans of time (and not the other way around), thus marking a stark difference to McTaggart’s basic assumptions; cf. Roark, Aristotle on Time, who develops a reading of Aristotle along these lines, recommending Aristotle’s theory as a source of “inspiration” (207) for contemporary A-­theorists. 9  Cf. Lammer, The Elements of Avicenna’s Physics, 432–51 for the former aspect and 451–62 for the latter. 10  Ultimately, the motion in question is, of course, that of the outermost sphere. So, time is actually a magnitude of a circular motion and, thus, without beginning and end; q.v. below, 109ff.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  83 number. Moreover, Avicenna is adamant—as a result of his novel account of motion—that the magnitude that is time does not exist in its entirety as actual: the past is no longer existent, the future is not yet existent, only the present moment exists as actual. Still, time does exist, it is extended, and is as such a feature of reality, even though it is not an extended reality, for the extended parts of time are non-­ existent. Nonetheless, we can picture time in its extended form in our imagination, e.g., as a memory from the past or a plan for the future.11 Since Avicenna generally emphasized the extended, quantitative nature of time and since, further, the prominent image of an extended line being drawn out by a point moving from a start to an end equally highlights the continuously extended nature of that line, Avicenna emerges as a prime suspect for the flowing now thesis. Accordingly, no one would be surprised if an investigation of Avicenna’s temporal theory would arrive at the conclusion that he approached time on the basis of the well-­ known image of a point-­ like now which flows from one moment to another and, thereby, draws out, and thus produces, time as a continuously extended quantity. Indeed, one would be even less surprised to hear that Avicenna himself mentioned the flow of the now in his own investigation and would treat this as further support. This is exactly what happened. That there are also strong reasons for denying Avicenna’s adherence to the flowing now, however, has so far escaped the attention of interpreters. What I shall do in this article is to discuss briefly the canonical ­passages in Aristotle that are also central to the interpretations by Alexander, Philoponus, and Simplicius (d. ~ 560), in order to provide the historical background to Avicenna’s later discussion.12 Thereafter, I shall turn to Avicenna and examine the claims advanced in the secondary literature about his alleged adherence to the flowing now as well as 11  On the manner of time’s existence in Avicenna, cf. esp. The Lecture on Nature II.13, §1, 167.4–10. 12  All Greek primary sources mentioned in this article were available in Arabic translation (except for Plato’s Timaeus, which was available only via Galen, and probably Simplicius’ commentary on the Physics). Apart from Aristotle’s texts and Alexander’s treatise on time, these Arabic translations are not known to be extant; cf. Lammer, The Elements of Avicenna’s Physics, 9–34 for a detailed survey. Translations of Greek authors will, thus, always be provided from their original Greek works, unless otherwise noted.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

84  Andreas Lammer the relevant passages from his works, concluding—against the current consensus—that for Avicenna, time is not produced by the now through its flow.13

1.  Aristotle and His Commentators on the Now and Its Relation to Time From the outset, it seems unlikely that Aristotle’s Physics in any way could provide textual support for the claim that time is nothing but the result of a flowing now. The reason for this lies in two claims Aristotle advanced in his Physics: first, he states that time is not a motion (and a flow would be a motion); and second, he shows that the now cannot move (and to flow would mean to move).14 Aristotle argues for the first claim right at the beginning of his discussion of time in Physics IV.10. There, he expounds a number of opinions that have been advanced by his predecessors on the subject of time, all of whom failed to state clearly what time is and to reveal its nature. The most common of these opinions, Aristotle writes, is the view that time is motion or some kind of change (κίνησις εἶναι καὶ μεταβολή τις ὁ χρόνος).15 Aristotle, however, states that time cannot be motion due to a number of essential differences that outweigh their admitted similarities. For example, motion can be fast or slow, whereas time cannot. He concludes the chapter saying that it is “evident” (φανερόν) that time is not motion.16 On that basis 13  One point that should no longer have to be emphasized, but I feel that it is still necessary, is that in most of his extant works, and in particular in his eight major summae of philosophy, Avicenna is no commentator on Aristotle. As should be clear immediately to anyone who reads these texts, and as we learn from his autobiography beyond doubt, Avicenna does not understand his own activity—both as a philosopher and as an author—as that of commenting on the text of Aristotle’s works. Consequently, any subsequent claims about whether or not Avicenna considers time to be the product of a flowing now are claims about Avicenna’s philosophy and not about how he interpreted Aristotle’s views or even less about how he thought they ought to be understood. For Avicenna’s own views on the purpose of his own philosophical activity, cf. Dimitri Gutas, Avicenna and the Aristotelian Tradition: Introduction to Reading Avicenna’s Philosophical Works (Leiden: Brill, 2014), 323–34. 14  Cf. Coope, Time for Aristotle, 133. 15  Phys. IV.10, 218b9f.; the Arabic translation, produced by Isḥāq ibn Ḥ unayn in the late third/late ninth century, was edited by Badawī. 16  Phys. IV.10, 218b18. Immediately afterwards, we read at the beginning of the next chapter that time “cannot be without” (οὐδ’ ἄνευ) motion either (Phys. IV.11, 218b21).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  85 it would seem to be unfit to claim that time is the flow of a now, for a flow is a kind of motion, which would, then, make time some kind of motion or change, which is precisely what he has just denied.17 The second point that seems to eliminate the possibility for viewing time as the flow of the now is that, in the sixth book of the Physics, Aristotle straightforwardly argues that what is indivisible cannot move—and the now, indeed, is indivisible.18 What is more, he even directly discusses the now in this context and concludes that it is impossible for something that is indivisible like a point to engage in motion. How, then, could the now, which Aristotle also explicitly and repeatedly relates to a point (στιγμή), be something that moves or flows? Despite these strong prima facie indications against time being the result of a flowing now, Aristotle’s discussion in Physics IV.10 features the following famous puzzle: It is not easy to see whether [the now] always remains one and the same or whether it is other and other (ἄλλο καὶ ἄλλο). (Phys. IV.10, 218a9–11)

Aristotle’s solution to this somewhat Heraclitean puzzle provides the textual basis for interpreting the now as something that flows, for he argues that the now is in fact both: in one way it remains the same and in another it does not. In doing so, Aristotle distinguishes between what the now is as such or in substrate or—as Aristotle likes to call it—as “that, whatever it is by being which the now is” (ὃ δέ ποτε ὄν ἐστι τὸ νῦν), and between what the now is in its being (τὸ δ’ εἶναι αὐτῷ) or in description (τῷ λόγῳ).19 The now in the former sense remains always the same, because the now in itself is simply always the present moment that is 17 Cf. also McGinnis’ final worry about the speed of the now’s flow in Time and Time Again, 311–15. 18 Cf. Phys. VI.10, 240b8–241a26; VI.3, 233b33–234a24; see also below, 113ff. 19  Phys. IV.10, 219b10–21. Different translations were offered for τῷ λόγῳ. Hardie/Gaye, Hussey, and Waterfield, for example, translate it as “definition,” whereas Wicksteed/Cornford speak about a “relation,” which seems to harmonize with Isḥāq ibn Ḥunayn’s Arabic rendering of the expression as bi-­l-­qiyās. For the meaning, and translation, of ὃ δέ ποτε ὄν ἐστι τὸ νῦν, cf. Coope’s discussion in the appendix of her Time for Aristotle, 173–7 and the remarks by McGinnis in Time and Time Again, 161f. as well as the comments by Themistius (In Phys., 150.3–7), Philoponus (In Phys., 726.27f.), and Simplicius (In Phys., 721.30f.).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

86  Andreas Lammer ‘now.’ In this sense it does not change, for whatever moment one may choose, this moment always is (or once has been) ‘now.’ In the latter sense, however, the now appears to be something that progresses or moves along or, indeed, flows, for whatever present moment one may choose, this moment, despite being always ‘now,’ is also always a different moment: there was a now seven minutes ago, there was a now five minutes ago, and there is in this very moment another now—all of these are the same insofar as they have been ‘now,’ but they are different insofar as they have been different moments in time. Aristotle’s strategy to answer the above-­quoted either-­or-­question “whether [the now] always remains one and the same or whether it is other and other” by saying, in his characteristic way, that actually both disjuncts are correct can certainly serve as a first textual basis for interpreting the now as something that flows. Accordingly, the notion of a flowing now is not based on the conception that at every moment a now comes into and passes out of existence, which in any case would invite the well-­known issues from Physics IV.10 about the precise moment of a now’s cessation.20 Instead, it employs the idea that there is one single now that is always the same as to what it is in itself, viz., being always ‘now,’ and always different as to its description or λόγος or “relation” (the Arabic translation has qiyās), viz., signifying an ever different moment. The passage that most strongly supports the idea of a flowing now is Aristotle’s analogy of the now and the thing-­in-­motion in Physics IV.11: Motion, as was said, follows (ἀκολουθεῖ) the magnitude and time [follows] motion, as we maintain. Likewise, the point [follows] the thing-­in-­motion by which we are aware of the motion and of the prior and posterior involved in it. That, whatever it is by being which [the  thing-­in-­motion] is, is the same, whether a point or a stone or something else of the kind, but in description it is different . . . The now follows the thing-­in-­motion, just as time follows the motion, for it is by means of the thing-­in-­motion that we become aware of the prior and posterior in the motion, and the prior and posterior as numerable is the now. Thus, also, that, whatever it is by being which the now is, is 20 Cf. Phys. IV.10, 218a3–25.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  87 the same, for it is the prior and posterior in motion, but its being is different, for the prior and posterior as numerable is the now . . . Thus, the now in one way is always the same, in another it is not the same, since the thing-­in-­motion is also [like this]. (Phys. IV.11, 219b15–33, tr. Hardie/Gaye, modified)

This long and often-­discussed passage is the very one that Philoponus and Simplicius later were commenting on when both suggested that time is the result of a flowing now, and we may well assume that it is also the passage that Alexander may have interpreted along the same lines in his lost commentary on the Physics.21 Aristotle’s invocation that motion “follows” the distance and that time, in turn, “follows” motion yields two important results. The first is that motion and time are essentially continuous and non-­atomic, because they ‘follow’ distance (which has been shown to be essentially continuous and non-­atomic). On that basis, Aristotle can conclude chapter IV.11 of his Physics with the following words: That time is, then, the number of motion in respect of the prior and posterior and that it is continuous, because it is of something continuous, is clear.  (Phys. IV.11, 220a24–6)

The second result Aristotle derives from the close relationship between distance, motion, and time is more concretely relevant for explaining the now and for his solution to the troubling question of how the now can be both always the same and always different. Just as time “follows” motion, there is also the now that “follows” the thing-­in-­motion (τῷ δὲ φερομένῳ ἀκολουθεῖ τὸ νῦν). More than anything else in Aristotle’s discussion it is this explicit correspondence between the now and the thing-­in-­motion together with the fact that the thing-­in-­motion is what is moving that fostered the idea of a flowing now: since there is no motion without something that performs this motion, there seems to be no time without something that ‘performs’ time. It is not too bold a move to interpret the relationship between the moving thing-­in-­motion 21  Alexander will be treated in more detail below, 89ff.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

88  Andreas Lammer and its motion, on the one hand, and the now and time, on the other, as a somewhat causal relationship, and to draw on terminology such as ‘to flow,’ ‘to produce,’ and ‘to generate,’ in order to explain and spell out this interpretation. This is exactly what Philoponus has done in the following comment on that passage: Indeed, it is the point that is productive (ποιητική) of magnitude, for the point generates (γεννᾷ) the line as it flows (ῥυϊσκόμενον) and the line is the primary magnitude, while the thing-­in-­motion is pro­duct­ ive of the motion . . . The now, too, is a productive cause of time (τὸ νῦν δὲ τοῦ χρόνου ποιητικὸν αἴτιον), since its flow generates time (ἡ γὰρ τούτου ῥύσις γεννᾷ τὸν χρόνον). So, if as the point stands to the magnitude and the thing-­in-­motion to the motion, so the now stands to time. (Philoponus, In Phys., 727.10–23, tr. Broadie, modified)

Simplicius signals his agreement with the interpretation of his contemporary and rival Philoponus and, commenting on the same Aristotelian passage, wrote the following: Since these things follow each other, the things that generate them will follow each other as well. It is the point which generates (γεννητική) the magnitude, when the line [is taken] to be the flow (ῥύσις) of a point . . . and the thing-­ in-­ motion is productive (ποιητικόν) of the motion and the now of time. As, then, the same point by flowing (ῥυϊσκομένη) produces (ποιεῖ) the magnitude . . . and as the identical thing-­in-­motion produces the motion, so the now, which is the same qua substrate, being numbered as prior and posterior, marks-­off and produces time (τὸ νῦν . . . τὸν χρόνον ὁρίζει τε καὶ ποιεῖ). (Simplicius, In Phys., 722.26–34, tr. Urmson, modified)

We can recognize two things here that are important. The first is that there is a strong tendency in the sixth century to employ the image of a flow (ῥύσις) in order to align the constant other-­and-­other-­ness of the now (ἄλλο καὶ ἄλλο) with its never-­changing identity in being ‘now’ by emphasizing the analogy between the now and the progressing activity of the thing-­in-­motion. This interpretation brings out the fact that

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  89 Aristotle, as mentioned, affirmed both seemingly mutually exclusive answers to the above question whether the now is always the same or always different.22 The now, he said, is simply both: in one way it is always the same and in another way it is always different. In support of his claim, Aristotle merely needed to point towards a thing undergoing motion: That, whatever it is by being which [the thing-­in-­motion] is, is the same . . . but it is different in description, as the sophists assume that Coriscus-­in-­the-­Lyceum is different from Coriscus-­in-­the-­marketplace. (Phys. IV.11, 219b18–21, tr. Hardie/Gaye, modified)

Only a sophist, and perhaps a Heraclitean evangelist, would claim that the one Coriscus is really different, i.e., different as an entity, from the other Coriscus. For Aristotle, the difference is merely a difference “in description,” as he would disagree with the sophist and reply that “that, whatever it is by being which” Coriscus is, remains the same throughout Coriscus’ stroll from the Lyceum to the marketplace. In other words, it is apparently the sameness of Coriscus that Aristotle wanted to emphasize here (rather than the motion or change of Coriscus).23 Consequently, it turns out that Aristotle employed the analogy of the now with the thing-­in-­motion not so much in order to illustrate the now as a moving

22  Another aspect that should not be overlooked is that the Neoplatonists’ attitude to read Aristotle’s discussion of the now in kinetic terms was textually influenced by Plato’s Timaeus and structurally by their efforts of reconciling Plato with Aristotle. In the Timaeus, the Demiurge intended to make “a moving likeness of eternity” and, thus, created time as an “eternal likeness moving according to number” (Tim. 37d5–7, tr. Cornford, modified). Consequently, a Neoplatonist inherited the notion of time as something moving and progressing already from Plato. Aristotle’s subsequent analogy between the now and the thing-­in-­ motion, then, constituted a more than convenient invitation to read Aristotle in Platonic terms; cf. Philoponus’ remark that time “has its being in becoming and in flowing” (ἐν τῷ γίνεσθαι καὶ ἐν ῥύσει τὸ εἶναι ἔχειν; In Phys., 735.24f.) and Damascius’ description of the “river of time” (apud Simplicium, In Phys., 798.17–24). On the meaning of the expression ἐν τῷ γίνεσθαι ἔχει τὸ εἶναι, see the comments in Lammer, The Elements of Avicenna’s Physics, fn. 37, 440; see also Avicenna’s remark in The Lecture on Nature II.13, §1, 167.4f. and in The Notes (al-­Taʿlīqāt), 142.16 (ed. Badawī)/§763, 423.11 (ed. Mousavian). 23 E.g. Hartmut Kuhlmann, “‘Jetzt’? Zur Konzeption des νῦν in der Zeitabhandlung des Aristoteles (Physik IV 10–14),” in E.  Rudolph (ed.), Zeit, Bewegung, Handlung: Studien zur Zeitabhandlung des Aristoteles (Stuttgart: Klett-­Cotta, 1988), 81; Sorabji, Time, Creation and the Continuum, 49.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

90  Andreas Lammer entity but to emphasize the essential sameness of the now at all points in time.24 In contrast, Philoponus and Simplicius took recourse to the powerful image of the flow, thus underscoring the motion of the now (rather than its sameness). In doing so, they arguably relied already on ancient authorities and, in particular, on the most important authority when it comes to commenting on Aristotle: Alexander of Aphrodisias. As is well known, Alexander’s commentary on the Physics is lost.25 However, in one of his treatises, the Maqāla fī l-­zamān (Treatise on Time), extant only in Arabic (and in a Latin translation from the Arabic), we read the following brief remark: The now, when it flows (sāla), makes (ʿamila) time. (Alexander of Aphrodisias, Maqāla fī l-­zamān, 21.13, 21.17)26

It is fair to assume that the Arabic verb sāla was the translator’s—the celebrated Ḥ unayn ibn Isḥāq (d. 260/873)—choice for translating words related to the Greek ῥεῖν and ῥύσις, which we have found in Philoponus and Simplicius.27 There is, thus, good reason to believe that it was Alexander who introduced, or at least propagated, the metaphor of the flow of the now as an illustration of Aristotle’s idea that the now is both always the same and always different. That Alexander understood the double nature of the now along kinetic terms is further borne out by a passage from Simplicius’ commentary reporting Alexander’s ex­plan­ ation of Aristotle’s statement that we are unable to recognize time when

24  We should keep in mind that Aristotle himself, as already mentioned, explicitly argued that no indivisible thing, such as the now, is able to move; cf. Phys. VI.10, 240b8–241a26; VI.3, 233b33–234a24; see also below, 113ff. 25 Surviving fragments have been edited by Rashed as Alexander of Aphrodisias, Commentaire perdu à la Physique d’Aristote or extracted from Arabic comments by Elias Giannakis, “Fragments from Alexander’s Lost Commentary on Aristotle’s Physics,” Zeitschrift für Geschichte der arabisch-­islamischen Wissenschaften 10 (1996), 157–87. 26  Cf. also Sharples’ translation and remarks in Alexander of Aphrodisias, “On Time”, 63 and 70, respectively. 27  For a brief discussion of the attribution to Ḥ unayn ibn Isḥāq, see Fritz Zimmermann and H. Vivian B. Brown, “Neue arabische Übersetzungstexte aus dem Bereich der spätantiken griechischen Philosophie,” Der Islam 50 (1973), 315.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  91 we do not perceive the now “as the same but in relation to a prior and posterior.”28 According to this report, Alexander said the following: The now is one and . . . the same, yet it is in some prior and posterior when we think of it as passing (παραγόμενον), as it were, and occurring sometimes at one, sometimes at another, stage of the motion. (Alexander of Aphrodisias apud Simplicium, In Phys., 713.14–16, tr. Urmson, modified)29

Although Alexander does not use the picture of a flow explicitly, he describes the now as “passing.” This participle seems to serve as just another metaphor that, like the metaphor of the flow from the Arabic translation of his treatise on time, describes the now as something that moves along, passes by, or progresses forth. Apparently, and just like Philoponus and Simplicius four centuries later, Alexander emphasized the moving aspect of the now (rather than its sameness). The second point we recognize in Simplicius’ and Philoponus’ in­ter­ pret­ ations of Aristotle’s analogy is that although Aristotle did not describe the now as that which produces or generates time, Philoponus and Simplicius did so explicitly. In particular, they used words derived from ποιεῖν (“to make, to produce”) and γεννᾶν (“to generate”). The Arabic version of Alexander’s treatise further provides ʿamila, which corresponds perhaps to the Greek ποιεῖν or ποιητική rather than to γεννᾶν or γεννητική. In addition to that, the marginal glosses on the Arabic translation of Aristotle’s Physics by Ḥ unayn’s son Isḥāq ibn Ḥ unayn (d. 298/910–11) as preserved in Ms. Leiden or. 583 contain a comment of “Yaḥyā”—here probably referring to Philoponus (in Arabic: Yaḥyā al-­Naḥwī) and not to his Baghdādī fellow miaphysite Christian Yaḥyā ibn ʿAdī (d. 364/974)—most likely in the translation of Qust ̣ā ibn Lūqā al-­Baʿlabakkī (d. 300/912), saying that the now is “generative” of time. The Arabic term used here, muwallid, is a likely translation of the

28  Phys. IV.11, 219a30–3, tr. Hardie/Gaye, modified. 29  See also Sharples’ fn. 65, 80 to his translation of Alexander of Aphrodisias, “On Time.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

92  Andreas Lammer Greek γεννητική.30 Other glosses describe the now as muḥdithan li-­l-­zamān (“productive of time”) or yuḥdithu l-­zamān (“producing the time”) and the point as tuḥdithu l-­khat ̣t ̣ (“producing the line”).31 Finally, Themistius also provides both Greek expressions, paraphrasing the Aristotelian text as follows: It has been stated that motion follows magnitude, and time [follows] motion. Now, it is reasonable [to say] that the principles of these, such as being productive and being generative of one another (τὰ ποιητικὰ καὶ τὰ γεννητικὰ ἀλλήλοις), are entailed: the point produces a magnitude, the thing-­in-­motion [produces] a motion, and the now [prod­uces] time. (Themistius, In Phys., 150.12–15)

In light of this evidence, it is unmistakable that the greatest and most important Aristotelian commentators—from Alexander in the late second century through Themistius in the fourth up to Philoponus and Simplicius in the sixth—agreed on how Aristotle’s analogy ought to be understood, viz., in such a way that the now is that which produces time through its flow. With the exception of Simplicius, it is these commentators whose works have been translated into Arabic and whose commentaries have been an integral part of the philosophical curriculum in natural philosophy.32 Thus, the Arabic philosophical tradition received a unanimous Greek interpretation of Aristotle, which regarded the now both as flowing and as productive of time. Only Themistius lacks the explicit notion of a flow, but he nonetheless agreed that the now produces time just as a point produces a line—and a point can only produce a line when it is moving. 30 Aristotle, al-­Ṭabīʿa, 453.9; see also Paul Lettinck, Aristotle’s Physics and its Reception in the Arabic World (Leiden: Brill, 1994), 366. 31 Aristotle, al-­Ṭabīʿa, 432.16f.; 433.23; 434.13; cf. Ahmad Hasnawi, “La définition du mouvement dans la Physique du Shifāʾ d’Avicenne,” Arabic Sciences and Philosophy 11 (2001), fn. 46, 237. 32  The centrality of these Greek commentators in the curricula of the early Arabic philosophical tradition is borne out by hard facts, in this case by the contents of the already mentioned Ms. Leiden or. 583, which is a fantastic document of historic value. On the manuscript, and for further references, see esp. Elias Giannakis, Philoponus in the Arabic Tradition of Aristotle’s Physics (DPhil dissertation, University of Oxford, 1992); Lammer, The Elements of Avicenna’s Physics, 17–19.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  93

2.  The Now in Avicenna’s Philosophy of Time In Aristotle’s Physics, time and the now are discussed together. Both concepts are closely intertwined with one another, and so the now has rightly been said to be the core feature of Aristotle’s conception of time.33 This stands in stark contrast to Avicenna’s account, in which the now (al-­ān) fulfills only a peripheral role, as can already be surmised from considering the structure of Avicenna’s discussion of time (al-­zamān) in the second book of his Lecture on Nature (al-­Samāʿ al-­t ̣abīʿī) being the ‘Physics’ of his opus magnum The Cure (al-­Shifāʾ) and, thus, his most extensive and elaborate account of the foundational discipline of natural philosophy.34 Instead of discussing the now while defining time, Avicenna effectively outsourced the discussion of the now to a chapter of its own. In fact, Avicenna mentions the now not even once in the central chapter II.11, in which he presents his understanding of time.35 This absence is conspicuous, indicating that the now as a philosophical concept has no immediate bearing on the philosophical concept of time as developed in The Lecture on Nature II.11. If the now, for Avicenna, were really that which produces or generates time in the first place, as commonly has been maintained up to this day, then the fact that it is

33  See Broadie, “Aristotle’s Now,” 104; Coope, Time for Aristotle, 125. 34  My references to The Lecture on Nature will contain both the page and line numbers from Zāyid’s Cairo edition together with the paragraph numbers from McGinnis’ translation. Analogously, references to Avicenna’s Metaphysics (al-­Ilāhiyyāt) will add Marmura’s paragraph numbers to the page and line numbers from the Cairo edition by Qanawātī, Zāyid, Mūsā, and Dunyā. Moreover, I will cite Avicenna’s The Founts of Wisdom (ʿUyūn al-­ḥikma) by Badawī’s 1954 edition. It should also be noted that I have not made a separate mention of Avicenna’s purportedly mysterious—and often considered non-­extant—work The Eastern Wisdom (al-­ Ḥikma al-­mashriqiyya) in the main body of the article, because (as has been noted by Gutas and as I shall explore in a future article) it follows the elaboration in The Lecture on Nature so closely that my arguments about the latter should also be valid for the former; see Dimitri Gutas, “Avicenna’s Eastern (‘Oriental’) Philosophy: Nature, Contents, Transmission,” Arabic Sciences and Philosophy 10 (2000), 159–80. To facilitate comparison, I have furnished my references to the discussion of time in The Lecture on Nature with references to the corresponding (often identical) passages in The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya). 35  There are two exceptions at the beginning of The Lecture on Nature II.11, where Avicenna used al-­ān twice in the adverbial sense of ‘now’ or ‘presently.’ It is obvious that on these occasions, the term is not used for the concept of the now but merely as a grammatical temporal notion. Finally, he mentions the now in the final clause—indeed, as the last word—of the chapter, in order to announce the next topic treated in the following chapter, viz., the now as a division of time.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

94  Andreas Lammer nothing but absent from the core of Avicenna’s account of time should strike us as distinctly odd. I shall now investigate briefly Avicenna’s “explanation of the subject ‘the now’ ” (bayān amr al-­ān), as the title line of chapter II.12 has it, and analyze its purpose, themes, and argument. As will be shown, the discussion of the now in this chapter is best understood as an appendix to Avicenna’s already completed investigation of time, and contributes more to our human experience of time and to Avicenna’s didactic purposes of teaching natural philosophy than to time itself. I would like to begin with the question why Avicenna discussed the now at all, if it really is that peripheral as I am claiming. Apart from the obvious reason that Avicenna might have felt compelled to discuss it simply because Aristotle, and the Peripatetics, have done so, there is Avicenna’s own justification for dwelling on the now, which he provides in the last sentence of chapter eleven serving as a transition to chapter twelve. He writes: Because time, as we said, is a certain magnitude (miqdār) which is continuous, paralleling the continuity of motions and distances, it inevitably has an imagined division (faṣl mutawahhim) and this is what is called ‘the now’ (al-­ān). (The Lecture on Nature II.11, §6, 159.15f., tr. McGinnis, modified)36

Indubitably, there is some merit in discussing the now. Discussing the now belongs to a complete discussion of time, because the now belongs to time—and it does belong to time, because it was one of the main results of Avicenna’s preceding analysis that time is precisely a magnitude (miqdar) that is continuous (muttaṣil). Time as a magnitude is continuous, not only because ‘magnitude’ is the name for the species of ‘continuous quantity’ (as opposed to ‘multiplicity’ being the name for the species of ‘discrete quantity’) known from Aristotle’s and Avicenna’s works on the categories;37 but more specifically because Avicenna makes use of Aristotle’s analogy between distance, motion, and time. Distance 36 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 34.20f. 37  See Lammer, The Elements of Avicenna’s Physics, 445–7.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  95 is continuous, because Avicenna vehemently rejects atomism.38 Motion is continuous, because it “follows” (ἀκολουθεῖ) or—as Avicenna has it— “conforms” (t ̣ābaqa) to distance in structure. The result is that every motion occurs along a distance, thus covering the entire distance without leaving any untraversed gaps behind. Time, then, follows or conforms to motion, being its magnitude. Therefore, time is continuous on account of motion and motion is continuous on account of the distance.39 Consequently, Avicenna discusses the now, because time is a continuous magnitude and every continuous magnitude is naturally such that the imaginative faculty of the human mind is able to posit points in it, which hypothetically divide the magnitude. In other words, the now has its proper part within a complete discussion of what time is, because it concomitantly follows from the essence of time as a continuous magnitude that the mind can posit divisions (fuṣūl, sg. faṣl) in it—and it is these divisions that, in the case of time, are called ‘nows’ (ānāt, sing. ān). This is the primary reason for Avicenna to add a chapter on explaining the now. According to these considerations, the now is something that follows conceptually from the essence and ontologically from the existence of time. As a division of time, the now is conceptually dependent upon time, because time exists as a continuous magnitude. It is also onto­logic­al­ly dependent upon time, because time exists as a continuous magnitude. On both counts, it is clear that the now is not prior to time in the way required for claiming that the now is what provides, produces, or generates time.40 The now is effectively precluded from producing time, because 38 See The Lecture on Nature III.3–5; The Salvation (al-­Najāt) II.1.2, 197.17–203.5 ≈ The Philosophy for ʿArūḍī (al-­Ḥikma al-­ʿArūḍiyya) II.1.1, 117.7–120.2; The Founts of Wisdom (ʿUyūn al-­ḥikma) II.7, 24.17–26.5; The Philosophy for ʿAlāʾ al-­Dawla (Dāneshnāme-­ye ʿAlāʾī) II.5–7, 15.2–23.14; The Guidance (al-­Hidāya) II.1, 143.3–147.7; The Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt) II.1.1–4. 39  This is most explicit in The Salvation (al-­Najāt) II.2.9, 231.5f. ≈ The Philosophy for ʿArūḍī (al-­Ḥ ikma al-­ʿArūḍiyya) II.2.7, 132.21f., translated below, 96. Note also that The Lecture on Nature II.13, §§4–5, 169.13–170.10 clarifies that motion and distance do not cause the continuity of time but cause the existence of time as something that is essentially continuous; see The Eastern Wisdom (al-­Ḥ ikma al-­mashriqiyya), 40.8–13. 40 The sixth/twelfth-­seventh/thirteenth-­century theologian Fakhr al-­Dīn al-­Rāzī, who in opposition to Avicenna ultimately accepts the flow of the now as the cause of time in his late writings, explicitly distinguishes between a now that is prior to time (and causes the existence

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

96  Andreas Lammer without time, there simply and plainly is no now: the now is a division of—and, thus, presupposes—time (and does so irrespective of the precise nature of time’s existence).41 Thus, it would seem to be more than adequate to describe the twelfth chapter of the second book of Avicenna’s Lecture on Nature as an appendix to his account of time, which Avicenna, as such, has already completed in the preceding eleventh chapter. It is only after time had been established in full that the now became a topic worthy of discussion. We encounter a similar—if not to say: the same— structure in all of Avicenna’s other main works, as we shall see now. In his Salvation (al-­Najāt), and in its earlier version The Philosophy for ʿArūḍī (al-­Ḥ ikma al-­ʿArūḍiyya), Avicenna provides a clear and well-­ structured account of the concept of time. He discusses the possibility for traversing a distance at different speeds and, on that basis, shows that time is the magnitude of motion. He goes on to argue that time and motion have not been created in time, and that, thus, the circular motion of the heavens is the most appropriate motion for time to be its magnitude. He finally adds: Therefore, time is a magnitude belonging to the circular motion with respect to the prior and posterior but not with respect to distance. The motion is continuous, and so time is continuous, because it conforms of time) and a now that is posterior to time (and divides time); see al-­Rāzī, The Higher Pursuits in the Divine Science (al-­Mat ̣ālib al-­ʿāliya min al-­ʿilm al-­ilāhī) V.7, 87.15–88.3. A similar distinction may be at play in Avicenna’s rejection of the idea that a point could produce a line in his Metaphysics (al-­Ilāhiyyāt) III.4, §11, which is discussed below, 115ff. 41  The now’s dependence on time is also one of the major points which Fakhr al-­Dīn al-­Rāzī (see the preceding footnote) raises against the Peripatetic position in his Commentary on the Founts of Wisdom (Sharḥ ʿUyūn al-­ḥikma): “If they claim that the now is the limit belonging to time and an attribute subsisting through it, then how can they claim that the now is the prin­ ciple (al-­aṣl wa-­l-­mabdaʾ) [of time] and that time only comes-­to-­be from its [sc. the now’s] motion?” (Commentary on the Founts of Wisdom (Sharḥ ʿUyūn al-­ḥikma) II.8, 144.16–18). It is not entirely clear whether Fakhr al-­Dīn attributes these two conflicting doctrinal points to Avicenna in particular or to the Peripatetics in general and, as a consequence, it is not entirely clear whether Fakhr al-­Dīn regards Avicenna as having defended the idea that time comes to be through the motion of a flowing now. While his earlier remarks in the Commentary on the Founts of Wisdom (Sharḥ ʿUyūn al-­ḥikma) seem to suggest that he did (143.21–144.2), he usually attributes the idea of a flowing now to Plato—not to Aristotle or the Aristotelian Avicenna (e.g., al-­Rāzī, The Higher Pursuits in the Divine Science (al-­Mat ̣ālib al-­ʿāliya min al-­ʿilm al-­ilāhī) V.7, 87.19–88.3; Commentary on the Founts of Wisdom (Sharḥ ʿUyūn al-­ḥikma) II.8, 144.10–15). On all this, see also Peter Adamson and Andreas Lammer, “Fakhr al-­Dīn al-­Rāzī’s Platonist Account of the Essence of Time,” in A. Shihadeh and J. Thiele (eds.), Philosophical Theology in Medieval Islam: The Later Ashʿarite Tradition (Leiden: Brill, 2020).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  97 to something continuous, and all that which conforms to something continuous is [itself] continuous. Therefore, time is so disposed that it is divisible by the imagination (yanqasima bi-­l-­tawahhum), because all that is continuous is such. So, when time is divided, endpoints are established for it in the imagination and we call them ‘nows’ (ānāt). (The Salvation (al-­Najāt) II.2.9, 231.4–8 ≈ The Philosophy for ʿArūḍī (al-­Ḥ ikma al-­ʿArūḍiyya) II.2.7, 132.20–3)

Thereafter, the word ‘now’ occurs only two more times, once when the nows are again said to be the limits of time and once when Avicenna states that the now in time is like the unit in number, somewhat echoing a remark from Physics IV.11, 220a1–4.42 That is all he has to say about the now in The Salvation (al-­Najāt) and in The Philosophy for ʿArūḍī (al-­Ḥ ikma al-­ʿArūḍiyya). Just as in The Lecture on Nature, he derives the now from the fact that time is continuous and explains the now after having completed his account of time, i.e., as an appendix that follows from one of the core features of time, viz., its magnitudinal continuity. The now does not emerge as a feature that could in any way be central to the discussion of time. The structure of the presentation in The Salvation (al-­Najāt) and in The Philosophy for ʿArūḍī (al-­Ḥ ikma al-­ʿArūḍiyya) is similar to that pursued in The Founts of Wisdom (ʿUyūn al-­ḥikma), where the now is only mentioned twice, once as a division of time and a limit of its assumed parts (faṣl al-­zamān wa-­t ̣araf ajzāʾihī l-­mafrūḍa fīhi), and once as time’s “origin” or “principle” (mabdaʾuhū) in analogy to the one (al-­wāḥid) being the origin, i.e., the unit and principle, of number.43 Otherwise, the now is entirely absent. In his chapter on time in The Guidance (al-­Hidāya), Avicenna gives again a brief, though complete, survey of his account. He discusses the possibility for traversing a distance at different speeds, the magnitude of motion, the notions of before and after, and continuity. He concludes his analysis with a restatement of the definition of time as the magnitude of

42  Cf. also Philoponus’ comments ad loc. 43  The Founts of Wisdom (ʿUyūn al-­ḥikma) II.8, 27.12; 28.9–11.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

98  Andreas Lammer motion and, then, adds a few remarks about the eternity of motion and time, which he introduces with the following statement: The dissection of time by the imagination is called a ‘now’ (wa-­qaṭʿ ­al-­zamān bi-­l-­wahm yusammā ānan) just as the dissection of the body is called a ‘surface’ and the dissection of the surface is called a ‘line’ and the dissection of the line is called a ‘point.’ Every now is posited (kull ān yufraḍu), and so is something that originates (ḥādith). (The Guidance (al-­Hidāya) II.1, 158.5–7)

Most interesting in this regard is the case of Avicenna’s Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt). There, Avicenna “drops all mention” of the now, as already Mayer pointed out.44 Admittedly, this work contains only two relatively brief chapters on time, which do not constitute a full discussion of the subject, but it is still surprising that the now is not even once mentioned within these two chapters. The first chapter, Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt) II.5.4, even discusses the existence of time (alongside treating its eternity). The central terms in the argument are ‘the before’ and ‘the after,’ their abstract notions ‘beforeness’ and ‘afterness,’ ‘continuity,’ the ‘renewal’ of a ‘state,’ ‘change’ and ‘motion,’ and finally ‘magnitude’ and ‘quantity.’ All of these terms are most relevant for Avicenna’s account of time, most of them have been central to Peripatetic discussions of time already for centuries. What is missing, though, is any mention of the now.45 The result is that in none of his eight major summae—leaving aside his Lecture on Nature from The Cure for a moment—does Avicenna come even close to describing the now as that which produces or generates time. What he instead reiterates in all these works—apart from the 44  Mayer, “Avicenna against Time Beginning,” 142. 45  It has been suggested that the absence of the now in Avicenna’s Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt) is due to a “development” in his works that “shows a retreat from an earlier accommodation . . . of the Platonist, flowing model, to a more purely Aristotelian model” (Mayer, “Avicenna against Time Beginning”, 143). Mayer attributes this suggestion to a personal communication with McGinnis. If my understanding of the status of the discussion of the now in the works of Avicenna is correct, a developmental hypothesis is no longer required, as Avicenna’s stance towards the now, and towards its relevance for his account of time, was always equal in that he never regarded the now as in any way important for constituting the essence or causing the existence of time.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  99 Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt), where the now is entirely absent, and the physics of The Philosophy for ʿAlāʾ al-­Dawla (Dāneshnāme-­ye ʿAlāʾī), where time is entirely absent—is that the function of the now consists in dividing and uniting a continuous time as a result of an act of the imaginative faculty of the human mind. Before I turn to an analysis of those passages from The Lecture on Nature that are usually quoted in support of the thesis that the now produces time through its flow, I would, first, like to examine more closely what, indeed, can be said about Avicenna’s concept of the now positively and with certainty—and this is that it functions as a fāṣil (“dividing”) and a wāṣil (“connecting”).46

3.  The Now as wāṣil and fāṣil We have just seen above how Avicenna announces the investigation of the now in the final sentence of The Lecture on Nature II.11. With the first sentence of II.12, he begins this investigation: We say that the now is known from the knowledge of time, for time, since it is continuous, undoubtedly has an imagined division (faṣl mutawahhim)47 and this is what is called ‘the now’ (al-­ān). (The Lecture on Nature II.12, §1, 160.4f.)

As before, Avicenna introduces the now as something derived from the continuity of time, as something imagined, and as a division. In particular the last aspect, that the now is a division (faṣl) of time, is his object of inquiry in the following analysis. He writes: This now, in comparison to time itself, is not actually existent at all (laysa mawjūdan al-­battata bi-­l-­fiʿl), for otherwise, then, there would 46  Avicenna’s discussion of these two points corresponds to Aristotle’s remarks in Physics IV.11, 220a4–14; IV.13, 222a10–20; for a first insightful analysis, see McGinnis, “Ibn Sīnā on the Now,” 83–97; Time and Time Again, 260–82. 47 Reading mutawahhim with Mss. Leiden or. 4 and or. 84 as well as Āl Yāsīn and McGinnis for yatawahhamu in Zāyid.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

100  Andreas Lammer be a severance of the continuity of time. On the contrary, its existence is only on the basis that the imaginative faculty imagines it as a con­nect­or (bal innamā wujūduhū ʿalā an yatawahhamahū l-­wahm wāṣilan) in a linear extension. The connector is not actually existent in the linear extension insofar as it is a connector, for otherwise, then, as we shall prove later, there would be infinitely many connectors. On the contrary, it would be actual only if time were severed through some sort of severance, yet it is absurd that the continuity of time should be severed.  (The Lecture on Nature II.12, §1, 160.5–8)48

It is an essential hallmark of the continuous to be divisible and, in fact, infinitely so.49 For both Aristotle and Avicenna, this divisibility is only a potential ability or aptness for division on the part of the potentially divided but essentially continuous object. If any such division were realized, the continuous object would cease to be, i.e., if one cuts through a single unified whole, the whole ceases to be and the process of division results in two individual objects corresponding to two halves of the initial whole, each of which is, again, continuous, i.e., susceptible to further division. So, if any such division occurred actually in time, the continuity of time would be severed and time, as such, could not be a unified continuous whole anymore, even though time essentially is continuous through and through. For that reason, the division is in reality only potential, whereas in the mind, we can divide and sever time as we want. All this has no bearing on the external reality of time, and is merely rele­ vant for our subjective purpose of measuring motion and comparing demarcated time-­spans with one another—the now as a faṣl, Avicenna 48 Cf. The Eastern Wisdom (al-­Ḥ ikma al-­mashriqiyya) III.11, 34.22–35.4. Interestingly, the text transmitted of Avicenna’s Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) replaces the word wāṣil with the word fāsil in this passage. This could be a scribal error introduced in the transmission of either The Lecture on Nature or The Eastern Wisdom (al-­Ḥ ikma al-­mashriqiyya). Yet, it could also have been a deliberate decision on the part of Avicenna to shift the emphasis of his account in The Eastern Wisdom (al-­Ḥ ikma al-­mashriqiyya). This latter hypothesis receives some support by the subsequent passage from The Lecture on Nature II.12, §1, 161.1–4 ≈ The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 35.14–18 (quoted below, 102f.), in which the phrase fa-­l-­ān wāṣil lā fāṣil is reduced to fa-­l-­ān laysa fāṣilan; cf. also fāṣila at The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 36.21 instead of wāṣila at The Lecture on Nature II.12, §5, 163.14. 49 See the recent analysis of continuity, extensionality, and divisibility in Lammer, The Elements of Avicenna’s Physics, 122–39.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  101 said right at the beginning of the above passage, “in comparison to time itself, is not actually existent at all” (laysa mawjūdan al-­ battata bi-­l-­fiʿl).50 Avicenna adds another notion to the discussion when he says that the now as a division (faṣl) exists “only on the basis that the imaginative faculty imagines it as a connector (yatawahhamahū l-­wahm wāṣilan) in a linear extension.” The participle wāṣil and the noun faṣl—Avicenna will later also use the participle fāṣil—at first sight appear to be contraries, one of them referring to an act of separating a whole into parts, the other to an act of uniting two parts to form a greater whole. For Avicenna, however, they are not meant to be contrary here. Indeed, his subsequent argument, which is modeled on the beginning of chapter IV.13 of Aristotle’s Physics, shows that an imagined fāṣil of time turns out to be a wāṣil, thereby revealing the essential continuity, rather than an actual dividedness, of time.51 The argument assumes an actual division of time, which then marks the beginning of a time. Having a beginning, this time did not yet exist before its beginning, and so was non-­existent before it came into existence. Therefore, there must have been a certain ‘before’ before the existence of this time, at which it did not yet exist, because this time had its 50  Emphasis added. One might say that the now is like a potential cut imagined in a cor­por­ eal magnitude as opposed to an actual limit of that magnitude. Finite magnitudes, indeed, have actual limits—in the case of bodies, for example, such an actual limit would be the surface of the body. It is clear that any such actual limit of magnitudes does not sever that magnitude of which it is a limit. One might think that there is something analogous to such an actual limit of a magnitude also in time. One might, perhaps, like to say that the objective moment in which two or more concurring events come together is such an actual limit. For example, if the assassin shot his victim at the moment of the final crash of cymbals, then both events—the crash and the shot—together might mark out a moment that is not merely imagined. Avicenna himself comes close to formulating something like this in The Lecture on Nature II.12, §1, 161.1–4 (translated below, 102f.) emphasizing, however, that such “does not really (bi-­l-­ḥaqīqa) bring about a division in the essence of time itself but only in its relation to motions.” At any rate, it is clear that even such a concurrence of events presupposes time (rather than constituting it); cf. The Lecture on Nature II.10, §§9–10, 152.11–153.15. 51  For Aristotle’s remarks, cf. esp. Phys. IV.13, 222a10–20, beginning with the words that the now is “the link of time (συνέχεια χρόνου) . . . and it is a limit of time” (πέρας χρόνου). Isḥāq ibn Ḥunayn’s Arabic translation renders the two expressions here as wuṣlat al-­zamān and t ̣araf li-­l-­ zamān, respectively. In the Arabic glosses that accompany Isḥāq’s translation in Ms. Leiden or. 583, we find what is apparently an Arabic quotation from Philoponus’ commentary on this passage (most probably going back to the translation of Qust ̣ā ibn Lūqā). This Arabic version of Philoponus provides a different terminology, speaking of rābit ̣ li-­l-­zamān and jāmiʿ (468.1, for κοινὸν πέρας at 760.24) and fāṣil li-­l-­zamān (468.4, for διαίρεσις . . . τοῦ χρόνου at 761.5).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

102  Andreas Lammer beginning in the assumed actual division of time. As Avicenna has already shown in the previous chapter, this beforeness must be predicated of something to which beforeness and afterness belongs through itself, and this can only be time. Thus, before the time which begins with the assumed division, there already had been another time.52 Accordingly, we get two times—one before the other—which are divided at a now assumed to mark an actual division. However, if there was time before the now as well as after the now, and if the now is basically surrounded by time having a time on each of its two sides, then the now does not really function as an actual division but merely marks the point at which these two times are united. Consequently, it serves as a con­nect­or, rather than a division, of time.53 The important point in Avicenna’s argument, then, is that although a now may mark a moment or the beginning of a motion, it does not split time actually into two halves; it divides time only potentially. What Avicenna wants us to recognize is that every now which has been posited as fāṣil (“dividing”) turns out to be really wāṣil (“connecting”). Consequently, far from representing an actual division, the now— entirely a potential and imaginary division—confirms only the inherent continuous unity of time.54 Avicenna puts it as follows: Before this time, then, there would be a time that is continuous with it—that one before, this one after—and this division would be uniting the two, although it was posited to be something that divides—but this is a contradiction. (The Lecture on Nature II.12, §1, 160.12f., tr. McGinnis, modified)55

The twist in Avicenna’s argument is in the seemingly contradictory phrase “and this division would be uniting the two” (wa-­hādhā l-­faṣl 52 See The Lecture on Nature II.11, §4, 157.9–15; §5, 158.11–17; The Eastern Wisdom (al-­ Ḥikma al-­mashriqiyya) III.11, 33.9–16. See also Phys. VIII.1, 251b19–26. 53  See also Avicenna’s Categories (al-­Maqūlāt) IV.1, 133. 11–14. 54  Avicenna adds another version of the argument, this time involving a now as an endpoint instead of a beginning, thus bringing about a time after a time instead of one before; see The Lecture on Nature II.12, §1, 160.13–161.1; see also Aristotle’s assessment that with regard to the now, “the dividing and the uniting are the same” (Phys. IV.13, 222a19f.). 55 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 35.9–11.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  103 yajmaʿuhumā): it is that which we posited as severing the continuity of time which actually emerges as safeguarding it. A little later, he concludes the argument as follows: The now, then, is something connecting, not something dividing (fa-­l-­ān wāṣil lā fāṣil). Thus, time does not have a now that exists as actual in relation to itself but only as a potentiality, I mean the p ­ otentiality close to actuality (al-­quwwa al-­qarība min al-­fiʿl).56 Time is so disposed that the now can always be posited in it, whether by someone’s simply positing [it] or by the motion’s arriving at some common indivisible limiting point, such as the beginning of a sunrise or of a sunset or something else.57 That does not really bring about a division in the essence of time itself but only in its relation to motions. (The Lecture on Nature II.12, §1, 161.1–4, tr. McGinnis, modified)58

The now, then, is primarily something that can be imagined as something connecting the past with the future or anything that, from the 56  One must not take the expression al-­quwwa al-­qarība min al-­fiʿl to mean that the now is more actual than simply being entirely potential, as though there was a special kind of potentiality that is almost actual by itself. What Avicenna intends to emphasize with this by no means rare phrase is that he is not talking about a potentiality as such but about that one potentiality that is directly adjacent, as it were, to an actuality. Thus, the now could immediately be actualized without there being any other further potentiality which first needed to be actualized. As an illustration of a potentiality that is close to actuality we may consider a man who is potentially writing as opposed to a healthy newborn who is potentially writing. It is true that both the man and the newborn are potentially writing, but only the former is close to this actuality, whereas the latter first has to learn how to write and, currently, does not possess “the potentiality close to actuality.” Another way of thinking about this is to consider a man as opposed to a mass of blood and flesh and bone. In this example, man is able to write not qua being a mass of blood and flesh and bone but qua being a man. Avicenna’s intention in using that phrase is reminiscent of Aristotle’s Metaphysics Θ.7 or, alternatively, the common distinction between a first and a second potentiality as well as a first and a second actuality, well known from, e.g., Aristotle’s De anima II.5. There are other examples one can choose to illustrate the idea; see Avicenna’s discussion in The Lecture on Nature II.1, §22, 91.9–92.4 and his remark at the end of Metaphysics (al-­Ilāhiyyāt) II.3, §8, 75.10f.; for yet another possible example in Avicenna’s works, see Avicenna’s Risāla fī l-­kalām ʿalā l-­nafs al-­nāt ̣iqa as translated by Gutas in his Avicenna and the Aristotelian Tradition, 289f.; see also Jon McGinnis, “Positioning Heaven: The Infidelity of a Faithful Aristotelian,” Phronesis 51 (2006), 153. All this may sound rather trivial, but it is worth noting in this context, especially in light of McGinnis’ claim that “[t]he flowing now is not conceived as some actual entity or object in its own right; nonetheless, it is a real state belonging to that which is borne along” (Time and Time Again, 286f.; “Ibn Sīnā on the Now,” 100f.); q.v. below, 107 and 124. 57  See above fn. 50, 100. 58 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 35.14–18; q.v. above fn. 48, 99f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

104  Andreas Lammer vantage point of the now, is past with whatever is, from the vantage point of the now, future.59 The upshot so far is that Avicenna presents the now as something that divides time potentially by being posited by the imaginative faculty. As such, “the now is known from the knowledge of time” and is a result of time’s inherent continuity, to which the now bears witness by the fact that it can best be described as a wāṣil, connecting a before with an after or a past with a future, even though it—considered as such—does not exist actually but merely potentially or imaginatively. In none of his major works does Avicenna integrate the now into his main discussion of time; it is and remains an afterthought or appendix that adds something, albeit nothing substantive, to the understanding of time. Consequently, it is hardly surprising that in the eclectic Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt) the supposedly central now is absent altogether. In other words, just as the now follows from time, a discussion of the now is no more than a possible follow-­ up—an ­appendix—to a discussion of time. All this provides some first reasons for beginning to doubt whether the now has any actual relevance for Avicenna’s account of time and, a fortiori, to contest the validity of the current consensus that the now is what is ultimately or immediately responsible for the existence of time. If the now were an important—even pivotal—component of his account of time, it would have been expedient for Avicenna to explicate this component and to clarify its importance somewhere in his major works. Admittedly, those who support this interpretation draw upon some passages from Avicenna’s discussion in The Lecture on Nature, in order to substantiate their claim. Before I shall begin to examine these very passages in the next section, two remarks about the terminology of ‘the now’ are in order, which should be heeded in the remainder of this 59  See Avicenna, Categories (al-­Maqūlāt) III.4, 118.4; IV.1, 133.11f.; see also Cat. 6, 5a5–7; Phys. IV.14, 223a4–8. It is striking that Avicenna’s present argument from The Lecture on Nature does not even rely on the concept of the now as such, as is clear from chapter four of the fifth namat ̣ of Avicenna’s Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt), which contains a similar reasoning to the same result—this time even without mentioning the now at all (II.5.4, 150.5–13). In fact, the upshot from this argument in Avicenna’s Pointers and Reminders (­al-­Ishārāt wa-­l-­tanbīhāt) is that it is the real existence of the real motion of the outermost sphere that, for Avicenna, causes, produces, or generates time (and not the now); q.v. below, 109ff.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  105 art­icle: (i) there is a distinction between ‘a now’ and ‘the now.’ While ‘the now’ may seem to us to designate the present moment only, Avicenna speaks of any now usually in terms of a moment in time currently ­stipulated or considered by a human being. In fact, human beings can stipulate—and hence imagine—several nows (ānāt), talk about them, and compare them to one another. So in Avicenna’s discussion, just as in Aristotle’s earlier, the expression ‘the now’ does not primarily mean ‘the present moment.’ Instead, ‘the now’ is simply the grammatically determinate expression for any specific, determinate moment in time that has been stipulated for whatever reason. In short, neither ‘a now’ nor ‘the now’ has to be ‘now.’ Furthermore, (ii) there is also a distinction between ‘the now’ (al-­ān) and ‘the present’ (al-­ḥāḍir).60 Although the question whether or not time is the product of a flowing now seems to be meaningful for us only if ‘the now’ means ‘the present,’ Avicenna nowhere uses the term ‘the present’ in this context or makes this connection. In fact, despite the lexical distinction between al-­ān and al-­ḥāḍir, Avicenna does not, to my knowledge, reflect on what constitutes ‘the present.’ Nonetheless it seems to be sufficiently clear that while ‘the now’ is meant to be a temporal term, Avicenna considers ‘the present’ as an ontological term in the sense that the present is (the sum total of) what is actual (al-­ḥāṣil).61 For our investigation about whether or not time is the result of a flowing now, this means that we cannot collapse the whole question into asking whether or not time is the result of a flowing present, for ‘the now’ is not ‘the present.’ All that can be said is that—in Avicenna’s own words—“the shared edge” between the past as (the sum total of) what is no longer actual and the future as (the sum total of) what is not yet actual “could be thought of ” (qad yafhamu) as “the now”—and this “now,” then, would be the now that may be (and historically has been) said to

60  See esp. The Lecture on Nature II.13, §9, 172.12–173.2. 61  One should be clear that for Avicenna, as for Aristotle, motion is the foundational concept and time the derivative. Any term like ‘the present’ arguably ought not to be understood along temporal lines but indicates the ontology of what is (i.e., what through motion has become) actual; cf., however, the dialectical context of The Lecture on Nature II.10, §2, 149.4f., where al-­ḥādir seems to be used as a temporal term.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

106  Andreas Lammer prod­uce time through its flow.62 The question now is: does this now exist—and if it exists, does it flow in such a way that its flow produces time?

4.  Is There a Flowing Now Producing Time? As mentioned, the secondary literature is unanimous with regard to the question at hand. Avicenna’s philosophy is said to allow principally the idea of a flowing now and his doctrine of time to rely fundamentally on that idea as the true cause of time’s existence. Yet, it is my sincere belief that the interpreters, some of them more than others, have treated the text in some instances perfunctorily, as they neglected elements of Avicenna’s overall intention which barred them from accurately understanding the function of the flowing now within Avicenna’s philosophy of time. An exceptionally oblique example of a careless reading of Avicenna’s text is offered by al-­ʿIrāqī in his book al-­Falsafa al-­t ̣abīʿiyya ʿinda Ibn Sīnā. In this book, we read the following: Our philosopher [i.e., Avicenna] holds the view that the now, when it remains through the prior and posterior of motion, produces (aḥdatha) time.  (al-­ʿIrāqī, al-­Falsafa al-­t ̣abīʿiyya ʿinda Ibn Sīnā, 251)

The crucial word in this quotation is aḥdatha, which I translated as “to produce.” In brief, what al-­ʿIrāqī asserts here is that Avicenna claims that “the now . . . produces time” (al-­ān . . . aḥdatha l-­zamān).63 This, however, is something Avicenna never intended to say anywhere in his works and actually never said—not even in The Lecture on Nature. It is true, the

62  See esp. The Lecture on Nature II.13, §9, 172.12f. 63  As confirmation, al-­ʿIrāqī cites Avicenna’s Salvation (al-­Najāt) and The Founts of Wisdom (ʿUyūn al-­ḥikma) but there is nothing to be found in either work that could support his reading, because neither work, as already noted, says much about the now anyway. What is more, al-­ʿIrāqī sees Avicenna in the tradition of his third/ninth-­century predecessor Abū Yaʿqūb ibn Isḥāq al-­Kindī, whom he claims to have maintained that “time is not identical with the now but with the continuity of nows” (al-­zamān laysa huwa l-­ān bal ittiṣāl al-­ānāt). This reading of al-­Kindī may have its origin in his Liber de quinque essentiis, 33.19–35.18; see also Jean Jolivet, “al-­Kindī, vues sur le temps,” Arabic Sciences and Philosophy 3 (1993), 56–62.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  107 above sentence is one of the many quotations which al-­ʿIrāqī draws from Avicenna’s Lecture on Nature without acknowledging his source (and so some of these words, indeed, were written by the hand of Avicenna), yet the sentence is both incomplete and out of context, so much so that in Avicenna’s text, the word al-­ān (“the now”) is not even the subject of the verb aḥdatha (“to produce”), as al-­ʿIrāqī makes us believe. I shall come back to this below.64 A more valuable contribution is McGinnis’ foundational doctoral dissertation Time and Time Again, in which McGinnis strongly, and rightly, emphasizes the Peripatetic background of Avicenna’s philosophy of time.65 He compares Aristotle’s and Avicenna’s texts, and pins down what he calls “the most striking difference” between their otherwise in many respects similar accounts: “Aristotle conceived of time as the static magnitude found between two different characterizations of an unchanging now. Ibn Sīnā on the other hand, envisioned time as the ‘flow’ of [a] constantly changing now.”66 McGinnis finds the incentive for Avicenna’s departure from Aristotle’s static conception in the influence from Alexander, whom he sees responsible for introducing the terminology of the flow, and from Philoponus, who is said to have “assimilated Aristotle’s static conception of time to Plato’s dynamic, [so that the] now, which for Aristotle indicated the terminus or boundary of time, has been converted into Plato’s moving present, generating time through its flow.”67 A few years earlier, Shayegan arrived at a similar conclusion. Time, she writes, is conceptualized by Avicenna as “always incomplete and constantly flowing.”68 Instead of the Peripatetic background, she highlights the correspondence between the now and the aspect of the medial motion, on the one hand, and the correlation between “[t]he moving point, the moving thing and the moving ‘now’ ” on the other, for these

64  See below, 132f. 65  See above, 84ff. I am grateful to Jon McGinnis and his wife (and IT assistant) Celina for, among other things, scanning and sharing this dissertation with me. 66 McGinnis, Time and Time Again, 307. 67 McGinnis, Time and Time Again, 179–86, 190; “Ibn Sīnā on the Now,” 97f.; cf. McGinnis, “The Topology of Time,” 8, 18f. 68 Shayegan, Avicenna on Time, 26.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

108  Andreas Lammer three being “indivisible in themselves . . . generate three divisible continua,” viz., distance, motion, and time.69 Both interpretations mention the analogy between the point, which through its motion produces a line; the thing-­in-­motion, which through its passage produces the motion; and the now, which through its flow produces time.70 Their agreement about time being the result of a flowing now notwithstanding, McGinnis’ and Shayegan’s accounts differ greatly in one important respect: Whereas Shayegan argues that “the moving ‘now’ following the moving ‘thing’ and the moving ‘point’ are pure conjectures and are references to geometrical objects, not to actual things,” McGinnis claims that “[t]he flowing now is not conceived as some actual entity or object in its own right; nonetheless, it is a real state belonging to that which is borne along.”71 Neither interpretation is fully convincing. McGinnis is forced to trivialize, or even to disregard, Avicenna’s repeated description of the now as an “imagined” (mutawahhim) entity—an aspect that Shayegan, of course, emphasizes. Her exegesis, however, fails to answer the question how the now, which is merely imagined (or “conjectural” as she puts it), is capable of producing time, which, after all, should be—and is—an actual, and not merely conjectural, feature of the concrete world—a shortcoming duly criticized in turn by McGinnis.72 Mayer also recognized this issue and suggests that although “an extramental now in act results in intractable problems . . . Avicenna prefers not to consign the now simply to mental status—given that it generates time” and refers to a passage which, similarly to the one quoted by al-­ʿIrāqī, seems to claim that the now, indeed, produces time.73 69 Shayegan, Avicenna on Time, 204; see Shayegan, Avicenna on Time, 28–30, 206. 70  See Shayegan, Avicenna on Time, 201–12; McGinnis, Time and Time Again, 190, 282–90. 71 Shayegan, Avicenna on Time, 205; McGinnis, Time and Time Again, 286f.; “Ibn Sīnā on the Now,” 100f. 72  That time is a real and actual feature of the concrete world does not entail that time is in its entirety real and actual. As already mentioned, of time only the present moment is actual; the remainder of time’s magnitude—the past and potentially the future—is non-­existent; see above, 82f. and fn. 22, 88f. 73  Mayer, “Avicenna against Time Beginning,” 142. Mayer, moreover, seems to find some justification for attributing the idea of a flowing now to Avicenna in Naṣīr al-­Dīn al-­Ṭūsī’s seventh/thirteenth-­century comments on the discussion of time in The Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt) and in Fakhr al-­Dīn al-­Rāzī’s commentary on that work. Interestingly, though, al-­Ṭūsī never mentions the flowing now (as also Mayer remarks) nor does his

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  109 What I shall do in the following is to show that the interpretation that a flowing now is responsible for the existence of time is untenable with regard to Avicenna’s philosophy. Thereupon, I shall examine some passages from Avicenna’s chapter on the now, in order to explain its purpose. There are three reasons why it is incorrect to claim that Avicenna conceives of time as the product of a flowing now in analogy to a point that produces a line: first, because Avicenna elsewhere argues—without any explicit (and even without any perceptible implicit) recourse to the concept of a flowing now—that the motion of the outermost sphere brings time into existence; second, because Avicenna, like Aristotle, denies that indivisibles are capable of moving; and third, because, according to Avicenna, a point cannot produce a line. I shall go through them in this order.

a.  The Outermost Sphere There is one particularly clear statement in Avicenna’s works about what it is that brings about—and, thus, is the cause of—the existence of time. We find this statement in his short work The Founts of Wisdom (ʿUyūn al-­ḥikma). There Avicenna writes: Motion74 is the cause for the occurrence of time (al-­ḥaraka ʿillat ḥuṣūl al-­zamān) and the mover is the cause of the cause of time (wa-­l-­ muḥarrik ʿillat ʿillat al-­zamān). Thus, the mover is the cause of time, yet not every mover but [only] the mover of circular [motion] (muḥarrik mustadīra) . . . [Circular motion, however, is due to soul.] . . . in­ter­pret­ation of Avicenna rely on that concept. Mayer seems to revert to the flowing now, in order to make sense of the elapsing and renewed character of time that both al-­Ṭūsī and Avicenna emphasized. As I have argued recently, however, the elapsing character of time is an inherent feature of time itself, because time is through itself the before and after (a point also emphasized by Mayer, 140f.); see Lammer, The Elements of Avicenna’s Physics, 473–6. If the flowing now were responsible for the elapsing and renewing character of time, then the before and after would belong to time through the flowing now and no longer through time itself. 74  I follow the punctuation as it is in the editions of ʿĀṣī and al-­Saqqā, in which the word al-­ḥaraka (“motion”) starts a new sentence. The editions of Ülken, Badawī, and al-­Jabr all lack punctuation here.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

110  Andreas Lammer Therefore, the motion that necessitates time (al-­mūjiba li-­l-­zamān) is due to soul and will (nafsāniyya irādiyya). Thus, the soul is the cause of the existence of time (fa-­l-­nafs ʿillat wujūd al-­zamān). (The Founts of Wisdom (ʿUyūn al-­ḥikma) II.8–9, 28.17–29.7)75

This passage contains two essential claims entailing a conclusion: first, circular motion is the cause for the occurrence (ḥuṣūl) of time, i.e., of its existence as an actual feature of reality; second, soul is the cause for circular motion. Therefore, soul is the cause of the cause of the existence of time. Later, in The Notes (al-­Taʿlīqāt), which arose from Avicenna’s discussions with his colleague and disciple Abū Manṣūr al-­Ḥ usayn Ibn Zayla (d. 440/1048), this claim is delivered in a nutshell: The sphere is the bearer of time (ḥāmil al-­zamān) and the moving power within it is the producer of time (fāʿil al-­zamān). (The Notes (al-­Taʿlīqāt) 142.17 (ed. Badawī)/§764, 424.1 (ed. Mousavian))

The argument underlying both of these passages must be read against the background of Avicenna’s lengthy argumentation in Avicenna’s Metaphysics (al-­Ilāhiyyāt) IX.2—which has been reused word for word in The Salvation (al-­Najāt) IV.2.27–30 and partially in The Procession and the Return (al-­Mabdaʾ wa-­l-­maʿād) I.39–41—that “the proximate mover of the heavens is neither a nature nor an intellect but a soul.” It is shown that the circular and eternal motion of the heavens is due to soul and, as we now have been told in The Founts of Wisdom (ʿUyūn al-­ ḥikma), it is their motion, and in particular the motion of the outermost sphere, which is the cause of time. This is also the gist of an argument in The Salvation (al-­Najāt) II.2.9. There, Avicenna remarks that time, which he has defined as “the magnitude that belongs to the circular motion,” is not created through a temporal creation (ghayr muḥdath ḥudūthan zamāniyyan). Even less so is motion created and, even less again, circular motion, because this type of motion is continuous and eternal. For that matter, the being (huwiyya) of this magnitude, viz., 75 Cf. The Lecture on Nature III.9, §6, 222.6–8; al-­Taʿlīqāt, 142.17 (ed. Badawī)/§764, 424.1 (ed. Mousavian).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  111 time, is due to circular motion. Moreover, Avicenna continues, time belongs primarily to only one circular motion (awwalan li-­shayʾ minhā) and secondarily to all other motions, which, then, are measured in accordance (bi-­l-­mut ̣ābaqa) with that one motion.76 In The Lecture on Nature, containing his most extensive treatment of time, Avicenna is somewhat sparing with references to the outermost sphere. It is merely mentioned in the first, somewhat ‘doxographical’ chapter as part of a dialectical engagement with a view commonly associated with the Muʿtazilī theologian Abū ʿAlī al-­Jubbāʾī (d. 303/915–16) and is briefly alluded to as the “positional motion” in the immediate vicinity of his definition of time in The Lecture on Nature II.11.77 Nonetheless, Avicenna makes up for this lack of references to the motion of the outermost sphere by emphasizing its relevance for the existence of time in the last chapter of his account. In this chapter, viz., chapter II.13, he sets out for a “solution to the doubts raised about time,” as the chapter title has it, and in particular those doubts that concern its existence. There, Avicenna reminds the reader that it has been demonstrated to us (tabarhana lanā) that time is dependent upon motion (mutaʿalliq bi-­l-­ ḥāraka) and that it is a disposition of it (hayʾa lahā). Yet, he immediately restricts this claim by saying that time does not depend (lā yataʿallaqu) on motions that have a beginning and an end (ibtidāʾ wa-­ntihāʾ): Indeed, when time exists by a motion having a certain description (­bi-­ḥaraka ʿalā ṣifa), it is correct [to say] that the existence of time is dependent upon it and that the rest of the motions will be measured by it. This motion is a motion of which it is true [to say that it is of] uninterrupted permanence (al-­istimrār) and that it is not actually determined by limits (lā yataḥaddadu lahā bi-­l-­fiʿl aṭrāf). (The Lecture on Nature II.13, §2, 168.7–9)78 76  The Salvation (al-­Najāt) II.2.9, 230.9–232.1 ≈ The Philosophy for ʿArūḍī (al-­Ḥikma al-­ ʿArūḍiyya) II.2.7, 132.16–133.2; cf. The Lecture on Nature II.13, §3, 168.9–169.1; The Founts of Wisdom (ʿUyūn al-­ḥikma) II.8, 28.3–5; The Guidance (al-­Hidāya) II.1, 160.2–163.6; The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 39.15–40.2. 77  The Lecture on Nature II.10, §§9–10, 152.17–153.4; II.11, §2, 156.16; The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya), 32.22f. On “positional motion” in Avicenna, cf. McGinnis, “Positioning Heaven”; Lammer, The Elements of Avicenna’s Physics, 334–42. 78 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 39.12–14.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

112  Andreas Lammer With this statement, Avicenna answers an obvious question, perhaps the same question that prompted Aristotle to introduce so hastily the motion of the outermost sphere as the ultimate motion of which time is the number.79 This question asks whether there exist multiple, perhaps even infinitely many, times: to each and every motion that happens anywhere in the universe one individual time. This would leave us with a vast number of uncoordinated times whose actual relation to one another is far from being clear. Avicenna’s answer to this is that there is one motion that is special in such a way that all other motions are ul­tim­ ate­ly measured by this motion and the time that comes-­to-­be from it. Yet, saying this instantly provokes a new and obvious objection: So, if it is said: “Do you think, if that motion did not exist, that then80 time would be lost, so that other motions different from it would be without priority and posteriority?” (The Lecture on Nature II.13, §3, 168.9f., tr. McGinnis, modified)81

Avicenna answers this objection with a twofold strategy. First, he argues that without “this motion,” viz., the circular motion of the outermost sphere, there would not be any other motion at all, because any subsequent circular as well as sublunar rectilinear motion could not exist in the absence of this first circular motion. Avicenna refers to a subsequent discussion, probably in The Lecture on Nature III.14 or The Heavens and the Earth (al-­Samāʾ wa-­l-­ʿālam) 2, in which he will explain that the directions of natural (and that is: rectilinear) motions are determined by a body at the periphery which moves in a circular motion, thus specifying— or as Avicenna usually says: “delimiting” (muḥaddid)—the direction and the extremities of up and down.82 Without this circular motion,

79 See Phys. IV.14, 223b18–23. 80 Reading la-­kāna with Mss. Leiden or. 4 and or. 84 as well as Zāyid for a-­kānā in McGinnis and Āl Yāsīn. 81 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 39.15f. It is clear that the objection is formulated from an entirely un-­ Avicennian perspective, thus also using ­un-­Avicennian vocabulary. 82 See The Lecture on Nature III.14, §§1–7, 251.4–255.6; The Heavens and the Earth (­al-­Samāʾ wa-­l-­ʿālam) 2, 11.17–12.3; see also Cael. I.2, 268b11–269b17; II.4, 286b10–287b21; IV.1, 307b28–308a33.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  113 there would not be any directionality nor any natural motion and, ­therefore, also no forced motion and, ultimately, no motion at all. So, in the first part of his reply, Avicenna argues that without a circular motion no other motion could occur in time, because there would be no other motion. In the second part, he argues that this is also true, because without a circular motion there would be no time: So, the existence of time, therefore, is dependent upon a single motion (mutaʿalliq bi-­ḥaraka wāḥida) which it [sc. time] measures and, also, the rest of the motions whose existence would be impossible without the motion of the body which, through its motion, is productive of time (ḥarakat al-­jism al-­fāʿil bi-­ḥarakatihī li-­l-­zamān)83 . . . That is like the magnitude existing in some body that measures [both] it [sc. this body] and whatever is parallel and juxtaposed to it. (The Lecture on Nature II.13, §3, 168.18–169.1, tr. McGinnis, modified)84

Time is dependent upon one single motion. This motion is the one without which all other motions could not exist. As we have just been told, this one motion is the motion that gives all motions their direction. Thus, the motion in question is the circular motion of the outermost sphere, whose body produces time “through its motion” (al-­jism al-­fāʿil bi-­ḥarakatihī li-­l-­zamān), as Avicenna writes here in The Lecture on Nature, and “preserves time” (yaḥfaẓu l-­zamān), as he remarks in The Heavens and the Earth (al-­Samāʾ wa-­l-­ʿālam).85 It is this motion that is the cause of the existence of time, while soul is the cause for this motion thereby being “the cause of the cause of time” (ʿillat ʿillat al-­zamān), as Avicenna puts it in The Founts of Wisdom (ʿUyūn al-­ḥikma), as we have just seen. No other cause (such as a flowing now) is mentioned; indeed, no other cause is needed. 83 Reading li-­l-­zamān with Mss. Leiden or. 4 and or. 84 as well as Zāyid for al-­zamān in McGinnis and Āl Yāsīn. 84 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 39.24–40.2. 85  The Heavens and the Earth (al-­Samāʾ wa-­l-­ʿālam) 4, 28.8–11; see also Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt) II.6.16, 165.1f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

114  Andreas Lammer

b.  Indivisibles Are Incapable of Moving In The Lecture on Nature III.6, Avicenna discusses the question “whether it is possible for something which has no parts to be in motion” (hal yajūzu an yakūna mā la juzʾ lahū yataḥarraku) and the contention found in the books of the Peripatetics that this has to be denied.86 In this discussion, we see Avicenna criticizing an Aristotelian argument to that conclusion as in no way satisfying, because it failed to observe the distinction between essential and accidental motion.87 This distinction, however, is vital in the present context, for an indivisible point may surely move in an accidental sense, for example, as a limiting point belonging to a larger body that itself moves essentially. With his criticism, Avicenna reminds Aristotle of a statement that he himself made at the beginning of Physics VI.10, when he stated that “that which is without parts cannot be in motion except accidentally” (τὸ ἀμερὲς οὐκ ἐνδέχεται κινεῖσθαι πλὴν κατὰ συμβεβηκός).88 Yet, Aristotle’s argument, as Avicenna reads it, tried to establish that whatever has no parts cannot move at all, i.e., not even accidentally. This, however, would prove too much, as both Avicenna and Aristotle agree that the part-­less (τὸ ἀμερὲς) can indeed move, albeit only accidentally. Therefore, Aristotle’s argument cannot be accepted as it stands. Despite Avicenna’s criticism, it would be rash to take this as evidence that Avicenna would, likewise, reject Aristotle’s position that part-­less things cannot move essentially. To the contrary, Avicenna strongly endorses Aristotle’s view, for after his criticism of the Aristotelian argument as unsatisfactory he provides another argument—this time one that “satisfies us.”89 86  The Lecture on Nature III.6, §§7–9, 206.13–208.10. 87  Aristotle’s argument is to be found at Physics VI.10, 241a6–14. It has been adapted by Philoponus in his corollary on place, where it serves as the final of his five arguments against Aristotle’s definition of place; see Philoponus, In Phys., 567.8–28; see also Proclus, Institutio physica I, prop. 31. 88  Phys. VI.10, 240b8f.; see Phys. VI.10, 240b12–20. 89  The Lecture on Nature III.6, §8, 207.10. It is to be noted that the argument just criticized by Avicenna is an additional second argument which Aristotle supplied after the intended conclusion that part-­less things cannot move had been already established by a first argument (see Phys. VI.10, 240b20–241a6). Avicenna does not say anything about this first argument and explicitly attacks only the second, but it seems that the argument which he himself devises is remotely inspired by that first argument

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  115 For Avicenna, the ultimate reason why something without parts cannot move is that it does not have a position that belongs individually to itself. Something that does not have a position of its own cannot move from that position to another position. A point does not have a position to itself and is merely located insofar as it belongs to a line or a surface and ultimately to a body. It is only this body, then, which has a position and a place, and so it is also only this body that is capable of moving essentially, while the part-­less point accidentally moves with it. With this argument, Avicenna establishes that a point, or anything indivisible like a point, cannot itself engage in local motion. Avicenna adds some reasons why, moreover, other kinds of motion, in particular alteration and growth, are also impossible for a point.90 In essence, his final conclusion matches that of Aristotle, who asserted the following: It is, thus, not possible for that which has no parts to be in motion or to change in any way (οὐδ’ ὅλως μεταβάλλειν). (Phys. VI.10, 240b30f., tr. Hardie/Gaye)

In this conclusion, Aristotle apparently expands the original argument, which relied exclusively on local motion, to all sorts of motion and change, thus including alteration and growth, as Avicenna does. So, Avicenna and Aristotle agree that something that has no parts, such as a point (or a now), cannot move, flow, or change in any way (except accidentally). It is, therefore, as unlikely that Avicenna would have allowed for a flowing now to be the generator of time as it was for Aristotle, for both deny that something like a point—and a now is like a point—could itself move or change or flow in any way.

c.  A Point Cannot Produce a Line This impression is confirmed by the third reason, which is Avicenna’s explicit denial, in Metaphysics (al-­Ilāhiyyāt) III.4, of the thesis that a

90  The Lecture on Nature III.6, §9, 208.2–5.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

116  Andreas Lammer moving point can bring about, and cause, the existence of a line. There, Avicenna argues as follows: That which is said that the point draws a line through its motion, it, indeed, is a matter said due to the imagination and has no possibility of [concrete] existence (amr yuqālu li-­l-­takhayyul wa-­lā imkāna wujūd lahū), not because one could not assume for the point a contact that locally moves [steadily with it] . . . but, since the contact is not stable and [since] the thing does not remain after the contact (except just as it was before the contact), so neither would there be a point that remains as the starting point of a line after the contact nor would there remain an extension (imtidād) between it and the parts of the contact, because that point became only one single point through the contact and nothing else, as you have learned in natural philosophy. So, if the contact ceases through motion, then how would it remain a point? Likewise, how could that which is a starting point for it remain as a stable drawing? Rather, this is something that exclusively belongs only to the estimation and imagination (innamā . . . fī l-­wahm wa-­l-­ takhayyul faqaṭ). (Metaphysics (al-­Ilāhiyyāt) III.4, §11, 115.5–12, tr. Marmura, modified)

The conclusion of this argument is clear: if we say that an indivisible point brings about a line through its motion, then what we say has only (innama, faqat ̣) relevance for, and obtains in, our imagination (fī l-­wahm wa-­l-­takhayyul). A point may surely be moving forward (accidentally, as we learned in the preceding section) and we can certainly imagine that it draws out a line through its motion, but this line remains an image with absolutely no impact on, or relevance for, what there is in actual reality and concrete existence (wujūd). It is, thus, simply untrue to say that an indivisible point could really bring about, and cause, the external existence of a line. All a single point can bring about is a single actualized point-­sized contact—and that is not enough for a line. One reason for this is that a point is no pen; it has no ink and, thus, leaves no trace.91 91  Perhaps an apposite image for a point’s motion could be a slowly moving flashlight in the absolute dark.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  117 Another reason is that a motion only exists as actual in one present point or moment but does not exist as an extended reality, as Avicenna famously states.92 Thus, the point also actualizes only one corresponding contacting point. This means that all that actually exists of the line that is assumed to be drawn out by the moving point is the one single indivisible spot that marks the current and instantaneous interim pos­ ition of the moving point. All other parts of the assumed line, including the starting point, at which the moving point once was, and the endpoint, at which it eventually will be, have no existence in concrete reality during the motion. Of course they may be retained in the memory or anticipated in the imagination, but as parts of that motion they do not exist in concrete reality—either any more (as points actualized in the past) or not yet (as those that have not yet been actualized)—nor does the assumed extension (imtidād) of the whole line exist. Consequently, the only thing that may perhaps be said to have been created by the moving point is the very point with which it currently is in contact.93 Regarding the analogous case of a now said to produce time through its flow, we should expect a similar result. All that actually exists of time is the present moment, because the past is no more and the future not yet existent.94 This present can be—but does not have to be—marked by a now.95 Time, then, which is the (extended) magnitude of motion, does not exist with all its parts together as actual. The same applies to the case of motion, for motion does not exist as an extended reality, either. The motion as extended exists only in the mind, whereas in concrete reality all that exists of motion is the present state of being in motion, which is

92  Indeed, this is one of the key points in Avicenna’s account of motion; see The Lecture on Nature II.1, §§5–6, 83.18–84.19; note also the parallel terminology in the present passage and in The Lecture on Nature II.1. §5, 83.18–84.8. On Avicenna’s account of motion, see the brief account in Lammer, The Elements of Avicenna’s Physics, 361 as well as the literature referred to there; see also the next paragraph as well as fn. 96, 117. 93 Avicenna does not deny that lines exist in concrete reality. The edge of a cube, for ex­ample, is a line and it surely exists, because the cube exists having a one-­dimensional edge. What Avicenna does deny is that an existing and indivisible point can produce a line through its motion, because motion does not exist as an extended whole in concrete reality. All that exists of a motion is its momentary state during the motion or its finished result after the motion. 94 See The Lecture on Nature II.13, §1, 166.6–167.14; see also Shayegan, Avicenna on Time, 20. 95 See The Lecture on Nature II.13, §9, 172.12.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

118  Andreas Lammer a first perfection existing in the moving thing.96 All this is a significant feature of motion and time (being the magnitude of motion), and constitutes their primary difference from spatial distance. In the case of distance, closer and more remote positions (such as the position next to this tree and the position next to that tree over there) do exist sim­ul­tan­ eous­ly together with each other, whereas in motion the prior does not exist together with the posterior nor does the before exist together with the after in time. In this way, distance can exist as an extended reality, but motion and time cannot, save as a figment imagined in the mind. To all this, Avicenna adds another consideration why the point cannot bring about a line through its motion: Moreover, without doubt [the point’s] motion indeed is when there is an [already] existing thing on which or in which (ʿalayhi aw fīhi) the motion occurs. This thing is receptive to having motion take place in it, and so it is a body or a surface or a dimension in a surface or a dimension which is a line. Thus, these things are existent before the point’s motion (hadhihī l-­ashyāʾ mawjūda qabla ḥarakat al-­nuqṭa), and so the point’s motion is not a cause for these to exist (fa-­lā takūnu ḥarakat al-­nuqṭa ʿillatan li-­an tūjada hiya). (Metaphysics (al-­Ilāhiyyāt) III.4, §11, 115.12–15, tr. Marmura, modified)

For us, this consideration is nothing new. Above it was explained that Avicenna introduces the now as something that is a result of time. If there were no time (and if time were not continuous), then one could not at will posit divisions and nows in it. The concept of the now, thus, presupposes the concept and the existence of time. Here, in that passage from the Metaphysics (al-­Ilāhiyyāt), we are told that, in order to move, the point requires something already existent on which or in which (ʿalayhi aw fīhi) to move. Rather than bringing about the existence of a 96  For the distinction between these two ways of speaking about motion, which in post-­ Avicennian literature are sometimes referred to as ḥaraka qat ̣ʿiyya vis-­à-­vis ḥaraka tawassut ̣iyya, see esp. Hasnawi, “La definition du movement,” 228–32; Asad Q. Ahmed, “The Reception of Avicenna’s Theory of Motion in the Twelfth Century,” Arabic Sciences and Philosophy 26 (2016), 215–43.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  119 line through its motion, a point requires the existence of a line (or at least a dimension), in order to be able to move on it. We might say, as an example, that a train could not move from one city to another without rails having been laid first. The rails, then, must exist before the train could move on them just as the line or the surface or the body must exist “before” (qabla) there could be a point’s motion on it. This harmonizes well with a remark in Avicenna’s discussion of place in The Lecture on Nature II.5, where he emphasizes that a point is a “privative meaning” (maʿnan ʿadamī), i.e., it does not have concrete existence. As such, it does not itself have a place or a position and, even less so, engage in local motion.97 All this corresponds to, and is forcefully formulated in, Avicenna’s Persian work The Philosophy for ʿAlāʾ al-­Dawla (Dāneshnāme-­ye ʿAlāʾī). There, while discussing quantity in the section on metaphysics, Avicenna writes the following: Whenever [it is] that we imagine that a point moves in a place, a line comes about from its motion in the imagination (andar wahm); whenever [it is] that we imagine that the line moves in a direction different than that [i.e., different from the direction of the point’s motion], its motion turns into surface; and whenever a surface moves in a direction different from these two, its motion turns into depth. Do not, however, think that this assertion is reality (ma-­pandār ke īn sokhanī ast be-­ḥaqīqat)! It is rather a simile (be-­mathal ast), for [some] people do think that the line really comes about through the motion of the point, while they do not know that this motion occurs in a place—and that place already has depth and magnitude before (pīsh az) that point brings about the line or the line the surface or the surface the depth. (The Philosophy for ʿAlāʾ al-­Dawla (Dāneshnāme-­ye ʿAlāʾī) II.10, 32.7–14)

This is a clear passage—and as if it was not yet clear enough, Avicenna immediately mentions time in the very next sentence, so as to say that all

97  The Lecture on Nature II.5, §5, 112.18–113.1.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

120  Andreas Lammer this pertains to time, too, because time, just as a line and a body, is a magnitude in the category of quantity, thus following the same rules: As for time, it is the magnitude of motion, as you shall come to know in natural philosophy. So, you have recognized continuous quantity and also that it is an accident. (The Philosophy for ʿAlāʾ al-­Dawla (Dāneshnāme-­ye ʿAlāʾī) II.10, 32.15f.)

The arguments here in Metaphysics (al-­Ilāhiyyāt) III.4 and in The Philosophy for ʿAlāʾ al-­Dawla (Dāneshnāme-­ye ʿAlāʾī) II.10 correspond with each other—just compare, for example, the former’s assertion that “these things [sc. the line, the surface, and the body] are existent before (qabla) the point’s motion” with the latter’s statement that “place has depth and magnitude before (pīsh az) that point brings about the line or the line the surface or the surface the depth.” In addition, both cor­res­ pond to what we have read in The Lecture on Nature. Consequently, regarding the question whether or not a now could produce time through its flow like a point may produce a line through its motion, it is clear that (i) before a point or a now could move and flow they require something in, on, or through which to move and flow. Without any spatial dimension already existing, no point could move (and much less could a point create a hitherto non-­existent dimension or magnitude by moving). The same pertains to the now: without time—i.e., without a temporal dimension—already existing, no now could even be posited (and much less could a now generate a hitherto non-­existent time by flowing). Moreover it is also clear that (ii) whatever a point and a now may produce when considered as moving and flowing—i.e., the extended result—exists in its extended form only in the imagination. If we take it all together, it emerges that, for Avicenna, there are not three different kinds of nows, i.e., the dividing now, the connecting now, and the moving now; nor are there two different kinds, i.e., the connecting/dividing now and the moving now. Instead, there is only one single concept of a now—and this concept is always the same and entirely potential. The now is purely a result of the mind’s activity of positing, assuming, stipulating, or imagining an indivisible moment within the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  121 continuity of time, be it as a dividing or connecting point, as a present point, or as a moving point—it always remains to be a point in time. Therefore, the now cannot be the cause of time’s existence in any true sense of the word.98

5.  Is There a Flowing Now Producing Time in The Lecture on Nature II.12? There are, however, a number of passages within chapter twelve of the second book of Avicenna’s Lecture on Nature that have been taken to indicate that the now is somehow responsible for the existence of time. The presumed expository strength of these passages, alongside their alleged clarity, have so far outweighed Avicenna’s perspicuity and determination in characterizing the existential status of the now as “im­agined.” Above, we have seen Avicenna introducing the now as a faṣl mutawahhim. In the next sentence, he wrote that the existence of the now as a wāṣil is ʿalā an yatawahhama l-­wahm—imagined by the im­agina­tive faculty.99 A few paragraphs later, he adds another possible understanding of the now: That which we have discussed [so far], then, is the now that is surrounded by the past and the future . . . Yet, another now with another description might be imagined (wa-­qad yatawahhamu ān ākhar ʿalā ṣifa ukhrā). So, just as through the motion and the flow of the limit of the moving thing100—and let it be some point (nuqṭa mā) —you assume (tafraḍu) some distance or, rather, some line—as if it (ka-­annahū), I mean this limit, were a thing-­in-­motion (al-­muntaqil) —and, then, in that line, you assume points which are not that which makes up the

98  The act of positing a now, be it a dividing, a connecting, or a moving now, is not entirely without purpose, though. It has a purpose—a didactic purpose, as it seems and as I shall argue below, 137ff. 99  The Lecture on Nature II.12, §1, 160.5f.; cf. The Eastern Wisdom (al-­Ḥikma al-­ mashriqiyya) III.11, 35.1. 100  Shayegan translates the text here as “just as [a ‘now’ which is] the limit of the moving thing . . . by means of whose movement and flux one assumes a certain distance” (68). There is nothing in the Arabic text corresponding to her addition, which, in turn, very much predetermines her interpretation. For Avicenna, though, a now is not the limit of the moving thing.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

122  Andreas Lammer line but rather are imagined (al-­mutawahhima) as con­nect­ors for it, like that it would seem (yashbahu) that in time and in motion, meaning traversal [motion], there was [also] something like this and something like the internal points in the line which do not make it up—and that is that a thing-­in-­motion and an edge101 is im­agined (yatawahhamu) in the distance and in time. (The Lecture on Nature II.12, §5, 163.11–16, tr. McGinnis, modified)102

With these words, Avicenna introduces the flowing now. Although he introduces this now as “another now,” it is not a new now, i.e., a now with a different ontological status, but only the already known now “under another description.” What is striking is the strong presence of references to the imagination in this passage. The now under the descriptions of being a connector and a separator, which Avicenna discussed before and which he here mentions again as the “now that is surrounded by the past and the future,” is once more characterized as imagined. Alongside these, Avicenna declares, one might be able to imagine (qad yatawahhamu) the now somewhat differently. He presents this now through an analogy with the thing-­in-­motion (al-­muntaqil) that we remember from Aristotle and that we are asked to visualize as some point (nuqt ̣a mā) and edge (ḥadd). The motion of the thing-­in-­ motion or, more precisely, the motion of a point on a moving thing’s limits allows us—Avicenna addresses the reader—to assume, posit, or mentally represent (tafraḍu) a distance or a line which the thing-­in-­ motion and the limiting point traverse (or have traversed). Considering 101 Reading wa-­ḥadd with Āl Yāsīn and McGinnis for wujida in Zāyid; Mss. Leiden or. 4 and or. 84 provide an undotted rasm. 102 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 36.18–37.3. Following “meaning traversal [motion],” the corresponding passage from The Eastern Wisdom (al-­Ḥikma al-­ mashriqiyya) has a different text, which does not seem to be safely transmitted. I would tentatively translate it as follows: “. . . like that it would seem (yashbahu) that in time and in motion, meaning traversal [motion], there is (i) something which produces through its un­inter­rupt­ed permanence (bi-­stimrārihī) a time and a motion, meaning traversal [motion], and (ii) something posited in it (shay yufraḍu fīhi) after the occurrence of time (baʿda ḥuṣūl al-­zamān), for motion is like a point internal in—not productive of—the line, imagined after the origination of the line (baʿda ḥuḍūth al-­khat ̣t).” This addition needs to be re-­evaluated after the text has been properly edited. In its current form, I take it as supporting my overall argument, especially when compared to subsection (c.) above, 115ff.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  123 this situation, it seems (yashbahu) that there is some sort of a temporal equivalent to the point at the limit of the thing-­in-­motion, i.e., something that is for a temporal duration what the indivisible point is for the traversed line and what the thing-­in-­motion taken as an indivisible point or edge is for the traversed distance. In other words, it would seem that a flowing now that corresponds to the moving thing and the moving point exists—exists in the imagination, that is. All this calls to mind what has been mentioned above already with regard to Aristotle and, in particular, to Philoponus’ and Simplicius’ interpretations of Aristotle. The commentators relied on Aristotle’s ana­ logy between the thing-­in-­motion (τὸ φερόμενον) and a point (ἡ στιγμή), and introduced the terminology of the flow (ἡ ῥύσις) through which, they argued, the now is productive (ποιητικόν) and generative (γεννητική) of time.103 This resemblance notwithstanding, it seems requis­ ite to emphasize the great difference between Avicenna and his predecessors, for Avicenna stresses—repeatedly—the importance of the imagination for any function of the now. As a philosophical concept, the now surely may indicate, or elucidate, certain real features of time, such as its continuity and infinite divisibility, but it does not establish or prod­uce these features. Further, what Avicenna relates about the assumed points which do not make up the line but are in the line as imagined connectors reminds us not only of Avicenna’s unequivocally clear description of the separating and connecting function of a posited now as the result of the im­agin­ ation but also recalls what we have read in his Metaphysics (al-­Ilāhiyyāt) about the moving point and the line, i.e., that the line is not made up of actualized points but that these points may be assumed in the line as connectors and as positions at which the moving point once was or will be.104 There is nothing in the way Avicenna introduces the idea of the flowing now in The Lecture on Nature (or the corresponding parts in The Eastern Wisdom (al-­Ḥ ikma al-­mashriqiyya)) nor is there anything in his 103  See above, 84ff., esp. 86f. 104  What is more, Avicenna’s assertion about the points which do not make up the line (but are within the line) aligns well with his explicit formulation in The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 36.18–37.3, translated above in fn. 102, 121 (points “internal in—not productive of—the line”).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

124  Andreas Lammer other works, such as the Metaphysics (al-­Ilāhiyyāt) and the Categories (al-­Maqūlāt) or in the physical sections of his Salvation (al-­Najāt), The Philosophy for ʿArūḍī (al-­Ḥ ikma al-­ ʿArūḍiyya), The Guidance (al-­ Hidāya), and The Founts of Wisdom (ʿUyūn al-­ḥikma), that could leave any doubt about that the flowing now, just as any ‘other’ now, is purely the result of the imagination. It is truly difficult to see how the now could be something that causes the concrete, actual, external existence of time.105 Accordingly, Shayegan’s claim that the now is merely a conjectural entity is entirely justified. It is, however, all the more surprising how she could, then, maintain that the imagined and conjectural now, nonetheless, “generates” time, as she wrote repeatedly. For her, “the dynamic elem­ent of the moving ‘now’ following the moving ‘thing’ and the moving ‘point’ are pure conjectures and are references to geometrical objects, not to actual things.”106 It may have been that what she, in fact, was arguing for is that the (extended) time which is produced by the flowing now is, like the now itself, merely conjectural and only existent in the imagination. If that is what she thinks, then I would agree with her analysis in this regard.107 It could also be that she just failed to mention that she thinks that the now as a conjecture refers to the moving body (and ultimately, of course, to the moving body of the outermost heavenly sphere) which, then, as she writes elsewhere, actually causes time, so that the conjectured now, insofar as it represents that body, could in a derivative way be said to generate time.108 However, she does not seem to formulate it that way. It may, however, also be the case that Shayegan simply overlooked the issue that an imagined entity could not cause something’s concrete existence. This is how McGinnis seems to have read her analysis. On that reading, then, it would be with good reason that McGinnis criticized Shayegan’s position. Yet, McGinnis’ own assertion, that the now, though 105 Avicenna’s Philosophy for ʿAlāʾ al-­Dawla (Dāneshnāme-­ye ʿAlāʾī) and Pointers and Reminders (al-­Ishārāt wa-­l-­tanbīhāt) are not mentioned here, because the former lacks a discussion of time, while the latter lacks the notion of the now. 106 Shayegan, Avicenna on Time, 205; see above fn. 100, 121. 107  See Shayegan, Avicenna on Time, esp. 16–20. 108 Shayegan, Avicenna on Time, 20: “As for infinite time, it is caused by the circular motion of the First Sphere.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  125 not being an actual or real object, is still somehow real enough as to be the producer of time, is no less enigmatic. It seems that McGinnis came to this conclusion, because he—like Mayer—felt the need to accommodate and acknowledge what he took to be the central idea of Avicenna’s now, viz., that it is precisely that which produces time through its flow. So, the flowing now, McGinnis seems to argue, just could not be fully imaginary, because otherwise it could not produce time, which he took to be a fact supported by the text itself. Thus, McGinnis describes the now, although it was expressly said to be “imagined,” as “a real state belonging to that which is borne along.”109 Again, it may be that what McGinnis wanted to say is precisely how I suggested Shayegan’s position could be saved, viz., that the now (though being an imagined entity) has a relation to the moving body and, since the moving body of the outermost sphere produces time, so the imagined now (insofar as it has a relation to that body) could likewise be said to generate time in a derivative, imaginative sense through an accidental motion. Yet, like Shayegan, McGinnis nowhere seems to formulate it that way.110 What needs to be done now in response to these interpretations is to investigate those passages by Avicenna which have so far been read as a confirmation of the claim that the now is the producer of time, and to analyze his analogy between the thing-­in-­motion and the now. This analogy is explained by Avicenna as follows:

109 McGinnis, Time and Time Again, 286f.; “Ibn Sīnā on the Now,” 100f.; on Shayegan’s and McGinnis’ views, see above, 106ff. 110  It may be added that Avicenna does not formulate it that way either, for he does not—in any way whatsoever and in any of his works—take recourse to the idea of a now when he establishes the rotation of the outermost sphere and describes how this motion causes the existence of time. What is more, it seems that if the now were an actual feature pertaining to the body of the outermost sphere (if that, indeed, is a reasonable thing to say), even then the now would engage only in accidental motion, so that, again, it simply would be more accurate to say that it is the essential motion of the outermost sphere—and not the accidental motion of the now which exists in, on, or with the outermost sphere—which brings about time. Finally, Avicenna could also point towards Metaphysics (al-­Ilāhiyyāt) III.4, §11, 115.5–15 (discussed above, 115ff.) where he argued that the point cannot be what produces a line, because it actually requires, and presupposes, a line, in order to engage in motion. Thus, the now, which perhaps may exist as an accidental feature of the body of the outermost sphere, could flow indeed—however, only through a time which the body of the sphere already (in an essential sense of priority) produced through its essential motion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

126  Andreas Lammer So, the thing-­in-­motion performs (yafʿalu) a continuous local motion over a continuous distance to which111 a continuous time conforms. So, it is as if (fa-­ka-­anna) the thing-­in-­motion, or rather its state (ḥālatahū) which belongs to it during the motion, were an indivisible limit producing (faʿʿāl) a continuity through its flow (bi-­sayalānihī), and [it is also as if] a point conforms to it in terms of the distance and a now in terms of time. Thus, together with it [sc. the thing-­in-­motion] is neither the line of the distance (for it has left it behind) nor the motion in the sense of ‘traversal [motion]’ (for it has been completed) nor time (for it has gone by). What is together with it [sc. the thing-­in-­motion] of each one [of these, i.e., the motion, the distance, and the time], is only an indivisible limit that belongs to them and divides them. (The Lecture on Nature II.12, §5, 163.16–164.1)112

The entire analogy between the point, the thing-­in-­motion, and the now is due to the intrinsic relation between distance, motion, and time, as has been mentioned before. While a thing moves, it covers a certain distance and requires some time to do so. On that basis, one might think that just as there is, on a kinetic level, a thing-­in-­motion which, being in the state of moving, performs the motion, so there should be, on a spatial level, a point which covers the distance and, on a temporal level, a now which flows through time. Immediately, Avicenna adds that this does not mean that there comes-­to-­be a real extended line, for all there is of the line is the one point at which the thing-­in-­motion currently is.113 Likewise, there does not come-­to-­be an extended motion in the sense of a traversal motion (al-­ḥaraka al-­qat ̣ʿiyya), for all there is of the motion is the “state” (ḥāla) of being in motion, i.e., the form of motion, that belongs to the thing-­in-­motion as a first perfection.114 Finally, there is also no extended time, for all there is of time is the present moment— which apparently can be marked in the imagination by a now—in which 111  The referent of the masculine singular suffix of the verb (yut ̣ābiquhū) is unclear. It is doubtful that it refers to the grammatically masculine thing-­in-­motion, because it is actually the now which corresponds to the thing-­in-­motion and the point, whereas time corresponds to motion (feminine) and distance (feminine). One would, thus, expect a feminine suffix. 112 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 37.3–7. 113  See the discussion of Metaphysics (al-­Ilāhiyyāt) III.4, §11 above, 115ff. 114  On these notions, see above fn. 96, 117.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  127 the thing-­in-­motion is in motion. Thus, it could be said that the moving thing corresponds to three indivisible limits: a kinetic state of being in motion, a spatial position, and a temporal now: So, together with it [sc. the thing-­in-­motion] is always of time a now, and of the traversal [motion] something which we have proven to be the motion in reality as long as the thing is in motion, and of the distance a limit [which is] either a point or something like it. Each one of these is a limit.  (The Lecture on Nature II.12, §5, 164.1f.)115

This, at least, is what might be said about motion or what seems to be the case or what could be imagined. What is required at this point, then, is an investigation of whether or not all this has any concern with the concrete reality and, in particular, whether this is really true for time, because time is what is currently under examination. Thus, Avicenna sets his agenda as follows: So, it is appropriate for us that we inquire whether – just as the thing-­ in-­motion in itself is one (dhātuhū wāḥida) and through its flow acts as (faʿala) that which is its edge and limit (ḥadduhū wa-­nihāyatuhū), and produces (faʿala) the distance – there is also in time a thing which flows which is the now (shayʾ huwa l-­ān yasīlu). (The Lecture on Nature II.12, §6, 164.4–6)

With this, Avicenna addresses the question of what we can learn about time on the basis of the offered analogy between the thing-­in-­motion, a point, and the now. He wants to find out whether there is something in time that serves as a flowing now just as there is a thing-­in-­motion that serves our imagination as a moving edge. The reader is asked to visualize or consider the moving thing as an indivisible point or limit or edge of a motion.116 This moving edge and limit, then, is taken to produce the 115 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 37.7–9. 116  It thus seems to be preferable to translate faʿala in the first instance along the lines of “to act as” rather than “to produce.” In the analogy, the thing-­in-­motion serves as its edge and limit (ḥadduhū wa-­nihāyatuhū), because we are allowing it to be an edge and limit in the motion, and present it as an edge and limit in our analogy; see also Dominicus Gundisalvi’s

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

128  Andreas Lammer distance that it was covering through the motion, like the common view that a point can be said to produce a line through its flow. We already know that, for Avicenna, there first has to be a line before a point can traverse the line. This seems to be true here, too, for it would be absurd to claim that the moving thing really brings about the existence of the distance it traverses, as if a train from Munich to Berlin would generate the distance between Munich and Berlin. This distance existed long before the train departed the station in Munich and is a spatial feature of the real world. As such, it does not need to be produced, because it already exists. Nonetheless, in planning a train journey, one may mentally represent the train as a moving edge and limit that, through its motion, brings about the distance it traverses, even though, in concrete reality, the spatial stretch exists already and even the rails have been laid decades ago. Thus, the question that Avicenna singles out for further investigation is already within the realm of imagination, because the analogy tries to compare the flowing now with something that is an image designed to help us grasp what it means to be in motion and to be traversing a distance (which, admittedly, are abstract enough concepts, especially in Avicenna’s philosophy, to deserve some visualizing aid).117 Overall, Avicenna’s question here is reminiscent of the following ­passage from Philoponus’ commentary on Aristotle’s Physics: The now, too, is a productive cause of time, since its flow generates time (ἡ γὰρ τούτου ῥύσις γεννᾷ τὸν χρόνον). So, if as the point stands to the magnitude and the thing-­in-­motion to the motion, so the now stands to time. If the point being one (ἓν ὂν) generates magnitude . . . and likewise the moving body being one (ἓν ὂν) generates the motion, it surely follows that also the now being one (ἓν ὂν) generates time. (Philoponus, In Phys. 727.21–6, tr. Broadie, modified)118

twelfth-­century Latin translation which, similar to my proposal, renders faʿala mā huwa ḥadduhū wa-­nihāyatuhū as qui fecit quod ipse est terminus et finis (Avicenna Latinus, Sufficientia: Liber primus naturalium II.12, 164). 117  Of course, the thing-­in-­motion as such is real—but imagining it as a moving edge which produces a distance is not. 118  See also above, 87f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  129 It seems that Avicenna’s dhātuhū wāḥida corresponds to Philoponus’ ἓν ὂν, as does his general outline of the situation. The major difference between the two passages, however, is that Philoponus seems to take the situation for granted, whereas Avicenna deems it appropriate, first, to ask whether the situation is, in fact, the case at all (fa-­ḥarīy bi-­nā an nanẓura hal . . .). Within the context of Avicenna’s investigation, this is a reasonable question given that Avicenna—unlike Philoponus—has not stopped emphasizing that the now is nothing other than an imagined division through which we mentally demarcate a point in time or, in this particular case, a temporal edge which our imagination can attribute to the thing-­in-­motion. There is, then, a considerable difference in tone and context between the passage in Avicenna and that in Philoponus. So, Avicenna asks whether or not there is something in time which is a now that is flowing (yaṣīlu): So, if a thing like this is existent, then it is true what is said (mā yuqālu) that the now produces time through its flow. (The Lecture on Nature II.12, §6, 164.10f.)119

With this, Avicenna acknowledges again that all this is worth an investigation. After all, the position that a flowing now is responsible for pro­ du­ cing time has been defended by respectable scholars within the Peripatetic tradition. Avicenna directly refers to “what is said” in the tradition before him and provides a formulation that resembles closely Alexander’s al-­ān idha sāla ʿamila zamānan (“the now, when it flows makes time”) and Philoponus’ τὸ νῦν δὲ τοῦ χρόνου ποιητικὸν αἴτιον· ἡ γὰρ τούτου ῥύσις γεννᾷ τὸν χρόνον.120 Yet, Avicenna does not yet say that this is true; he merely states that it would be true to say this if a thing like this existed. He does not assert that a thing like this exists (and there is, in fact, no place where he does—neither in his Lecture on Nature nor 119 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya) III.11, 37.9: “So, on the basis of this it is true what is said that the now produces time through its flow.” This sentence follows immediately upon a passage that corresponds to the above quotation from The Lecture on Nature II.12, §5, 164.1f., so that “on the basis of this” must be taken as referring to that very ex­plan­ ation. In other words, I take it that it is only in light of the above explanation that it is possible to say—as tradition has done—that the now produces time through its flow. 120  Alexander of Aphrodisias, Maqāla fī l-­zamān, 21.13, 21.17; Philoponus, In Phys., 727.21.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

130  Andreas Lammer anywhere else in his major works). To the contrary, he repeatedly stresses the conditional character of the investigation. Had Avicenna really wanted to say that the now produces time through its flow, he would have been well advised to frame his discussion differently, to explain how something imagined could produce something concrete, and perhaps even to announce his allegiance to the view of his predecessors. In fact, however, he does nothing of that sort.121 This brings us back to the passage that usually is taken as direct evidence for Avicenna’s adherence to the flowing now being the producer of time and that we have above seen partially reproduced in al-­ʿIrāqī’s book on Avicenna’s physics. In full, Avicenna writes the following: If existence belongs to this thing1 (fa-­in kāna li-­hādhā l-­shayʾ wujūd), then [the existence] is the existence of the thing2 as associated (maqrūnan) with the meaning which we earlier verified as motion without taking122 what is prior and what is posterior and what is simultaneous. Just as its [sc. thing2’s] being something that has a ‘where’ (kawnuhū dhā ayn), when it [sc. thing2] remains while flowing through the distance (istamarra sāʾilan fī l-­masāfa), produces motion (aḥdatha l-­ḥaraka), so its [sc. thing2’s] being something that has that meaning which we called ‘the now’ (kawnuhū dhā dhālikā l-­maʿnā lladhī sammaynāhu l-­ān), when it [sc. thing2] remains through the prior and posterior of motion (istamarra fī l-­mutaqaddim al-­ḥaraka wa-­mutaʾakhkhirihā), produces time (aḥdatha l-­zamān). So, the relation of this thing2 to the prior and posterior is in its [sc. thing2’s] being

121 An argumentum ex silentio surely cannot be much more than a hint. It is duly admitted that one could easily object and ask—against the here defended interpretation—why, conversely, Avicenna was not more straightforward in rejecting the idea of a flowing now pro­du­ cing time. A possible answer to this is that Avicenna may have thought that he was already straightforward enough with his constant reminder of the imaginative status of the now and the omnipresent hesitancy expressed by the conditional “if a thing like this is existent.” After all, if my interpretation is right and chapter II.12 is, indeed, a mere appendix to the actual discussion of time, issued by the fact that time, just as any continuous magnitude, is potentially divisible, then Avicenna was very clear already, as he made a clear distinction between what is immediately relevant to his discussion of time (chapter II.11) and what is not (chapter II.12). 122 Reading min ghayr akhadha with Mss. Leiden or. 4 and or. 84 as well as Āl Yāsīn and McGinnis for min ghayr in Zāyid.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  131 a now (fī kawnihī ānan),123 which in itself is something which makes time and numbers time (wa-­huwa fī nafsihī shayʾ yafʿalu l-­zamān wa-­ yaʿuddu l-­zamān) through what comes-­to-­be when we take a now from among the edges in them [sc. the prior and posterior]. (The Lecture on Nature II.12, §6, 164.12–16)124

Following upon Avicenna’s declaration that “if a thing like this is existent, then it is true what is said that the now produces time through its flow,” Avicenna now begins to investigate whether existence belongs to “this thing” (marked as thing1) and to explain what the existence in question is. His answer is that the existence in question is the existence of another thing (marked as thing2). This other thing is apparently significant, because it produces (aḥdatha) motion as well as time—the former insofar as it “has a ‘where,’ ” the latter insofar as it “has that meaning which we called ‘the now.’ ” So, the grammatical subject of the verb aḥdatha is undoubtedly the “thing” (i.e., thing2) and not the “now” (as al-­ʿIrāqī made us believe)125 or the “where” (which would be a consequence for the first occurrence of aḥdatha on al-­ʿIrāqī’s reading). What is more, it is not difficult to see what that thing2 is that serves as the grammatical subject of aḥdatha: it is the thing-­in-­motion. The very fact that the thing-­in-­motion has a place or position (a “where”) together with (maqrūnan bi-) its being in a state of motion creates our mental representation of a motion as an extended whole with the thing-­in-­ motion being that which “remains” (istamarra) what it is while it continues to be “flowing through the distance” (sāʾilan fī l-­masāfa). This is not unlike Aristotle’s example of Coriscus who, though walking from the Lyceum to the marketplace, remains one and the same Coriscus.126 In other words, motion occurs through (i) a thing-­in-­motion that (ii) is in fact currently moving and that (iii) is spatially located between here (where it was) and there (where it will be), and that (iv) endures throughout the motion. Therefore, it is the thing-­in-­motion as such that

123  Āl Yāsīn and McGinnis read fī here, which is omitted in Zāyid and not attested in Mss. Leiden or. 4 and or. 84. 124 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya), 37.11–14. 125  See above, 105f. 126 Cf. Phys. IV.11, 219b18–25; q.v. above, 89f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

132  Andreas Lammer performs a local motion and, thereby, can be said to produce that motion, especially if we envisage that motion as an extended whole in our minds. Moreover, the thing-­in-­motion is also that which, con­ sidered as a spatial limiting point, can be said to produce the distance as mentioned above, even though, in reality, it produces neither the motion nor the distance as extended realities; it rather performs the motion and, thereby, traverses the already existing distance, as we have seen. That motion, or distance, is not the product of a flowing place (or a flowing ‘where’) is immediately plausible, both philosophically and from the vantage point of an Aristotelian definition of place, which Avicenna emphatically shares, as I have shown elsewhere.127 A thing that is undergoing local motion precisely changes its place throughout its motion. In fact, this is what it means to undergo local motion. Moreover, there is no single place that is borne along with the thing-­in-­motion. If Coriscus’ place would be a remaining feature that moves with Coriscus (as, for example, his garments do), then he would be unable to move spatially, i.e., to change his place, because wherever he goes, he brings the same place with him (just as he wears the same garments). This would make Coriscus unable to go anywhere, because he would essentially be staying in the same place (just as he is wearing the same garments).128 Admittedly, however, the image of a flowing place, or more precisely of a moving thing insofar as it has a ‘where’ (kawnuhū dhā ayn), can serve as a didactic tool for someone who has trouble understanding the meaning of local motion, because through this image one can mentally represent, i.e., imagine, motion as an extended whole. This, however, does not constitute how reality ought to be described and it is already known that, for Avicenna, motion does not exist as an extended reality. The situation about time is parallel to the situation with place. Once more Avicenna refers to thing2, i.e., to the thing-­in-­motion, which is now considered not insofar as it “has a ‘where’ ” but insofar as it “has that meaning which we called ‘the now’ ” (kawnuhū dhā dhālikā l-­maʿnā

127  See Lammer, The Elements of Avicenna’s Physics, cf. 5. 128  See Lammer, The Elements of Avicenna’s Physics, 313f.; see also The Lecture on Nature II.5, §2, 112.3–6; see further Keimpe Algra, Concepts of Space in Greek Thought (Leiden: Brill, 1995), 16f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  133 lladhī sammaynāhu l-­ān).129 Just as the thing-­in-­motion by having a ‘where’ was considered as spatially located between here (where it was) and there (where it will be), so the thing-­in-­motion by having a ‘now’ is here considered as located between what is prior and what is posterior. That is to say, it refers to the thing-­in-­motion qua being in motion, i.e., qua having the form of motion or, as Avicenna put it above, in its “relation . . . to the prior and posterior.” Considered as such, this thing-­ in-­motion exists at one particular moment, and qua being in motion it is such that it never has been at this moment before nor will it ever be there again (which is one of the core characteristics of motion, according to Avicenna).130 Once this has been understood, it emerges that this is also precisely how Avicenna introduced the above passage, for he wrote: “If existence belongs to this thing1, then [the existence] is the existence of the thing2 as associated with the meaning which we earlier verified as motion without taking what is prior and what is posterior and what is simultaneous.” The “thing2” that Avicenna is talking about, and the thing that has a ‘where’ and a ‘now,’ is precisely the thing-­in-­motion insofar as it is currently engaging in motion. The reference to the ‘where’ elucidates that the thing-­in-­motion is moving from ‘here’ to ‘there,’ i.e., from one position to another position, currently being situated in-­between, while the reference to the ‘now’ clarifies that the motion of the thing-­in-­motion is advancing from ‘prior’ to ‘posterior,’ likewise currently being situated in-­between. Thus, all in all, Avicenna says that the thing-­in-­motion is moving now, i.e., is currently undergoing motion, has the form of motion in this very moment, and is, thus, somewhere between what is prior and what is posterior, where ‘prior’ and ‘posterior’ have both a spatial and a temporal dimension (alongside their kinetic import). On this reading, Avicenna expressed himself clearly in this discussion, for all he says and writes about is the thing-­in-­motion. It is the thing-­in-­motion that performs a motion, traverses a distance, and prod­ uces time. It is not that the where and the when produce motion and

129 See The Lecture on Nature II.12, §6, 164.9, where Avicenna already referred to “the thing which, for whatever reason, became a now” (al-­shayʾ alladhī li-­amr mā ṣāra ānan). 130 See The Lecture on Nature II.1, §6, 84.9–19, esp. 84.15f., 84.18f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

134  Andreas Lammer time, respectively—it is only the thing-­in-­motion that does so, while it remains what it is during its flow over the distance (istamarra sāʾilan fī l-­masāfa) and through the prior and posterior of motion (istamarra fī l-­mutaqaddim al-­ḥaraka wa-­mutaʾakhkhirihā). Only in this sense does it produce motion and time (aḥdatha l-­ḥaraka and aḥdatha l-­zamān)— but even so it does not produce them as actually extended, for neither exists in actuality as an extended reality. They exist punctually qua their present state, i.e., all that exists in actuality is the thing-­in-­motion having a ‘where’ and a ‘now’ (and that is the thing-­in-­motion qua having the form of motion), whereas, in the mind, motion and time can be im­agined as extended. It should now have become clear that it is a painful misunderstanding to treat, and a severe distortion to quote, the above sentence the way al-­ʿIrāqī has done as if it said “the now, when it remains through the prior and posterior of motion, produces time.”131 In short, I am proposing that Avicenna offers a reductivist approach towards Aristotle’s analogy between the thing-­in-­motion and the now by saying that all there is is the thing-­in-­motion—and this creates time. When Avicenna above asked whether “existence belongs to this thing,” i.e., the flowing now, he gave the answer: this existence is “the existence of the thing as associated with the meaning which we earlier verified as motion”: it is the thing-­in-­motion. Yet, on the reading offered here, it needs to be explained why the same passage above also states that the now is “in itself something which makes time” (fī nafsihī shayʾ yafʿalu l-­zamān). The answer is that this formulation, too, must not be taken out of context. There is no reason to break up the sentence between yafʿalu l-­zaman and wa-­yaʿuddu l-­zamān, as McGinnis did in his edition, presumably following Simone van Riet’s edition of Gundisalvi’s Latin translation.132 What Avicenna said in full is this: So, the relation of this thing2 to the prior and posterior is in its being a now, which133 in itself is something which makes time and numbers 131 al-­ʿIrāqī, al-­Falsafa al-­t ̣abīʿiyya ʿinda Ibn Sīnā, 251. 132  Gundisalvi, however, rearranged the structure of the following clause which may have prepared van Riet’s decision to introduce a paragraph break in the Latin text. 133  The pronoun huwa could also refer to the thing-­in-­motion instead of the now, so that it would then be the thing-­in-­motion that is “in itself something which makes time.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  135 time (wa-­huwa fī nafsihī shayʾ yafʿalu l-­zamān wa-­yaʿuddu l-­zamān) through what comes-­to-­be when we take (idha akhadhnā) a now from among the edges in them [sc. the prior and posterior]. (The Lecture on Nature II.12, §6, 164.15f.)134

With this, Avicenna does nothing other than to emphasize once more that the now is something stipulated and posited by us. The counting or numbering of time is an operation we perform in our imagination, because it is only in the imagination that the whole motion as a process from a starting edge to an ending edge can be represented both spatially and temporally. Reality is only instantaneous, but the mind is capable of providing an impression of the whole extended process—and it does so, when “we” (!) retain the whole extension, mark-­off nows, and, subsequently, compare the thus numbered stretch of time with its parts or with other stretches from other motions. Therefore, it seems that “making time” and “numbering time” are one and the same function of our acts of positing nows. Or, put differently, “making time” does not mean ‘bringing about the existence of time’ (because time already exists as the unnumbered magnitude of the celestial sphere’s motion) but means precisely ‘numbering time’ in the Aristotelian sense.135 Further, we should recognize that in this clause Avicenna does not mention the flow, i.e., he does precisely not say “the now is in itself something which makes time through its flow” (yafʿalu bi-­sayalānihī). Instead, he writes that the now makes time and numbers time “through what comes-­to-­be when we take” (yafʿalu . . . wa-­yaʿuddu . . . bi-­mā yaḥduthu idha akhadhnā) two extremities. Apparently, the now is only something that through itself or “in itself ” (fī nafsihī) makes and numbers time when we take one now for what is prior and another now for what is posterior and, thereby, obtain a period of time in-­between these two nows that we can represent in our minds. We do the same when we think about the moments of departure of a train and its arrival, attribute a now to each moment, and compare the thus made period of 134 Cf. The Eastern Wisdom (al-­Ḥikma al-­mashriqiyya), 37.13f. 135  Cf. also Phys. IV.11, 219b11f.: τὸ δὲ νῦν τὸν χρόνον ὁρίζει, ᾗ πρότερον καὶ ὕστερον, translated into Arabic as wa-­l-­ān miqdār al-­zamān min jiha annahu yaḥudduhu bi-­l-­mutaqaddim wa-­l-­mutaʾakhkhir.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

136  Andreas Lammer time in-­ between with what we have seen on our watch: the train departed at twelve o’clock (one designated now) and arrived at 1:14 (another designated now), i.e., 74 minutes later, at the next station. In essence, what Avicenna describes here is the same as what Aristotle had explained in the following passage about the apprehension (not about the existence or the creation) of time: But we apprehend (γνωρίζομεν) time only when we mark-­off motion, marking it off by the prior and posterior. It is only when we have perceived the prior and posterior in motion that we say that time has passed. Now we mark-­off by taking them as other and other and some third thing as between them (μεταξύ τι αὐτῶν ἕτερον). When we conceive of (νοήσωμεν) the limits as different from the middle (τοῦ μέσου) and the soul pronounces (εἴπῃ ἡ ψυχὴ) that the nows are two, one prior and one posterior, then it is that we say that there is time, for what is marked-­off by the now is time, as it seems. (Phys. IV.11, 219a22–30, tr. Hardie/Gaye, modified)136

Avicenna clearly takes this up and stresses the mental or imaginative aspect involved in the marking-­off of motion and time by conceiving of ‘wheres’ and ‘nows’ as different. Neither time nor motion is itself already marked-­off by certain points, as we know from Aristotle and Avicenna that a continuous line does not consist of points, so how could there be marked-­off points existing already in them? The marking-­off is an operation of the mind and, ultimately, it is the soul that pronounces two posited nows as different from each other (δύο εἴπῃ ἡ ψυχὴ τὰ νῦν). The now, once again, emerges as an imagined thing that is superimposed on an already existing feature of reality, viz., time. The Greek commentators, however, did not emphasize that the now is the result of a mental operation as much as Avicenna does later. Nonetheless, they used the same picture of the flow in order to explain the constant other-­and-­ other-­ness of the now (ἄλλο καὶ ἄλλο) alongside its essential never-­ changing identity. 136  See Philoponus, In Phys., 721.15–19, 722.11, 737.27f.; Philoponus’ and Ibn al-­Samḥ’s comment in Aristotle, al-­Ṭabīʿa VI.2, 624.10; see also The Lecture on Nature II.13, §1, 167.4f.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  137 What we witness in Avicenna, then, is that he is actually rather s­keptical about the real import of formulations such as al-­ān yafʿalu l-­zamān (“the now produces time”). They only obtain if such a thing as an independent flowing now exists, but it is never shown or said that it exists, and even if it were to exist, Avicenna makes it more than clear that all this is due to an operation in the imaginative faculty (fī l-­wahm wa-­l-­takhayyul), as he wrote, for example, in the Metaphysics (al-­ Ilāhiyyāt), as we have seen.137 The thing-­in-­motion is real, but the now—whether flowing or not—is described by Avicenna as something posited by the imagination. The flowing now is a means for illustrating the temporal dimension of a thing. When we posit the now as belonging to a thing and when that thing is presently moving over a distance from a starting point to an endpoint, we may say that the now produces time through its flow—even though all there is is the thing-­in-­motion. In order to understand time, we do not depend on, or require, the now and its flow. In fact, we can dispense with the now entirely, as has been demonstrated by Avicenna himself when he decided not to mention the now anywhere in his central eleventh chapter of the second book of The Lecture on Nature, in which he discusses and establishes time, nor anywhere prominently in his other major works when discussing time. Moreover, time is a reality, whereas the now is a figment—therefore, the now cannot be the producer of time. Consequently, time, for Avicenna, is not the product of a flowing now or itself the flow of a now.

6. Conclusion In light of this result, we need to ask again why Avicenna, then, discussed the now and its flow at all, if it was not meant to contribute anything to what time actually is in reality. To answer this we should keep in mind that Avicenna discusses the now in a separate chapter after he has provided a full analysis of time. The now should be discussed in an extensive discussion of time, because it is something that belongs to time. The Cure, Avicenna’s most extensive and complete philosophical 137  Metaphysics (al-­Ilāhiyyāt) III.4, §11, 115.12.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

138  Andreas Lammer masterpiece, contains The Lecture on Nature as its foundational work in the field of natural philosophy. It was a reasonable decision by Avicenna to treat the now in The Lecture on Nature, his most extensive and complete treatment of physics, especially in light of the fact that the now is treated in none of his other seven summae, apart from The Eastern Wisdom (al-­Ḥ ikma al-­mashriqiyya), which is closely based on The Cure. Yet, his treatment reads like an after-­thought or appendix—not an unimportant appendix, given the prominence of the now in the tem­ poral theories of his predecessors, but not an essential subject either. Time as an actual feature of reality can be comprehended without taking recourse to the now. Nonetheless, the flow of the now is a powerful image, especially because it seems to capture our own human awareness of time; it is an image for how we represent and imagine time in particular instances of motion. Within the context of The Lecture on Nature, it is didactically expedient to expound the concept of a flowing now—while not all that is didactically expedient is also essential—for explicating our own ex­peri­ence of time’s passing and its constant elapsing and renewal. We may say that although Avicenna does not adopt Alexander’s, Themistius’, and Philoponus’ idea of a flowing now as that which produces time through its flow as a metaphysical story about what there is in concrete reality, he regards the idea as conducive and helpful for teaching purposes insofar as it helps to visualize something that does exist, but not wholly, and which, for that reason, is difficult to understand, viz., time. In The Lecture on Nature, Avicenna takes the idea of a flowing now seriously, and in devoting quite some space to discussing it, he displays, first and foremost, his didactic abilities. Students who already grasped what time is after reading The Lecture on Nature II.11 could perhaps even skip chapter twelve, because they would be able to work out its derivative contents by themselves. Those, however, who still have ­trouble understanding—or visualizing—such abstract concepts as time and motion, are aided by the explanations Avicenna offered in his chapter on the now. In this regard, it is a remarkable detail that the analogy about the flow of the now exclusively relies on local motion, concerning a corporeal object moving from here to there. Thus, it excludes other kinds of motions (such as alteration and growth), even though these likewise, as Avicenna would emphasize, take time. Furthermore, the analogy also

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  139 excludes moving objects that do not have a place (such as the outermost sphere, as Avicenna famously argues, even though it is nothing other than the outermost sphere which produces time through its motion). While this severely impairs the explanatory strength of the flowing now for the reality of time (which exceeds the realm of rectilinear local motion), it does not eliminate the didactic power such an image has on any mental visualization of what time is like. In addition, Avicenna’s discussion of the flow of the now always seems to focus on particular instances of local motion of particular objects. He is not talking about the one, universal time that is generated by the eternal motion of the outermost sphere. It is clear that without the existence of time as the magnitude of this sphere’s motion, one could not consider the temporal extent of particular motions, because neither time as such nor any particular temporal duration would be existent in the first place. This detail brings us full circle, back to Avicenna’s assertion that a thing-­ in-­motion can have a ‘now’ and a ‘where’ only insofar as time and distance are already there. Far from causing the existence of time and dimension through its local motion, the particular thing-­in-­motion presupposes time and distance, and thus has a ‘where’ and a ‘now’ that we can imagine as flowing, thereby creating the image of an extended motion and an extended time alongside the wholly real distance. Taking it all together, Avicenna’s chapter on the “explanation of the subject ‘the now’ ” is designed specifically for all those who still have trouble understanding this abstract notion that is time, in particular in light of time’s rather “tenuous existence,” which both Aristotle and Avicenna explicitly emphasize.138 In this chapter, Avicenna uses the theories of his predecessors about the flowing now, but he deconstructs them and reinterprets them to serve his own purposes. Overall, it seems that Avicenna was successful in explaining the now and in underscoring its didactic value—maybe too successful, because he accidentally misled modern interpreters to this day.139 Trier University 138  Phys. IV.10, 217b32f.; The Lecture on Nature II.13, §1, 166.14–16; The Notes (al-­Taʿlīqāt), 142.16 (ed. Badawī)/§763, 423.11 (ed. Mousavian). 139  I would like to express my gratitude to Peter Adamson, Fedor Benevich, and Jens Ole Schmitt as well as to the anonymous reviewers for helpful comments on earlier drafts of the paper.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

140  Andreas Lammer

Bibliography a. Primary Sources Alexander of Aphrodisias. Commentaire perdu à la Physique d’Aristote (Livres IV–VIII): Les scholies byzantines, ed. M.  Rashed (Berlin: De Gruyter, 2011). Alexander of Aphrodisias. De motu et tempore, ed. G.  Théry, in G.  Théry (ed.), Autour du décret de 1210, Volume 2: Alexandre d’Aphrodisias: Aperçu sur l’influence de sa noétique (Kain: Le Saulchoir, 1926), 92–7. Alexander of Aphrodisias. On Time, tr. R.  W.  Sharples, Phronesis 27 (1982), 58–81. Aristotle. al-Ṭabīʿa: Tarjamat Isḥāq ibn Ḥ unayn maʿa shurūḥ Ibn al-Samḥ wa-Ibn ʿAdī wa-Mattā ibn Yūnus wa-Abī Faraj ibn al-Ṭayyib, ed. ʿA.  al-R.  Badawī (Cairo: al-Dār al-Qawmiyya li-l-Ṭibāʿa wa-lNashr, 1964). Aristotle. Categoriae et Liber de interpretatione, ed. Lorenzo Minio-Paluello (Oxford: Clarendon Press, 1949). Aristotle. De anima, ed. W. D. Ross (Oxford: Clarendon Press, 1961). Aristotle. Metaphysics, ed. W. D. Ross (Oxford: Clarendon Press, 1924). Aristotle. Physics, ed. W. D. Ross (Oxford: Clarendon Press, 1936). Aristotle. Physics: Books III and IV, tr. E.  Hussey (Oxford: Clarendon Press, 1983). Aristotle. Physics, tr. R.  P.  Hardie and R.  K.  Gaye, in J.  Barnes (ed.), The Complete Works of Aristotle: The Revised Oxford Translation, vol. 1 (Princeton, NJ: Princeton University Press, 1995), 315–446. Aristotle. Physics, tr. R. Waterfield (Oxford: Oxford University Press, 1996). Aristotle. Physics, tr. P. H. Wicksteed and F. M. Cornford (Cambridge, MA: Harvard University Press, 1957). Avicenna. al-Mabdaʾ wa-l-maʿād, ed. ʿA.  A.  Nūrānī (Tehran: Institute of Islamic Studies, McGill University, Tehran Branch, 1984).

Avicenna. al-Samāʿ al-t ̣abīʿī min Kitāb al-Shifāʾ, ed. J. Āl Yāsīn (Beirut: Dār al-Manāhil li-l-Ṭibāʿa wa-l-Nashr wa-l-Tawzīʿ, 1996).

Avicenna. al-Shifāʾ, al-Mant ̣iq, al-Mujallad 2: al-Maqūlāt, ed. J.  Sh. Qanawātī, M.  M.  al-Ḥ uḍayrī, A.  F.  al-Ahwānī, and S.  Zāyid (Cairo: al-Hayʾa al-ʿĀmma li-Shuʾūn al-Mat ̣ābiʿ al-Amīriyya, 1959).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  141 Avicenna. al-Shifāʾ, al-Ilāhiyyāt, ed. J. Sh. Qanawātī, S. Zāyid, M. Y. Mūsā, and S. Dunyā (Cairo: al-Hayʾa al-ʿĀmma li-Shuʾūn al-Mat ̣ābiʿ al-Amīriyya, 1960). Avicenna. al-Shifāʾ, al-Ṭabīʿiyyāt, al-Mujallad 2–4: al-Samāʾ wa-l-ʿālam, alKawn wa-l-fasād, al-Afʿāl wa-l-infiʿālāt, ed. M.  Qāsim (Cairo: Dār al-Kitāb al-ʿArabī li-l-Ṭibāʿa wa-l-Nashr, 1969).

Avicenna. al-Shifāʾ, al-Ṭabīʿiyyāt, al-Mujallad 1: al-Samāʿ al-t ̣abīʿī, ed. S. Zāyid (Cairo: al-Hayʾa al-Miṣriyya al-ʿĀmma li-l-Kitāb, 1983).

Avicenna. al-Taʿlīqāt, ed. ʿA.  al-R.  Badawī (Cairo: al-Hayʾa al-Miṣriyya al-ʿĀmma li-l-Kitāb, 1973). Avicenna. al-Taʿlīqāt, ed. S.  H.  Mousavian (Tehran: Moʾassese-ye Pazhūheshī-ye Ḥ ekmat-o Felsefe-ye Īrān, 2013). Avicenna. Dāneshnāme-ye ʿAlāʾī: Elāhiyyāt, ed. M.  Moʿīn (Tehran: Enteshārāt-e Anjoman-e Āṯār-e Mellī, 1952). Avicenna. Dāneshnāme-ye ʿAlāʾī: Ṭabīʿiyyāt, ed. M.  Meshkāt (Tehran: Enteshārāt-e Anjoman-e Āṯār-e Mellī, 1952). Avicenna. al-Ḥ ikma al-mashriqiyya: al-Ṭabīʿiyyāt, ed. A.  Özcan, in A. Özcan, İbn Sīna’nın el-Hikmetuʾl-Meşrikiyye adlı eseri ve tabiat felsefesi (Istanbul: Marmara Üniversitesi, 1993). Avicenna. Kitāb al-Hidāya, ed. M.  ʿAbduh (Cairo: Maktabat al-Qāhira al-Ḥ adīṯa, 1974). Avicenna. Kitāb al-Ishārāt wa-l-tanbīhāt: Le livre des théorèmes et des avertissements publié d’après les mss. de Berlin, de Leyde et d’Oxford, ed. J. Forget (Leiden: Brill, 1892). Avicenna. Kitāb al-Majmūʿ aw al-Ḥ ikma al-ʿArūḍiyya, ed. M. Sāliḥ (Beirut: Dār al-Hādī li-l-Ṭibāʿa wa-l-Nashr wa-l-Tawzīʿ, 2007). Avicenna. Kitāb ʿUyūn al-ḥikma, ed. M.  F.  al-Jabr (Damascus: Dār al-Yanābīʿ, 1996). Avicenna. Kitāb ʿUyūn al-ḥikma wa-yushtamilu ʿalā ṯalāṯat funūn al-mant ̣iqiyyāt wa-l-t ̣abīʿiyyāt wa-l-ilāhiyyāt, vol. 1, ed. H.  Z.  Ülken, in H.  Z.  Ülken (ed.), Ibn Sina Risâleleri (Ankara: Türk Tarih Kurumu Basımevi, 1953), 1–55. Avicenna. Liber primus naturalium: Tractatus secundus de motu et de consimilibus, ed. S.  van Riet, J.  Janssens, and A.  Allard (Brussels: Académie Royale de Belgique, 2006). Avicenna. al-Najāt min al-ġaraq fī baḥr al-ḍalālāt, ed. M. T. Dāneshpazhūh (Tehran: Enteshārāt-e Dāneshgāh-e Tehrān, 1985).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

142  Andreas Lammer Avicenna. Risāla fī l-kalām ʿalā l-nafs al-nāt ̣iqa, ed. A.  F.  al-Ahwānī, in A.  F.  al-Ahwānī (ed.), Aḥwāl al-nafs (Cairo: Dār Iḥyāʾ al-Kutub al-ʿArabiyya, 1952), 195–9. Avicenna. al-Ṭabiʿiyyāt min ʿUyūn al-ḥikma, ed. Ḥ . ʿĀṣī, in Ḥ . ʿĀṣī (ed.), Tisʿ rasāʾil fī l-ḥikma wa-l-t ̣abīʿiyyāt wa-fī ākhirihā qiṣṣat Salāmān wa-Absāl (Cairo: Dār al-ʿArab, 1989), 2–38. Avicenna. The Metaphysics of The Healing, ed. and tr. M.  E.  Marmura (Provo, UT: Brigham Young University Press, 2005). Avicenna. The Physics of The Healing, ed. and tr. J. McGinnis (Provo, UT: Brigham Young University Press, 2009). Avicenna. ʿUyūn al-ḥikma, ed. ʿA.  al-R.  Badawī (Cairo: al-Maʿhad al-ʿIlmī l-Faransī li-l-Āṯār al-Sharqiyya, 1954). al-Ġazālī, Abū Ḥ āmid. The Incoherence of the Philosophers, ed. and tr. M. E. Marmura (Provo, UT: Brigham Young University Press, 2000). al-Kindī, Abū Yaʿqūb ibn Isḥāq. Liber de quinque essentiis, ed. M.  ʿA.  al-H.  Abū Rīda, in M.  ʿA.  al-H.  Abū Rīda (ed.), Rasāʾil al-Kindī l-falsafiyya, vol. 2 (Cairo: Dār al-Fikr al-ʿArabī, 1950–3), 9–35. Philoponus. In Aristotelis Physicorum libros quinque posteriores commentaria, ed. G. Vitelli (Berlin: Georg Reimer, 1888). Philoponus. In Aristotelis Physicorum libros tres priores commentaria, ed. G. Vitelli (Berlin: Georg Reimer, 1887). Philoponus. On Aristotle “Physics” 4.10–14, tr. S. Broadie (London: Bristol Classical Press, 2011). Plato. Plato’s Cosmology: The Timaeus of Plato, tr. F. M. Cornford (London: Kegan Paul, Trench, Trübner, 1937). Plato. Platonis Opera, ed. J. Burnet (Oxford: Clarendon Press, 1900–7). Proclus. Στοιχείωσις φυσική: Institutio physica, ed. and tr. A.  Ritzenfeld (Leipzig: Teubner, 1912).

al-Rāzī, Fakhr al-Dīn. al-Mat ̣ālib al-ʿāliya min al-ʿilm al-ilāhī, ed. A.  Ḥ . al-Saqqā (Beirut: Dār al-Kitāb al-ʿArabī, 1987). al-Rāzī, Fakhr al-Dīn. Sharḥ al-Ishārāt wa-l-tanbīhāt, ed. ʿA. R. Najafzādeh (Tehran: Anjoman-e Āṯār-o Mafākher-e Farhangī, 2005). al-Rāzī, Fakhr al-Dīn. Sharḥ ʿUyūn al-ḥikma, ed. A.  Ḥ . al-Saqqā (Cairo: Maktabat al-Anjilū l-Miṣriyya, 1986). Simplicius. In Aristotelis Physicorum libros quattuor posteriores commentaria, ed. H. Diels (Berlin: Georg Reimer, 1895).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  143 Simplicius. In Aristotelis Physicorum libros quattuor priores commentaria, ed. H. Diels (Berlin: Georg Reimer, 1882). Simplicius. On Aristotle “Physics” 4.1–5, 10–14, tr. J. O. Urmson (London: Duckworth, 1992). Themistius. In Aristotelis Physica paraphrasis, ed. H. Schenkl (Berlin: Georg Reimer, 1900). Themistius. On Aristotle’s “Physics” 4, tr. R.  B.  Todd (Ithaca, NY: Cornell University Press, 2005). al-Ṭūsī, Naṣīr al-Dīn. Sharḥ al-Ishārāt wa-l-tanbīhāt, ed. Ā. Ḥ. al-Āmolī (Qom: Būstān-e Ketāb, 2012).

b. Secondary Sources ʿAbd al-Mutaʿāl, ʿAlāʾ al-Dīn Muḥammad. Taṣawwur Ibn Sīnā li-l-zamān wa-uṣūluhū l-yūnāniyya (Alexandria: Dār al-Wafāʾ li-Dunyā l-Ṭibāʿa wal-Nashr, 2003).

Adamson, Peter. “The Existence of Time in Faḫr al-Dīn al-Rāzī’s al-Mat ̣ālib al-ʿāliya,” in D. N. Hasse and A. Bertolacci (eds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Physics and Cosmology (Berlin: De Gruyter, 2018), 65–99.

Adamson, Peter, and Andreas Lammer. “Fakhr al-Dīn al-Rāzī’s Platonist Account of the Essence of Time,” in A.  Shihadeh and J.  Thiele (eds.), Philosophical Theology in Medieval Islam: The Later Ashʿarite Tradition (Leiden: Brill, 2020). Ahmed, Asad  Q. “The Reception of Avicenna’s Theory of Motion in the Twelfth Century,” Arabic Sciences and Philosophy 26 (2016), 215–43. Algra, Keimpe. Concepts of Space in Greek Thought (Leiden: Brill, 1995). Alpina, Tommaso. Subject, Definition, Activity: Framing Avicenna’s Science of the Soul (Berlin: De Gruyter, 2021). Azadpur, Mohammad. Analytic Philosophy and Avicenna: Knowing the Unknown (London: Routledge, 2020). Benevich, Fedor. Essentialität und Notwendigkeit: Avicenna und die aristotelische Tradition (Leiden: Brill, 2018). Bertolacci, Amos. The Reception of Aristotle’s Metaphysics in Avicenna’s Kitāb al-Šifāʾ: A Milestone of Western Metaphysical Thought (Leiden: Brill, 2006).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

144  Andreas Lammer Broadie, Sarah. “Aristotle’s Now,” The Philosophical Quarterly 34 (1984), 104–28. Broadie, Sarah. “A Contemporary Look at Aristotle’s Changing Now,” in R. Salles (ed.), Metaphysics, Soul, and Ethics in Ancient Thought: Themes from the Work of Richard Sorabji (Oxford: Oxford University Press, 2005), 81–93. Coope, Ursula. Time for Aristotle: Physics IV.10–14 (New York: Oxford University Press, 2005). De Haan, Daniel. Necessary Existence and the Doctrine of Being in Avicenna’s Metaphysics of the Healing (Leiden: Brill, 2020). Di Vincenzo, Silvia. Avicenna, The Healing, Logic: Isagoge: A New Edition, English Translation and Commentary of the Kitāb al-Madḫal of Avicenna’s Kitāb al-Šifāʾ (Berlin: De Gruyter, 2021). Giannakis, Elias. “Fragments from Alexander’s Lost Commentary on Aristotle’s Physics,” Zeitschrift für Geschichte der arabisch-islamischen Wissenschaften 10 (1996), 157–87. Giannakis, Elias. Philoponus in the Arabic Tradition of Aristotle’s Physics (DPhil dissertation, University of Oxford, 1992). Gutas, Dimitri. Avicenna and the Aristotelian Tradition: Introduction to Reading Avicenna’s Philosophical Works (Leiden: Brill, 2014). Gutas, Dimitri. “Avicenna’s Eastern (‘Oriental’) Philosophy: Nature, Contents, Transmission,” Arabic Sciences and Philosophy 10 (2000), 159–80. Hasnawi, Ahmad. “La définition du mouvement dans la Physique du Shifāʾ d’Avicenne,” Arabic Sciences and Philosophy 11 (2001), 219–55. Hassan, Laura. Ashʿarism Encounters Avicennism: Sayf al-Dīn al-Āmidī on Creation (Piscataway, NJ: Gorgias Press, 2020). Hasse, Dag Nikolaus. Avicenna’s De anima in the Latin West: The Formation of a Peripatetic Philosophy of the Soul, 1160–1300 (London: Warburg Institute, 2000). al-ʿIrāqī, Muḥammad ʿĀt ̣if. al-Falsafa al-t ̣abīʿiyya ʿinda Ibn Sīnā (Cairo: Dār al-Maʿārif, 1983).

Janos, Damien. Avicenna on the Ontology of Pure Quiddity (Berlin: De Gruyter, 2020). Jeck, Udo Reinhold. Aristoteles contra Augustinum: Zur Frage nach dem Verhältnis von Zeit und Seele bei den antiken Aristoteleskommentatoren,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Now Is Not the Time  145 im arabischen Aristotelismus und im 13. Jahrhundert (Amsterdam: B. R. Grüner, 1994). Jolivet, Jean. “al-Kindī, vues sur le temps,” Arabic Sciences and Philosophy 3 (1993), 55–75. Kalbarczyk, Alexander. Predication and Ontology: Studies and Texts on Avicennian and Post-Avicennian Readings of Aristotle’s Categories (Berlin: De Gruyter, 2018). Kalbarczyk, Nora. Sprachphilosophie in der islamischen Rechtstheorie: Zur avicennischen Klassifikation der Bezeichnung bei Faḫr ad-dīn ar-Rāzī (gest. 1210) (Leiden: Brill, 2018). Kretzmann, Norman. “Aristotle on the Instant of Change (II),” Proceedings of the Aristotelian Society: Supplementary Volumes 50 (1976), 91–114. Kuhlmann, Hartmut. “‘Jetzt’? Zur Konzeption des νῦν in der Zeitabhandlung des Aristoteles (Physik IV 10–14),” in E. Rudolph (ed.), Zeit, Bewegung, Handlung: Studien zur Zeitabhandlung des Aristoteles (Stuttgart: KlettCotta, 1988), 63–96. Lammer, Andreas. “Anzahl und Ausmaß: Die griechisch-arabisch-lateinische Rezeption der aristotelischen Zeitdefinition,” Das Mittelalter: Perspektiven mediävistischer Forschung 23 (2018), 109–27. Lammer, Andreas. The Elements of Avicenna’s Physics: Greek Sources and Arabic Innovations (Berlin: De Gruyter, 2018). Lammer, Andreas. “Time and Mind-Dependence in Sayf al-Dīn al-Āmidī’s Abkār al-afkār,” in D.  N.  Hasse and A.  Bertolacci (eds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Physics and Cosmology (Berlin: De Gruyter, 2018), 101–61. Lettinck, Paul. Aristotle’s Physics and its Reception in the Arabic World. (Leiden: Brill, 1994). McGinnis, Jon. “Ibn Sīnā on the Now,” American Catholic Philosophical Quarterly 73 (1999), 73–106. McGinnis, Jon. “Making Time Aristotle’s Way,” Apeiron 36 (2003), 143–69. McGinnis, Jon. “Positioning Heaven: The Infidelity of a Faithful Aristotelian,” Phronesis 51 (2006), 140–61. McGinnis, Jon. “The Topology of Time: An Analysis of Medieval Islamic Accounts of Discrete and Continuous Time,” The Modern Schoolman 81 (2003), 5–25.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

146  Andreas Lammer McGinnis, Jon. Time and Time Again: A Study of Aristotle and Ibn Sīnā’s Temporal Theories (Ph.D. dissertation, University of Pennsylvania, 1999). McGinnis, Jon. “Time to Change: Time, Motion and Possibility in Ibn Sīnā,” in N. Bayhan, M. Mazak, and A. Süleyman (eds.), Uluslararası İbn Sīnā Sempozyumu Bildirileri, vol. 1 (Istanbul: İstanbul Büyükşehir Belediyesi Kültür A.Ş. Yayınları, 2009), 251–7. Mayer, Toby. “Avicenna against Time Beginning: The Debate between the Commentators on the Ishārāt,” in P.  Adamson (ed.), Classical Arabic Philosophy: Sources and Reception (London: Warburg Institute, 2007), 125–49. Miller, Jr., Fred D. “Aristotle on the Reality of Time,” Archiv für Geschichte der Philosophie 56 (1974), 132–55. Noble, Michael. Philosophising the Occult: Avicennan Psychology and “The Hidden Secret” of Fakhr al-Dīn al-Rāzī (Berlin: De Gruyter, 2021). Nusseibeh, Sari. Avicenna’s al-Shifāʾ: Oriental Philosophy (New York: Routledge, 2018). al-Rahim, Ahmed H. The Creation of Philosophical Tradition: Biography and the Reception of Avicenna’s Philosophy from the Eleventh to the Fourteenth Century A.D. (Wiesbaden: Harrassowitz, 2018). Roark, Tony. Aristotle on Time: A Study of the Physics (Cambridge: Cambridge University Press, 2011). Shayegan, Yegane. Avicenna on Time (Cambridge, MA: Harvard University Press, 1986). Sorabji, Richard. Time, Creation and the Continuum: Theories in Antiquity and the Early Middle Ages (Chicago: University of Chicago Press, 1983). Riccardo Strobino, Avicenna’s Theory of Science: Logic, Metaphysics, Epistemology (Berkeley: University of California Press, 2021). Wieland, Wolfgang. Die aristotelische Physik: Untersuchungen über die Grundlegung der Naturwissenschaft und die sprachlichen Bedingungen der Prinzipienforschung bei Aristoteles (Göttingen: Vandenhoeck & Ruprecht, 1992). Zimmermann, Fritz, and H.  Vivian  B.  Brown. “Neue arabische Übersetzungstexte aus dem Bereich der spätantiken griechischen Philosophie,” Der Islam 50 (1973), 313–24.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-­Century Aristotelian Logic The Study of Scientific Method Ana María Mora-­Márquez

Thirteenth-­century logic developed in close connection with the teaching delivered at the intellectual institutions of medieval Europe,1 of which the University of Paris was the studium generale—an inter­nation­al academic center—with the most influential faculty of arts.2 While ­terminist logic—the one that survives in the logical textbooks of authors such as Peter of Spain and William of Sherwood—was perhaps taught in some provincial studia (in southern France, Italy, and Spain),3 the official logical curriculum of Paris’s Faculty of Arts focused exclusively on Aristotelian logic. According to the official curriculum from 1255,4 the teaching of Aristotelian logic included Porphyry’s Isagoge, Aristotle’s Organon, 1 For a comprehensive study of the development of medieval logic from this socio-­ institutional perspective, see Julie Brumberg-­Chaumont, A l’école de la logique (Habilitation à diriger des recherches, France, 2019); see also Sten Ebbesen, “What Counted as Logic in the Thirteenth Century?,” in M.  Cameron and J.  Marenbon (eds.), Methods and Methodologies: Aristotelian Logic East and West, 500–1500 (Leiden: Brill, 2011), 93–108. 2 For the history and organization of the universities of Paris and Oxford, see Laurence W. B. Brockliss, The University of Oxford: A History (Oxford: Oxford University Press, 2016), part I; Jacques Verger, L’essor des universités au XIIIe siècle (Paris: Cerf, 1997). 3 For the place of terminist logic within medieval academic institutions, see. H. A. G. Braakhuis, “Logica Modernorum as a Discipline at the Faculty of Arts of Paris in the Thirteenth Century,” in O. Weijers and L. Holtz (eds.), L’enseignement des disciplines à la faculté des arts (Paris et Oxford, XIIIe–XVe siècles) (Turnhout: Brepols, 1997), 129–45; Brumberg-­ Chaumont, A l’école de la logique, chapter 3; Lambert Marie De Rijk, “Introduction,” in Peter of Spain, Tractatus Called Afterwards Summule Logicales (Assen: Van Gorcum, 1972), xcv–c. 4 See Heinrich Denifle (ed.), Chartularium Universitatis Parisiensis, Vol. 1 (Cambridge: Cambridge University Press, 2014), §246: “Statute of the faculty of arts about the mode of teaching and governing in the arts, and about what books must be read.” Ana María Mora-­Márquez, Thirteenth-­Century Aristotelian Logic: The Study of Scientific Method In: Oxford Studies in Medieval Philosophy Volume 9. Edited by: Robert Pasnau, Oxford University Press. © Ana María Mora-­Márquez 2021. DOI: 10.1093/oso/9780192844637.003.0004

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

148  Ana María Mora-Márquez Boethius’s De divisione and De topicis differentiis, and the anonymous Sex principia. This set of treatises (henceforth AL) is the text­ual basis of what I call thirteenth-­century Aristotelian Logic (henceforth AL-­13). The prominence of the logical tradition that developed on the basis of AL, AL-­13, is statistically attested to by the large amount of commentaries on AL produced during the thirteenth century5 in comparison to the much smaller amount of thirteenth-­century glossulae and commentaries on terminist textbooks. If we take the most influential thirteenth-­ century terminist textbook, Peter of Spain’s Tractatus, the glossulae on it are most likely not of Parisian origin, and the only thirteenth-­century commentary known to us that can be related to a Parisian context— Simon of Faversham’s—is the result of extra-­curricular work.6 This prominence notwithstanding, the attention of experts in medi­ eval logic has largely focused on the study of the terminist tradition, and hence turns mostly to the seminal thirteenth-­century treatises and their fourteenth-­century development, which is both eventful and fascinating.7 But fourteenth-­century logic also builds on AL-­13, so a more complete understanding of the latter will naturally result in an improved ­understanding of the former, and in general of the development of lo­gic­al theories in the late medieval period. To date, there is no systematic and exhaustive study of AL-­ 13, although its parts have been studied extensively.8 Thus, the question 5  If we take just Aristotle’s Topics, for instance, there are 34 extant commentaries from the 13th and 14th centuries, against a handful of glossulae and two commentaries on, for instance, Peter of Spain’s Tractatus. For a list of medieval commentaries on the Topics and on Boethius’s De topicis differentiis, see Niels-­Jørgen Green-­Pedersen, The Tradition of the Topics in the Middle Ages: The Commentaries on Aristotle’s and Boethius’ Topics (Munich: Philosophia Verlag, 1984), 381–432. 6  See Brumberg-­Chaumont, A l’école de la logique, ch. 3; de Rijk, “Introduction”, xcv–c. 7  For a recent systematic reconstruction, see Terence Parsons, Articulating Medieval Logic (Oxford: Oxford University Press, 2014). 8 For partial studies of 13th-­century Aristotelian logic, see, for instance, for the Topics: Green-­Pedersen, The Tradition of the Topics; for the Posterior Analytics, Amos Corbini, La teoria della scienza nel XIII secolo (Florence: SISMEL, 2006); for the Isagoge, Alain de Libera, La querelle des universaux: de Platon à la fin du Moyen Âge (Paris: Seuil, 1996), David Piché, Le problème des universaux à la faculté des arts de Paris entre 1230 et 1260 (Paris: Vrin, 2005); for its semantic aspects: Costantino Marmo, La semiotica del XIII secolo (Florence: Bompiani, 2010), Ana María Mora-­ Márquez, The Thirteenth-­ Century Notion of Signification: The Discussions and Their Origin and Development (Leiden: Brill, 2015), Giorgio Pini, Categories and Logic in Duns Scotus: An Interpretation of Aristotle’s Categories in the Late Thirteenth Century (Leiden: Brill, 2002); for seminal punctual studies, see the work of Sten Ebbesen,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  149 emerges whether the lack of such a systematic study reflects the lack of a well-­determined ‘system’ in the period. For the terminist tradition, the systematic approach is warranted at least by the fact that one can trace the fortune of self-­contained and well-­structured textbooks, for instance Peter of Spain’s Tractatus in the case of the tradition to which John Buridan belongs, or the Logica “cum sit nostra”9 in the case of the trad­ ition to which Walter Burley belongs. For AL-­13, on the contrary, there is no such textbook but rather a set of treatises at best controversially considered as a systematic unity even when we narrow it only to Aristotle’s Organon.10 The aim of this study is, first, to show that a systematic approach to AL-­13 is warranted by the thirteenth-­century masters’ own systematic approach to AL. Thus, I will show that AL-­13 is a logical tradition that considers AL as a more or less well-­structured system that is more or less coherently organized around a more or less unified subject matter, with a consistent tendency towards the more rather than the less. Now, how should we characterize AL-­13? Could we characterize it as understanding logic to be structurally equivalent to its understanding in the post-­Fregean, or even post-­Kantian, tradition, that is, as the formal study of reasoning? In his PhD thesis from 2000, John MacFarlane convincingly shows that the first historical attempt to demarcate logic as the formal study of reasoning is found in Kant (this in some relevant sense of formality).11 His claim certainly holds for medieval logic, which, whose complete bibliography (1966–2012) is collected in J. Fink, H. Hansen, and A. M. Mora-­ Márquez (eds.), Logic and Language in the Middle Ages: A Volume in Honour of Sten Ebbesen (Leiden: Brill, 2013), 455–67; notably, for the Sophistical Refutations: “The Way Fallacies Were Treated in Scholastic Logic,” CIMAGL 55 (1987), 107–34, and for the Prior Analytics: “The Prior Analytics in the Latin West: 12th–13th Centuries,” Vivarium 48 (2010), 96–133. 9  Logica “cum sit nostra,” text in Lambert Marie de Rijk, Logica modernorum: A Contribution to the History of Early Terministic Logic, Vol. 2, Part II (Assen: Van Gorcum, 1967), 413–51. 10  For the complex question of the relation between the treatises of the Organon in its text­ ual, historical, and doctrinal dimensions, see Michel Crubellier, “L’unité de l’Organon,” in J.  Brumberg-­Chaumont (ed.), Ad notitiam ignoti: l’Organon dans la “translatio studiorum” à l’époque d’Albert le Grand (Turnhout: Brepols, 2013), 37–62. 11  Cf. John MacFarlane, What Does it Mean to Say that Logic is Formal (PhD Dissertation, University of Pittsburgh, 2000), 1–30 and 95–113. The relevant senses of formality he considers are: (i) abstraction of the rules of reasoning as such (as in Kant); (ii) abstraction from objects, i.e., permutation invariance (as in Tarski); and (iii) abstraction from any content whatsoever (as in Hilbert). For another analysis of formality in relation to logic, where abstraction from subject matter is also considered, see Catarina Dutilh Novaes, Formal Languages in Logic: A Philosophical and Cognitive Analysis (Cambridge: Cambridge University Press, 2012); for the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

150  Ana María Mora-Márquez although including formal analyses of reasoning (in the sense of abstraction from subject matter, notably in commentaries on Prior Analytics12), cannot be generally characterized as the formal study of reasoning. For instance, the dialectic argument, which crucially relies on non-­formal warrants (such as the different predicational relations between genus and species), is a key part of medieval logic.13 The second aim of this study is to show that AL-­13 can be generally characterized as a tradition that takes logic to be the study of scientific method. To be sure, this study includes some sort of formal analysis as one of its parts (notably the analysis of validity in abstraction from subject matter in commentaries on Prior Analytics) but this part is by no means the crucial one. In fact, if, as I intend to show, the main aim of logic is to provide methods for knowledge production and dis­sem­in­ ation, its crucial parts must instead be the methods for scientific proof provided in commentaries on the Posterior Analytics and Topics.14 In the first part of the paper (1),15 I will argue for the possibility of talking of medieval ‘science’, ‘scientific knowledge’, and ‘scientific method’ from a modern perspective, and discuss how the modern perspective I take relates to the Latin ‘scientia’ in its different senses. In the second part (2–5) I will show the progression from Nicholas of Paris (1240s) and Albert the Great (1250s), who see AL as a systematic study of scientific method where syllogistic argumentation is fundamental, but who struggle to coherently organize AL around syllogistic argumentation, to Radulphus Brito (1290s) who, still seeing AL as a systematic study of scientific method, uses the notion of ‘second intention’ in order to coherently structure AL around syllogistic argumentation. The progression from Nicholas of Paris to Radulphus Brito will emerge as a medieval case specifically, see her Formalizing Medieval Logical Theories: Suppositio, Consequentiae and Obligationes (Dordrecht: Springer, 2007). 12  See Dutilh Novaes, Formal Languages. 13  See Eleonore Stump, Dialectic and Its Place in the Development of Medieval Logic (Ithaca, NY: Cornell University Press, 1989). 14  This is, by the way, consistent with Aristotle’s taking the Prior Analytics to be parasitic on the theory of demonstration in Posterior Analytics; see Prior Analytics (I.24a10–11): “Our first duty is to state the scope of our inquiry, and to what science it pertains: that it is concerned with demonstration, and pertains to a demonstrative science” (tr. Tredennick, 199). 15  The need for this part of the paper was pointed out to me by an anonymous reader. I am grateful for the prompt to fine-­tune the methodological commitments of this study.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  151 diachronic dialectical process where authors ‘discuss with’16 their predecessors in the production of their commentaries on AL, keep salient structural aspects, and challenge and develop problematic points.

1.  Preliminaries I take a theory of scientific method17 to include minimally: (a) a pre­sup­ pos­ition of the aim(s) of science; and (b) a determination of the tool(s) by means of which success with respect to the aim(s) can be achieved and of the rules that govern the application of said tools to the specific sciences. Consequently, a theory of scientific method is fundamentally normative inasmuch as its objective is not merely to describe the scientific enterprise, but to determine how to go about it so as to maximize success regarding its presupposed aim(s). The study of scientific method assures that the correct application of the method will bring about success with respect to the aim(s) of science and the possibility of demarcating science from non-­science, bad science, and pseudo-­science.

a.  Problem 1—‘Science’ across Historical Periods First, however, the question arises of whether talking about science, scientific knowledge, and scientific method in the medieval context makes any sense at all.18 Faced with the same difficulty, the historian David Lindberg said that “[h]istorians . . . require a very broad working def­in­ ition of ‘science’ – one that will permit investigation of the vast range of practices and beliefs about the operations of nature that preceded the

16  Most likely by using earlier commentaries in the production of their own courses and commentaries; for instance, Simon of Faversham and Radulphus Brito frequently quote Albert the Great’s paraphrases in their commentaries, and Albert himself heavily relies on the commentaries of Robert Kilwardby; see Sten Ebbesen, “Albert (the Great?)’s Companion to the Organon,” in S.  Ebbesen, Topics in Latin Philosophy from the 12th–14th Centuries: Collected Essays of Sten Ebbesen, vol. 2 (Aldershot: Ashgate, 2008), 95–108. 17  For theories of scientific method, see Robert Nola and H.  Sankey (eds.), After Popper, Kuhn, and Feyerabend: Recent Issues in Theories of Scientific Method (Dordrecht: Springer, 2000). 18  A question that was rightly raised by an anonymous reader of this paper.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

152  Ana María Mora-Márquez modern scientific enterprise,”19 but he was not explicit about any working definition that could justify talking about scientific method in the medieval context. I think, however, that a fairly recent theory of science, Helen Longino’s, provides us with exactly the inclusive definition that is needed. In Science as Social Knowledge,20 Longino puts forward her contextual empiricism, an account of science that tries to avoid the most serious pitfalls of positivism, holism, and scientific realism. Positivist accounts of science fall short in their claim that hypotheses can and should be confirmed exclusively by means of syntactic relations between the observational sentences that are taken as evidence and the hypothesis sentence. The inadequacy of such accounts has been shown repeatedly in the past forty years. As Longino reminds us, positivism is problematic not only because of its idealized empiricism but also, and most im­port­ ant­ly, because it misconceives the logical relations between evidence and hypothesis.21 In turn, holism—roughly the view that any observational sentence is theory-­laden so as to make the positivist ideal in fact unattainable—collapses into relativism, and hence into the incommensurability of scientific theories that are arguably comparable, for instance classical and quantum physics.22 Finally, scientific realism fails to account for the epistemic processes that lead to scientific knowledge. Longino focuses mainly on realist accounts that appeal to the principle of inference to the best explanation—abduction—in order to defend the claim that scientific hypotheses are true and must be read literally. As Longino rightly points out, scientific realism accounts at best for the inference of a hypothesis’ plausibility.23 Moreover, none of these theories provides an appropriate framework for historical studies: positivism rules out the possibility of counting as science many relevant past

19  Mark Shank and David Lindberg, “Introduction,” in D.  C.  Lindberg and M.  H.  Shank (eds.), The Cambridge History of Science, Volume 2: Medieval Science (Cambridge: Cambridge University Press, 2013), 6. 20  Helen Longino, Science as Social Knowledge: Values and Objectivity in Scientific Inquiry (Princeton, NJ: Princeton University Press, 1990); with further elaboration in her The Fate of Knowledge (Princeton, NJ: Princeton University Press, 2002). 21 Longino, Science, 25; she focuses on Carl Hempel. 22 Longino, Science, 26; she focuses on Thomas Kuhn and Paul Feyerabend. 23 Longino, Science, 29; she focuses on Hilary Putnam and Richard Boyd.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  153 practices that we should count as such (for instance, medieval cosmology); holism makes problematic even the possibility of talking about commonalities between different historical scientific traditions; and scientific realism uses strategies (notably abduction or the principle of inference to the best explanation) that, to the best of my knowledge, are alien to, for instance, the medieval context. Longino’s strategy is to broaden the epistemic notions most crucial to a theory of scientific method and scientific knowledge (notably the notions of evidential reasoning, rationality, and objectivity) in a way that avoids both the collapse into relativism and a misrepresentation of the epistemic processes leading to the production of scientific know­ ledge. For us, the most interesting feature of her account is that it is broad enough to allow us to count as science scientific traditions as distant in time and method as, say, Aristotelian physics and modern physics. According to contextual empiricism, the primary aim of science is the truthful representation24 of the external world, that is, knowledge construed as accurate and evidence-­based belief of the external world.25 Contextual empiricism is, then, a theory about scientific knowledge and the methods by means of which it is produced. However, scientific knowledge is here understood not as an epistemic attitude of some individual but as a body of knowledge—“putative truths”26—that belongs to a community and is the result of a number of specifically organized and controlled social practices. Accordingly, scientific method aims to determine the socio-­epistemic processes that yield the best product possible. Among those processes evidential reasoning is key, so a central part of the theory is an account of it based on social notions of rationality and objectivity. Evidential reasoning is a scientific community’s defense or rejection of hypotheses on the basis of evidence, and consists in the presentation or rebuttal of evidential relations between evidence and

24  With truth understood as adequacy and not as strict correspondance; cf. Longino, The Fate of Knowledge, ch. 6. 25  This includes both the natural and the social worlds. Nancy Cartwright has also devoted a great deal of her work to debunk the idea that claims in the hard sciences are literally true. See for instance How the Laws of Physics Lie (Oxford: Clarendon Press, 1983). 26 Longino, Science, 68.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

154  Ana María Mora-Márquez hypothesis. Longino argues that these relations are never sufficiently grounded in observational statements and their syntactic relations to hypothesis statements alone, but are always established on the basis of background assumptions that fill the logical gap between evidence and hypothesis. The social notion of rationality that underpins this account of evidential reasoning describes rationality as “the acceptance or rejection of beliefs on the basis of evidence”27 not by some individual (individual rationality is a necessary condition but not a sufficient one) but by a community. The notion of objectivity that prevents the theory from collapsing into relativism is also social: science relies on a minimal realism that assumes that the objects of inquiry have mind-­independent existence, and uses public language and public reasoning processes. These two features combined make scientific criticism possible. Scientific criticism in turn “permits objectivity in spite of the context dependence of evidential reasoning” and provides ways to “block the influence of subjective preferences at the level of background beliefs.”28 One of the many strengths of contextual empiricism is that it is broad enough so as to be inclusive of different historical scientific practices, such as the ancient and medieval ones. What drives scientific change are changes in social interests and needs rather than a conception of scientific progress that renders the practices of the past obsolete. So, in Longino’s own words “[w]hat Aristotle said of Protagoras still holds”29 insofar as their claims were rational and based on evidence, even though the interests and needs of the scientific communities have changed. As long as a community, past or present, has rational and objective ways of producing evidence-­based knowledge about some aspect of the external world, with rationality and objectivity taken as explained above, we are warranted in counting that body of knowledge as science, and such ways as scientific method. By these standards, then, it is unproblematic to count as science the academic practices that took place in the context of medieval universities, the disciplines that were critically discussed there, and the resulting bodies of knowledge.

27 Longino, Science, 59.

28 Longino, Science, 73.

29 Longino, Science, 222.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  155 It is from this broad approach that I use ‘science’ and its cognate expressions in this study. My claim is that AL was precisely the corpus that provided medieval scholars with a scientific method, that is to say with the different ways in which it was possible for them to produce knowledge that is rational, objective, and based on evidence. The widely accepted view that medieval accounts of scientific knowledge are found exclusively in commentaries on the Posterior Analytics has the her­men­ eut­ic­al disadvantage of leaving unexplained the role that medieval interpreters themselves explicitly gave to other parts of Aristotle’s Organon as tools for knowledge production and dissemination in scientific settings. My contention is that Longino’s theory of science provides a framework that allows us to account better for what medieval scholars themselves understood as logic, scientific knowledge, scientific method (via or modus sciendi), and the relations between them.

b.  Problem 2—The Polysemy of ‘Scientia’ A second problem for my purposes is linguistic and relates to the polysemy of the Latin ‘scientia.’ First, it is the Latin translation of the Greek ‘epistêmê’ in its specific sense of the individual knowledge of a conclusion obtained demonstratively. Second, ‘scientia’ is often synonymous with looser terms that refer to non-­demonstrative knowledge in scientific settings, for instance, with some Latin cognates of cognoscere (in turn related to the Greek gnosis) such as notum, notitia, and cognitio—as we shall see, this use is recurrent in Nicholas of Paris and Albert the Great. Third, ‘scientia,’ like the Greek ‘epistêmê,’ also often refers to a body of knowledge the production of which happens in the social framework of academic institutions.30 The main focus of this study is on the second and third uses of ‘scientia.’ For the sake of clarity in what fol-

30 See for instance, Plato, Sophist, 257c–d; Aristotle, De caelo I.1.268a1–6; De anima II.5.417b26; Prior Analytics I.24a21 and 46a23; Metaphysics A.1–2. Medieval scholars inherited their multiple senses of ‘scientia’ not only from the Latin translations of Aristotle’s works but also from other late ancient sources such as Isidore of Seville’s Etymologies, who says: “A dis­cip­ line (disciplina) takes its name from ‘learning’ (discere), whence it can also be called science (scientia)” in Etymologies, tr. S. Barney et al. (Cambridge: Cambridge University Press, 2006), 9.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

156  Ana María Mora-Márquez lows whenever ‘scientia’ is used in the first sense I will leave it untranslated; for its second sense I will use ‘scientific knowledge’ or simply ‘knowledge’; and for the third sense I will use the modern ‘science.’

c.  Thirteenth-­Century Aristotelian Logic (AL-­13) and Medieval Science The different versions of AL-­13 I will present commonly presuppose that the aim of science is the production of knowledge, either with a view to action (in the practical sciences) or for its own sake (in the the­ or­et­ic­al sciences). This knowledge includes both the scientia that is achieved through demonstration and the scientific knowledge that must be achieved by other means, of which metaphysical knowledge is a paradigmatic case. Medieval science includes the medieval bodies of scientific knowledge and the practices conducive to their production, dissemination, and development (so logic both provides tools for the sciences and is itself a science, that is, it is utens and docens). The practices in question include notably the interpersonal processes of teaching, learning, and disputing, which consist mostly in providing arguments in favor of a thesis (“S est P” problematized as “utrum S est P”) and rebutting the arguments that support its contradictory claim (“S  non est P”). Accounts in AL-­13, then, aim to provide the tools to achieve success with respect to knowledge production and dissemination through interpersonal argumentation in an academic setting. Furthermore, these accounts are pluralist in that they propose different methods that contribute to the aim in different ways, although, as we shall see, the details of their co-­existence within the same account vary from author to author. For this study, which is more focused on providing a general characterization of AL-­13 as a logical tradition than on a detailed account, the prologues in thirteenth-­century commentaries on Porphyry’s Isagoge are the sources of choice. As a rule of thumb, these prologues address the questions: (i) what is logic; (ii) what is it useful for; and (iii) what is its subject matter. Despite specific points of disagreement in the replies to these questions, I will show in what follows that medieval masters

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  157 commonly construe AL as the theoretical study of scientific method.31 According to their construal, Aristotle’s Categories and De interpretatione provide analyses of the constituent parts of arguments, that is, predicates (observational predicates such as ‘white,’ non-­observational predicates such as ‘form,’ and theoretical predicates such as ‘species’) and statements. De interpretatione, in particular, provides a method for the production of statements that are susceptible of truth and falsity, and hence of contradiction. Prior Analytics provides a method for valid inference. Posterior Analytics and Topics provide different methods for scientific proof, demonstrative and dialectical respectively. Finally, Sophistical Refutations provides a method for refutation and hence is the foremost tool for scientific checks and balances in the medieval setting. The accounts differ in various ways, for instance, in the kind of science they take logic to be and in the role that they give to the auxiliary treatises (Porphyry’s Isagoge, Boethius’s De divisione and De topicis differentiis, and the Sex principia). But most importantly, they have different degrees of theoretical sophistication, with Radulphus Brito’s logic of second intentions being the most accomplished.

2.  The 1230s, Nicholas of Paris One of the earliest commentaries on the Isagoge32 from the university period is that by Nicholas of Paris.33 For Nicholas, logical theory is entirely covered in Aristotle’s Organon, the other texts in the logical 31  There are very few general histories of theories of scientific method (cf. Laurens Laudan, “Theories of Scientific Method from Plato to Mach: A Bibliographical Review,” History of Science 7 (1968), 1–63; the situation does not seem to be much better today than in 1968). The few historical studies almost entirely neglect the Middle Ages, covering them in a few pages in the best-­case scenario. A notable exception is Marshall Clagett (ed.), Critical Problems in the History of Science (Madison, WI: University of Wisconsin Press, 1969), which contains a paper on the Middle Ages by A. C. Crombie. The focus on the 13th century, in particular, is limited to the few figures who took an interest in experimental science such as Robert Grosseteste and Roger Bacon. 32  Edition of the prologue (=InPhil.) in Claude Lafleur and J. Carrier, “L’Introduction à la philosophie de maître Nicolas de Paris,” in C. Lafleur and J. Carrier (eds.), L’enseignement de la philosophie au XIIIe siècle (Turnhout: Brepols, 1997), 447–65. 33  For the life and work of Nicholas of Paris, the Latin and Neo-­Platonic sources of his logic, and his influence on later Parisian masters, see Lafleur and Carrier, “L’Introduction à la philosophie de maître Nicolas de Paris,” 447–53.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

158  Ana María Mora-Márquez curriculum aiming at its better understanding (its bene esse) and not containing any additional doctrine.34 In his prologue to the Isagoge, Nicholas places logic, alongside grammar and rhetoric, among the sciences of reason (scientiae rationales), which can also be called sciences of speech (scientiae sermonicales) insofar as they study speech as a product of reason grounded in natural things as signified by speech.35 For Nicholas, logic is a science of speech in that it provides a method (via) for other sciences—moral and natural—with respect to instruction: Logic . . ., since it is a way . . . to natural and moral [science] (and this with respect to instruction), is not a science regarding which there is a state of the theoretical intellect,36 nor is what is immediately taught in logic the cognition of things, but rather the cognition of the ways through which the things and their structures can be known.37

Hence, logic is not a theoretical science (scientia speculativa) but rather a science of reason and speech (insofar as its subject matter is a product of reason), which provides methods (viae) for the other sciences, the­or­et­ic­al and practical, specifically with respect to instruction and learning. Doctrina and disciplina are two sides of the same coin: knowledge as imparted by the master, knowledge as received in the student.38 This interrelation suggests that Nicholas takes instruction to be crucial to the scientific enterprise, for, as he says, one can in principle acquire know­ledge on one’s own but cannot transmit it by means of instruction without speech: Speech is the most efficient instrument for the production of instruction (doctrina); for we very efficiently form diverse speech by our will . . . And Aristotle agrees with this, saying that “animals that do not 34 Cf. InPhil §46. 35 Cf. InPhil §29. 36  This means that the intellect does not acquire the status of ‘theoretical’ while studying logic, which is surely related to Averroes’s four states of the intellect in his Long Commentary on De anima. In the Arabic tradition a quadripartite division of intellectual powers and states is already found in Al-­Kindī. For the different forms that this division took in Arabic philo­ sophers from Al-­Kindī to Averroes, see Deborah L. Black, “Psychology: Soul and Intellect,” in P. Adamson and R. Taylor (eds.), The Cambridge Companion to Arabic Philosophy (Cambridge: Cambridge University Press, 2005), 308–26. 37  InPhil §38. All translations are mine. 38  For the interrelated character of the words ‘doctrina’ and ‘disciplina,’ see Sten Ebbesen, “Anonymous D’Orvillensis’ Commentary on Aristotle’s Categories,” CIMAGL 70 (1999), 229–423, at 242.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  159 have hearing with memory cannot be taught,” suggesting with this that instruction and learning (disciplina) are made by means of an object of hearing which leaves something in the memory. Hence, rational phil­ oso­phy, which is instituted for the sake of instruction, is about speech, through which instruction is made.39

In other words, we cannot produce knowledge as a community (i.e., knowledge that is produced, transmitted, and preserved within a given social group) without speech, and hence the processes of interpersonal production of scientific knowledge are more fundamental in the study of scientific method than individual reasoning. Masters exclusively transmit propositional knowledge. Therefore, the speech used in instruction must replicate the internal structure of propositional knowledge so as to be truth-­preserving. In order to transmit the knowledge that S is P, it is not enough that the masters utter ‘S is P’ but they must also give evidence for the claim so that it can be accepted as true: Everyone who teaches teaches the truth, but the truth is twofold, that is, of what is compound and of what is non-­compound. The truth of what is non-­compound can be known (sciri)40 in two ways: either through what is prior or through what is posterior. If through what is prior, thus there is definition; if through what is posterior, there is div­ ision. If he teaches the truth of what is compound, similarly this is known either through what is prior, and thus there is syllogism, or through what is posterior, and thus there is induction. And these relate to both natural and linguistic [science] as instruments for teaching possessed by the logician. In fact, the natural [philosopher] qua nat­ ural does not know what are the syllogism and the induction of the philosopher of speech but they belong to the consideration of the logician.41

39  InPhil §27. 40  Note that here ‘sciri’ cannot refer exclusively to knowing demonstratively because it also points explicitly to knowing through definition, division, and induction, hence my more inclusive translation ‘to be known.’ It will be so translated henceforth unless it specifically refers to knowing demonstratively. 41 Cf. InPhil §49.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

160  Ana María Mora-Márquez Natural philosophers as natural philosophers do not learn the methods for instruction because nothing in their training as natural philosophers involves learning those methods. But being a master of Arts involves, other than training as a natural philosopher, strict training as a logician so that masters can apply their logical knowledge in their work as nat­ ural and moral philosophers. Thus, logic aims to provide medieval scholars with different tools that can prompt assent in their interlocutors (in public disputations and in the classroom) to scientific theses precisely by means of evidential reasoning based on the methods of definition, division, deduction, and induction.42 As Nicholas says, the method of definition leads to the knowledge of non-­compound things by using as evidence things that are prior in predicational lines, for instance the genus under which a species is subsumed. The method of division leads to the knowledge of non-­compound things by using as evidence things that are posterior in predicational lines, for instance the differentiae through which a genus is divided into species. The method of induction leads to the inference of universal statements by using particular instances as evidence. Finally, the syllogistic method establishes a conclusion by taking causes or indications as evidence or rebuts conclusions by pointing to shortcomings (fallaciae) in syllogistic reasoning. Accordingly, Nicholas claims that the subject matter of logic are these four methods: From what has been said, it is evident that logic, which is necessary for instruction, is about the four ways of knowing (de iiiior viis ad connoscendum43) things and their structures, that is definition, div­ision, syllogism, and induction . . .44

Which he makes correspond to the texts of the logical curriculum as follows:45 42  The cruciality of these four methods in the understanding of logic as a study of scientific method has historical precedents as early as Sextus Empiricus and his criticism of the dogmatists; see Justin Vlasits, “Pyrrhonism and the Dialectical Methods: The Aims and Arguments of Outlines of Pyrrhonism II,” History of Philosophy and Logical Analysis 23 (2020), 225–52. These ideas are transmitted to Nicholas indirectly via Boethius. 43  I take ‘cognoscere’ in this passage as synonymous with ‘scire’ above, hence the same translation as ‘to know.’ 44  InPhil §47. 45  InPhil §39. It is unclear to what Islamic sources Nicholas had access, but he suggests here the idea we also find in Avicenna that knowledge is not just the possession of a composition

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  161 non-compound object

through the prior

art of definition

through the posterior

art of division art of judgment

ways of knowing through the prior

Nicholas of Paris, Division of Logic 1

compound object

art of discovery through the posterior

preparatory texts

syllogistic way

art of argumentation inductive way

through the cause of inference, Prior Analytics through the cause of inference and being Posterior Analytics through probable signs, Topics through apparent signs, Sophistical Refutations

Categories Perihermeneias

As explained above, the four methods are the four ways to arrive at propositional knowledge, which can be accepted as true, of compound and non-­compound things. Non-­compound things are those to which simple words can be applied, for instance, actions such as running, to which the words ‘runs’ and ‘running’ can be applied. Compound things must be combinations of non-­compound things, for instance the com­ bin­ation of Socrates and the action of running, which is expressed with ‘Socrates runs.’ We can know non-­compound things, say, the human being, by applying the method of definition so as to obtain the ac­cept­ able proposition ‘a human being is a rational animal’ (where ‘rational’ and ‘animal’ are predicationally prior to ‘human being’, that is, have a wider extension); or, say, animal, by applying the method of division so as to obtain the acceptable proposition ‘animal is rational and non-­ rational’ (where ‘rational’ and ‘non-­rational’ are predicationally posterior to ‘animal’, that is, have a narrower extension). As for compound things, we can know them through the syllogistic or inductive methods, say, when the acceptable proposition ‘butterflies have a four-­stage life cycle’ is the conclusion of a syllogism (with a middle term prior to the terms of the conclusion) or the result of an inductive process (through particular, hence posterior, instances of the conclusion). However, without explaining why, Nicholas provides an alternative division of logic that disregards the Roman division between an art of discovery (dialectic) and an art of judgment (analytic) and is based only on the syllogism, the expression of “the product proper to reason”:46 and division but also, and fundamentally, the justified assent knowers give to it. Cf. Avicenna, Logica, ed. F.  Hudry (Paris: Vrin, 2018), 124. At any rate, Nicholas relies mainly on Roman sources via Boethius, here notably on the Ciceronian idea of an ars inveniendi as opposed to an ars iudicandi, and the identification of the latter with Aristotle’s topical theory; cf. Cicero, Topica, tr. T. Reinhardt (Oxford: Oxford University Press, 2003), §6. For the medieval reception of the Cicero-­Boethian theoretical framework, see Joël Biard, “Produire la confiance: la fides dans la tradition topique médiévale,” in C. Grellard, P. Hoffmann, and L. Lavaud (eds.), Génèses antiques et médiévales de la foi (Paris: Institut d’études augustiniennes, 2020), 295–313. 46  InPhil §42–§43.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

162  Ana María Mora-Márquez only inferring, Prior Analytics syllogism

complete knowledge through the cause, Posterior Analytics inferring and proving

Nicholas of Paris, Division of Logic 2

constituent parts

Categories

incomplete knowledge through signs, Topics imagination through apparent signs, Sophistical Refutations

Perihermeneias Isagoge

bene esse

Sex principia Boethius’ Topica and De divisione

This division puts forward the idea that the knowledge produced through the syllogistic method is of different kinds—certain and complete (that is, scientia) and probable and incomplete (that is, opinio)— and that the syllogism is the main, if not the proper, subject matter of logic. Accordingly, the Prior Analytics studies the inferential power of syllogisms in general. Here it is noteworthy that the Prior Analytics is not given any particular pride of place in the system; rather it is presented as dealing with one of the aspects with which logic must deal without making that aspect more or less crucial than the aspects that the other treatises of the Organon study. This approach to the formal aspect of AL-­13 will continue in Albert the Great and Radulphus Brito. The Posterior Analytics, in turn, provides the demonstrative method for the production of scientia, certain and complete knowledge through demonstration of the cause. The Topics provides the dialectical method for the production of opinio, incomplete knowledge through the indication of signs. The Sophistical Refutations deals with sophistical arguments through apparent signs, most likely in order to provide a method for defeating them. Finally, the Categories and Perihermeneias study predicates and statements, the constituent parts of syllogisms. As mentioned before, the remaining texts of the curriculum do not offer new logical doctrine but help to clarify some of the obscurities in Aristotle’s Organon. Both divisions reflect a systematic approach to AL. We also find in both divisions the construal of AL as a pluralist logic inasmuch as different methods outside and within the syllogistic method have an im­port­ ant place in the system. Nonetheless, Nicholas also puts forward demonstration as the foremost method in the scientific enterprise

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  163 without saying much about why we should accept less than scientia as its result (altogether or at any stage thereof): From what has been said, it is evident that logic, which is necessary for instruction, is about the four ways of knowing things and their structures, namely definition, division, syllogism, and induction – and maximally and principally the demonstrative syllogism, which generates a complete disposition in those capable of learning.47

Here the relation between the different methods and the way they must co-­operate in the scientific enterprise is far from clear but demonstration is presented as the foremost method, although the reason and purpose of its co-­existence with the remaining methods are left unexplained. To sum up, for Nicholas, AL is: (i) a psycho-­linguistic science, which (ii) provides methods for the specific sciences with regard to instruction, and hence is (iii) generally about the four methods for the evidence-­based production of propositional knowledge that can prompt assent in teaching and learning (definition, division, syllogism, and induction), but principally about the syllogistic method of which the foremost kind is the demonstrative. In other words, AL provides ­theoretically a plurality of scientific methods, although without ex­plan­ation demonstration is understood as prominent. Not that such an explanation was necessary in Nicholas’s context, given the venerable tradition—as ancient as Neo-­Platonic commentators such as Simplicius, and well known to medieval masters—of organizing AL with demonstration as the focal point.

3.  The 1250s, Albert the Great Authors as early as Albert the Great start to introduce modifications to accounts like Nicholas’s.48 Crucial to this stage is the arrival of Avicenna’s 47  InPhil §47. 48 Prominent in the period between Nicholas and Albert is Robert Kilwardby, whose account of logical inference was recently reconstructed and characterized as intensional in Paul Thom, Robert Kilwardby’s Science of Logic: A Thirteenth-­Century Intensional Logic (Leiden:

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

164  Ana María Mora-Márquez Logica (the first part of the Book of Healing) to Paris, most likely in the hands of Dominican friars coming from Spain.49 The Logica transmits three fundamental ideas to the Latin Middle Ages: first, that we know the external world through epistemic perspectives—mental accidents (universality, particularity, etc.) that ­ befall essences only as they exist in the intellect (in intellectu); second, that logic is the science that teaches the method for establishing the know­ledge of something unknown by taking something already known as evidence; and third, that logic, even though it necessarily relies on some sort of linguistic structure (locutio interior),50 is primarily about the mental accidents that befall essences as they are in intellectu.51 Logic, then, is not primarily a linguistic science but rather a science about the accidents of essences qua understood. Thus, with the added pre­sup­pos­ ition that the aim of science is knowledge, either in itself or with a view to action,52 Avicenna’s Logica provides the Latin Middle Ages with the broad outlines of a theory that studies the properties of things qua understood, with the aim of devising a general method for the production and dissemination of scientific knowledge.

Brill, 2019). Kilwardby stands closer to Nicholas than to Albert in holding that logic is a science of speech (scientia sermonicalis), principally concerned with the syllogism. Kilwardby also anticipates Radulphus Brito’s use of the notion of intentio in the analysis of some crucial logical notions although Kilwardby’s source is most likely Averroes. For Kilwardby’s general account of logic, see Thom, Robert Kilwardby’s Science of Logic, 8–21. 49  For the transmission of the Logica in 13th-­century Paris and the role that Albert the Great played therein, see Alain de Libera, “Entre Tolède et Hamadan: La Logica d’Avicenne,” in Avicenna, Logica, 9–22; F.  Hudry, “La tradition manuscrite de la Logica d’Avicenne,” in Avicenna, Logica, 94–7. For Avicenna on the subject matter of logic, see Abdelhamid I. Sabra, “Avicenna on the Subject-­Matter of Logic,” Journal of Philosophy 77 (1980), 746–64. 50 Cf. Logica, 136–7: “If it were possible to express logic only with the intellect, only intellectual things would be considered and this would be sufficient.” 51 Cf. Logica, 127: “When we would want to consider [things] so as to know them, it will be necessary to collect them in the intellect. And then dispositions that are proper only to the intellect will necessarily befall them, and principally when we will attempt to grasp things that are unknown from things that are known by thinking . . . The disposition, that is, that which befalls things . . . is a disposition and accident that befalls them in the intellect . . . Therefore, it is necessary to have a science (scientia) for knowing those dispositions: how many are they, of what kind, and in what way that accident is considered . . . And this kind of speculation is called logical science, which is a speculation of the things mentioned through which we come to the knowledge of what is unknown . . . .” 52 Cf. Logica, 127: “The end of theoretical philosophy is the acceptance (agnitio) of the truth, and the end of practical philosophy is the cognition of goodness (bonitatis).”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  165 Most masters from about the 1250s onwards adopt Avicenna’s description of logic as a study of the method by means of which we arrive at the knowledge of something unknown by taking what is already known as evidence.53 They also side with Avicenna in rejecting Nicholas’s classification of logic as a science of speech (scientia sermonicalis) because logic relies first and foremost on an analysis of epistemic processes. However, the idea that logic builds on epistemic perspectives and is ultimately about some kind of being (mental being, modes of being, etc.) will not be pushed to its logical conclusions before Radulphus Brito in the 1290s. In the 1250s, Albert the Great characterizes AL as a rational science that aims to improve the quality of our natural capacity to proceed from something already known to the knowledge of something unknown.54 In fact, against his predecessors (e.g., Nicholas and Kilwardby) Albert stresses Avicenna’s point that the focus of logic must be on the epistemic processes of knowledge production rather than on their linguistic expression: There are some who understand ‘logical’ as being the same as ‘linguistic,’ saying that general logic is the same as linguistic science. . . . Whence, they say that the subject matter of general logic is speech, insofar as it designates things that are signified by it. This is an opinion that Avicenna rejects in the beginning of his Logica, saying that speech “signifies nothing of itself.”55

The production of knowledge necessarily arrives at a mental com­pos­ ition or division as at a conclusion. So, Albert takes up the idea, present both in Nicholas and Avicenna, that the mere act of composing and dividing thoughts is not sufficient for bringing about knowledge – unjustified 53 For a collection of articles tackling different aspects of this idea, see Brumberg-­ Chaumont, Ad notitiam ignoti. 54 Cf. De v universalibus, tr. 1, c. 1: “And thus here the method of logic starts from nature, advances towards art, and receives perfection in use and practice.” De v universalibus is Albert’s paraphrase on Porphyry’s Isagoge. Accessed online at Alberti Magni opera. Editio digitalis, Institutum Alberti Magni Coloniense, https://www-­albertusmagnusonline-­de.ezproxy.ub.gu. se/pages/impressum. 55  De v universalibus, tr. 1, c. 4.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

166  Ana María Mora-Márquez composition and division is not knowledge. However, a complete ­scientific method must put forward not only the correct ways of concluding on the basis of evidence but also the correct ways of problematizing (that is, of asking relevant questions), of rejecting wrong conclusions, and of transmitting knowledge as knowledge: Indeed, logic teaches the principles of the how, that is the principles and argumentations whereby we must argue about a knowable (scibile): . . . Since scientific knowledge56 (scientia) is the possession of a conclusion, those who acquire scientific knowledge must know how to draw conclusions. But those who don’t know what they must draw conclusions about don’t know how to draw conclusions. And since nothing must be concluded but a question, they must know the what and the how of the act of asking about that about which they intend to draw a conclusion . . . Therefore, [logic] is not only useful and instrumental for all the sciences but also necessary. That’s why those who don’t know logic . . . don’t know how anything must be known and how one must prove and disprove. Moreover, they don’t know because of what they give assent to the things known, or how they must respond to those contradicting [them]. But logicians know this, if they have the science of arguing about anything. . . . However, those who don’t know how to expound a reason cannot teach.57

Albert’s account58 is peculiar in that it stresses the usefulness of logic in more than just providing methods of scientific proof. Logic must provide tools for other practices key to the production of scientific know­ 56  Albert is talking in general about the different kinds of argument that logic teaches, so I take ‘scientia’ here to refer to the possession of a conclusion that is based on evidence regardless of whether the conclusion was obtained through demonstration, dialectic proof, or induction, hence the translation as ‘scientific knowledge.’ This will be the case in the remaining uses of ‘scientia’ as an epistemic disposition of the intellect in the passages that follow. 57  De v universalibus, tr. 1, c. 3. 58  For other analyses of Albert’s general account of logic (and its internal coherence or lack thereof), see Julie Brumberg-­Chaumont, “Les divisions de la logique selon Albert le Grand,” in Brumberg-­Chaumont, Ad notitiam ignoti, 336–416 (this rich study is an exhaustive account of Albert’s general account of logic); Aurélien Robert, “Le débat sur le sujet de la logique et la réception d’Albert le Grand au Moyen Âge,” in Brumberg-­Chaumont, Ad notitiam ignoti, 467–512; Bruno Tremblay, “Albert le Grand et le problème du sujet de la science logique,” Documenti e studi sulla tradizione filosofica medievale 22 (2011), 301–45.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  167 ledge: the accurate formulation of scientific problems, the accurate interpersonal transmission of scientific knowledge, and the effective refutation of contrary positions. This brings to the fore the pragmatic character of his view on logic. From this pragmatic perspective, the scientific practice extends beyond proving theses to problematizing, refuting, and teaching, and hence logic must theorize about all these actions. Albert’s view of logic as a study of scientific method is, thus, strongly practice-­oriented. Accordingly, for Albert, logic is a full-­fledged theory of argumentation. The subject matter of logic is argumentation, the aim of which is to provide the mind of knowers with the evidence required to give assent to some conclusion by means of the relations of its terms to what they already know: Since logic is a science teaching theoretically how and by means of what one comes through what is known to the knowledge of what is unknown, it is necessary that logic be about such an instrument of reason by means of which the scientific knowledge (scientia) of what is unknown is acquired through what is known regarding everything that can be made known about what is unknown. But this is argumentation inasmuch as argumentation is a reasoning showing to and convincing the mind about the scientific knowledge of what is unknown by means of a relation of what is known to what is unknown. Among the species of argumentation, the main one is the syllogism; that is why some said that logic in its entirety is about the syllogism and the parts of the syllogism, not determining the general subject matter of logic but its main subject matter.59

In establishing a distinction between a general and a principal subject matter of logic (the latter referring to divisions of logic like Nicholas’s), Albert reveals the tension, common in thirteenth-­century accounts of logic,60 caused by the opposition between a standard and an extended 59  De v universalibus, tr. 1, c. 4. 60  For other 13th-­century authors (for instance, Giles of Rome and Simon of Faversham) who adopt the extended Organon, see Costantino Marmo, “Logique élargie et sémiotique: Albert le Grand, Roger Bacon et Gilles de Rome,” in Brumberg-­Chaumont, Ad notitiam ignoti,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

168  Ana María Mora-Márquez Organon. The latter had an impact on Latin scholasticism through the Latin translations of important Islamic treatises, notably al-­Fārābī’s De scientiis, a division of the sciences that considers Rhetoric and Poetics as parts of Aristotle’s Organon.61 In the Islamic tradition, the use of logic, and more specifically the use of the syllogistic method, is extended to social enterprises where persuasion through rhetorical or poetic devices is central.62 Albert accepts this idea that logic is also useful for enterprises that are not specifically theoretical, and puts forward a general logic (logica generalis), which follows the Islamic tradition in including Rhetoric and Poetics as parts of the Organon,63 and a main logic (logica principalis), which focuses exclusively on the scientific method that is based on the standard Organon. Albert suggests that the latter takes the syllogistic proof as the spine of scientific argumentation: [T]hose teaching how to acquire scientific knowledge (scientia) of the compound teach the syllogism, which is the instrument proper to that knowledge, and teach other species of argumentation and the prin­ ciples of the syllogism and the things that surround it, and its prin­ ciples, parts, and the matter in which can be put the form of the syllogism and the form of other argumentations that imitate the syllogism. And hence, the things with which logicians must deal are divided and multiplied according to them.64

His consequent division of logic,65 instead of logica generalis, fits the logica principalis centered around the syllogistic method:66 447–65, and “Suspicio: A Key Word to the Significance of Aristotle’s Rhetoric in Thirteenth-­ Century Scholasticism,” CIMAGL 60 (1990), 145–98. 61  For its Latin reception, see Jean-­Marc Mandosio, “La place de la logique et ses subdiv­ isions dans l’Énumération des Sciences d’al-­Fârâbî et chez Dominicus Gundissalinus,” in Brumberg-­Chaumont, Ad notitiam ignoti, 285–310. 62  For the Islamic case, see Deborah L. Black, Logic and Aristotle’s Rhetoric and Poetics in Medieval Arabic Philosophy (Leiden: Brill, 1990); Mandosio, “La place de la logique,” 298–306. 63 Cf. De v universalibus, tr. 1, c. 2. Albert places peirastic alongside dialectic and sophistic and as distinct from them; cf. Brumberg-­Chaumont, “Les divisions de la logique selon Albert le Grand,” 366. 64  De v universalibus, tr.1, c. 7. 65 Cf. De v universalibus, tr. 1, c. 7. 66  This is Albert’s division in De v universalibus, but other of his paraphrases put forward different divisions. This apparent incoherence in Albert must be attributed to the early stage of development of 13th-­century philosophy rather than to poor philosophical acumen, with commentators from the first half of the century being more exegetes than independent

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  169 non compound

through definition (no text) hypothetical

Albert the Great, Division of Logic

something Unknown

through syllogism

categorical

for discovery

through topical relations, Topics, Boethius’ Topica

for judgment

through analysis

compound

into formal principles, Prior Analytics into material principle, Posterior Analytics

deceptive constituent parts

Categories, Isagoge, De divisione Perihermeneias

It is typical of Albert’s philosophical exegeses to try to make use of all the authoritative sources available to him. This division of logic is no exception: like Nicholas, Albert uses the Roman division of logic into an art of discovery (ars inveniendi) and an art of judgment (ars iudicandi) as corresponding to the division between the dialectical and the demonstrative methods (note that, like Nicholas, Albert gives no particular pride of place to the Prior Analytics); he also adopts Avicenna’s idea that the part of AL dealing with knowledge of non-­compound things (where compound and non-­compound things take exactly the same sense as in Nicholas), the method of definition, was either not developed or did not survive;67 and he includes the hypothetical syllogism in the division, which is systematically ignored by most scholars before and after him.68 More fascinating, though, is the suggestion that the demonstrative method (in the Prior and Posterior Analytics) is subsequent to the method of discovery (in the Topics) in order to elevate to the level of certainty,69 through the presentation of causal evidence, the probable knowledge that is established by means of indications. So, the thinkers; for Albert, see Brumberg-­Chaumont, “Les divisions de la logique selon Albert le Grand,” part 4. The 13th-­century transformation of Aristotelian logic into a well-­structured system is a dialectical process that spans the whole century and won’t be satisfactorily achieved before the end of the century in the work of scholars such as Radulphus Brito. And yet the co-­ existence of multiple and not always coherent Aristotelian and non-­Aristotelian sources surfaces here and there in Brito and later medieval authors, which betrays the tension between the exegete and the philosopher that medieval authors typically bear. For Albert’s incoherencies, see Ebbesen, “Albert (the Great?)’s Companion to the Organon”; Marmo, “Suspicio.” 67 Cf. Logica, 130. 68  Albert, however, reduces it to the categorical syllogism. The tradition that considers a role for the hypothetical syllogism in Aristotelian logic is at least as old as Alexander of Aphrodisias; see Laura Castelli, “Introduction,” in Alexander of Aphrodisias, On Aristotle, Topics 2 (London: Blomsbury Academic, 2020), 1–30. 69  The idea that through the methods of logic one can progressively certify belief, i.e., make it more certain, is also present in Avicenna; cf. Logica, 133: “In his conceptions, humans have some things to which they attend more and some things to which [they attend] less. But they who have the doctrine of logic can study in order to certify their truth by visiting frequently the intention of their action according to its rule.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

170  Ana María Mora-Márquez demonstrative method, rather than the foremost means to achieving the­or­et­ic­al knowledge, as suggested by Nicholas of Paris as per the Neo-­ Platonic tradition, is here presented as the method suitable for the final stage of the scientific enterprise. In this sense, Albert’s understanding of AL as a study of method suggests a plurality of methods that correspond to different stages of the scientific enterprise, so that the different methods must be applied at their corresponding stage: dialectics, for instance, when the genus of a scientific object must be set; definition when its definition must be determined; dialectics again in order to establish the belief that some accident befalls it; and demonstration in order to improve the epistemic quality of this belief by making it certain through causal evidence. However, Albert also suggests that some sciences, because of the nature of their subject matter, have specific methods that suit such a nature. So, the plurality of methods may also be grounded in different kinds of objects: But here [i.e., in al-­Fārābī] the way . . . is varied according to the diversity of the matter regarding which scientific knowledge (scientia) is looked for . . . And in real sciences, it goes one way in probable [matter], otherwise in necessary and standing [matter], and otherwise in conjectural [matter], which are the sciences of divination . . . For some things are known in themselves, but some are taken from others as known, and maximally in conjectural sciences, regarding which nothing is known in itself according to the thing.70

This suggestion anticipates the later claim, by for instance Boethius of Dacia, that different things determine a specific scientific method that best suits them. To sum up, Albert continues the tradition (of which Nicholas of Paris represents an earlier stage) that considers knowledge (understood as the acceptance of a conclusion on the basis of evidence) to be the aim of the scientific enterprise, and that approaches LA as a systematic, pluralist, account of the corresponding method. But the introduction of 70  De v universalibus, tr.1, c. 7.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  171 Avicenna’s Logica brings about important breakthroughs. Albert departs from earlier accounts in holding, like Avicenna, that logic is a rational science, and hence in strengthening its focus on the epistemic processes that lead to knowledge production and dissemination. In line with the Islamic tradition, Albert’s account is strongly practice-­ oriented: he understands logic in general as a theory of argumentation that serves purposes beyond the strictly theoretical. However, in keeping with earl­ ier views, Albert also puts forward a logica principalis that revolves around the syllogism, which is the foremost tool for scientific inquiry. Finally, his most significant leap forward is his suggestion of a plurality of methods instrumental both for different stages of one scientific enterprise and for different kinds of scientific object. Later authors are explicit about this idea that different objects of science determine a method that best suits them, among whom Boethius of Dacia (1270s) and Simon of Faversham (1280s) are noteworthy. Boethius of Dacia claims71 that the methods that AL provides are causally reduced to certain properties of external things, their modes of being (modi essendi). This is why humans, regardless of their origin or of the language they speak, have the same patterns of reasoning.72 So, thing a, through its mode of being m1 (modes of being are, for instance, being endowed with perception, being white, being in Copenhagen, etc.) determines the method for knowing it, say dialectic, that best corresponds to it, while thing b, through its mode of being m2, determines another method, say demonstration. Boethius gives substance as an example of something that determines the method of definition. He adds that certain knowledge is possible only with respect to things with some specific modes of being, so as to make the capacity to choose the method that corresponds to a given mode of being something akin to an epistemic 71 In his Topica, in Boethii Daci opera VI.1, ed. N.  J.  Green-­Pedersen and J.  Pinborg (Copenhagen: Gad, 1976). 72 Cf. Topica, 4: “Every mode of the thing is reduced to the thing itself as to its cause and is taken from it; in fact, it cannot be a pure figment of the intellect. That which logic teaches or about which logic is is a mode of the thing, as it was proved. Hence, that about which logic is is reduced to the thing itself and is taken from it . . . And since things and their properties or modes of being are similar everywhere and cannot be changed . . . hence logic is the same for everyone and cannot be changed.” For Boethius of Dacia’s understanding of logic as a science, see Sten Ebbesen, “The Man Who Loved Every: Boethius of Dacia on Logic and Metaphysics,” Modern Schoolman 82 (2005), 235–50 (reprinted in Ebbesen, Collected Essays, vol. 2, 163–78).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

172  Ana María Mora-Márquez virtue that the prudent knower possesses.73 The same goes for Simon of Faversham, who writes his logical commentaries in the 1280s.74 For Simon, it is a shortcoming of the mind that does not know logic to think that it can reach certainty regarding all things, so those who know logic also know that certainty regarding all things is not possible and apply the method that corresponds to the nature of a given object.75 This idea that different scientific objects determine different scientific methods remains implicit in the analyses put forward by the later master Radulphus Brito, with whose account this study ends.

4.  1290s, Radulphus Brito AL-­13 reaches its pinnacle in Radulphus Brito’s logic of intentions, a study of scientific method influenced by Avicenna to a much greater extent than what was the case in his predecessors. Brito stresses more boldly than Boethius of Dacia76 the theoretical support of logic in the external world, where it has its causes and prin­ ciples, so as to characterize it as a scientia speculativa on a par with

73 Cf. Topica, 4: “. . . you must know that any thing through its property and mode of being determines a way of knowing (modus sciendi), and several [things] several [ways]; and if humans do not proceed according to that way, it is impossible for them to know the thing, as when they want to demonstrate what cannot be demonstrated, or know topically (topice scire) what can be demonstrated. In fact, in any matter it is prudent to search for certainty as much as the nature of the thing allows it, as it is said in Ethics I.” 74  Quaestiones super libro Porphyrii (= QPor), in Magistri Simonis Anglici sive de Faverisham Opera omnia, Vol. I, ed. P. Mazzarella (Padua: Cedam, 1957). 75 Cf. QPor, 12: “Some believe that there is demonstration of everything and hence they don’t want to accept anything that is [not] [(my revision)] said to them through the mathematical way. In fact, this happens because of an illnes of the mind, because as the Philosopher says in Metaphysics IV it is an illness of the mind to look for demonstration of that of which there is no demonstration. Therefore, this happens because of the ignorance of logic . . . They must be instructed so as to ask for as much certainty in something as the nature of the thing allows. Logic is the science that instructs humans in the cognition of anything. It provides the way (modum) according to which in the singular sciences one must arrive to the cognition of something unknown from something known so as to not look for the same level of certainty in all things but according to the exigence of the nature that is its subject matter. And hence, logic, even though it is a science in itself, is also a way of knowing (modus sciendi) for the other [sciences].” See also Antonino Tiné, “Simon of Faversham: Proemi alla logica,” in J.  Marenbon (ed.), Aristotle in Britain during the Middle Ages (Turnhout: Brepols, 1996), 211–31. 76  Boethius of Dacia, even though he claims that logic is about modes of being, still classifies logic as a rational science.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  173 natural philosophy, mathematics, and metaphysics. Unlike them, however, logic is not a principal (scientia principalis) but rather an instrumental science (scientia adminiculativa).77 Insofar as logic is instrumental, it teaches in a theoretical way the scientific method that should rule ­scientific practice in general: . . . any science has some way of knowing (modus sciendi) and in order not to repeat it in every science there must be some science that considers, through its own principles, the affections and properties of the way of knowing; such a science is logic. Hence logic, which teaches the way of knowing in any science . . . is instrumental principally for the theoretical sciences and consequently for every other science.78

Recall the claim in Boethius of Dacia and Simon of Faversham that logic teaches different methods, which are not all syllogistic, and that are appropriately suited to the nature of the thing problematized. Brito preserves this pluralism79 but within a unified theory, which takes the syllogism as the focal point to which every scientific method is subordinated: And because every way of knowing is subordinated to the syllogism, hence it is said that the syllogism is the subject matter of logic in its entirety as the main way of knowing to which all the other ways of knowing are subordinated.80

Thus, logic in its entirety (tota logica) is about the syllogism “in itself and absolutely . . . or according to its being in its constituent and sub­ject­ ive parts.”81 The constituent parts of a syllogism are predicates (its remote, mediate parts) and the statements (its proximate, immediate parts) of which predicates are parts. Its subjective parts are its different 77  The prologue of Brito’s commentary on the Isagoge (=InPor) is edited in Sten Ebbesen and Jan Pinborg, “Gennadios and Western Scholasticism: Radulphus Brito’s Ars Vetus in Greek Translation,” Classica et mediaevalia 33 (1981–2), 263–319. 78  InPor, 299. 79  Although Brito, contrary to Simon, does not adopt the general logic that corresponds to the extended Organon. 80  InPor, 301. 81  InPor, 301.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

174  Ana María Mora-Márquez kinds of instances, that is, demonstrative, dialectical, and sophistical. Accordingly, his first division of logic organizes the texts that form AL on the basis of the relation they hold to the syllogism: in itself and without qualification, Prior Analytics

Radulphus Brito, Division of Logic 1

in its constituent parts

syllogism in its parts

remote, Categories, Isagoge, Sex principia proximate, Perihermeneias regarding necessary matter, Posterior Analytics, De divisione

in its subjective parts

regarding probable matter, Topics, Boethius’ Topics regarding apparent matter, Sophistical Refutations

With this division, Brito inscribes himself in the tradition, going back to Nicholas of Paris, which considers the syllogistic method as the foremost tool for the acquisition of evidence-­based knowledge of the external world. Brito’s achievement in this tradition is to provide us with a perfectly structured way to divide the study of logic according to col­ lect­ ive­ ly coherent and mutually exclusive aspects of the syllogistic method. Noteworthy here are, first, the lack of any judgment on the scientific value of the different methods; hence, like Nicholas and Albert, Brito does not consider the formal aspect of logic (Prior Analytics) as particularly key, but unlike Nicholas and like Albert, Brito does not consider the demonstrative method (Posterior Analytics) as key either. The second key point is that the auxiliary books (i.e., Isagoge, Sex principia, De divisione, and Boethius’ Topica) are clearly subsumed under the standard Organon (Isagoge and Sex principia under Categories, De divisione under Posterior Analytics, Boethius’s Topica under Topics) and taken as supplementing what it lacks rather than as shedding light on its obscurities (this again unlike Nicholas).82 In Brito, then, AL-­13 reaches a level of internal coherence that his predecessors struggled to achieved. However, Brito’s major breakthrough is his unprecedented account of the epistemic foundations of the syllogistic production of scientific knowledge.

82 Cf. InPor, 305: “. . . it would be more properly said that they contribute to the being of logic rather than contributing to its well being.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  175

a.  Brito’s Logic of Intentions Brito’s sophisticated logic of second intentions83 builds directly on Avicenna’s epistemic perspectives and accidents of essences as understood.84 This logic aims to understand: (i) what is an intention; (ii) what are its kinds; and (iii) how do they relate to each other so as to produce new ones by establishing relations between already acquired ones. An intention is a cognitive relation to a thing that is directed to some aspect thereof, say, the cognitive relation to the human essence with a focus on its having perception, that is, its being an animal, and hence its being a biological species. Thus, intentions can be described as perspectival cognitions. In a direct reference to Avicenna, Brito tells us that the subject matter of logic are second intentions grounded on first intentions: . . . second intentions are certain secondary cognitions of the thing itself or certain notions for understanding the thing taken from certain common modes of being of the thing, and they presuppose the first cognition of the thing as the respective cognition presupposes the absolute cognition.85

83 For other analyses of Brito’s account of intentions, see Ana María Mora-­ Márquez, “Radulphus Brito on Common Names, Concepts and Things,” in Fink et al., Logic and Language in the Middle Ages, 357–72; Jan Pinborg, “Zum Begriff der Intentio Secunda: Radulphus Brito, Hervaeus Natalis und Petrus Aureoli in Diskussion,” CIMAGL 13 (1974), 49–59; and Mary Sirridge, “The Universal Living Thing is Either Nothing or Posterior,” in P.  Bakker and J.  Thijssen (eds.), Mind, Cognition and Representation: The Tradition of Commentaries on Aristotle’s De anima (Aldershot: Ashgate, 2007), 45–68. 84  Brito continues a trend that started at least as early as Boethius of Dacia and that follows in Avicenna’s footsteps in considering logic as a science about being in some sense. For Boethius it is about modes of being. For Peter of Auvergne (perhaps in the late 1270s) it is about being insofar as it falls under intellectual perspectives; see Antonino Tiné, “Le questioni su Porfirio di Pierre d’Auvergne,” AHDLMA 64 (1997), 235–333. Peter either did not comment on the logica nova or his commentaries did not come down to us; for Peter’s logic, see Sten Ebbesen, “The Logical Writings of Peter of Auvergne,” in C. Flüeler (ed.), Peter of Auvergne: University Master of the Thirteenth Century (Berlin: De Gruyter, 2015), 71–88. 85  InPor, 307. Brito mentions Avicenna’s ‘logica’ but quotes his Metaphysics (Liber de philosophia prima sive scientia divina I–IV, ed. S. Van Riet [Louvain: Peeters, 1977], 10): “The subject matter of logic, as you know, are the intentions thought secondarily, which are applied to the intentions thought primarily.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

176  Ana María Mora-Márquez Let’s take a look, first, at what Avicenna (latinus) says in his Logica about the subject matter of logic. Avicenna’s reasoning starts with the essence, which, according to him, exists in concrete reality or in the intellect, and which we can approach in three ways: first, in itself; second, as it exists in concrete individuals; and third, as it exists in the intellect.86 Different accidents befall essences according to their different modes of existence. In particular, predicational accidents befall them according to their existence in the intellect, among which we find predicational particularity, universality, essentiality and accidentality, being subject, being predicate, etc.87 Logic, Avicenna tells us, is concerned with the accidents that essences have according to their mental existence.88 So, in the Metaphysics, Avicenna must understand ‘intentio intellecta primo’ as the simple mental apprehension of an essence, and ‘intentiones intellectae secundo’ as the accidents of mental essences, notably those that come about when the former are put in more or less complex predicational relations to each other; accidents which, Avicenna stresses, do not befall external essences at all.89 Avicenna compares the intentiones intellectae primo to the items that are needed to build a house (wood, clay, etc.), and the intentiones intellectae secundo to the accidents those items have exclusively when they are understood as (possible or effective) parts of a house.90 It is on the basis of the latter, the intentiones intellectae secundo, that we can investigate what is unknown to us by taking what is known to us as evidence. Brito’s understanding of these Islamic ideas carries noteworthy and distinctive marks. Like Avicenna, Brito understands intentiones, in this context, as cognitive, mental items,91 but introduces a distinction between intentiones concretae and intentiones abstractae, which allows for a direct grounding of logical notions—second intentions—in the 86 Cf. Logica, 127. 87 Cf. Logica, 127: “And then accidents proper to its being [in the intellect] follow [the essence], such as supposition and predication, universality and particularity in predication, essentiality and accidentality in predication, etc.” 88 Cf. Logica, 127–8. 89 Cf. Logica, 127: “In external things there is neither essentiality nor accidentality at all, nor something compound or non-­compound, proposition or argumentation, or anything else of the sort.” 90 Cf. Logica, 136. 91 Cf. Quaestiones in SE, in Ebbesen and Pinborg, “Gennadius and Western Scholasticism,” 290.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  177 external world. A first intention is defined as the “first cognition of the thing, or the first notion for understanding the thing, taken from the mode of being proper to the thing,”92 for instance, the cognition of the human being (the thing) as a being endowed with reason (its proper mode of being). Brito takes such a cognition to be a relation between (i) the soul through an intellectual operation and (ii) an external essence that has the mode of being to which the operation is directed. If we approach the relation from the point of view of (i) its mental aspect, what is at stake is an abstract first intention (prima intentio in abstracto), say, the concept of humanity; if we approach it from the point of view of (ii), the essence with its proper mode of being, what is at stake is a concrete first intention (prima intentio in concreto).93 The latter should be something akin to Avicenna’s mental essence (essentia in intellectu). Second intentions (intentiones secundae) are the subject matter of logic. They should be akin to what Avicenna calls intentiones intellectae secundo, in his Metaphysics, and the accidents of the essence in intellectu, in his Logica. Elsewhere, Brito characterizes these second intentions as: . . . secondary intention[s] added to the first intention, as when after understanding the human being as reasoning and intelligent, I understand it thereafter as capable of being found in many numerically different things.94

As first intentions, second intentions can also be abstract or concrete. An abstract second intention is, for instance, the concept of uni­ver­sal­ity.95 A concrete second intention, in turn, is closer to what Avicenna takes the subject matter of logic to be, the accidents of the essence in intellectu; except that Brito takes the concrete second intention to be not the 92  InPor, 308. 93 Cf. InPor, 307: “For instance, as I can understand the human first according to its being reasoning and intelligent, and this cognition is called ‘abstract first intention’ . . . And the thing thus cognized is called ‘a concrete first intention’, such as the human.” 94  Quaestiones in SE, 290. 95 Cf. InPor, 307: “. . . this cognition is said to be an abstract second intention, such as specificity and universality, which are cognitions or notions for understanding a thing as it exists without qualification (absolute) in several things.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

178  Ana María Mora-Márquez mental accident itself, but the external essence as it is subjected to the mental accident. A concrete second intention, then, is the external essence understood from a cognitive perspective other than a focus on its proper being, that is, the external essence understood as genus, as species, as subject, as predicate, etc.96 Moreover, contrary to Avicenna’s claim that the mental accidents of the essence in intellectu do not befall external essences, for Brito the cognitive perspective that determines a concrete second intention is ultimately caused by certain metaphysical constituents of external essences, their common modes of being. In Brito’s words: [T]hese intentions are taken from common modes of being, such as from the mode of being in several things specifically different in relation to a genus, or numerically different in relation to a species, etc.97

In particular, the subject matter of logic are concrete second intentions, that is, external essences as understood from some cognitive perspective grounded in one of their common modes of being.98 Logic is, then, a theoretical science on a par with those sciences that have real subject matters, like physics and mathematics, insofar as the principles of its subject matter exist in external reality. But logic is also an instrumental science, so that as a study of scientific method AL carves the world at its crucial joints, that is, at the common modes of being that allow us to classify reality by means of common terms and perspectival cognitions. The different methods logic provides correspond to the ways in which external things and their modes of being are related to each other, according to the following division: 96 Cf. InPor, 307: “. . . the thing thus understood is a concrete second intention, such as universal, species, or genus.” 97  InPor, 307–8. For modi essendi and their relation to the notion of apparens, see Silvia Donati, “Apparentia and modi essendi in Radulphus Brito’s Doctrine of the Concepts: The Concept of Being,” in Fink et al., Logic and Language in the Middle Ages, 337–55; Sten Ebbesen, “Radulphus Brito on the ‘Metaphysics’, ” in J. A. Aertsen, K. Emery, Jr., and A. Speer (eds.), Nach der Verurteilung von 1277: Philosophie und Theologie an der Universität von Paris im letzten Viertel des 13. Jahrhunderts (Berlin: De Gruyter, 2001), 456–92 (reprinted in Ebbesen, Collected Essays, vol. 2, 179–96). 98 Cf. InPor, 309: “. . . let’s divide logic according to the division of concrete second intentions logic is about.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  179 according to the first act of the intellect according to the second act of the intellect

Radulphus Brito, Division of Logic 2

predicates, Categories, Isagoge, Sex principia statements, Perihermeneias in itself and without qualification, Prior Analytics

second intentions

regarding necessary matter, Posterior Analytics, De divisione according to the third act of the intellect

different kinds of proof regarding probable matter, Topics, Boethius’ Topics regarding apparent matter, Sophistical Refutations

According to this division, the different parts of AL deal with col­lect­ ive­ly coherent and mutually exclusive sets of logico-­epistemic notions somehow linked to the syllogistic proof and whose principles are to be found in the external world. Brito has little to say about the multiplicity of methods that logic puts forward, not least, I think, because he takes for granted Boethius of Dacia’s and Simon of Faversham’s point that different kinds of things require different methods of scientific inquiry. This is, at any rate, what he suggests when he addresses the following objection: And you could say: “from where is it that you distinguish these logical books according to a distinction of objects and not according to a distinction of intentions?” We must say that the distinction of intentions is according to a distinction of objects, and hence inasmuch as the objects are diverse, the intentions attributed to them are diverse.99

Consequently, objects that can be understood as genus, because they have the relevant mode of being, can be dealt with dialectically, those that can be understood causally can be dealt with demonstratively, and so on, as long as the different arguments that are used in order to support claims about them ultimately rely on a syllogistic proof. Brito, then, pushes to their logical conclusions the steps that his close predecessors (Boethius of Dacia, Peter of Auvergne, and Simon of Faversham) did not push that far. The result is an account of AL as a pluralist study of scientific method (a scientia teaching a modus sciendi) that includes a thorough account of its dependence on logico-­ epistemic notions (second intentions), whose tools allow us to carve the world at some

99  InPor, 293.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

180  Ana María Mora-Márquez crucial joints (scientia speculativa grounded in communes modi essendi), that is systematically organized around the syllogistic proof as its backbone, and whose plurality is dictated by the different natures of scientific objects.

5.  Conclusion In this journey from Nicholas of Paris in the 1230s to Radulphus Brito in the 1290s, AL-­13 has noteworthy points of continuity and of dis­con­ tinu­ity. The main points of continuity are: (1) a shared presupposition that the aim of scientific inquiry is knowledge, either for its own sake or for the sake of action, and an analysis of knowledge as the knower’s acceptance of a given thesis on the basis of solid evidence; (2) the general understanding of logic as a science instrumental for the aim of knowledge production and dissemination in any other science, that is, as a study of scientific method; and (3) the prominent position given to the syllogistic argument (which is an epistemic and not a formal tool) among the different scientific methods. The main points of dis­con­tinu­ ity are: (a) the gradual shift from understanding logic as a linguistic dis­ cip­line to understanding it first as a rational discipline and then as a theoretical discipline, which reflects the movement towards a scientific method through which the different sciences can carve the world at its crucial joints; (b) the gradual shift towards an understanding of logic as a methodological discipline coherently structured around the syllogistic proof, so that in its final stage of development all parts of the logical curriculum are clearly subordinated to the syllogism as the foremost method of knowledge production and dissemination; (c) the gradual shift towards a theoretical discipline that meets central requirements of Aristotle’s Posterior Analytics, notably a specific and unified subject matter (second intentions) and external causal principles (common modes of being); and (d) the gradual shift towards a study of scientific method that posits a plurality of tools clearly determined by different sorts of scientific objects. Evidently, many aspects of this logical tradition are in need of further inquiry, among which are the detailed accounts of each of the methods

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  181 it proposes, that is, of definition, division, syllogism (demonstrative, dialectical, and sophistical), and induction: what is their exact contribution to the production of scientific knowledge in the medieval setting? To what kind of objects do they correspond? And how exactly do they co-­exist coherently within one and the same scientific inquiry?100 Gothenburg University, Department of Philosophy, Linguistics, and Theory of Science (FLoV)

Bibliography Albert the Great. Super Porphyrium de V universalibus, in M. Santos-Noya (ed.), Alberti Magni Opera omnia, t. 1A (Monasterii Westfalorum: Aschendorff, 2004). Aristotle. Prior Analytics, tr. H.  Tredennick (Cambridge, MA: Harvard University Press, 1938). Avicenna. Liber de philosophia prima sive scientia divina I–IV, ed. S.  Van Riet (Louvain: Peeters, 1977). Avicenna (Ibn Sina). Logica (logique du Šifā), ed. F.  Hudry (Paris: Vrin, 2018). Biard, Joël. “Produire la confiance: la fides dans la tradition topique médiévale,” in C.  Grellard, P.  Hoffmann, and L.  Lavaud (eds.), Genèses antiques et médiévales de la foi (Paris: Institut d’études augustiniennes, 2020), 295–313. Black, Deborah  L. Logic and Aristotle’s Rhetoric and Poetics in Medieval Arabic Philosophy (Leiden: Brill, 1990). Black, Deborah  L. “Psychology: Soul and Intellect,” in P.  Adamson and R.  C.  Taylor (eds.), The Cambridge Companion to Arabic Philosophy (Cambridge: Cambridge University Press, 2004), 308–26. Boethius of Dacia. Boethii Daci Opera: Topica-Opuscula, ed. N.  J.  GreenPedersen and J. Pinborg (Copenhagen: Gad, 1976). 100  This paper was produced with the financial support of the Knut and Alice Wallenberg Foundation (Sweden) and the Swedish Research Council. It benefited greatly from the criticism and comments by an anonymous referee, Gustavo Fernández Walker, Jenny Pelletier, Justin Vlasits, and some participants at the Cornell Colloquium in Medieval Philosophy 2020. I am very thankful to all of them. Jenny Pelletier also made sure that the English is correct and reads smoothly, so I would like to thank her twice.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

182  Ana María Mora-Márquez Braakhuis, H. A. G. “Logica Modernorum as a Discipline at the Faculty of Arts of Paris in the Thirteenth Century,” in O. Weijers and L. Holtz (eds.), L’enseignement des disciplines à la faculté des arts (Paris et Oxford, XIIIe–XVe siècles) (Turnhout: Brepols, 1997), 129–45. Brockliss, L.  W.  B. The University of Oxford: A History (Oxford: Oxford University Press, 2016). Brumberg-Chaumont, Julie. A l’école de la logique (Habilitation à diriger des recherches, France, 2019). Brumberg-Chaumont, Julie (ed.). Ad notitiam ignoti: l’Organon dans la translatio studiorum à l’époque d’Albert le Grand (Turnhout: Brepols, 2013). Brumberg-Chaumont, Julie. “Les divisions de la logique selon Albert le Grand,” in J.  Brumberg-Chaumont (ed.), Ad notitiam ignoti: l’Organon dans la translatio studiorum à l’époque d’Albert le Grand (Turnhout: Brepols, 2013), 335–416. Cartwright, Nancy. How the Laws of Physics Lie (Oxford: Clarendon Press, 1983). Castelli, Laura M. “Introduction,” in Alexander of Aphrodisias, On Aristotle, Topics 2, tr. L. M. Castelli (London: Bloomsbury Academic, 2020). Cicero. Topica, tr. T. Reinhardt (Oxford: Oxford University Press, 2003). Clagett, Marshall (ed.). Critical Problems in the History of Science (Madison, WI: University of Wisconsin Press, 1969). Corbini, Amos. La teoria della scienza nel XIII secolo: i commenti agli analitici secondi (Florence: SISMEL, 2006). Crubellier, Michel. “L’unité de l’Organon,” in J. Brumberg-Chaumont (ed.), Ad notitiam ignoti: l’Organon dans la translatio studiorum à l’époque d’Albert le Grand (Turnhout: Brepols, 2013), 37–62. Denifle, Henrich and Emile Chatelaine (eds.). Chartularium Universitatis Parisiensis, Vol. 1 (Cambridge: Cambridge University Press, 2014). Donati, Silvia. “Apparentia and modi essendi in Radulphus Brito’s Doctrine of the Concepts: The Concept of Being,” in J.  L.  Fink, H.  Hansen, and A.  M.  Mora-Márquez (eds.), Logic and Language in the Middle Ages: A Volume in Honour of Sten Ebbesen (Leiden: Brill, 2012), 337–55. Dutilh Novaes, Catarina. Formal Languages in Logic: A Philosophical and Cognitive Analysis (Cambridge: Cambridge University Press, 2012). Dutilh Novaes, Catarina. Formalizing Medieval Logical Theories: Suppositio, Consequentiae and Obligationes (Dordrecht: Springer, 2007).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  183 Ebbesen, Sten. “Anonymous D’Orvillensis’ Commentary on Aristotle’s Categories,” Cahiers de l’Institut du Moyen-Âge Grec et Latin 70 (1999), 229–423. Ebbesen, Sten. “The Logical Writings of Peter of Auvergne,” in C.  Flüeler (ed.), Peter of Auvergne: University Master of the Thirteenth Century (Berlin: De Gruyter, 2014), 71–88. Ebbesen, Sten. “The Man Who Loved Every: Boethius of Dacia on Logic and Metaphysics,” The Modern Schoolman 82 (2005), 235–50. Ebbesen, Sten. “The Prior Analytics in the Latin West: 12th–13th Centuries,” Vivarium 48 (2010), 96–133. Ebbesen, Sten. “Radulphus Brito on the ‘Metaphysics’, ” in J.  Aertsen, K.  Emery, Jr., and A.  Speer (eds.), Nach der Verurteilung von 1277: Philosophie und Theologie an der Universität von Paris im letzten Viertel des 13. Jahrhunderts (Berlin: De Gruyter, 2001), 456–92. Ebbesen, Sten. Topics in Latin Philosophy from the 12th–14th Centuries: Collected Essays of Sten Ebbesen, vol. 2 (Aldershot: Ashgate, 2009). Ebbesen, Sten. “The Way Fallacies Were Treated in Scholastic Logic,” Cahiers de l’Institut du Moyen-Âge Grec et Latin 55 (2001), 107–34. Ebbesen, Sten. “What Counted as Logic in the Thirteenth Century?,” in M.  Cameron and J.  Marenbon (eds.), Methods and Methodologies: Aristotelian Logic East and West, 500–1500 (Leiden: Brill, 2011), 93–108. Ebbesen, Sten and Jan Pinborg. “Gennadios and Western Scholasticism: Radulphus Brito’s Ars Vetus in Greek Translation,” Classica et mediaevalia 33 (1981–2), 263–324. Fink, Jakob, H.  Hansen, and A.  M.  Mora-Márquez (eds.). Logic and Language in the Middle Ages: A Volume in Honour of Sten Ebbesen (Leiden: Brill, 2013). Green-Pedersen, Niels Jørgen. The Tradition of the Topics in the Middle Ages: The Commentaries on Aristotle’s and Boethius’ Topics (Munich: Philosophia Verlag, 1984). Hudry, Françoise. “La tradition manuscrite de la Logica d’Avicenne,” in F. Hudry (ed.), Logica (logique du Šifā) (Paris: Vrin, 2018), 94–7. Isidore of Seville. Etymologies, tr. S. A. Barney, W. J. Lewis, J. A. Beach, and O. Berghof (Cambridge: Cambridge University Press, 2006). Lafleur, Claude and Joanne Carrier. “L’Introduction à la philosophie de maître Nicolas de Paris,” in C. Lafleur and J. Carrier (eds.), L’enseignement de la philosophie au XIIIe siècle (Turnhout: Brepols, 1997), 447–65.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

184  Ana María Mora-Márquez Laudan, Laurens. “Theories of Scientific Method from Plato to Mach: A Bibliographical Review,” History of Science 7 (1968), 1–63. Libera, Alain de. “Entre Tolède et Hamadan. La Logica d’Avicenne,” in F. Hudry (ed.), Logica (logique du Šifā) (Paris: Vrin, 2018), 9–22. Libera, Alain de. La querelle des universaux: de Platon à la fin du Moyen Âge (Paris: Seuil, 2014). Longino, Helen  E. The Fate of Knowledge (Princeton, NJ: Princeton University Press, 2002). Longino, Helen  E. Science as Social Knowledge: Values and Objectivity in Scientific Inquiry (Princeton, NJ: Princeton University Press, 1990). MacFarlane, John. What Does It Mean to Say That Logic Is Formal (PhD dissertation, University of Pittsburgh, 2000). Mandosio, Jean-Marc. “La place de la logique et ses subdivisions dans l’Enumération des Sciences d’al-Fârâbî et chez Dominicus Gundissalinus,” in J.  Brumberg-Chaumont (ed.), Ad notitiam ignoti: l’Organon dans la translatio studiorum à l’époque d’Albert le Grand (Turnhout: Brepols, 2013), 285–310. Marmo, Costantino. La semiotica del XIII secolo: tra arti liberali e teologia (Milan: Bompiani, 2010). Marmo, Costantino. “Logique élargie et sémiotique: Albert le Grand, Roger Bacon et Gilles de Rome,” in J. Brumberg-Chaumont (ed.), Ad notitiam ignoti: l’Organon dans la translatio studiorum à l’époque d’Albert le Grand (Turnhout: Brepols, 2013), 447–65. Marmo, Costantino. “Suspicio: A Key Word to the Significance of Aristotle’s Rhetoric in Thirteenth-Century Scholasticism,” Cahiers de l’Institut du Moyen-Âge Grec et Latin 60 (1990), 145–98. Mora-Márquez, Ana María. “Radulphus Brito on Common Names, Concepts and Things,” in J. Fink, H. Hansen, and A. M. Mora-Márquez (eds.), Logic and Language in the Middle Ages: A Volume in Honour of Sten Ebbesen (Leiden: Brill, 2013), 357–72. Mora-Márquez, Ana María. The Thirteenth-Century Notion of Signification: The Discussions and Their Origin and Development (Leiden: Brill, 2015). Nola, Robert and Howard Sankey (eds.). After Popper, Kuhn and Feyerabend: Recent Issues in Theories of Scientific Method (Dordrecht: Springer, 2001). Parsons, Terence. Articulating Medieval Logic (Oxford: Oxford University Press, 2014).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

Thirteenth-Century Aristotelian Logic  185 Peter of Spain. Summulae Logicales. Tractatus Called Afterwards Summule Logicales: First Critical Edition from the Manuscripts with an Introduction by L. M. de Rijk (Assen: Van Gorcum, 1972). Piché, David. Le problème des universaux à la faculté des arts de Paris entre 1230 et 1260 (Paris: Vrin, 2005). Pinborg, Jan. “Zum Begriff der Intentio Secunda: Radulphus Brito, Hervaeus Natalis und Petrus Aureoli in Diskussion,” Cahiers de l’Institut du Moyen-Âge Grec et Latin 13 (1974), 49–59. Pini, Giorgio. Categories and Logic in Duns Scotus: An Interpretation of Aristotle’s Categories in the Late Thirteenth Century (Leiden: Brill, 2002). Rijk, L. M. de. “Introduction,” in Peter of Spain, Tractatus Called Afterwards Summule Logicales: First Critical Edition from the Manuscripts with an Introduction by L. M. de Rijk (Assen: Van Gorcum, 1972). Rijk, L. M. de. Logica Modernorum: A Contribution to the History of Early Terminist Logic (Assen: Van Gorcum, 1962). Robert, Aurélien. “Le débat sur le sujet de la logique et la réception d’Albert le Grand au Moyen Âge,” in J.  Brumberg-Chaumont (ed.), Ad notitiam ignoti: l’Organon dans la translatio studiorum à l’époque d’Albert le Grand (Turnhout: Brepols, 2013), 467–512. Sabra, Abdelhamid I. “Avicenna on the Subject Matter of Logic,” Journal of Philosophy 77 (1980), 746–64. Shank, Michael H. and David C. Lindberg. “Introduction,” in D. C. Lindberg and M.  H.  Shank (eds.), The Cambridge History of Science, Volume 2: Medieval Science (Cambridge: Cambridge University Press, 2013), 1–26. Simon of Faversham. Magistri Simonis Anglici sive de Faversham Opera logica: 1. Quaestiones super libro Porphyrii; 2. Quaestiones super libro Praedicamentorum; 3. Quaestiones super libro Perihermeneias, ed. P. Mazzarella (Padua: Cedam, 1957). Sirridge, Mary. “The Universal Living Thing is Either Nothing or Posterior,” in P. J. J. M. Bakker and J. M. M. H. Thijssen (eds.), Mind, Cognition, and Representation: The Tradition of Commentaries on Aristotle’s De anima (Aldershot: Ashgate, 2007), 45–68. Stump, Eleonore. Dialectic and Its Place in the Development of Medieval Logic (Ithaca, NY: Cornell University Press, 1989). Thom, Paul. Robert Kilwardby’s Science of Logic: A Thirteenth-Century Intensional Logic (Leiden: Brill, 2019).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 24/06/21, SPi

186  Ana María Mora-Márquez Tiné, Antonino. “Le questioni su Porfirio di Pierre d’Auvergne,” Archives d’histoire doctrinale et littéraire du Moyen Âge 64 (1997), 235–333. Tiné, Antonino. “Simon of Faversham: Proemi alla logica,” in J. Marenbon (ed.), Aristotle in Britain during the Middle Ages (Turnhout: Brepols, 1996), 211–31. Tremblay, Bruno. “Albert le Grand et le problème du sujet de la science logique,” Documenti e studi sulla tradizione filosofica medievale 22 (2011), 301–45. Verger, Jacques. L’essor des universités au XIIIe siècle (Paris: Cerf, 1998). Vlasits, Justin. “Pyrrhonism and the Dialectical Methods: The Aims and Arguments of Outlines of Pyrrhonism II,” History of Philosophy and Logical Analysis 23 (2020), 225–52.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, and the Being of Universal Forms Franziska van Buren

About 50 years ago, there was a flurry of debate around the question of the influence of Aristotle on Bonaventure’s philosophy. At one end of the extreme, Gilson claimed that Bonaventure was entirely opposed to Aristotelianism. At the other end, van Steenberghen—indeed, alone in this position—maintained not only that Bonaventure was wholly committed to integrating Aristotelianism into his own thought, but that Aristotle was the key influence on Bonaventure’s philosophical views, to a greater extent even than Augustine. Over time, these two extreme positions arrived at a kind of middle ground in the minds of most ­scholars: Bonaventure uses Aristotle, but his philosophical thought is not properly speaking Aristotelian in nature. Why? Certainly, one of the most important reasons is that Bonaventure and Aristotle have fundamentally opposed ontologies due to the differences between their respective understandings of form.1 Indeed, with the exception of van Steenberghen, this is a point that all scholars of Bonaventure’s thought agree is a major rift between the two thinkers—perhaps one even more foundational than the question of the eternity of the world. Bonaventure’s forms have esse in the sense of being ontologically prior to the hylo­morph­ic composite, while Aristotle’s forms are immanent in and dependent on

1  David Knowles, for example, cites this as the major rift between the two thinkers. See The Evolution of Medieval Thought (New York: Vintage Books, 1962), 245. Elders comes to a similar conclusion in Leo Elders, “Les citations d’Aristote dans le ‘Commentaire sur les Sentences’ de Saint Bonaventure,” in A. Pompei (ed.), San Bonaventura maestro di vita Francescana e di sapienza Cristiana (Rome: Pontificia Facoltà Teologica San Bonaventura, 1976), I: 831–42. Franziska van Buren, Bonaventure, Aristotle, and the Being of Universal Forms In: Oxford Studies in Medieval Philosophy Volume 9. Edited by: Robert Pasnau, Oxford University Press. © Franziska van Buren 2021. DOI: 10.1093/oso/9780192844637.003.0005

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

188  Franziska van Buren their composition in sensible things. Taking the ontological priority of the forms to imply separateness of the forms, it seems then that, on this pivotal metaphysical position, Bonaventure must adhere to Platonism rather than Aristotelianism. In turn, if Bonaventure is at odds with Aristotle when it comes to a metaphysical position as basic as the nature of forms, it seems difficult to see how one could be justified in calling him an Aristotelian. This characterization of Bonaventure’s thought, however, is precisely what I will bring into question in this paper. First of all, Bonaventure certainly would not agree with the above assessment of the difference between Plato and Aristotle. For Bonaventure, both Plato and Aristotle consider forms to be universals and ontologically prior to sensible things. The difference, in Bonaventure’s eyes, is that Plato considers his forms to be separate from (or to transcend) particulars, insofar as they exist in a world distinct from the sensible world around us. Aristotle, however, considers the forms to be an inseparable part of nature. When it comes to this choice between the Platonic view of forms and the Aristotelian, Bonaventure quite clearly and emphatically takes the side of Aristotle. In the course of this paper, I will show that Bonaventure’s understanding of form, despite being in accordance with Plato’s view that forms are not ontologically dependent on sensibles, is based entirely on his reading of Aristotle’s texts, such as we find it primarily in his Commentary on the Sentences. Thus, with respect especially to the concept of form, Bonaventure is indeed an Aristotelian—even if he is advocating an Aristotelianism quite different from our contemporary notions. In order to demonstrate Bonaventure’s unique brand of Aristotelianism, I will highlight the following positions in Aristotle’s thought upon which Bonaventure bases his understanding of form: (1) Aristotle’s prohibition on universals being in sensible particulars, as Bonaventure finds it in the Categories and the Posterior Analytics; (2) Aristotle’s distinction between the form and the particular in de Caelo and the Metaphysics.2 The first point establishes the foundational paradox for Bonaventure: universal 2  Particularly on the first point, we will see Bonaventure’s arguments are quite similar to those of Ockham (Ordinatio 1.2.7), insofar as Bonaventure argues against both a naive realism which places the whole universal in each particular, and the conceptualist position which claims that the particularized form allows the mind to attain knowledge of a universal.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  189 forms cannot be “in” particulars without being deprived of their universality; however, they also cannot be separate from their particulars, lest we end up with all the same problems that Plato had with his separate forms. The second point directs us to Bonaventure’s solution: to posit a “particularized” form (i.e. the sem­inal reason) that is distinct from the universal form—a distinction that Bonaventure does not name explicitly, but one that lies between the real and the conceptual. On the one hand, Bonaventure avoids understanding the forms as separate, inasmuch as the forms are not really distinct from the sensible world, as the forms of the Platonists are—yet, on the other hand, he also does not reduce the forms to being dependent on the sensible composite inasmuch as they are also not merely conceptually distinct from the particularized form.3 This distinction between the universal form and the particular form (i.e. seminal reason) Bonaventure bases on Aristotle’s understanding of the relationship between the form itself (the universal) and the form-­ in-­ the-­ matter (the particularized form), as found in de Caelo. However, before we delve into the details of the above two points, which will comprise the second and third sections of this paper, I would like briefly to lay out some basics about Bonaventure’s notion of form, and how it has been received in modern scholarship. Here, we will look at the texts in Bonaventure that have been most often discussed, including Bonaventure’s infamous critique of Aristotle in the Collationes— quite clearly the biggest challenge to my thesis—before turning to my own selection of passages from Bonaventure, ones that are often overlooked.

1.  The Current Consensus: Bonaventure’s Forms and his Attitude towards Aristotle When scholars first became interested in Bonaventure at the end of the nineteenth century and into the twentieth, the general consensus was that Bonaventure should be characterized as an “Augustinian” in 3  To a certain extent, this is similar to Duns Scotus’s distinction between the individual form and the common nature. Yet, in contrast to Scotus, Bonaventure clearly places being (esse) among the forms instead of among sensible things.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

190  Franziska van Buren opposition to the “Aristotelians” of his day, such as Aquinas.4 Indeed, this earlier characterization was compounded by Gilson’s in-­depth study of Bonaventure’s thought, which in a similar manner portrayed Bonaventure as a defender of the traditional Augustinian sources against the perils of Aristotelianism.5 It was in this climate that van Steenberghen made the claim that his fellow scholars found so provocative: that Bonaventure is an Aristotelian. As van Steenberghen points out, the only real evidence that Bonaventure was critical of Aristotle comes from the Collationes6—a text that no doubt has already been in the forefront of the reader’s mind. In the Collationes, Bonaventure criticizes (or appears to criticize) Aristotle not only for maintaining that the world was eternal and for developing a philosophy that could be used to defend the unity of the intellect, but— most worrisome for our purposes here—for denying Platonic forms. Wanting to mitigate Bonaventure’s anti-­Aristotelian sentiment in the Collationes, van Steenberghen rightly notes that the Collationes sermons were written after the Commentary on the Sentences, during the period (1267–73) when the validity of using the works of Aristotle was being questioned.7 Van Steenberghen’s position is that because of Bonaventure’s generally positive attitude towards Aristotle in the Commentary, these comments made about Aristotle in the Collationes are simply a concession made to an increasingly anti-­ Aristotelian audience—and thus should be taken with a grain of salt. However, van Steenberghen’s claim, insofar as it did not go into the details of what Bonaventure says in the Commentary as opposed to what he says the Collationes, only served to mitigate the earlier characterization of Bonaventure as an

4  This view of Bonaventure was also very much tied to the characterization of Bonaventure’s student, John Peckham, as an anti-­Aristotelian. See Franz Ehrle, “Der Augustinismus und der Aristotelismus in der Scholastik gegen Ende des 13 Jahrhunderts,” Archiv für Literatur und Kirchengeschichte des Mittelalters 5 (1889), 603–35, and “John Peckham über den Kampf des Augustinismus und Aristotelismus in der zweiten Hälfte des 13 Jhs,” Zeitschrift für katolische Theologie 13 (1889), 172–93. 5  Etienne Gilson, La philosophie de Saint Bonaventure (Paris: Vrin, 1924), 98–100. In agreement with Gilson, see Daniel Callus, “The Philosophy of St. Bonaventure and St. Thomas,” Blackfriars 21 (1940), 151–64, and Knowles, Evolution, 243–5. 6 Specifically, Collationes in Hexaemeron 6–7. 7  Fernand van Steenberghen, The Philosophical Movement in the Thirteenth Century (Belfast: Thomas Nelson & Sons, 1955), 59.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  191 anti-­ Aristotelian—leading us to the now standard position that Bonaventure uses some Aristotle, but is not properly an “Aristotelian.” Here, it is necessary for us to turn to the Collationes themselves insofar as this is the main (and indeed often only) text that scholars mine for evidence of Bonaventure’s “anti-­Aristotelianism.” While it is almost universally accepted that the Collationes confirm Bonaventure’s hostility towards the task of incorporating Aristotelianism into a Christian phil­ oso­phy, I will show that by looking at the text of the Collationes, as well as comparing it to parallel discussions from the Commentary, we will find quite a different picture. First of all, we will see that while the Collationes seems critical of Aristotle on a number of issues, when each of these issues is discussed in the Commentary, Bonaventure speaks rather quite positively of Aristotle. Secondly, in the Collationes themselves, Bonaventure includes in each of his critiques of Aristotle a caveat that he is targeting not Aristotle himself but a particular interpretation of Aristotle, citing either the Arabic philosophers or fellow Christian interpreters. In the Collationes, there are three positions to which Bonaventure seems to accuse Aristotle of being “blind.” The most easily dismissed is the criticism concerning the unity of the intellect, a position that is certainly not maintained by Aristotle himself but only defended using Aristotle—and Bonaventure is aware of this in the Collationes as well as in the Commentary, and so this critique in reality is not aimed at Aristotle, but at Ibn Rushd. The second “blindness” on the part of Aristotle is his dismissal of Platonic ideas. Indeed, this one seems most important for our purposes in this paper and so we will wait to address it in greater detail in the following section when we turn to the intricacies of Bonaventure’s view of forms. We can say, however, in a pre­lim­in­ ary way, that while the Collationes praises Plato’s quasi-­anticipation of a Christian notion of divine ideas, the Commentary quite explicitly rejects the application of Platonic forms to divine ideas and even states that Aristotle was right to reject the Platonic notion of separate forms.8

8  It is also good to note that in the Collationes, Bonaventure is not speaking about divine ideas in general, but specifically targets ideas of virtues.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

192  Franziska van Buren Leaving this second blindness aside for the moment, let us move directly on to the third—and indeed most infamous criticism of Aristotle: that he was wrong to maintain that the world was eternal.9 First of all, one should note that in the Collationes Bonaventure says that the position that the world is eternal is one that only seems to be in Aristotle (not dissimilar to the issue of the unity of the intellect). Bonaventure makes clear that in attributing such a position to Aristotle, he is relying on the interpretation of Aristotle provided by the “Greek Doctors” and the Arabs, while he himself concludes only that “Aristotle’s words seem to sound like this.”10 Secondly, in the Collationes, there is not the extensive list of arguments for the position that the world began at one point in time that we find in the Commentary. Turning then to the Commentary, we find that each of Bonaventure’s arguments in favor of a temporal beginning to the world, barring the last which is based on the definition of “creation,” begins with a premise from Aristotle—either explicitly naming Aristotle or quoting him.11 This, of course, means nothing more than that Bonaventure was using Aristotle—a point that a few (although oddly not all) scholars concede—not that Bonaventure necessarily thinks Aristotle is on his side in this debate.12 9 Most scholars have highlighted this as one of the main points of division between Bonaventure and Aristotle. See Angelo Marchesi, “L’atteggiamento di S. Bonaventura di fronte al pensiero di Aristotele,” in A. Pompei (ed.), San Bonaventura maestro di vita Francescana e di sapienza Cristiana Vol. I (Rome: Pontificia Facoltà Teologica San Bonaventura, 1976), 843–59; Bernandino Bonansea, “The Question of an Eternal World in the Teaching of St. Bonaventure,” Franciscan Studies 34 (1974), 7–33; and Christopher Cullen, Bonaventure (Oxford: Oxford University Press, 2006), 42. 10  Collationes VI.4. “ . . . as Aristotle seems to say, according to all of the Greek Doctors, as Gregory of Nyssa, Gregory of Nazianzus, Damascus, Basil, and all of the Arabic commentators . . . .” 11  The references to Aristotle are as follows: Argument 1 (impossibility of adding to the in­fin­ite) contains a direct quote: “it is impossible that infinity be increased” (de Cael. I.12 [283a4–11], cf. Phys. III.6 [206a9–207a14]). Argument 2 (impossibility of traversing the in­fi­ nite) references Phys. VIII.5 [257a32–258b9], de Cael. I.V [272a22–30], and Meta. II.2 [994b28–35]. Argument 3 (impossibility of knowing the infinite) contains a direct quote: “It is impossible to traverse the numerically infinite” (Post. Anal. I.22 [83b7], cf. Phys. VI.7 [238a32–238b22]). Argument 4 (impossibility of ordering the infinite) references Meta. VII.3 [1043b33–1044a11], and Phys. VII.1 [241b33–242b54]. Argument 5 (impossibility of an infinity existing at one time, with respect to an infinite number of souls) contains a direct quote: “the proper act is in the proper matter” (de An. II.2 [414a26–27]), and references Phys. II.2 [194a1–194b8], cf. Phys. III.5 [205b7–206a6], and de Cael. I.5 [271b1–27]. 12  Boehner and Gilson both mention his use of Aristotle, but do not draw any conclusions. See Gilson, La philosophie, 179–95 and Philotheus Boehner, The History of the Franciscan School: John of Rupella and Saint Bonaventure (St. Bonaventure, NY: Franciscan Institute

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  193 However, after these arguments, Bonaventure explicitly addresses Aristotle’s own position—Aristotle, whom he here calls the “most excellent of philosophers.”13 Bonaventure entertains two different readings of Aristotle on this point. The first is that Aristotle does in fact think that the world is eternal. The other is the position held by the quite obviously “Aristotelian” thinker Albert the Great.14 Albert maintains that Aristotle considers the world to be eternal only with reference to physical causes. Bonaventure concedes that one cannot say with absolute certainty that the reading à la Albert is right, because Aristotle himself never explicitly stated such. However, Bonaventure says that a reading such as Albert’s is the better reading because it makes better sense within Aristotle’s wider metaphysics—as evidenced by the fact, as Bonaventure notes, that he himself has just used basic Aristotelian positions to argue that the world began in time.15 Bonaventure then concludes that if Aristotle himself were to have thought that the world did not begin in time, he would simply have been contradicting himself. Or put another way, despite the fact that Aristotle might have made a mistake, for Bonaventure, Aristotelianism, taken as a philosophical system, supports Bonaventure’s Publications, 1943). Quinn and Kovach mention only a few of Bonaventure’s references to Aristotle. See John Francis Quinn, “St. Bonaventure and the Arabian Interpretations of Two Aristotelian Problems,” Franciscan Studies 37 (1977), 219–28, and Francis Kovach, “The Question of the Eternal World in St. Bonaventure and St. Thomas: A Critical Analysis,” The Southwestern Journal of Philosophy 5 (1974), 141–72. 13  In Sent. II, d. 1, p. 1, a. 1, q. 2, p. 22. 14 For a summary of Albert’s position, see David Twetten, Steven Baldner, and Steven C. Snyder, “Albert’s Physics,” in I. M. Resnick (ed.), A Companion to Albert the Great: Theology, Philosophy, and the Sciences (Leiden: Brill, 2013), 173–219. Interestingly, both Albert and Bonaventure make reference to the fact that Aristotle seems to indicate that even the fifth element which engages in eternal motion needs a cause—i.e. God. See footnote 26. And see de Cael. I.4 [271a34–5]. 15  The assessment of the second position is as follows: “However, some moderns say that the Philosopher did not himself feel nor intend to prohibit that the world began in time at all, but only to prohibit that it began by a natural motion. Which of these is more true, I do not know; I know that if he held the first, that the world did not begin according to nature, then he maintained the right position and his reasons which I have summarized above from motion are efficacious. If, however, he felt that the world did not begin at all, obviously he was in error, according to those reasons I have put forth above. And it was necessary, putting him at a contra­dic­tion which should be avoided, that either the world was not made or it was not made from nothing.” This point about the contradiction in Aristotle seems to reference the fact that Bonaventure thinks Aristotle considered the world to be made from nothing (see footnote 26), which Bonaventure thinks implies a temporality to creation. He then says that attributing the eternity of the world to Aristotle would also contradict the position—which Bonaventure considers Aristotle to maintain—that souls are immortal, and there is no transmigration of the soul, a point he made in his fourth argument. In Sent. II d. 1. p. 1. a. 1. q 2, pp. 22–3.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

194  Franziska van Buren own position that the world began in time. Most importantly, the fact that Bonaventure considers Aristotle’s own position not to be clear makes Aristotle quite emphatically not Bonaventure’s opponent in this debate. At best, he is an ally; at worst, neutral.16 How then are we to react when presented with two contradictory statements, one made under the influence of being politically correct, and the other made free from such an influence? There seems to be a few possible options. The first is that Bonaventure simply changed his mind at some point between writing the Commentary and the Collationes—in which case we can amend our thesis that Bonaventure is an Aristotelian to the claim that the “early” Bonaventure is an Aristotelian. Or, one could conclude that Bonaventure is simply lying about his feelings towards Aristotle in one text or the other, as van Steenberghen asserted of the Collationes. However, there is also a third option that is not as drastic as these other two. Here, I would like to highlight again the fact that in each of the critiques of Aristotle in the Collationes, Bonaventure is sure to work in a comment that this criticism is thrown at an interpretation of Aristotle, e.g. that of the Greek Doctors or of the Arabs. Thus, the contradiction between the Commentary and the Collationes is perhaps not as stark as it appears. Rather than being critical of Aristotle, Bonaventure presents himself as critical of certain readings of Aristotle and rather unenthusiastic about defending Aristotle from such readings—a lack of enthusiasm that one might find understandable considering also that, as the head of the Franciscan order, Bonaventure, unlike his contemporary Thomas Aquinas, had to maintain a certain political figura. Thus, it is fair to conclude that these apparent critiques of Aristotle are in fact critiques of only certain interpretations of Aristotle, and they neither indicate a hostility towards Aristotle nor stand in contradiction to his positive attitude in the Commentary, where he was freer to express his enthusiasm for Aristotle.

16 van Steenberghen takes a different position here regarding how Bonaventure viewed Aristotle, maintaining that Aristotle only thinks the world is eternal because it would not have occurred to him not to think so. See Fernand van Steenberghen, “Le mythe d’un monde éternel: note complémentaire,” Revue philosophique de Louvain 80 (1982), 497.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  195 Returning now to the reception of van Steenberghen’s original claim about Bonaventure’s Aristotelianism, which resulted in a moderation of the original opposing positions of van Steenberghen and Gilson, we find ourselves today having arrived at the “standard” position that Bonaventure is not entirely hostile to Aristotle, but also that he is not an Aristotelian fundamentally. However, despite the many disagreements regarding Bonaventure’s notion of forms, and indeed his philosophy generally, we find one point of commonality among all scholars of Bonaventure’s thought, from Gilson to today: for Bonaventure, forms possess esse. As I have indicated in the introduction, I do not dispute this point, insofar as this position is repeated by Bonaventure in a number of places throughout his corpus, which we will discuss in the following section.17 But what does it mean for Bonaventure to say that forms have esse? The answer to this question is usually to explain this statement as though it came from the mouth of a Neoplatonist—if this point is treated at much length at all. If forms have being that is ontologically prior to  ­sensibles, does this not also mean that Bonaventure’s forms are transcendent à la Plato? Are these forms then hypostasized like Platonic ideas? Are they exemplars? Or are they something else? To answer these questions, I will focus on two key points that have not yet been fully addressed in Bonaventure’s thought. The first is that, although the forms, for Bonaventure, are ontologically prior to sensibles, they quite clearly are not transcendent, as they are often understood to be; for Bonaventure, forms are neither exemplars in the mind of God nor are they hypostasized as in a Plotinian Intellect. Here, recalling the point I made in the introduction, for Bonaventure the claim that forms are ontologically prior to sensibles and have “being” in themselves is a pos­ ition shared by both Plato and Aristotle. For Bonaventure, the position that makes his own account “Aristotelian” is his emphatic assertion that these forms do not exist separate from sensibles—neither in a separate 17  See for example the discussion of spiritual matter: In Sent. II, d. 3, p. 1, a. 1, q. 1, p. 90. Similarly, when Bonaventure responds to the question of whether the image of God is greater in man than in woman, his answer—thankfully a “no”—is that “they are produced from the image with regard to their primum esse,” i.e. the form of the soul (In Sent. II, d. 16, a. 2, q. 2, p. 403).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

196  Franziska van Buren “world” nor in the mind of God. Rather, forms exist only in the natural world, as they do for Aristotle. However, and this is indeed the difficulty of understanding Bonaventure, these forms nonetheless cannot be conceived as dependent on their composition in sensible things—as he makes clear in a number of arguments that he also derives from Aristotle. A second point that has been neglected in Bonaventure’s thought, but is crucial to his understanding of forms, concerns the role of the seminal reasons and how they are understood as distinct from the universal forms, as well as how they allow Bonaventure to place the forms qua esse into his wider ontology. While it is clear enough that Bonaventure adopts the Augustinian term as well as the use of seminal reasons to explain occurrences such as new species coming to be after the seven days of creation,18 I wish to highlight another aspect of seminal reasons: they take on the role of the form in the sensible composite, i.e. the form as present in, ontologically dependent on, and particularized in the composite qua operative principle, which signifies the potency from which the composite is generated—this in contrast to the universal form in itself which maintains its status as ontologically prior. What is particularly important here is the way in which Bonaventure understands the relationship between the seminal reasons and the universal form, i.e. in a manner that indicates neither a conceptual nor a real distinction.19 On the one hand, because this distinction is not real, Bonaventure avoids separating the universal form from the composite; on the other hand, because the distinction is not conceptual, he avoids equating the universal with the seminal reason, which would render it ontologically dependent and thereby particularized, i.e. would make it no universal at all.

18  However, he attributes such an account of seminal reasons to Aristotle just as much as to Augustine (In Sent. II, d. 15, a. 1, q. 1. p. 374). He cites ad Orosium (Quaest. 65, q. 37) and On the Generation of Animals II.3 [736a25–737b10]. 19  Boehner considers Alexander of Hales’s distinction between quo est and quod est to be a quasi-­formal distinction. Bonaventure similarly uses quo est and quod est—which Boehner himself does not mention. See The History of the Franciscan School: Alexander of Hales (St. Bonaventure, NY: Franciscan Institute Publications, 1943), 64, as well as Alexander of Hales, SH II, n. 60, p. 75.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  197

2.  Forms: Universal or Individual? In order to see where Bonaventure stands on the question of the universality of forms, let us turn to his discussion of angels, where he first of all argues that angels cannot be pure form because form is pure actuality or pure being. Angels, however, are not pure act, thus they must have a component of potentiality, i.e. spiritual matter.20 There is, however, a perhaps more important reason why angels cannot be forms: angels cannot be forms because they are not universal, whereas forms are. In this discussion, Bonaventure lists a number of different kinds of compos­itions which angels have, e.g. quo est and quod est or quis est and quod est. It is not necessary to summarize them here, but what they all indicate is that angels are particular things. It would be absurd to say of something universal, namely a form, that it has, e.g., a composition of quo est and quod est—since “what” a form is does not differ from the “by what” it is. Bonaventure supports this point by turning to de Caelo where Aristotle makes a distinction between the form itself (i.e. the universal) and the compound of form and the matter (i.e. the individual thing)—a distinction that we see Bonaventure quite frequently shorthand by quoting the line, as he does here: “when I say heaven I mean the form, and when I say this heaven I mean the matter.”21 Again, using Aristotle to substantiate what it means for the form to be universal, he cites another preferred text, the Posterior Analytics, that “the universal form is by nature ‘always and everywhere’ ”—angels, by contrast, are not always and everywhere, but here and now.22 This indicates that we must have some principle that distinguishes the individual angel from the universal form, and for Bonaventure this is matter or, more precisely in the case of angels, spiritual matter. The only way to get around the position that angels must have matter is to assert that forms are not universal species, and then angels could each be one “individual form” (and thereby, as Aquinas maintains, could be said to be composed simply of act and potency but not of form and matter). However, as we will see 20  In Sent. II, d. 3, p. 1, a. 1, q. 1, p. 90. 21  In Sent. II, d. 3, p. 1, a. 1, q. 1, p. 90; de Cael. I.9 [277b27–278a26]. 22  In Sent. II, d. 3, p. 1, a. 1, q. 1, p. 90; Post. Anal. I.31 [87b29–33].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

198  Franziska van Buren Bonaventure argue momentarily (again using Aristotle), an individual or particular form is a contradiction in terms: forms insofar as they are intelligible must be universals. Given now that Bonaventure is clear in his statement that forms not only have being (esse), but also are considered strictly as universals, a question arises concerning the existence of sensible things relative to these forms. In a question considering the hylomorphic composition of creatures, Bonaventure writes: “Metaphysics considers the nature of all creatures and the highest substance being in itself (per se entis), in which is considered the act of being (actus essendi), and this is what the form gives [to composite substances].”23 In contrast, the matter gives “existence (existere)” and “stability for things existing (per se existendi).”24 On the one hand, the form gives, and therefore presumably itself has, the act of being (actus existendi) that it provides for the composite thing, i.e. makes it to be (esse) what it is, makes it intelligible as one kind of thing, e.g. a horse or a dog. However, there is a kind of being that the form cannot provide to the composite—being that is not intelligible and that is not “always and everywhere,” but rather is a kind of “existential” stability—a stability that endures through the many “heres and nows” in the life of a creature existing in space and time. This is what the matter provides: the existential stability of being the substratum for generation and change as this composite leads out its spatio-­temporal existence.25 This is to say, the being (esse) that the form provides is not quite the same as the being (existere) that the matter provides—which perhaps is why Bonaventure uses two different terms loosely to indicate the two different ways of being. Thus far, we know that the forms are universals, have being in themselves, and that in some way their being is distinct from the being of matter, as well as from the being of composite things that partake in both matter and form. But this advances us little in substantiating my claim that Bonaventure’s forms are not “separate” from sensible things— in fact, none of this eliminates the possibility that Bonaventure 23  In Sent. II, d. 3, p. 1, a. 1, q. 2, p. 96. 24  In Sent. II, d. 3, p. 1, a. 1, q. 2, p. 96. 25  Accordingly, Bonaventure writes, extending this concept to include both kinds of matter: “for, just as the matter of corporeal things sustains and gives to the forms existence (existere) and subsistence (subsistere), so also does spiritual matter” (In Sent. II, d. 1. p. 1. a. 2, q. 2, p. 29).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  199 will  endorse some kind of Platonic separateness of these forms. Yet, Bonaventure is well aware of the problems that arise in positing forms that are separate, either in the manner of Plato’s transcendent forms or in the manner of the Neoplatonic hypostasis of the Intellect. Indeed, he argues against both the notion of transcendent forms in general,26 as well as against the Neoplatonic theory of emanation.27 Here, to aid in his critique of the Platonic positions, Bonaventure draws on the discussion of Platonic forms in Aristotle’s Metaphysics. Bonaventure first targets the obvious implication involved in the Platonic understanding of participation, i.e. “it seems absurd to posit a third man . . . .”28 Bonaventure then turns to the perhaps more important issue—again drawing on Aristotle’s critique of Platonic forms—that if forms are separate, they do little to account for the existence, generation, and motion of sensible things. Bonaventure first explains that matter is pure potentiality, for “matter, which in itself is imperfect, is never perfected except through the form,”29 i.e. matter never exists unless it receives (or is composed with) some principle of being, a form. If then the forms were always separate from matter, matter would never attain any level of perfection, i.e. it would never exist in any way. Or, put another way, were matter never to have form, it would never attain any kind of existence, and no material thing would ever exist. But material things do exist. Hence, Bonaventure comes to the same conclusion as Aristotle that it is absurd to say that forms could ever be separate from matter, or as Aristotle puts it in the form of a question: “how can the Ideas, if they are the substances of things, exist in separation from them?”30 Returning to our Collationes issue, it is clear that the above assessment of Platonic forms is far from a ringing endorsement. Moreover, we find Bonaventure expressing doubt about Augustine’s claim that Platonic

26  In Sent. II, d. 1, p. 1, a. 1, q. 1, p. 16. It is interesting to note that this explicit dismissal of the Platonic position takes place within Bonaventure’s wider discussion of the position that the world was created from nothing—a position which he sees evidence for in Aristotle, citing Meteor. II.3 [354b1–356b1], and “the beginning of de Caelo.” Bonaventure adds that there are so “many other places” where Aristotle indicates that the world was in fact made that he does not need to provide the references himself. 27  In Sent. II, d. 1. p. 1. a. 2, q. 2, p. 29. 28  In Sent. II, d. 1, p. 1, a. 1, q. 1, p. 16. 29  In Sent. II, d. 1, p. 1, a. 1, q. 1, p. 16. 30  Meta. I.9 [991b3–5].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

200  Franziska van Buren forms are like divine ideas.31 Indeed, in a different question, Bonaventure quite explicitly rejects the equation of forms with divine ideas insofar as forms are themselves (a) a plurality and (b) each intelligible, and so positing forms, such as Plato’s, in the mind of God would imply both plurality and intelligibility in God.32 Given that Bonaventure carefully spells out the problems with Platonic forms, it is clear that his intention is not to take the route of positing separate or transcendent forms, despite the fact that he wants the forms to be universal and ontologically prior. But if he wants the form to be immanent in the composite, as the forms of Aristotle are, we arrive at our central paradox: how can the form be in the composite without sacrificing its universality and ontological priority? His own answer to this question is based on a distinction that he draws out of de Caelo between “the form” and “the form in the matter”—or, to use Bonaventure’s alternative terminology, “the form of the whole” and “the form of the part.”33 The former, for Bonaventure, is the universal form that is ontologically prior to the composite, and properly speaking this is what he means by the form. The latter, which refers to the particularized “form,” is rather dependent for its existence on its own composition within a certain hylomorphic composite—and, for Bonaventure, this is indeed not itself a form at all, but what he rather calls a seminal reason or, alternatively, the natural form or the singular form.34 However, before we answer this first question, we need to ask the more foundational question—why does Bonaventure think it is actually impossible to posit the universal form in the sensible? Why does he dismiss the position that there is just the one form, immanent and inseparable from a hylo­morph­ic composite, which accounts for the existence and intelligibility of the sensible thing?35 Quite useful for our purposes, this very question is asked in the Commentary on the Sentences: “are the seminal reasons universals?”

31  In Sent. II, d. 1, p. 1, a. 1, q. 1, p. 16. 32  In Sent. II, d. 1. p. 1. a. 2, q. 2, p. 29. 33  de Cael. I.9 [277b27–278a26]. 34  In Sent. II, d. 18, a. 1, q. 2, p. 435. “. . . seminal reasons are singular forms.” In Sent. II, d. 18, a. 1, q. 3, p. 439. “. . . therefore this white (haec albedo) signifies the singular form . . . .” 35  This is precisely the distinction which, as I noted above, many scholars miss, e.g. Boehner, History of Franciscans: Bonaventure.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  201 And Bonaventure’s answer to the question is “yes and no”—loosely “yes,” but more properly “no.”36 In his respondeo, he first takes a step back and clarifies his terms: “the seminal reason is an active power, inserted into the matter . . . and this active power is the essence of the form.”37 The seminal reason, however, is “incomplete,” i.e. esse in potentia, and thereby dependent on the sensible things in order to be actualized, while the universal form is “complete,” i.e. esse in actu and, in this sense, independent of the sensible thing.38 It is important to note that by “in act,” it is clear that Bonaventure means that the universal form is not only independent of this particular composite but from any composition; e.g. even though particular dodo birds no longer exist (existere), the form of the dodo bird still does exist (esse). While Bonaventure is not aware of species going out of existence, he is aware of a phenomenon which happens in the opposite direction: new species appearing after the world was created. He explains such a phenomenon by saying, “all things were made to be (esse) at once, but were not all made at once,” i.e. all things were (esse) but did not yet happen to be made into existing (existere) sensible particulars. Thus, forms exist (esse) regardless of whether or not they have spatio-­temporal existence (existere).39 Applying such a principle to our dodo bird: what has happened with the form of the dodo bird is that even though it does not exist (existere) anymore, it still is (esse)—it always is, and always is everywhere. In a certain sense, that the dodo bird still is (esse) is evidenced by the fact that I can still know what a dodo bird is, i.e. we still have a concept of “dodo bird,” and so it must still be in order for it to be the object of knowledge—and indeed it exists (existere) whenever someone thinks about it, or describes it, draws it in a book, or even recognizes it from fossil remains. Thus, Bonaventure asserts that the status 36  Here in this “yes” and “no” we already begin to see an indication of a distinction which is neither real nor conceptual, as I mentioned above. 37  In Sent. II, d. 18, a. 1, q. 3, p. 439. 38  In Sent. II, d. 18, a. 1, q. 3, p. 439. “. . . the seminal reason is the essence of the productive form, differing from it according to the distinction between being incomplete and complete, i.e. according to being in act and being in potency.” Also interesting to note is that Bettoni attributes this definition of seminal reasons as “incomplete” to Gilson’s interpretation of Bonaventure instead of to Bonaventure himself. See Efrem Bettoni, S. Bonaventura: gli aspetti filosofici del suo pensiero (Milan: Biblioteca Francescana Provinciale, 1973), 147. 39  In Sent. II, d. 12, a. 1, q. 2, p. 296.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

202  Franziska van Buren of the forms as actuality and being is not taken away from them by the fact that they enter into composition with matter: “matter does not remove actuality from the form . . .,” i.e. the fact that the form is instantiated by matter has nothing to do with the actuality of the form considered in itself.40 Bonaventure then turns more explicitly to the question of why there needs to be a distinction between the form which is in the sensible thing as a potency, i.e. the seminal reason, and the universal form, which exists as an actuality. Why can we not simply say, contrary to the above, that the universal form and the seminal reason are identical—that the universal form is in and dependent on the sensible thing—and so when I know Socrates, I can know the universal form, humanity? Bonaventure indeed takes this option as his starting point: “For there are some who want to say that, given that the universals are not made up (fictiones), they really and according to truth exist not only in the mind but also in nature; and since all things that are in nature have been inserted into matter, so universal forms, and singulars, have being (esse) in matter.”41 This is clearly a realist position, in the sense that it does not deny being to the universals, but rather indicates that they exist and have being only insofar as they are singular, i.e. are in matter—or, put another way, that humanity exists whenever there is a man. The form of humanity, according to this position, would exist in both Socrates and Callias, although only as a singular, and this “singular” form is sufficient for us to know the universal.42 This brand of realism seems either not to make a clear distinction between a singular form and a universal form, or explicitly clarifies that only the singular form exists external to the mind, whereas the universal exists only in the mind, abstracted. Bonaventure admits that this account seems prima facie to make the most sense, insofar as we usually think about things in this way: “ ‘­singular’ indicates being in act and ‘matter’ being in potency, and the universal 40  In Sent. II, d. 13, a. 2, q. 1, p. 317. 41  In Sent. II, d. 18, a. 1, q. 3, p. 439. 42  This is not dissimilar to Scotus’s position that the forms only exist when they are contracted into individuals. Interestingly, Bonaventure himself makes use of the term contraction: “But composite creatures are not really simple, since they have in their being a mixture of ac­tual­ity and potency, and likewise in species and genus through a contracted addition (per additionem contractum) . . .” (In Sent. II, d. 18, a. 1, q. 3, p. 439). Bonaventure’s student, John Peckham, also makes use of a notion of contraction. See Summa de Esse et Essentia 7.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  203 form means in one sense being in actuality [i.e. when the composite is actualized] and in another sense being in potency [i.e. when the composite is not actualized].”43 Indeed, this brand of realism sounds very much like a certain reading of Aristotle,44 which Bonaventure himself notes, saying that it finds support not only in the Commentator45 but also in Aristotle’s Physics, where we find Aristotle provide an account of our experience of sensible reality in which we move from a particular sensible to knowledge of a universal.46 However, Bonaventure sees two main problems with the above pos­ition—problems that, we should note, will also be pointed out by Ockham.47 Bonaventure targets these problems in a two-­horned critique of the above—what one might want to call—quasi-­realist, or perhaps proto-­conceptualist, position. The first horn targets the position that considers the universal form to be in a sensible thing in the sense of being ontologically dependent upon the sensible thing, by showing that this position is self-­contradictory—the universal cannot be in a particular, insofar as this would render the universal form itself particular and therefore not universal. For the second horn of his critique, Bonaventure targets the position that tries to ground knowledge of a universal form in the sense perception of a particularized form. Here, he argues that a particularized form is not sufficient to ground (1) any kind of human knowledge (even of the particular sensible thing), or (2) univocal predication. Indeed, Bonaventure’s arguments are quite similar to Ockham’s arguments that individualized forms are not sufficient to ground our knowledge of a universal. But, of course, while Ockham uses these arguments as a platform for turning to nominalism, Bonaventure goes on to defend a realism that will withstand his own critiques. Looking now to the first horn, how does Bonaventure develop his argument that the universal form cannot ever really be in a sensible particular? Here, he refers the reader back to his responses to the opposing

43  In Sent. II, d. 18, a. 1, q. 3, p. 439. 44  I.e., Aquinas’s reading of Aristotle. 45  Metaph. cap. quoniam autem in fundamento. 46  In Sent. II, d. 18, a. 1, q. 3, p. 439; Phys. I.1 [184a10–184b14]. 47 Ockham similarly uses Aristotle, although he quotes instead the Metaphysics—like Bonaventure, Ockham argues against the existence of universals as being immanent in sensible things (Ordinatio 1.2.7).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

204  Franziska van Buren positions where he argues, on the basis of Aristotle’s texts, for the impossibility of a universal form (qua universal) being “in” a particular. Bonaventure provides five arguments, all of which aim to demonstrate that one cannot place the universal form into a particular composite as dependent on this composite without rendering this universal particular— thereby negating the very universality of the form. Here, we will examine only the fourth and fifth arguments in depth insofar as they are the strongest and most elaborate. It is to be noted that the premises from the first three arguments are all taken from Aristotle— the first from de Anima,48 the second from Posterior Analytics,49 and the third from De Interpretatione.50 These three arguments aim to show: (1)  the universal cannot be in the sensible insofar as the universal— precisely because it is actual being—cannot be the potency from which, e.g., a baby becomes a man; (2)  rather, it is the seminal reason which is in the sensible thing as the thing’s own potency, e.g. Socrates’s own particular potency to become a man, and therefore it is not—as the universal is (according to Aristotle)—“always and everywhere,” e.g. in Socrates in Athens in 430 bc and in Cicero in Rome in 60 bc; (3)  in a similar vein, if the universal is applicable to the many (i.e. predicated of many), it cannot belong to one particular—e.g. we do not predicate Socrates’s humanity of Socrates, but the humanity common to Socrates and Callias. Bonaventure’s fourth argument (and fifth as well) demonstrates the absurdity of claiming that the whole of the universal is present in 48  “The Philosopher writes in de Anima: ‘the universal is either nothing or it is posterior.’ ” In Sent. II, d. 18, a. 1, q. 3, p. 439; de An. I.1 [402b7]. Here, “posteriority” is clearly meant with reference to temporal posteriority, i.e. a man is born with the potency to become a man, not with the actuality of humanity. John Peckham takes this position, almost word for word, from Bonaventure (along with the quote from Aristotle), and develops the same interpretation. See Summa de esse et essentia 7. 49  “ ‘The universal is always and everywhere’; but the seminal reason is with respect to this matter, in which it is made determinately: therefore, the seminal reason cannot be the universal form” (In Sent. II, d. 18, a. 1, q. 3, p. 439; Post. Anal. I.31 [87b29–33]). 50  “The seminal reason is not predicated of that of which it is a seminal reason . . . but ‘the universal is predicated of singulars’: therefore, the seminal reason cannot be a universal” (In Sent. II, d. 18, a. 1, q. 3, p. 439; de Int. VII [17a39–17b5]).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  205 the particular. We should note that Bonaventure takes the whole of the universal to mean not only the universal species, but also the universal genus—i.e. anything which corresponds to secondary substance in Aristotle’s Categories, the text that Bonaventure uses as his guide in these passages. Here, he is focusing on Aristotle’s prohibition in the Categories against a universal (i.e. a secondary substance—universal or specific) being present in a substrate. Bonaventure begins by taking the universal genus as his example: “if the seminal reason means the universal form, it is therefore either the form of the genus or the form of the species. If the form of the genus, then in man there is the seminal reason with respect to a donkey.”51 This is to say that, if some particular animal, e.g. a man, had the genus of animality as potency within itself, this particular man would have the potency to become any other animal, e.g. a donkey— which is obviously not the case. Bonaventure next considers the possibility that the universal form of the species (rather than the form of the genus) is present in the sensible composite: But this is the seminal reason of something (aliquid), which preexists in matter, before it may be the complete thing in act (res completa in actu): therefore, there is the form of humanity in matter, before there is the complete thing. But this is false and unintelligible, that the form of humanity be in a particular, and that it not be the complete thing. Therefore, one may not think that the universal form is the seminal reason.52

Bonaventure’s position here is that the universal means the complete thing in act—the form of humanity is the whole of humanity—and if the whole of humanity is in one man, this is unintelligible because one man is not the whole of humanity. Socrates is not humanity—he is a man. This seems moreover to be a clear reference to the Categories, where Aristotle gives the same examples: “man is said of the particular man as substrate, but is not in a substrate: man is not in the particular man.”53 51  In Sent. II, d. 18, a. 1, q. 3, p. 439. 53  Cat. V.5 [3a10–12].

52  In Sent. II, d. 18, a. 1, q. 3, p. 439.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

206  Franziska van Buren Likewise, Aristotle substantiates Bonaventure’s claim above that the universal genus cannot be in the particular things: “Similarly, animal also is said of the particular man as substrate, but animal is not in the ­particular man.”54 Bonaventure’s fifth argument also demonstrates the absurdity of positing a universal form in a particular, this time focusing on the attempt to posit the universal form as an operative principle in a particular, and using as his example the accident of whiteness being posited a universal present in the particular:55 If the universal form, with respect to the particular, were the seminal reason, then, while ‘this white’ (haec albedo) means the singular form, and ‘white’ (albedo) means the form of the species, and ‘color’ the form of the genus, and ‘sensible quality’ furthermore the more universal form, then nature, in producing this white (haec albedo), would proceed through the mediation of all of these predicates: therefore, quality would be made before corporeal quality, and corporeal quality before sensible quality, and sensible quality before color, which is not intelligible.56

Here, Bonaventure shows the absurdity of a position that would place a universal within a particular as a principle of operation or of generation. In order to arrive at the “this white,” which is indeed in this particular thing, there would also have to occur the generation of every universal 54  Cat. V.5 [3a13–15]. 55  It is admittedly unclear whether Bonaventure thinks that the seminal reason/universal form relationship should also include accidents or qualities, e.g. whiteness or smallness— indeed, as well the more general “color” or “sensible quality.” This is to ask whether Bonaventure thinks that there is a universal form of, e.g. not only “whiteness” but also “quality,” which would then have to have a correlate seminal reason within each particular, or if is he only speaking counterfactually, i.e. “if there were a form a whiteness . . . .” It seems to me more likely that he is speaking counterfactually because of his fondness for the Categories—in which secondary substances (i.e. universals) are understood as those predicates whose definitions are said of a substrate (i.e. both species and genus). However, the above argument would work by taking any universal one would like, regardless of whether or not it happens really to be a universal—it would work with “tallness” as well as “humanity,” i.e. one would have to generate “the whole of animal” before “the whole of man” before “this particular humanity” in this particular man. It seems moreover that Bonaventure chooses this example of whiteness because it is Aristotle’s example in the Categories (see Cat. V.1 [2a27–34]). 56  In Sent. II, d. 18, a. 1, q. 3, p. 439.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  207 form, from the most general to the most specific. This would entail first the generation of the universal form of sensible quality, then the universal form of color, then the universal form of white—all within one particular white thing. From Bonaventure’s arguments above, it is clear that he maintains the impossibility of the universal form being in (i.e. being ontologically dependent on) a particular thing, precisely because what is in a particular thing in such a manner must itself be particularized. Whatever is dependent upon Socrates belongs to him and is particularized in him. This must be the case with reference to the generation of the particular since the cause of generation must itself be particular, not universal—i.e. it must be Socrates’s humanity, in a proximate sense, which causes him to grow up to be a man, not humanity in general.57 Then again, with regard to knowledge or predication, what is in the particular sensible thing is itself particular, and thus it is no real predicate or object of knowledge. As Aristotle writes: “Sense-­perception must be concerned with particulars, whereas knowledge depends upon recognition of universals.”58 Thus, the humanity that exists in Socrates belongs to Socrates—it is not the humanity in Callias, it is not the universal, and thereby it is only perceived by the senses, not known by the intellect.59 Indeed, for Bonaventure, the quasi-­realist position cannot respond to the above objections and therefore must relinquish its realism and turn explicitly to a conceptuality, i.e. accept the position that the being of the universal qua universal—not qua singular instantiated in matter (in which case it is no universal at all)—exists only in the soul: “if one wishes to maintain this position [i.e. the quasi-­realist position], one would have to avoid reasons brought up to the opposition, saying, that

57 Of course, Bonaventure considers that humanity (i.e. the universal) is ultimately the cause of Socrates, but more proximately he considers it necessary that the incomplete seminal reason be present in Socrates as the principle from which he actually grows—i.e. it is the potentiality to become humanity which is more proximately the cause of him becoming a man, than the actuality towards which he is aiming, because, temporally speaking, he has potentiality before actuality. 58 Aristotle, Post. Anal. I.31 [87b29–33]. Bonaventure cites the lines just preceding this passage, but not this passage itself. 59  This is very similar not only to the problem brought out in the Categories, but to Plato’s sail problem in the Parmenides.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

208  Franziska van Buren he is speaking of the universal as it is abstracted in the soul.”60 Thus, the universal would exist only in the mind—and such universals are, as Bonaventure puts it, quite close to being “fictiones.” Or, to make the situation sound less dire, we could say that this position, insofar as it does not provide a response to Bonaventure’s objections, would have to abandon its strong realism in favor of a conceptualism. And, indeed, this is still an option on the table insofar as none of the above shows that the extra-­mental particularized form in Socrates or Callias would be insufficient to allow the mind to abstract the universal form, humanity—i.e. that there is not an ontological grounding for the universal in the mind, namely this particularized form. This brings us to the second horn of Bonaventure’s wider argument, which targets not only the implicitly conceptualist position, but also the position of the “self-­aware” conceptualist. Here, Bonaventure’s argument against particularized forms grounding knowledge of universals is itself two-­fold: “it is necessary to posit universal forms for the sake of (1) cognition and (2) univocal predication.”61 These arguments show that a universal is necessary not only for knowing the particular, i.e. by connecting the particular with a universal, but also for being able to know that different particulars are of the same kind.62 For his first argument, Bonaventure begins by stating that, when we know, we know the universal: “it is not complete cognition if the whole being of the thing (totum esse rei) is not cognized; and it is not cognition unless it is through the form.”63 This position is taken quite clearly from Aristotle, whom he quotes here: “Sense-­perception must be concerned with particulars, whereas knowledge depends upon recognition of universals.”64 Put another way, we only know what a thing is when we know its whole essence, i.e. the universal form, which in some way must be “in” the many particulars of which we predicate it in order to justify 60  In Sent. II, d. 18, a. 1, q. 3, p. 439. 61  In Sent. II, d. 18, a. 1, q. 3, p. 439. 62  One could say that whereas Ockham does not consider a form which has esse and is onto­ logic­al­ly independent from sensible things and which is also not transcendent à la Platonic forms, Bonaventure does just that. 63  In Sent. II, d. 18, a. 1, q. 3, p. 439. 64  Post. Anal. I.31 [87b29–33]. This point is found in many places in Aristotle. The Quaracchi editors cite de An. II.2 and III.8.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  209 such a predication. We do not know what something is when we know only the essence which is in the one particular, i.e. particularized essence or seminal reason; rather we know what something is when we know the essence which is in many particulars, without itself ever being particularized: “the thing abstracted from matter [i.e. the universal] itself may very well be in other things, which each have their own matter and form [i.e. seminal reason], just as the similitude of color in a mirror.”65 The fact that cognition is dependent upon the universal, which is in the many, is made evident by the fact that one knows what a particular is not by knowing particular thing alone, but only by knowing the particular thing’s universal form. With respect to cognition, we seem, at least to ourselves, to be working from the particular to the universal simply because we encounter the particular first—with the senses.66 However, sense perception of the particular does not constitute knowledge. We do not know what the particular is before we have connected it to the universal, and so it is the universal that we know first, despite the fact that we immediately connect it to the particular, i.e. name the universal of the particular. For example, it is necessary to know what humanity is more primarily than to know what Socrates is, because one can know what Socrates is, i.e. a man, only once one knows what humanity is. Thus, we cannot say that we know the universal from the particular.67 Rather, quite the opposite is the case: knowledge of the particular has to be grounded in knowledge of the universal, i.e. in the connecting of the universal with the particular. Thus, Bonaventure concludes that if we want to have cognition of anything (particular or universal) that is grounded extra-­mentally, “it is necessary that there be some form that embraces the whole esse; but this we call the essence and this is the universal form.”68

65  In Sent. II, d. 18, a. 1, q. 2, p. 414. 66  Aristotle makes this point in Phys. I.1 [184a10–184b14]. While the particular is what we encounter first, what we know and what is prior is the universal. 67  Ockham would agree here, and—like Bonaventure—takes this argument to indicate that the conceptualist position is absurd. However, of course, Ockham—in contrast to Bonaventure— does not take this point to indicate that we have to attribute a stronger sense of being to universal forms. 68  In Sent. II, d. 18, a. 1, q. 3, p. 439.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

210  Franziska van Buren We now come to Bonaventure’s second line of argument: without the universal, it is impossible to make univocal predications between ­particulars. While the first argument concerns the issue of knowing the particular and the universal, this second argument concerns our ability to identify two particulars as being of the same kind, i.e. predicating of two particulars the same quality or essence, where this quality or essence means the same thing for both particulars. Bonaventure writes: “similarly, there is no true univocation except when something (aliquid) is really assimilated to a common form, which is then essentially predicated of it (i.e. of the aliquid).”69 He continues: “but this form, to which many assimilate, can only be the universal form; for what is essentially predicated of them can only be the form that embraces them all (forma totum complectens).”70 This is to say, if we have access only to the particular humanity in Socrates and the particular humanity in Callias, it is impossible to say that they are both humans. Even if one could know Socrates’s humanity without knowing first the universal (which as we saw above was not possible either), one still would be knowing a particularized humanity, i.e. the humanity that I would abstract from Socrates would not be applicable to Callias, but only to Socrates. Moreover, when we also take into consideration the fact that the form is universal and “being in act,” it seems plainly absurd to say that what Callias and Socrates assimilate to is already complete (completus), i.e. is already the totum esse rei, in Callias and Socrates. In particular, if Callias and Socrates are each assimilating to one and the same thing, how could it already be particularized in each of them? Bonaventure’s conclusions thus far are (1) that universals as universals cannot exist as ontologically dependent upon sensible particulars, i.e. that one cannot simply identify the immanent particularized form (i.e. seminal reason) with the universal form, and (2) that this particularized form is not sufficient to ground knowledge of the universal. Given that Bonaventure does not want to posit separate forms, as the Platonists do, he is left with two options: (1) to assert that the universal neither exists in nor is ontologically grounded in particulars, but is just a

69  In Sent. II, d. 18, a. 1, q. 3, p. 439.

70  In Sent. II, d. 18, a. 1, q. 3, p. 439.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  211 convention or a name we attribute to things, or (2) to find a way of affirming the existence of the universal without positing it as onto­logic­ al­ly dependent on the particular, or as something that exists only in the mind. Naturally, wanting to defend a realist account, Bonaventure decides on the second option. Here, it is quite clear that the main ­difficulty Bonaventure has to face is the apparent contradiction, or paradox, between the two claims to which Bonaventure has committed himself: (1) universals qua universals cannot exist “in” particular sensible things, but (2) they also cannot exist separately from them, as the forms of Plato do—both of these claims based in his reading of Aristotle. In  order to see how Bonaventure resolves this paradox, we need now  to  examine the relationship between the universal form and the ­seminal reason.

3.  The Relationship between Seminal Reasons and Universal Forms The preliminary discussion of angels, earlier, showed Bonaventure speaking of forms as universals. In a different question he also explicitly denies that there can be anything like a particularized form that is a form in the true sense of the word—i.e. while the preceding section showed that the seminal reason and the universal are not identical, Bonaventure also explicitly denies that we can conceive of seminal ­reasons as being a second “kind” of form in addition to universal forms. We find this argument in Bonaventure’s account of individuation, in which he objects to a position quite similar to that of Duns Scotus, namely, one that would posit a forma individualis as the principle of individuation. Bonaventure’s reasoning for dismissing this individual form as the cause of individuation is that an individual form is quite clearly no form at all—such would be a contradiction in terms.71 Were it a form, the individual form would be included in the universal def­in­ ition; for example, Socrates’s humanity would be part of the definition of humanity. Bonaventure bases this position on an example given in the 71  In Sent. II, d. 3. p. 1, a. 2, q. 3, p. 109.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

212  Franziska van Buren Metaphysics, where Aristotle entertains the idea that the definition of the sun or the moon may be individual because there is only one individual sun and one individual moon. Aristotle’s conclusion is that even though there is only one (individual) sun or one (individual) moon, this does not mean that when we define the sun or the moon, the definition itself is individual. These definitions are still universal and potentially applicable to many suns and many moons. It just happens to be the case that there is only one individual instance of the universal.72 The form is the definition of the thing, while the individualized form, e.g. Socrates’s humanity or this sun’s sun-­ness, is not part of the definition.73 Therefore, the individual form is no form in the strict sense of the word. If then the seminal reason and the universal form cannot be distinguished as two different kinds of forms, what is the nature of the distinction between the two? In answering this question, we will also clarify why Bonaventure responded to the original question above of whether the universal is the seminal reason with a “yes” and a “no.” Bonaventure writes: “if the universal form is said properly, according to the thing that is ordered into a genus, which metaphysics considers, the seminal reason is not the universal form. If, however, the universal form is said to be existing (existens) according to an incomplete being in matter and indifferent and able to be produced in many, so may one call the seminal reason the universal form.”74 This is to say, they can be equated only insofar as the universal form is “existing (existere) in potency” in some composite.75 The crux of the distinction brings us back to our original esse/existere distinction. The universal form really exists (esse), while the seminal reason has merely a contingent existence (existere). Or put another way, insofar as the universal form exists (esse) it is distinct from (or better, “indifferent to”) the fact that it happens to exist (existere) as operative in some composite. However, this does not mean that we have 72  The Quaracchi edition gives the reference as Meta. VI.15, which must be a typo. The correct citation is VII.15. The editors are, however, right to cross-­reference de Cael. I.9, a chapter which we have seen Bonaventure cite very frequently. 73  For a study of a similar reception of this passage in Plotinus and Porphyry, see Peter Adamson, “One of a Kind: Plotinus and Porphyry on Unique Instantiation,” in R. Chiaradonna and G.  Galluzzo (eds.), Universals in Ancient Philosophy (Pisa: Edizioni della Normale, 2013), 329–52. 74  In Sent. II, d. 18, a. 1, q. 3, p. 439. 75  In Sent. II, d. 18, a. 1, q. 3, p. 439.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  213 two entities, nor again that we have something which is merely conceptually distinct. The distinction is correlate neither to the ontological distinction between “Rye the horse” and “Alejandro the horse,” nor the conceptual distinction between calling Rye “the horse from Germany” or “the horse who won the competition.” We rather have a distinction which is neither real nor conceptual, i.e. the distinction between Rye’s equinity and equinity in general. Indeed, it is clear, although he does give a particular name for this distinction, e.g. a “formal distinction,” that Bonaventure must intend a distinction which is neither real nor conceptual. Otherwise, he would be contradicting himself with regard either to (1) his position that the universal cannot be in, or inhere in, the particular in any way that would render it ontologically dependent, or to (2) his position that the universal cannot be separate. If the distinction were real, the universal would be separate from the particular; if the distinction were conceptual, the universal would be dependent on the particular.76 It is important to mention here that Bonaventure has a well-­developed account of what a conceptual and a real distinction are. He indicates a conceptual distinction with the phrase, secundum rationem (e.g. he applies a conceptual distinction to God as final end and God as beginning)—a phrase that is emphatically absent from Bonaventure’s discussion of the distinction between the seminal reason and the universal.77 Of course, on the other hand, if he had intended a real distinction, he would simply have answered the original question of whether the seminal reason is the universal form with a “no,” and not, as he does, with a “yes” and a “no.” We can also bring this above distinction to bear on Bonaventure’s understanding of the hylomorphic composite. Bonaventure’s well known doctrine of a universal hylomorphism is usually described as the position that all things are a composition of form and matter—but this is only correct very generally speaking, just as it is only correct generally 76  Bonaventure also draws a parallel between God’s relationship to his exemplar causes and the universal form’s relationship to the singular form in his Disputed Questions on the Knowledge of Christ (III con.). Just as the distinction between God and the exemplar causes is neither real nor conceptual, so is the distinction between the universal and the singular form neither real nor conceptual. 77  For a study of the different types of distinction in Bonaventure, see: Sandra Edwards, “St. Bonaventure on Distinctions,” Franciscan Studies 38 (1978), 194–212.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

214  Franziska van Buren speaking to say that the universal form and the seminal reason are one and the same. More properly speaking, it is the seminal reason, not the universal, which is compounded with matter—and the seminal reason is thereby individualized by its dependence on such a composition, while the universal retains its independence, being formally distinct from the seminal reason. This distinction between the universal and the seminal reason provides now a more nuanced understanding of hylomorphism in Bonaventure’s thought. Bonaventure’s hylomorphism is not simply the form and the matter, but the seminal reason and matter—while the universal form is the third thing, not separate from the matter, insofar as it is not ontologically distinct from the seminal reason, but not particularized by it either, insofar as it is also not identified with the sem­ inal reason. Turning now back to the texts, we see that Bonaventure draws the two-­fold understanding of form as (1) the universal and (2) the seminal reason (the latter of which is compounded with the matter), out of the texts of Aristotle: For this position [i.e. the preceding account of universals and their relationship to seminal reasons] agrees with authority. For the Philosopher says, “when I say heaven, I mean the form; when I say this heaven, I mean the matter”: therefore, the individual does not add form beyond the universal, but only adds matter. For Boethius says “that the species is the whole being of the individual”: therefore, the universal form, which is the species, is the form of the whole, which embraces complete being, and which is the sufficient grounding of knowledge (ratio cognoscendi) with regard to substantial being.78

It is unclear what Boethian text Bonaventure has in mind, but the passage from Aristotle is one that we have seen him often utilize, and indeed with good reason, insofar as it expresses the two-­fold understanding of forms: the form itself and the form in the matter (or here, we could say qua operative principle in the matter). If we look at the full

78  In Sent. II, d. 18, a. 1, q. 3, p. 439.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  215 passage in Aristotle from which Bonaventure is quoting, we see even more clearly what Bonaventure has in mind: Suppose for instance only one example of a circle were apprehended, the distinction would nonetheless remain between (1) the essential nature of the circle (τὸ κύκλος) and (2) the essential nature of this particular circle (τῶδε τῶ κύκλω). The former is simply the form (εἶδος), and the latter is the form-­in-­matter (τὸ δ’εἶδος ἐν τῆ ὓλη) and must be counted among the particulars (καθ’ἓκαστον).79

From this Aristotle concludes: “This heaven and heaven in general are therefore two different things, the latter being distinguishable as form or shape and the former as something compounded with matter.”80 Prima facie, one would think that here by “this heaven,” Aristotle means the composite of form and matter—the individual sensible hylomorphic composite. But this is not precisely what he says. He rather says that “this heaven” is that which is compounded with matter, not the resulting compound of form and matter. This is to say, “this heaven” is one half of our composite sensible thing, with matter, of course, being the other— and indeed “this heaven” is not the universal form but “τὸ δ’εἶδος ἐν τῆ ὓλη” or, as Bonaventure would take it, the particularized form or sem­ inal reason.81 Continuing in a similar vein, Aristotle writes: “In all formations and products of nature and art alike a distinction can be drawn between the shape in and by itself and the shape as it is combined with the matter.”82 This dual way of thinking about a (certainly not separate) form in Aristotle’s thought seems to be what Bonaventure is drawing on here. “Essential nature of the circle” would be the form in itself and “essential nature of this circle” would be the form considered as existing in this particular. Moreover, what Aristotle is calling the “essential nature of the circle” and the “essential nature of this circle” also correspond in Bonaventure

79  de Cael. I.9 [278a7–11]. 80  de Cael. I.9 [278a13–15]. 81  Here, we see why he states that Augustine and Aristotle both agree when it comes to seminal reasons (In Sent. II, d. 15, a. 1, q. 1. p. 374). 82  de Cael. I.9 [277b30–5].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

216  Franziska van Buren to the notions of the forma totius and the forma partius.83 The universal is “the form of the whole (forma totius),” in contrast to “the form of the part (forma partius),” which is rather the seminal reason.84 Bonaventure gives an example of the difference between the forma totius and the forma partius: with respect to the forma totius, “soul is not said with respect to one man, but rather with respect to man.”85 Here, “soul said with respect to one man” is the particularized form (forma partius), while “soul said with respect to man” is the universal (forma totius). The former is applicable to only part of the set of members of this kind (i.e. only to one man), while the latter is applicable to the whole set of members (i.e. to every man). When one designates “soul” as the form of man, “soul” itself is known as something that is related to many men, or to man in general, i.e. to the whole. As we have already seen Bonaventure argue, we cannot even know a particular soul if we cannot connect it to a universal. This universal form is indeed the form “that gives being to all [i.e. the whole set of members of a species], and this is called the  essence of the thing (essentia rei), which embraces the complete being (esse).”86

4. Conclusion By way of conclusion, let us first of all relate Bonaventure’s account of forms back to his use of Aristotle, and make a few comments about the way in which Bonaventure has interpreted Aristotle. Here, it is probably also a good idea, to forestall any objection to the way in which Bonaventure has used Aristotle to defend his notion of forms, to stress that of course Bonaventure is assuming a certain reading of these texts. In particular, as we saw above, he assumes that these texts have an ontological and not only a logical import (particularly with regard to the 83  The “whole-­to-­part” relationship is said with respect to a one particular related to the whole set of particulars. This is to say the “whole” is the whole set of particulars and the “part” is any one particular member of the set. We then arrive at a distinction between the form of the whole (i.e. of the whole set), which is universal, and the form of the part (i.e. of one particular), which is particular. 84  In Sent. II, d. 18, a. 1, q. 3, p. 439. 85  In Sent. II, d. 18, a. 1, q. 3, p. 439. 86  In Sent. II, d. 18, a. 1, q. 3, p. 439.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  217 Posterior Analytics and the Categories). This, of course, has been questioned by contemporary scholars of Aristotle. Indeed, one could also object to Bonaventure’s reading of de Caelo by saying that the “essential nature of the circle” exists only in the mind, while the “essential nature of this circle” is the only thing that exists extra-­mentally. However, this objection, insofar as it would appeal to a more conceptualist interpretation of Aristotle, would have to respond to the many arguments that Bonaventure puts forth against conceptualism, himself relying on Aristotle’s texts. Indeed, Bonaventure’s interpretation of Aristotle certainly has its merits—particularly insofar as he attempts to respond to his own arguments against alternative realist and conceptualist positions, and in doing so to eliminate any contradiction from the understanding of form that Aristotle develops over his entire corpus. Given that, for Bonaventure, Aristotle wishes to posit the universals as existing extra-­ mentally, if Aristotle posits them as existing in sensibles, such that they are dependent on sensibles, he is contradicting his own position in the Categories where he says that universals cannot be “in” a substrate. Moreover, if Aristotle posits the forms as existing only as particularized in sensible composites, i.e. as a singular or individual form, then his account is incoherent insofar as such forms would not provide us with knowledge of the universal. Thus, if we want to read Aristotle as a realist, Bonaventure provides us with a solid interpretation, particularly because it anticipates and addresses the critiques that someone such as Ockham would raise. We can also highlight the main points that Bonaventure has taken up from Aristotle’s thought in developing his understanding of forms. The first of course is the position that although the form is universal and ontologically prior, it nevertheless cannot be separate. Or, put another way, the form must be conceived of as being in the composite, but not in any way that would deprive it of such universality or priority. Maintaining these two seemingly contradictory positions has indeed proven to be a fine line for Bonaventure to walk. But just as he took this original paradox concerning the simultaneous ontological priority and inseparability of forms from Aristotle, in Aristotle he also found its solution: a distinction between the two modes of form, the form’s esse and

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

218  Franziska van Buren the form’s existere, or the form considered as the universal and the form considered as the seminal reason—a distinction that straddles the real and the conceptual. What then of the Platonic or Augustinian character of Bonaventure’s notion of forms? It is quite clear that Bonaventure takes from Augustine both the term “seminal reason” and the general notion that the seminal reason is a kind of potency within the sensible thing. However, insofar as Augustine does not develop a detailed understanding of the relationship between these seminal reasons and universal forms, we can hardly say that Augustine is the main influence on Bonaventure’s forms. With regard to the Platonists, it seems apparent that Bonaventure would certainly agree that forms primarily exist and are universals. However, this does not make Bonaventure’s notion of forms Platonic in the way that many scholars have considered. Bonaventure does not see the position regarding the form’s existence and universality as the dividing point between Plato and Aristotle. The dividing point is whether the forms are separate or not, and here he quite clearly takes the position of Aristotle over Plato—regardless of the fact that his reading of Aristotle does not correspond to the more standard interpretations of both his day and ours. As a final point, showing the Aristotelian character of Bonaventure’s notion of form makes Bonaventure’s metaphysics more significant not only from a conceptual but also from a historical perspective. If we see the Aristotelianism in Bonaventure’s understanding of form, we arrive at a more accurate picture of the development of Franciscan thought and the kinds of conversations occurring among scholastic Franciscans. This is to say, we can see Bonaventure not as the last defender of the trad­ ition­al Augustinianism of scholastics before him, but as a figure who, precisely like the later Franciscans, e.g. Scotus and Ockham, was highly critical of alternative medieval accounts of universal forms, Platonic and Aristotelian alike. As we saw, in a manner very similar to Ockham, Bonaventure argues both against the naive realism of his day and against conceptualism—using Aristotle’s texts to do so. He also arrives, as I have mentioned above, at a relationship between universals and particularized forms quite similar to, yet nonetheless different from, the account provided by Scotus. My reading of Bonaventure thus allows us to see a

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  219 clearer continuity between his writings and those of the later Franciscans, precisely insofar as he anticipates so much of later Franciscan thought. Moreover, because Bonaventure anticipates (and quite possibly influences) the positions of these later thinkers, his view of forms, from our point of view, is in a better position to hold strong against the concerns and critiques they would later raise.87 Ludwig-­Maximilians-­Universität, München

Bibliography Adamson, Peter. “One of a Kind: Plotinus and Porphyry on Unique Instantiation,” in R.  Chiaradonna and G.  Galluzzo (eds.), Universals in Ancient Philosophy (Pisa: Edizioni della Normale, 2013), 329–52. Aristotle. Metaphysics, Volume I: Books 1–9, tr. H. Tredennick (Cambridge, MA: Harvard University Press, 1933). Aristotle. Physics, Volume II: Books 5–8, tr. P.  H.  Wicksteed and F. M. Cornford (Cambridge, MA: Harvard University Press, 1934). Aristotle. Metaphysics, Volume II: Books 10–14. Oeconomica. Magna Moralia, tr. H.  Tredennick and G.  C.  Armstrong (Cambridge, MA: Harvard University Press, 1935). Aristotle. Categories. On Interpretation. Prior Analytics, tr. H. P. Cooke and H. Tredennick (Cambridge, MA: Harvard University Press, 1938). Aristotle. On the Heavens, tr. W. K. C. Guthrie (Cambridge, MA: Harvard University Press, 1939). Aristotle. Generation of Animals, tr. A. L. Peck (Cambridge, MA: Harvard University Press, 1942). Aristotle. On the Soul. Parva Naturalia. On Breath, tr. W. S. Hett (Cambridge, MA: Harvard University Press, 1957). Aristotle. Physics, Volume I: Books 1–4, tr. P.  H.  Wicksteed and F. M. Cornford (Cambridge, MA: Harvard University Press, 1957).

87  I would like to thank the anonymous reviewer of this paper for their comments, which were extremely helpful in improving its quality. I would also like to thank Peter Adamson for likewise providing invaluable feedback.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

220  Franziska van Buren Aristotle. Posterior Analytics. Topica, tr. H.  Tredennick and E.  S.  Forster (Cambridge, MA: Harvard University Press, 1960). Augustine. Arianism and Other Heresies, tr. R. J. Teske (Hyde Park, NY: New City Press, 1995). Bettoni, Efrem. S. Bonaventura: gli aspetti filosofici del suo pensiero (Milan: Biblioteca Francescana Provinciale, 1973). Boehner, Philotheus. The History of the Franciscan School: Alexander of Hales (St. Bonaventure, NY: Franciscan Institute Publications, 1943). Boehner, Philotheus. The History of the Franciscan School: John of Rupella and Saint Bonaventure (St. Bonaventure, NY: Franciscan Institute Publications, 1943). Bonansea, Bernandino. “The Question of an Eternal World in the Teaching of St. Bonaventure,” Franciscan Studies 34 (1974), 7–33. Bonaventure. Opera Omnia (Quaracchi: Editiones Collegii S. Bonaventurae, 1882–1902). Bougerol, J. Guy. Introduction à l’étude de S. Bonaventure (Paris: Desclée & Cie, 1961). Bougerol, J. Guy. “Dossier pour l’étude des rapports entre Saint Bonaventure et Aristote,” Archives d’histoire doctrinale et littéraire du Moyen Age 40 (1973), 135–222. Callus, Daniel. “The Philosophy of St. Bonaventure and St. Thomas,” Blackfriars 21 (1940), 151–64. Callus, Daniel. The Condemnation of St. Thomas at Oxford (London: Blackfriars, 1946). Crowley, Theodore. “John Peckham, O.F.M., Archbishop of Canterbury, Versus the New Aristotelianism,” The Bulletin of the John Rylands Library 33 (1951), 242–55. Cullen, Christopher. Bonaventure (Oxford: Oxford University Press, 2006). Edwards, Sandra. “St. Bonaventure on Distinctions,” Franciscan Studies 38 (1978), 194–212. Ehrle, Franz. “Der Augustinismus und der Aristotelismus in der Scholastik gegen Ende des 13 Jahrhunderts,” Archiv für Literatur und Kirchengeschichte des Mittelalters 5 (1889), 603–35. Ehrle, Franz. “John Peckham über den Kampf des Augustinismus und Aristotelismus in der zweiten Hälfte des 13 Jhs,” Zeitschrift für katolische Theologie 13 (1889), 172–93.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Bonaventure, Aristotle, And the Being of Universal  221 Elders, Leo. “Les citations d’Aristote dans le ‘Commentaire sur les Sentences’ de Saint Bonaventure,” in A.  Pompei (ed.), San Bonaventura maestro di vita Francescana e di sapienza Cristiana, Vol. I (Rome: Pontificia Facoltà Teologica San Bonaventura, 1976), 831–42. Gilson, Etienne. La philosophie de Saint Bonaventure (Paris: Vrin, 1924). Hayes, Zachary. “Beyond the Prime Mover of Aristotle: Faith and Reason in the Franciscan Tradition,” Franciscan Studies 60 (2002), 7–15. Knowles, David. “Some Aspects of the Career of Archbishop Peckham,” The English Historical Review 57 (1942), 178–201. Knowles, David. The Evolution of Medieval Thought (New York: Vintage Books, 1962). Kovach, Francis. “The Question of the Eternal World in St. Bonaventure and St. Thomas:A Critical Analysis,” Southwestern Journal of Philosophy 5 (1974), 141–72. Marchesi, Angelo. “L’atteggiamento di S. Bonaventura di fronte al pensiero di Aristotele,” in A.  Pompei (ed.), San Bonaventura maestro di vita Francescana e di sapienza Cristiana Vol. I (Rome: Pontificia Facoltà Teologica San Bonaventura, 1976), 843–59. Quinn, John  F. “St. Bonaventure and the Arabian Interpretations of Two Aristotelian Problems,” Franciscan Studies 37 (1977), 219–28. Twetten, David, Steven Baldner, and Steven C. Snyder. “Albert’s Physics,” in I. M. Resnick (ed.), A Companion to Albert the Great: Theology, Philosophy, and the Sciences (Leiden: Brill, 2013), 173–219. van Steenberghen, Fernand. The Philosophical Movement in the Thirteenth Century (Belfast: Thomas Nelson & Sons, 1955). van Steenberghen, Fernand. “Le mythe d’un monde éternel: note complémentaire,” Revue Philosophique de Louvain 80 (1982), 486–99. William of Ockham. Opera philosophica et theologica (St. Bonaventure, NY: Franciscan Institute Publications, 1974).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command Demystifying Ockham’s Quodlibet III.14 Eric W. Hagedorn

1. Introduction When it comes to William Ockham’s writings on ethics, the single ­passage most discussed in the secondary literature might well be ques­ tion 14 in the third set of his Quodlibeta septem.1 Here Ockham gives perhaps the last discussion of his academic career on issues such as whether the moral status of every human action is merely contingent and the extent of God’s power over moral obligation. This passage has generated a volume of scholarly discussion greatly disproportionate to its size; despite comprising a mere fourteen paragraphs and less than five pages in the critical edition, nearly every modern discussion of Ockham’s ­ethics includes some sustained analysis of the passage. Yet despite the expansive literature on this brief question, there is no clear consensus on what Ockham’s answer is to the issues raised in it. In this text, Ockham responds to an objection regarding what would be morally right for one to do if one were to receive a divine command to 1  Throughout, I use the following abbreviations when discussing Ockham’s works: Ordinatio in librum primum Sententiarum [Ord.], Reportatio in libros II, III, IV Sententiarum [Rep.], Quodlibeta septem [Quod.], Quaestiones Variae [Var. Ques.], Brevis summa libri Physicorum [BrevPhys], Summa logicae [SL]. All references to Ockham’s works are to the standard critical editions, Opera theologica [OTh] and Opera philosophica [OPh]. All translations are my own unless noted otherwise, though I have closely consulted the translation in William of Ockham, Quodlibetal Questions, tr. A. J. Freddoso and F. E. Kelley (New Haven: Yale University Press, 1991), 211–15. Eric W. Hagedorn, On Loving God Contrary to a Divine Command: Demystifying Ockham’s Quodlibet III.14 In: Oxford Studies in Medieval Philosophy Volume 9. Edited by: Robert Pasnau, Oxford University Press. © Eric W. Hagedorn 2021. DOI: 10.1093/oso/9780192844637.003.0006

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  223 not love God; his response involves discussing which actions could be performed by an individual who was given such a command. But the precise nature of this hypothetical divine command and the ultimate resolution of the objection are subjects of significant disagreement. Some think that Ockham’s considered view is that the command in question cannot be obeyed, others that the command cannot be dis­ obeyed. Some cast the passage as Ockham recognizing an insoluble paradox that endangers his previous ethical commitments, others as a logical puzzle that is easily dismissed. Just a small sampling of the dis­ pute here should be sufficient to convince the reader of the range of the debate. One group of interpreters take the main point of this quodlibet to be that a supposed divine command to not love God cannot be obeyed. Terence Irwin is one who gives this reading: It is logically possible for God to give us a command that we could not obey. . . . God might command us not to love God; we could not be moved to obey this command unless we loved God, but if we loved God we would not be obeying this command. Hence we could not obey God’s command not to love God.2

Tania Holopainan says something rather similar: According to Ockham, if God gives the kind of command that He be not loved, the will cannot then obey it. In that case, no act of will can be an act which would fulfill the divine law . . ..3

And  A.  S.  McGrade says much the same (although he speaks of this quodlibet as if the case in question concerns the possibility of a creature being commanded to hate God, rather than being commanded to just not love God).

2 Terence Irwin, The Development of Ethics, vol. 1 (Oxford: Oxford University Press, 2007), 720. 3 Taina  M.  Holopainen, William Ockham’s Theory of the Foundations of Ethics (Helsinki: Luther-­Agricola Society, 1991), 141.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

224  Eric W. Hagedorn .  .  .  although it seems God can without contradiction command ­someone to hate him, such a command could not be obeyed. . . . God presumably cannot wish for the impossible. This would imply that God cannot will the fulfillment of an impossible command, but as far as Ockham’s general principle is concerned, God could wish us to attempt the impossible and hence order us to do so.4

(For those not previously familiar with the debate, note that McGrade is running together two different passages here: in the early Rep. IV, q. 16, Ockham considers a possible command to hate God, while in his much later Quod. III.14 he discusses a possible command to not love God. McGrade’s discussion treats the latter passage as an elaboration of the former; this leads him to conflate these two distinct hypotheticals and treat the quodlibet’s discussion as if it concerned a command to hate God. I will say more later about the case of being commanded to hate God.) Thomas Ward also holds that the command to not love God cannot be obeyed, although he sees this problem as arising only if the individual being commanded is one who already completely loves God: But as long as your love for God inspires you to do what he tells you, you can’t do what he tells you. So the command can’t be fulfilled. . . . The paradox here is that the person most disposed to follow divine com­ mands is the person unable to do so in this case, but there is nothing contradictory about the paradox. Someone who has a lower-­level love for God, or no love for God at all, or even no belief in God, could per­ form the action commanded . . . [A] divine command not to love God cannot be obeyed by someone who loves God above all.5

So, there are a significant number of scholars who all agree that the cen­ tral thesis of the quodlibet is that God can give a command that is 4  A. S. McGrade, “Natural Law and Moral Omnipotence,” in P. V. Spade (ed.), The Cambridge Companion to Ockham (Cambridge: Cambridge University Press, 1999), 280, emphasis in the original. 5  Thomas Ward, “The Incoherence of Ockham’s Ethics,” in M. Roques (ed.), Grounding in Medieval Philosophy (Leiden: Brill, forthcoming).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  225 impossible to obey. Though no one points it out, this reading seems to commit Ockham to a denial of the “ought implies can” principle (assum­ ing, as is plausible, that Ockham takes a divine command to be at least sufficient for generating moral obligation).6 However, a fair share of the literature takes Ockham to be making a strikingly different move here, with this second group of scholars taking the passage to affirm that it is indeed possible to act in accordance with this command to not love God. Thus Marilyn Adams: In Quodl. III.14, Ockham observes that if God were to command us not to love him, we would not be able to love God above all and for God’s own sake. For the latter involves a generalized commitment to obey all God’s commands, and in this case one of these commands forbids love of him.7

Peter King agrees with Adams. King argues that Ockham’s point is not that one cannot obey a command to not love God, but rather that one cannot positively love God once given the command to not do so. Thus the problem is not in God giving a command that cannot be obeyed, but rather that a command of this sort curtails what other actions are pos­ sible for the agent: Ockham’s response is to hold that in such circumstances [viz., having been given a command to not love God] the agent cannot in fact elicit the act of loving God above all else for his own sake . . ..8

So, according to King and Adams, if anything the agent cannot disobey this command; once commanded to not love God, it is impossible for the agent to elicit such love. 6  Whether Ockham thinks a divine command is necessary for moral obligation is of course the central question in the ongoing scholarly debate whether his ethic is just a thoroughgoing divine-­command theory, and reasonable cases can be made both for and against. But I cannot think of any reason for one to claim that a divine command is not sufficient for obligation on Ockham’s view. 7  Marilyn McCord Adams, “Ockham on Will, Nature, and Morality,” in P. V. Spade (ed.), The Cambridge Companion to Ockham (Cambridge: Cambridge University Press, 1999), 261. 8  Peter King, “Ockham’s Ethical Theory,” in P. V. Spade (ed.), The Cambridge Companion to Ockham (Cambridge: Cambridge University Press, 1999), 232.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

226  Eric W. Hagedorn Thomas Osborne agrees with Adams and King that Ockham’s central response is to hold that one cannot love God in this situation, although he thinks that this conclusion only holds if there is already a standing command that God should be loved: Ockham is supposing that a command to love God is in effect, and he is considering what would happen if God were to command that for a time someone study rather than love Him. The conclusion is that it would be impossible for the person to love God because such love requires obedience to God’s commands.9

So, the first set of scholars holds that Ockham’s position is that the com­ mand to not love God cannot be obeyed, while the latter set instead hold that it cannot be disobeyed (a somewhat unusual case where “ought” seems to imply “must”). Additionally, in an apparent attempt to have their cake and eat it too, in a recent addition to the literature J. Caleb Clanton and Kraig Martin suggest that Ockham affirms that it is both impossible to obey and impossible to disobey such a command: After all, if someone, S, were to obey the divine command to not love God, then S would be both not loving God and loving God (i.e., by willing what God wills for S to will) in the same act. Alternatively, if S were to disobey the divine command to not love God in an effort to love God above all else, then S would be both loving God and not ­loving God  (i.e., by refusing to will what God wills for S to will).10

Though they do not note it, this third sort of interpretation presents a particularly paradoxical situation. On Clanton and Martin’s reading, it seems that God can give a command that is such that no action would count as either obedience or disobedience to that command! 9  Thomas  M.  Osborne, “Ockham as a Divine-­Command Theorist,” Religious Studies 41 (2005), 15. 10  J. Caleb Clanton and Kraig Martin, “William of Ockham, Andrew of Neufchateau, and the Origins of Divine Command Theory,” American Catholic Philosophical Quarterly 94 (2020), 413.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  227 More such citations could be presented here, but suffice it to say that the literature on this short question is somewhat of a mess. (More sur­ prisingly, out of all the scholars mentioned here only Osborne explicitly points out that there is a disagreement in how to understand this ­passage.11) This is not a large interpretive debate where the two sides differ on the relative importance of disparate passages that suggest somewhat different accounts; rather, this is a case where there seems to be no shared understanding of what is even being asserted in one short discussion of limited dialectical complexity. (The morass is all the more puzzling given that Ockham’s prose is not especially muddled or dense; we are not trying to interpret Hegel here, or even Scotus.) In what fol­ lows, I argue that, so long as one keeps the dialectic between Ockham and his imagined objector straight, the points that Ockham makes in this question are actually quite straightforward. In hopes of forestalling confusion over this passage for future generations, I will step through the core of the text paragraph-­ by-­ paragraph, explaining Ockham’s claims according to his own order of presentation.

2.  Five Claims of Quodlibet III.14 It will help to be clear from the start precisely what I am arguing. I intend to show in the remainder of this paper that in Quodlibet III.14 Ockham affirms the following five claims: 1. The Contingency of Moral Status: No act of a creaturely will is ne­ces­sar­ily virtuous, strictly speaking. As we will see, he holds this for two main reasons: first, since no such act exists necessarily, the moral status of the acts cannot be necessary; second, every act of a creaturely will could be caused by God alone, in which case the act would have no moral status at all.12

11  See Osborne, “Divine-­Command Theorist,” 15. 12  At least, the act would have no moral status for the creature; it is less clear whether the act would have any moral status for God. See fn. 17 below for further discussion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

228  Eric W. Hagedorn 2. Loving God Above All is Actually Always Virtuous: As a matter of fact, given God’s actual commands, whenever any creature elicits an act of loving God above all else, that act is virtuous. As is well known, Ockham says in this text that loving God above all else is “necessarily virtuous,” but only in some qualified sense of that term; that qualified sense, I will argue, turns out to be the limited modal claim that loving God is always virtuous given God’s actual commands. 3. Possible to be Commanded to Not Love God: God can, in the actual world, consistent with his actual commands, also command some individual to not elicit an act of loving God at some time t. 4. Impossible to Disobey a Divine Command to Not Love God: If God commands an individual to not elicit an act of loving God at some time t, it is impossible for that individual to disobey that command by eliciting an act of loving God above all else at t (though it is pos­ sible for the individual to disobey the command by eliciting a lesser kind of love). 5. Possible to Obey a Divine Command to Not Love God: If God com­ mands an individual to not elicit an act of loving God at some time t, it is possible for that individual to obey the command by not eliciting an act of loving God at t. Ockham puts in the mouth of his objector the claim that God can com­ mand an individual to not elicit an act of loving God at a certain time, but he accepts that this is indeed a coherent and possible command. But whether the command can be obeyed or disobeyed is somewhat compli­ cated: I will argue that Ockham’s view is that, once given the command, the highest kind of love for God is no longer possible for the agent, but a lesser kind of love is still possible, and it is within the agent’s power to actively obey the command by not eliciting the proscribed act. With respect to the scholarship mentioned in the introduction, then, I hold that Adams, King, and Osborne largely have the right of it in holding that Ockham’s point is that, once given a command to not love God, it is then impossible to love God above all and thus impossible to disobey God’s command to not do so. But none of their treatments fully

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  229 explicate the entirety of Ockham’s point in this question, none of them clearly show how and why the other lines of interpretation have gone awry, and none of them make entirely clear why he mentions a second, lesser kind of love at the end of the passage.

3.  On the Contingency of Moral Status and the Goodness in Loving God The main line of argument13 of Quod. III.14 begins as follows: Regarding the affirmative claim [that some act of the will is necessarily virtuous], I say first that no act is necessarily virtuous according to the literal meaning of the words. This is proved as follows: No act is ne­ces­ sary, and consequently no act is necessarily virtuous. Also, because every act can be brought about by God alone, and consequently such an act is not necessarily virtuous because an act of this sort [namely, one brought about by God alone] is not within the will’s power.14

This is the first bit of the question that seems to have tripped up some interpreters.15 But Ockham’s point here is quite simple: speaking strictly and precisely, no act of a creaturely will is necessarily virtuous, because no act of a creature’s will exists necessarily. Every creaturely act exists only contingently, and so, Ockham the logician reminds his reader, the moral status of all such acts is also contingent, strictly speaking. (Though I do not believe any other interpreter has noted it, this claim precisely 13  In the first seven paragraphs of the question, Ockham argues that all actions other than acts of the will—that is, acts of the intellect, acts of the sensory appetites, external movements, and the like—cannot be intrinsically virtuous, so that the moral status of all such actions always depends upon the moral status of acts of the will. This is all relatively clear and uncon­ troversial in the literature, so I pass by it without comment. 14 “Circa affirmativam exponentem [scilicet quod aliquis actus voluntatis est necessario virtuosus] dico primo quod de virtute sermonis nullus actus est necessario virtuosus. Hoc probatur, tum quia nullus actus necessario est, et per consequens non est necessario virtuosus; tum quia omnis actus potest fieri a solo Deo, et per consequens non est necesssario virtuosus, quia talis actus non est in potestate voluntatis” (Quod. III, q. 14, OTh IX, 253–7, lines 35–41). 15  Ward, for one, seems to overlook that Ockham’s use of the phrase ‘necessarily virtuous’ invokes some looser notion of necessity, as he criticizes Holopainen for holding that loving God is necessarily virtuous only in some conditional sense. See Ward, “Incoherence,” fn. 20.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

230  Eric W. Hagedorn parallels Ockham’s frequently made point that no sentence is necessarily true, strictly speaking, because all sentences are contingent entities.16) Furthermore, even assuming the relevant act of will exists, it does not necessarily follow that the action has any moral status at all, because acts that God directly brings about in the will of a creature (without any activity of the creature’s will) have no moral status.17 So then, if Ockham is going to call a creaturely act necessarily virtuous in any sense whatso­ ever, it must be under the assumption that (1) the act exists and (2) the act was partly caused by that very creature’s will. Thus Ockham affirms claim #1, the Contingency of Moral Status. Ockham does continue on to give a qualified sense of what it is to call an act necessarily virtuous: Nevertheless, an act can be understood to be necessarily virtuous in another sense, namely, that the act could not be vicious given the established divine command. Similarly, that the act cannot be caused by a created will without being virtuous. Understanding ‘a virtuous act’ in this sense, I say, second, that some act can be necessarily virtu­ ous in this sense.18

16  Ockham holds that, since all sentences are contingent, no sentence is necessarily true in a strict sense. However, a sentence can be said to be necessarily true in a loose sense if and only if it is not possible for the sentence to exist and be false. See, e.g., Ord., Prologue, q. 8 (OTh I, 222); SL II.9 (OPh I, 275), and BrevPhys, Prologue, ch. 2 (OPh VI, 6). Note the similarity between this loose sense of a sentence’s being necessarily true and his loose sense of an act’s being necessarily virtuous, as explained below. 17  At least, an act caused by God alone will not be morally good or bad for the creature; this is a claim Ockham implicitly relies on in multiple places, but he states it most explicitly in Rep. III, q. 11: “If an act in the will were totally caused by God, it would not be called morally good or bad, since those names connote an activity of the will . . . If God were to make in my will an act conforming to right reason while my will is not acting in any way, then that act would be neither meritorious nor virtuous. So, it is required for the goodness of an act that it be in the power of the will having that act” (Rep. III, q. 11; OTh VI, 389). As for whether an act caused by God alone is morally good or bad for God, that is less clear, but Ockham does repeatedly insist that God is indebted to no one, and it seems to me he thereby intends to imply both that God has no moral obligations and also that God’s acts are immune to ordinary moral appraisal. See, e.g., Ord. d. 47, q. un., where Ockham argues that God does not act wrongly when he brings about a bad state of affairs because there is nothing that God is obligated to not bring about (OTh IV, 680–5). Do note, however, that Ockham somewhat distances himself from this claim, expressing concern about whether it might border on heresy. 18 “Tamen aliter potest intelligi actum esse necessario virtuosum, ita scilicet quod non possit esse vitiosus stante praecepto divino; similiter non potest causari a voluntate creata nisi sit virtuosus. Et sic intelligendo actum virtuosum, dico secundo quod sic potest aliquis actus esse virtuosus necessario” (Quod. III.14, lines 43–5).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  231 Ockham here gives additional conditions for a creaturely act being ne­ces­sar­ily virtuous. In addition to (1) the act’s existing and (2) the act’s being partly caused by that creature’s will, Ockham claims that (3) it must be the case that “the act could not be vicious given the established divine command,” and (or else?) that (4) “the act cannot be caused by a created will without being virtuous.” Here is where the confusion of the passage really begins, and, in all fairness to Ockham’s interpreters, here the fault mostly rests with Ockham. First, is he speaking of some specific divine command or commands here—perhaps the command to love God with one’s whole heart, or perhaps the whole Decalogue—or is he speaking of just any arbitrary divine command, whether actual or merely possible? Second, are conditions 3 and 4 meant to be disjunctive or conjunctive—is meeting either of them (together with conditions 1 and 2) sufficient for an act to be necessarily virtuous, or must the act meet all of conditions 1–4? Finally, what could possibly lead Ockham to say that conditions 3 and 4 are similar? I do not think answering these questions is necessary for untangling the scholarly confusion about the paragraphs that follow and so I do not want to get bogged down here. But my best guess is that what Ockham means here is that conditions 3 and 4 are “similar” because we are meant to read stante praecepto divino as applying to both clauses. That is, his claim is that an act is necessarily virtuous if, given the actual divine commands, that act cannot be vicious; or else, likewise, given the actual divine commands, that act cannot fail to be virtuous. Now, being vicious and not being virtuous are generally not identical for Ockham; he rec­ ognizes that it is entirely possible for an act to be morally neutral. In particular, every act whose moral status is determined extrinsically (i.e., every act whose moral status depends upon the moral status of some other act) is morally neutral in itself; even some acts of the will are neu­ tral in this way. But those acts of the will that have their moral status intrinsically cannot be neutral; they are intrinsically either virtuous or vicious, with no middle ground.19 So, for such acts of the will, failing to be virtuous turns out to be just the same as being vicious. And so, if we assume that stante praecepto divino is intended to apply to both 19  For Ockham’s account of morally indifferent acts, see the final dubium in Rep. III, q. 11 (OTh VI, 383–90).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

232  Eric W. Hagedorn condition 3 and condition 4, those two conditions turn out to be ­coextensive, just slightly different ways of describing the same criterion. (A further reason to think condition 4 is just a redescription of condi­ tion 3 is that a parallel text only mentions the third condition.20) But all of this is just speculation: Ockham never applies these condi­ tions in a way that would further elucidate them and the Latin as written does not obviously apply stante praecepto divino to both clauses. But even if my guess is wrong, what I take to be clear from this passage is that what Ockham means by “necessarily virtuous” is more akin to nat­ ural necessity than to metaphysical necessity. Just as for a thing to be naturally necessary is for it to always come about given the actual laws of nature that obtain, so for an act to be necessarily virtuous in this limit­ed sense only requires it to always be virtuous given the actual divine commands that hold. The unstated implication seems to be that if the relevant divine commands were to change, the moral status of the “necessarily virtuous” act might change as well, just as what is naturally ne­ces­sary would change if the laws of nature were to change.21 In the next bit of text, then, Ockham affirms claim #2, that Loving God Above All is Actually Always Virtuous: Third, I say that an act that is necessarily virtuous in the aforemen­ tioned sense is an act of the will, for an act by which God is loved above all things and for his own sake is such an act, for this act is virtu­ ous in such a way that it cannot be vicious, nor can this act be caused by a created will without being virtuous.22

Here Ockham affirms that loving God above all else is necessarily virtu­ ous (in his loose sense of being necessarily virtuous); that is, an act of loving God above all else is always virtuous when the act is caused by 20  In the parallel text Ockham just says that a necessarily virtuous act is one that “is virtuous in such a way that it cannot become vicious given the established divine command” (Var. Ques., q. 7, a. 1; OTh VIII, 328). 21  My thanks to an anonymous referee for helping me see the connection to natural neces­ sity here. 22 “Tertio dico quod ille actus necessario virtuosus modo praedicto est actus voluntatis, quia actus quo diligitur Deus super omnia et propter se, est huiusmodi; nam iste actus sic est virtuosus quod non potest esse vitiosus, nec potest iste actus causari a voluntate creata nisi sit virtuosus . . .” (Quod. III.14, lines 60–6).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  233 the creature’s will, so long as the divine command to love God holds. He immediately goes on to give five brief arguments why, given the actual divine commands that are in force, the love of God cannot be elicited without being virtuous: This is so, first, because everyone is obligated to love God above all things at some23 time and place, and consequently the act cannot be vicious. Second, because this act is foremost of all good acts. Also, only an act of the will is intrinsically praiseworthy and blameworthy. Also, according to the saints no act is praiseworthy or blameworthy except because of a good or bad intention, and intention is an act of the will; therefore, etc. Also, according to Anselm only the will is punished because only the will sins; therefore, etc.24

Note that the first of these arguments is just that everyone as a matter of fact has an obligation to love God (perhaps as a result of Christ’s first commandment, that one must love God with all one’s heart, mind, and soul); the thrust of the argument is that an act of loving God above all else cannot be vicious because of the standing general obligation to love God above all else. (Before moving on, I should note that there is a scholarly debate whether Ockham thinks loving God above all else is the necessarily ­virtuous act or merely a necessarily virtuous act.25 Perhaps curiously, 23  How precisely to translate pro loco et tempore is a bit of a mystery. In their edition of Ockham’s quodlibets, Freddoso and Kelley render it “everyone in his own time and place is obliged to love God above all things” (Quodlibetal Questions, 213). King’s translation of the passage reads, “everyone, no matter where or when, is obligated to love God above all else” (King, “Ockham’s Ethical Theory,” 232). Near the end of the paper I explain why I think the translation “at some time and place” (or, perhaps, “at a certain time and place”) is indeed what Ockham intends here. 24  “. . . tum quia quilibet pro loco et tempore obligatur ad diligendum Deum super omnia, et per consequens iste actus non potest esse vitiosus; tum quia iste actus est primus omnium actuum bonorum. Praeterea solus actus voluntatis est intrinsece laudabilis et vituperabilis. Praeterea secundum Sanctos nullus actus est laudabilis vel vituperabilis nisi propter intentionem bonam vel malam; intentio autem est actus voluntatis; igitur, etc. Praeterea secundum Anselmum sola voluntas punitur, quia sola peccat; igitur etc.” (Quod. III.14, lines 66–72). 25  King translates using the definite article, ascribing to Ockham the view that “the act of the will that is intrinsically virtuous is an act of loving God” (“Ockham’s Ethical Theory,” 232). Ward appears to agree, as he claims this quodlibet implies that “the love of God above all [is] an unusual or extra important moral obligation” (“Incoherence,” fn. 20). Freddoso and Kelley’s translation does not explicitly make this commitment, translating the passage much as I do;

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

234  Eric W. Hagedorn I have not seen anyone in this debate point to the parallel passage in Var. Ques., where Ockham says that “willing to do something because it is a divine command” is necessarily virtuous in the qualified sense.26 But whether both kinds of acts—intentionally obeying a divine command and loving God above all else—are necessarily virtuous, whether any other acts are such, and/or whether Ockham changed his mind about the virtuous act between writing these two texts is largely irrelevant for my purposes here.) One additional strength of my interpretation—namely, that here Ockham is only affirming that loving God above all is always virtuous in the actual world, rather than some stronger modal claim—is that it helps clear up a seeming contradiction in Ockham’s texts. As I noted earlier, in Rep. IV, q. 16, while arguing that even the wills of the blessed are still in some sense free to turn away from God, Ockham discusses the possibil­ ity of God commanding some creature to hate God. In that text Ockham clearly affirms that such a command is possible and even holds that in such a case it would be morally right for the creature to hate God: Every created will can conform itself to a divine command. But God can command a created will to hate him, and so a created will can do this. Furthermore, everything that can be a right action in this life can also be a right action in the next life. But hating God can be a right action in this life (e.g., if it were commanded by God); therefore, it can also be a right action in the next life.27

Now, if one understands the argument of Quod. III.14 as holding that loving God is necessarily virtuous in such a way that such love is their Ockham says that “an act that is necessarily virtuous in the way just explained is an act of the will, since an act by which God is loved above all things and for his own sake is an act of this sort” (Freddoso and Kelley, 213, my emphasis). Osborne explicitly notes that “although Ockham emphasizes that the love of God is intrinsically virtuous, in many passages he suggests that there can be many such acts,” although he does not say what passages he has in mind (“Divine-­Command Theorist,” 7). 26  Var. Ques., q. 7, a. 1 (OTh VIII, 328). While doing final revisions of this paper, I became aware of Matthew Dee’s recent Ph.D.  dissertation, William of Ockham’s Divine Command Theory (University of South Florida, 2019); Dee rightly notes the parallel text and discusses it alongside Quod. III.14 throughout his chapters 2–3. 27  Rep. IV, q. 16 (OTh VII, 352).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  235 virtuous in all possible worlds, then these two passages seem to contradict each other; how can hating God be possibly morally right if loving God is necessarily morally right?28 But recognizing that in Quod. III.14 Ockham is only affirming that loving God above all else is actually always virtuous clears up this tension. As a matter of fact, given the actual divine commands, loving God is always morally right; but were God to give a command to hate him, then the act of loving God would no longer be morally right.

4.  On the Command to Not Love God Next, Ockham considers a potential objection to his position, and here is where much of the confusion in the secondary literature arises. Recall that Ockham has asserted that loving God above all is actually always virtuous; so, naturally, the imagined objector tries to fashion a counter­ example according to which an act of loving God above all else would fail to be virtuous: You might object as follows: God can command that he not be loved for some time, since he can command an intellect (and likewise a will) to be so intent on studying that it could in no way think about God for that time. Now suppose that the will elicits an act of loving God at that time. Then that act is either virtuous (but this cannot be said, because the act is elicited contrary to a divine command) or else the act is not virtuous, and then the claim [of the objection] is established, namely, that some act of loving God above all things is not virtuous.29

28 Lucan Freppert, for one, explicitly states that Ockham must have changed his mind between the Reportatio and the Quodlibeta on the basis that it would be impossible to rightly hate God if the love of God is necessarily virtuous. See his The Basis of Morality According to William of Ockham (Chicago: Franciscan Herald Press, 1998), 121–5. 29 “Si dicis quod Deus potest praecipere quod pro aliquo tempore non diligatur ipse, quia potest praecipere quod intellectus sit sic intentus circa studium et voluntas similiter, ut nihil possit pro illo tempore de Deo cogitare. Tunc volo quod voluntas tunc eliciat actum diligendi Deum; et tunc aut ille actus est virtuosus, et hoc non potest dici, quia elicitur contra praeceptum divinum; aut non est virtuosus, et habetur propositum, quod actus diligendi Deum super omnia non est virtuosus” (Quod. III.14, lines 74–81).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

236  Eric W. Hagedorn The objection is this: suppose, for the sake of argument, that Ockham is right that loving God above all is actually always virtuous. Then, what if God commands Brennan to not elicit an act of loving God on Thursday morning at 10 AM so that she can intensely focus on her math class instead? Well, the objector suggests, then, Ockham’s claim is in danger: if Brennan does elicit an act of loving God above all else at 10 AM, she is disobeying a divine command and so her act is not virtuous, and so it is not the case that loving God above all else is actually always virtuous. Ockham, in his response, does not take any issue with the possibility of God giving a command to not think about God at 10 AM on Thursday: I respond that if God could command this – as it seems that he can without contradiction – then I say that the will cannot elicit this sort of act at that time.30

Ockham affirms (or at least sees no reason to doubt) that God can give Brennan this command, and so holds claim #3, that it is Possible to be Commanded to Not Love God. The core of his response to the objection is that there is some act that Brennan cannot elicit at 10 AM on Thursday (“the will cannot elicit this sort of act at that time”). But what act is he talking about that cannot be elicited? As I see it, it is misreading this sentence that leads to many interpretive missteps. Some seem to think that Ockham means the will cannot elicit an act of hating God,31 others that the will cannot elicit an act of loving God,32 and yet others that what cannot be done is refraining from loving God.33 To correctly understand what Ockham is arguing here, we must remember the objector’s scenario: the objector is trying to figure out what happens if Brennan loves God above all else at 10 AM on Thursday.

30 “Respondeo: si Deus posset hoc praecipere, sicut videtur quod potest sine contradictione, dico tunc quod voluntas non potest pro tunc talem actum elicere . . .” (Quod. III.14, lines 83–5). 31  E.g., “[A]lthough it seems God can without contradiction command someone to hate him, such a command could not be obeyed” (McGrade, “Natural Law,” 280). 32  E.g., “For this reason, Ockham holds that ‘the will is not able to elicit an act of [loving God above all things] during that time’—namely, the time when God commands S to hate or not love God” (Clanton and Martin, “Origins of Divine Command Theory,” 414). 33  E.g., “[S]uppose that, loving God above all, you seek to obey the command [that you not love him] . . . . [But] the command can’t be fulfilled” (Ward, “Incoherence,” 6).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  237 It is this act of loving God that the objector suggested would serve as a counterexample to Ockham’s claim that loving God above all is actually always virtuous, and it is this act that Ockham now claims is impossible for Brennan to elicit: For if the will were to elicit an act of this sort [i.e., an act of loving God above all else], then it would love God above all things and conse­ quently it would carry out the divine command, for loving God above all things is to love whatever God wills to be loved. But from the fact that it would love him in this way, the will would not carry out the divine command from the example. Consequently, by loving God in this way, the will would both love God and would not love God; it would carry out God’s command and not carry it out.34

But why does Ockham think that Brennan cannot elicit an act of loving God at 10 AM? Well, he claims loving God above all else entails willing what God wills for one to will. If one loves God, one obeys God’s ­commands.35 (As Osborne very rightly points out, the entailment is not bi­dir­ec­tion­al: if one loves God above all else, then that individual will obey God’s commands; but obedience to God’s commands does not entail that one loves God.36 I say more about this below.) So, if Brennan elicits an act of loving God above all else at 10 AM on Thursday, then Brennan will follow God’s commands for her at 10 AM on Thursday. But God commanded her to not love God at that time and study her math instead; so, following God’s command would involve not eliciting an act of loving God. So, by hypothetical syllogism, if Brennan elicits an act of loving God above all else at that time, then Brennan will not elicit that act of loving God at that time. But now we have a direct contradiction: her having an act of loving God would imply that that very act was not 34  “. . . quia ex hoc ipso quod talem actum eliceret, Deum diligeret super omnia, et per consequens impleret praeceptum divinum, quia hoc est diligere Deum super omnia: diligere quidquid Deus vult diligi; et ex hoc ipso quod sic diligeret, non faceret praeceptum divinum per casum; et per consequens sic diligendo, Deum diligeret et non diligeret, faceret praeceptum Dei et non faceret” (Quod. III.14, lines 85–91). 35  Cf. John 14:23 in the Vulgate: “If someone loves me, they will follow my words.” 36  “Divine-­Command Theorist,” 15–16. There are also epistemological worries in the back­ ground that Ockham is aware of but usually elides; assume for the sake of argument that Brennan knows what God has commanded her to do.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

238  Eric W. Hagedorn elicited. Thus, Ockham concludes claim #4, that it is Impossible to Disobey a Divine Command to Not Love God. Why does Ockham keep specifying that he’s speaking about an act of loving God above all else? Because he believes that Brennan could elicit a different kind of love of God instead, one that does not generate the same paradox: However, the will could love God with a simple and natural affection, which is not the love of God above all things.37

The unstated implication here must be that the simple and natural “affection” for God does not entail obeying all God’s commands the way that the love of God above all else does so entail. So, one can consistently have this simple affection for God without generating any contradiction in action, and so it is not entirely impossible to disobey God’s command: Brennan could successfully disobey God’s command by eliciting just this simple kind of fondness of God.

5.  On Obeying the Command to Not Love God But can Brennan successfully obey the command to not love God? Here Ockham is not quite as explicit, but note that the command here is not a difficult one: the command is to just sufficiently focus one’s mind on studying that one’s will does not elicit an act of loving God, and this is an entirely ordinary sort of occurrence. It would be quite strange if it were impossible to act in this way. Furthermore, Ockham points out that even an atheist is able to act in such a way as to fulfill the command: Supposing that someone does not believe that God exists, that person cannot love him, because nothing can be loved except what exists or can exist.38 37 “Posset tamen Deum diligere simplici amore et naturali, qui non est dilectio Dei super omnia . . .” (Quod. III.14, lines 91–2). 38  “. . . posito quod aliquis non credat Deum esse, non potest eum diligere, quia nihil potest diligi nisi quod est vel potest esse” (Quod. III.14, lines 92–4).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  239 So, if the atheist can fulfill the command to not love God, it is hard to see why Brennan could not do so. After all, to fulfill the command, all she needs to do is simply refrain from eliciting an act of love at 10  AM, which she can do just by focusing on her math instead. So, presumably, it follows that it is possible to obey the command to not love God. But there is a potential complication here.39 What is clear is that the atheist is capable of not eliciting an act of loving God. And, likewise, it seems clear that Brennan is capable of not eliciting an act of loving God. But is doing what is commanded—not eliciting the act of loving God at 10 AM—the same as obeying the command to not love God? Surely Brennan can do the former. Can she do the latter? This concern seems to motivate at least some of the scholars I dis­ cussed at the beginning. Irwin in particular clearly states that one can only obey God’s command if one first loves God, that in order for an act to count as obedience it must be that the relevant act is performed out of love. Indeed, I think Irwin takes this to be the key point of the whole passage, glossing Quod. III.14 as follows: In Ockham’s view, we obey God’s commands because we follow the precept of non-­positive morality that enjoins the love for God. . . . God might command us not to love God; we could not be moved to obey this command unless we loved God, but if we loved God we would not be obeying this command. . . . This argument presupposes that the only motive for obeying God that we need to consider is love for God. Ockham is not considering obedience out of ‘servile’ fear of God’s power to harm us, but obedience on moral grounds, and hence obedi­ ence based on the love of God.40

Irwin’s claim that moral obedience must be founded in the love of God seems to derive from Ockham’s discussion of the fourth degree of virtue in On the Connection of the Virtues:

39  My thanks to an anonymous referee for helping me see this concern much more clearly. 40 Irwin, Development of Ethics, 720, emphasis added.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

240  Eric W. Hagedorn The fourth degree of a virtue is when someone wills to perform this sort of deed according to the previously discussed conditions and cir­ cumstances, and does so just on account of a love of God, e.g., because the intellect has dictated that deeds of this sort should be done just on account of a love of God. Only this degree of virtue is the perfect and true moral virtue about which the saints speak.41

And Ockham also agrees that fear of harm might give one a pragmatic reason but not necessarily a moral reason to obey a command, when he argues that one does not have a moral obligation to obey commands given by earthly tyrants: [A]lthough sometimes permitted power is to be obeyed to avoid the power’s wrath in case greater evil or damage results, it is nevertheless not to be obeyed for conscience’s sake. If there were no motive for obedi­ence but its wrath alone, disobedience would be permissible – someone captured by a robber or pirate (a little one or a great!) rightly escapes, if he can.42

Nevertheless, I do not think Irwin establishes the point here that Brennan can only obey God’s command if she already loves God. The fact that only the perfect degree of moral virtue requires the love of God indicates that Ockham accepts that one can act morally without the love of God; likewise, Ockham regularly notes that the pagan philosophers performed virtuous acts, albeit acts of a different kind and degree than the virtuous acts performed by Christians.43 So, love of God is not ne­ces­sary for doing what is right or doing what one ought. 41  Var. Ques. q. 7, a. 2 (OTh VIII, 335–6). The sourcing from here to Irwin is indirect: Irwin cites Adams on this point, who cites Ockham. For the intermediary source, see Marilyn McCord Adams, “The Structure of Ockham’s Moral Theory,” Franciscan Studies 46 (1986), 27–31. 42  William Ockham, A Short Discourse on Tyrannical Government, ed. A.  S.  McGrade, tr. J. Kilcullen (Cambridge: Cambridge University Press, 1992), 79. 43  For just one example, see Rep. IV, qq. 3–5 (OTh VII, 58): “When one abstains from an act of sexual intercourse for the sake of God and because God commanded one to abstain in this way, then God (or God’s command) is the final cause of this abstinence; and it is thus for all other virtues acquired by a good Christian, because God is always the chief intended end. However, a philosopher, even though they might abstain from such acts, nevertheless does so totally because of another end: either for the sake of preserving one’s nature for proficiency in knowledge or for the sake of some other such end. Therefore, there was a different partial

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  241 Further, just because disobeying a pirate’s command is morally licit does not entail that Jim is not obeying Long John Silver when Silver tells Jim to get in the rowboat and Jim complies. Acting in conformity to a received command just is obeying that command, regardless of the motive for the obedience or the question of whether disobedience is morally permitted. Indeed, when Ockham speaks about the obedience that creatures owe to the divine will, he centers his inquiry on the claim that creatures are obligated to conform to the divine will by willing what God wills those creatures to will; the question of the motive for the crea­ ture’s conformity does not arise in that context.44 So long as Brennan acts in conformity with God’s command, I argue, she is obeying God’s command, regardless of whether her motivation to obey comes from love, fear, just being the highly suggestable type, or whatever.45 Additionally, there seems to be another way to respond to Irwin here which largely sidesteps the whole paradox, one I have avoided discuss­ ing heretofore because Ockham does not mention it in this passage nor do any of the commentators seem to consider it. But elsewhere Ockham regularly distinguishes between habitual love and occurrent acts of love, and he holds that although some acts are incompatible with other acts and some habits incompatible with other habits, no act is naturally incompatible with any habit.46 Now, the habitual love of God might be sufficient on its own to motivate one to obey divine commands, and having that habitual love at a given time is compatible with not eliciting an act of love at that same time. So, if one is convinced with Irwin that love for God is the only thing that could motivate obedience to God’s command, it seems that Brennan could readily obey God’s command as a result of maintaining her habitual love of God, even while she obeys the command by refraining from any occurrent act of loving God at 10 AM.

object of abstinence for the philosopher and for the good Christian; consequently, there was a different virtue of a different nature.” 44 See Ord., d. 48, q. un (OTh IV, 686–91). 45  Freppert claims that, for Ockham, “love and obedience are equivalent” (Basis of Morality, 124). I think the considerations in the main text are sufficient to rebut this point; but for fur­ ther argument against Freppert’s equivalency thesis, see Dee, William of Ockham’s Divine Command Theory, 94–100. 46  See especially the discussions of the connections of the virtues at the end of Rep. III, q. 12 (OTh VI, 421–8) and the end of Rep. IV, qq. 3–5 (OTh VII, 59–60).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

242  Eric W. Hagedorn Either way, I suggest we have good reason to think Ockham endorses claim #5, that it is Possible to Obey a Divine Command to Not Love God. Now, Adams agrees that it is possible to obey the command to not love God, but she has a further worry about Brennan’s obedience. She is concerned that, by refraining from loving God, Brennan will be violat­ ing her other moral norms. According to Adams’s interpretation, hating God or refraining from loving God would violate an existing moral obli­ gation, since she reads Ockham as holding that right reason always gen­ erates a standing obligation to love God: . . . Ockham’s God could also thwart such attempted sacrificial loyalty to himself by commanding rational creatures not to love or command­ ing them to hate Him. This external circumstance would put rational creatures in the bind of not being able to conform to their norms . . .47

But, contra Adams, obeying a command to concentrate on one’s math homework instead of God is not itself sinful or contrary to right reason. Now, she may well be correct that being given a command to positively elicit an act of hating God would place a creature in such a moral dilemma, but she is mistaken to think that the command to not elicit love for God at a certain time fails under the same umbrella. Not elicit­ ing an act of loving God at 10 AM on Thursday is not itself a sin accord­ ing to Ockham, since one is only obligated to have such an act at some time and place, not at every time and place. As he clearly affirms else­ where, Ockham believes that divine commands to positively perform a given action generally only bind in a particular situation or at a particu­ lar place and time; the only commands that bind one at all times are the negative commands to refrain from a given action: Someone sins mortally when they do not conform themselves to the divine will at a time and place at which they are obligated to do so in order for them to be saved. But, as was said before, it is not the case that everyone is obligated to do so at all times; this is especially so with 47  Adams, “Will, Nature, and Morality,” 265. See also her “Structure of Ockham’s Moral Theory,” 27–31.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  243 respect to affirmative commands, although everyone is always obligated by the negative commands.48

The command to love God, being an affirmative command, does not bind one to elicit an act of loving God at all times; rather, one is only obligated to love God at some time or other, and perhaps at some spe­ cific particular times. (Perhaps one is obligated to elicit the act of loving God while praying in church, say.) Further, it simply cannot be the case that everyone is always obligated to elicit acts of loving God, for, as Ockham points out in a different con­ text, there are times at which one has no acts of will at all. For example, one has no acts of will while asleep,49 and so the command to love God above all things cannot be an obligation to elicit that sort of act of will at every moment. So, contra Adams, simply not eliciting an act of love for God cannot, in general, be a sin or a breach of ethical norms, especially not in the case where God has commanded Brennan to instead concen­ trate on her math homework for a given stretch of time.

6. Conclusion It is difficult to see how progress can be made on large-­scale interpretive issues in Ockham’s ethics when such confusion reigns regarding the central thrust of such a widely discussed passage. In order to discern whether Ockham is an unmitigated divine-­command theorist, a pro­ ent of some heterodox natural-­ law view, or something else pon­ al­together, we must first come to some accurate and grounded under­ standing of the key texts in his corpus. Recognizing that Ockham thinks one can in fact not love God for a given time and that doing so violates no eth­ic­al norms is a key building block for an accurate understanding of his ethical theory. Seeing that he believes that loving God above all

48  Ord., d. 48, q. un. (OTh IV, 690). 49  “If [two men] are sleeping, there will not be an act of intellect or an act of will in either of them” (Rep. IV, qq. 10–11; OTh VII, 196).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

244  Eric W. Hagedorn else entails obeying all God’s commands, but that such obedience does not entail such love, is another crucial piece of that puzzle.50 St. Norbert College

Bibliography Adams, Marilyn McCord. “The Structure of Ockham’s Moral Theory,” Franciscan Studies 46 (1986), 1–35. Adams, Marilyn McCord. “Ockham on Will, Nature, and Morality,” in P.  V.  Spade (ed.), The Cambridge Companion to Ockham (Cambridge: Cambridge University Press, 1999), 245–72. Clanton, J.  Caleb and Kraig Martin. “William of Ockham, Andrew of Neufchateau, and the Origins of Divine Command Theory,” American Catholic Philosophical Quarterly 94 (2020), 405–29. Dee, Matthew. William of Ockham’s Divine Command Theory (Ph.D. Dissertation, University of South Florida, 2019). Freppert, Lucan. The Basis of Morality According to William of Ockham (Chicago: Franciscan Herald Press, 1988). Holopainen, Taina M. William Ockham’s Theory of the Foundations of Ethics (Helsinki: Luther-Agricola Society, 1991). Irwin, Terence. The Development of Ethics, vol. 1 (Oxford: Oxford University Press, 2007). King, Peter. “Ockham’s Ethical Theory,” in P. V. Spade (ed.), The Cambridge Companion to Ockham (Cambridge: Cambridge University Press, 1999), 227–44. McGrade, A. S. “Natural Law and Moral Omnipotence,” in P. V. Spade (ed.), The Cambridge Companion to Ockham (Cambridge: Cambridge University Press, 1999), 273–301. Osborne, Thomas M. “Ockham as a Divine-Command Theorist,” Religious Studies 41 (2005), 1–22. Ward, Thomas. “The Incoherence of Ockham’s Ethics,” in M. Roques (ed.), Grounding in Medieval Philosophy (Leiden: Brill, forthcoming).

50  Thanks to Sonja Schierbaum for the conversation that inspired this paper.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

On Loving God Contrary to a Divine Command  245 William Ockham. Opera philosophica, ed. G. Gál et al. (St. Bonaventure, NY: Franciscan Institute, 1967–86). William Ockham. Opera theologica, ed. G. Gál et al. (St. Bonaventure, NY: Franciscan Institute, 1974–88). William Ockham. Quodlibetal Questions, tr. A. J. Freddoso and F. E. Kelley (New Haven: Yale University Press, 1991). William Ockham. A Short Discourse on Tyrannical Government, ed. A.  S.  McGrade, tr. John Kilcullen (Cambridge: Cambridge University Press, 1992).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts Suárez on Exemplar Causes Dominik Perler

1.  Introduction: Two Problems Suppose you have saved some money to build a house, and now you sit down with an architect to think about what the house should look like. You make a detailed plan indicating all the rooms and their location, and then you use this plan to build the house. Clearly, the plan does not exist only on paper. First and foremost, it exists in your mind; in fact, it is because of its existence in your mind that you are able to draw a plan on paper. Moreover, the plan in your mind does not appear just for a short moment and then disappear without any consequences. Rather, it is something that persists and makes you active: it incites you to contact a construction company, to talk to your bank about the financing, and to actually build the house. Clear and simple as this story looks at first sight, it gives rise to at least two deep problems. The first concerns the intentionality of the plan in your mind. How can it be directed at the house and present it as a build­ ing with a specific structure, distinguishable from other buildings? After all, the house does not yet exist. Consequently, the plan cannot be directed at the house in virtue of standing in a similarity relation to it, since there is simply no material thing to which it could be similar. Nor can the plan be directed at the house in virtue of a causal relation; since the house has not yet been built, it cannot have acted upon you. How then can the plan be about the house without an appropriate relation? Does it magically point towards the future house? Dominik Perler, Productive Thoughts: Suárez on Exemplar Causes In: Oxford Studies in Medieval Philosophy Volume 9. Edited by: Robert Pasnau, Oxford University Press. © Dominik Perler 2021. DOI: 10.1093/oso/9780192844637.003.0007

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  247 The second problem concerns the efficacy of the plan in your mind. The plan seems to exert some kind of causal power over you so that you engage in a number of activities. But how can that be? After all, the plan as such does not have any intrinsic force or power. It is simply a design of a building that presents it with many distinctive features, and there can very well be such a presentation without any activity following from it. For instance, when you are daydreaming and imagining a fancy ­castle, you also have a presentation of a building with a distinctive structure, but this presentation does not have any practical consequences: it does not incite you to build a castle. Why then does the plan of the house make you become active? Why does it have a causal power that is missing in other cases of mental presentation? As scholastic philosophers clearly noticed, these two problems need to be closely examined. It is not enough to observe that we have pro­ duct­ive thoughts, that is, thoughts that guide us in our actions and make us produce the things that are mentally present. We also need to give an  account of why and how this is possible. Many medieval authors attempted to do so in debates on divine ideas. Inspired by Augustine, they asked how God could create the world by having ideas of all crea­ tures.1 In what sense were his ideas directed at creatures before their creation? And how did these ideas incite God to really create them? An examination of these questions served as a springboard to an analysis of the human case: for if one understands how divine ideas can be about something that does not yet exist and why they give rise to an act of creation, one has a model for explaining human ideas.2 1  Augustine’s influential Quaestio de ideis (in his De diversis quaestionibus octoginta tribus, q. 83) shaped debates up to the seventeenth century. For helpful overviews, see Lambert M. De Rijk, “Quaestio de ideis: Some Notes on an Important Chapter of Platonism,” in J. Mansfeld and L. M. de Rijk (eds.), Kephalaion: Studies in Greek Philosophy and Its Continuation (Assen: Van Gorcum, 1975), 204–13; Olivier Boulnois, Jacob Schmutz, and Jean-­Luc Solère (eds.), Le contemplateur et les idées: modèles de la science divine, du néoplatonisme au XVIIe siècle (Paris: Vrin, 2002); Jacopo Falà and Irene Zavattero (eds.), Divine Ideas in Franciscan Thought (XIII– XIVth Century) (Canterano: Aracne Editrice, 2019). For a thorough analysis of late scholastic debates, see Tobias Hoffmann, Creatura intellecta: Die Ideen und Possibilien bei Duns Scotus mit Ausblick auf Franz von Mayronis, Poncius und Mastrius (Münster: Aschendorff, 2002). 2  To some extent the analysis of human ideas was also inspired by Augustinian debates about the inner word (verbum), which was supposed to be a mental state that presents an external thing, in some cases even a thing that does not yet exist. On this debate, see Claude Panaccio, Le discours intérieur: de Platon à Guillaume d’Ockham (Paris: Seuil, 1999); Joël Biard (ed.), Le langage mental du moyen âge à l’âge classique (Louvain: Éditions de l’institut supérieur de philosophie, 2009).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

248  Dominik Perler Admittedly, human ideas do not fit the divine model perfectly. For one thing, they are present in minds that are neither eternal nor fully independent of matter, and hence are subject to restrictions that do not apply to God. Nevertheless, the divine model can be used as a starting point for an analysis of the problems of intentionality and efficacy in the human case, for we humans do not produce thoughts only after we have been in touch with material things, as is the case with thoughts based on perception. Just like God, we are also able to do so before we come into contact with material things, and even before these things come into existence. We are thus similar to the divine creator, inasmuch as we have thoughts that make us produce things. Francisco Suárez was fully aware of this similarity. Looking back at a long Augustinian debate, he evaluated the positions taken in this debate not only in order to explain the structure and the functioning of divine ideas. He was equally interested in human ideas; he also wanted to know how human beings could somehow reach out for things that do not yet exist and thereby trigger actions that make them exist. In his monumen­ tal Metaphysical Disputations, he devoted an entire disputation to this problem.3 Adopting traditional terminology, he called it the problem of “exemplar causes.” This is quite understandable, since this problem con­ cerns an exemplar, that is, an idea or template of a thing in the mind; it also concerns a cause, since it deals with something that gives rise to an act of production.4 What makes Suárez’s analysis of exemplar causes particularly intriguing is his attempt to integrate it into a comprehensive theory of intentionality and causation. In analyzing this special case, he wants to show how mental states can be about something and how they can have an effect on other mental states, as well as bodily states. In par­ ticular, he intends to clarify how this type of causation can be accounted for within an Aristotelian framework. 3  He opens the disputation by stating that ideas are primarily in the divine mind, but sec­ ondarily also in human minds; see Disputationes Metaphysicae (= DM) 25.1.4. (The first num­ ber refers to the disputation, the second to the section, the third to the paragraph in his Opera omnia. At present there is no English translation of DM 25. The translations from this text and all other Latin texts are mine.) 4  Suárez concedes that one can also talk about an external exemplar, that is, a material model that is used as a blueprint for other material things to be made. But every external exem­ plar presupposes an internal one, because one cannot construct the material model unless one first has an idea of it in the mind (DM 25.1.2).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  249 In recent research, much work has been done to elucidate Suárez’s general notion of causation and his theory of the four Aristotelian ­causes.5 But surprisingly, his analysis of exemplar causes has been almost completely neglected.6 In what follows, I intend to fill this gap by first looking at Suárez’s critique of traditional theories of exemplar causes (section 2). I will then examine his own theory, spelling out his account of the intentional character of exemplar causes and explaining how this account relates to his general cognitive theory (section  3). Next, I will explore how he accounts for the causal aspect, and here again my aim will be to explain how this account fits into his general theory of causation (section  4). Finally, I will sketch out some conse­ quences of Suárez’s analysis of exemplar causes for the characterization of thoughts—consequences that go beyond the narrowly defined debate about ideas (section 5). Hopefully, this will make clear that it is worth paying attention to the special case of exemplar causes, since it sheds light on Suárez’s broader picture of cognitive activity.

2.  The Critique of Objective Existence Theories Suárez develops his theory of exemplar causes by first presenting and evaluating alternative theories. It is important to look at his critique of these theories, which can be traced back to thirteenth- and fourteenth-­ century debates, in order to see what kind of analysis he takes to be problematic and what impasses he wants to avoid. The the­or­ies he 5  See, for instance, Vincent Carraud, Causa sive ratio: la raison de la cause, de Suarez à Leibniz (Paris: Presses Universitaires de France, 2003); Dennis Des Chene, “Suárez on Propinquity and the Efficient Cause,” in B.  Hill and H.  Lagerlund (eds.), The Philosophy of Francisco Suárez (Oxford: Oxford University Press, 2012), 89–100; Tad M. Schmaltz, “Efficient Causation: From Suárez to Descartes,” in T. M. Schmaltz (ed.), Efficient Causation: A History (Oxford: Oxford University Press, 2014), 139–64; Jakob Leth Fink (ed.), Suárez on Aristotelian Causality (Leiden: Brill, 2015); Jacob Tuttle, “Suárez’s Non-­Reductive Theory of Efficient Causation,” Oxford Studies in Medieval Philosophy 4 (2016), 125–58; Jacob Tuttle, “Suárez on Creation and Intrinsic Change,” American Catholic Philosophical Quarterly 93 (2019), 29–51. 6  The only study I am aware of is Michael Renemann, Gedanken als Wirkursachen: Francisco Suárez zur geistigen Hervorbringung (Amsterdam: Grüner, 2010); for a short English version, see “The Mind’s Focus as an Efficient Cause: Francisco Suárez’s Re-­interpretation of the Traditional Understanding of the Idea,” American Catholic Philosophical Quarterly 84 (2010), 693–710. However, this study deals just with the intentionality problem and looks more at Suárez’s sources than at his own theory.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

250  Dominik Perler discusses mostly deal with the intentionality problem. In particular, they try to explain what specific feature of an idea, be it divine or human, makes it directed at a future material thing.7 The first theory that Suárez discusses explains intentionality in terms of an “objective concept” (DM 25.1.6). The exemplar cause or idea is taken to be a concept that exists as an object in the mind, and which is identical to the future material thing. Thus, when a craftsman thinks about an object that he wants to produce, “it is the same thing, which is to be produced by his art, that objectively exists in the mind of the craftsman” (DM 25.1.6). This theory clearly appeals to a special kind of existence a thing can have in the mind. Its core thesis can be sum­mar­ ized as follows: (1) An idea in the mind is about x because x has objective existence in the mind. The crucial point is that the very same thing that is to be produced also exists in a special way in the mind. There is no doppelganger in the mind, that is, there is no internal thing that merely corresponds to or refers to an external thing; hence there is no gap that needs to be filled between an internal and an external thing. Whenever one has an idea of x, x itself is present in the mind. How exactly is this thesis to be understood? One can make sense of it by looking at Thomas Aquinas’s theory of intentionality, which Suárez explicitly mentions as a possible source (DM 25.1.6). In some famous passages, Aquinas holds that one can think about a thing because the  mind assimilates it so that the thing comes to exist in the mind.8 The thing then has a double existence: it exists both inside and outside the mind. This means that there is a single thing with two modes of exist­ ence: immaterial existence in the mind and material existence outside the mind. Objective existence is simply immaterial existence in the 7  Note that Suárez uses the term ‘idea’ only for thoughts that are directed at creatable things. He explicitly says that ‘idea’ means the same as ‘exemplar’ (DM 25, preface). Unlike later ­thinkers, most famously Descartes, he does not take every thought to be an idea. I will follow him in this restrictive use. 8  See Thomas Aquinas, Summa theologiae (= STh) I, q. 14, art. 2, corp.; STh I, q. 85, art. 2, ad 1; Sentencia libri De anima III, ch. 7.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  251 mind.9 Of course, this appeal to double existence is far from harmless. It gives rise to the questions of how a thing can come to exist in the mind, and what the special existence in the mind consists in. However, these questions can be put aside.10 What matters here is the main thrust of Aquinas’s theory. His theory explains intentionality by positing a special mode of existence and avoids the introduction of mental representa­ tions. We can think about an external thing, this theory maintains, not because we have produced an inner representation that we apprehend and that somehow leads us to the external thing, but rather because we apprehend the thing itself, which exists in a special way in our mind. Is this a convincing account of the idea a person has when she thinks about something she wants to produce? Suárez is skeptical, and presents several objections. Let me focus on two of them. The first concerns the assumption that this person assimilates a thing in such a way that it comes to exist in her mind. If this were the case, Suárez argues, her mind would really become like that thing. Thus, a person assimilating a house would eventually have a house-­like mind. Of course, her mind would not turn into a material house, since the house would be in her mind only in an immaterial way; nevertheless, the immaterial form of the house would have to be assimilated in such a way that the mind would be somehow shaped by it. But this is not the case, as Suárez points out with respect to the divine mind. For when God thinks about future creatures, his mind is not shaped by them. Rather, it “is completely dissimilar to creatures” (DM 25.1.15) and remains a unique form. The same is true for human minds (DM 25.1.17): when they think about a house, they remain what they are, that is, distinctive forms with their own capaci­ ties, and they are not transformed. It would therefore be misleading to

9  Aquinas himself calls it “spiritual existence” rather than “objective existence”; see STh I, q. 78, art. 3, corp., and Quaestiones disputatae de veritate, q. 3, art. 1, ad 2. 10  For relevant discussion, see Robert Pasnau, Theories of Cognition in the Later Middle Ages (Cambridge: Cambridge University Press, 1997), 195–219; Dominik Perler, Theorien der Intentionalität im Mittelalter (Frankfurt am Main: Klostermann, 2002), 31–80; Jeffrey E. Brower and Susan Brower-­Toland, “Aquinas on Mental Representation: Concepts and Intentionality,” The Philosophical Review 117 (2008), 193–243; Han Thomas Adriaenssen, Representation and Scepticism from Aquinas to Descartes (Cambridge: Cambridge University Press, 2017), 13–39; Therese Scarpelli Cory, “Knowing as Being? A Metaphysical Reading of the Identity of Intellect and Intelligible in Aquinas,” American Catholic Philosophical Quarterly 91 (2017), 333–51.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

252  Dominik Perler say that the thinking mind becomes the thing it thinks about.11 Rather, the mind changes only insofar as it produces acts of thinking and per­ haps also a number of cognitive devices that make these acts possible. But it always keeps its own form, remaining dissimilar to all the things it thinks about. This objection shows that Suárez insists on a strict division between internal and external forms.12 Whatever it is that the mind does, it can­ not assimilate external forms and thereby change itself. But the prob­ lems with the theory run even deeper, as Suárez’s second objection shows. The assimilation theory was originally developed to explain the intentionality of thoughts that are based on perception. In this case it may make sense to speak of assimilation, since the perceiver is affected by a material thing, undergoes a change, and thereby receives that thing with immaterial existence. But in the case of the idea of a house in the builder’s mind, there is no material thing to start with. How then can a thing with immaterial existence be caused? Suárez dryly remarks that “one and the same thing cannot be the true and proper cause of itself ” (DM 25.1.22). A thing with immaterial existence cannot cause itself. All it can eventually cause is some other thing distinct from itself, but to do that it first needs to come into existence. And it remains a mystery how that happens if there is no process of affection and assimilation.13 In light of this difficulty, one could give an account of exemplar causes that does not invoke a double existence. Suárez presents such an account as the second theory of ideas. According to this theory, “there is an 11  This is the claim advocates of the assimilation theory made, quite often with reference to Aristotle’s authority; see De anima II.5 (418a3–6). On the widespread medieval presence of this  claim, see José Filipe Silva and Christina Thomsen Thörnqvist (eds.), Assimilation and Representation in Medieval Theories of Cognition, Vivarium 57:3–4 (2019). 12  It also shows that Suárez no longer accepts the basic idea that cognition requires assimila­ tion, although he still uses traditional vocabulary. In his analysis of standard cases of cognition, which are based on perception, he criticizes the assimilation model, claiming that the mind needs to form representations. See De anima, disp. 9, q. 2. For a detailed analysis, see Dominik Perler, “Súarez on Intellectual Cognition and Occasional Causation,” in D. Perler and S. Bender (eds.), Causation and Cognition in Early Modern Philosophy (London: Routledge, 2020), 18–38. 13  Of course, proponents of the double existence theory could respond that the thing with immaterial existence is based on earlier perceived things. Hence, they could say, there is an indirect causal relation that explains the existence of that thing. However, in making this claim they would give up the strict identity claim, for the thing in the mind would not be identical to any of the earlier perceived things. It would, as it were, only borrow elements from these things, and there would be at best a partial identity with each of them.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  253 objective concept not of the same thing, but of a different thing in imita­ tion of which another thing is to be made” (DM 25.1.6). This means that there is not just one thing that exists both inside and outside the mind. Rather, there are two things, namely, (i) a thing in the mind and (ii) another thing outside the mind that will eventually be produced and that stands in a similarity relation to the first thing. The main thesis of this theory can be summarized as follows: (2) An idea in the mind is about x because it is a mental thing to which x is similar. This theory avoids the thorny problem of double existence, since it posits two distinct things. It also avoids the problem of external affec­ tion. Since the idea is different from the material thing, it can very well exist before the material thing comes into existence and before that thing affects the mind. It can even exist without there being any material thing (e.g. when someone fails to build the house she has planned). Finally, this theory has the advantage that it does not posit a radical change or transformation of the mind. All it claims is that the mind undergoes a change insofar as it produces special things. But the mind remains a ­distinctive form that does not become like the things it produces. This second theory seems to present a successful solution to the prob­ lems posed by the first one. Suárez hesitates to attribute it to a specific author, especially with respect to ideas in the human mind,14 but a good example of a proponent of such a theory is Peter Aureol. Aureol posits that there are special things in the mind, which he takes to be different both from external things and from mental acts. He calls them “things with objective existence” or “intentional existence,” holding that they are the first things the mind produces and apprehends.15 Thus, when the 14  Suárez claims that Cajetan and “many Thomists” subscribe to it with respect to the divine mind but hastens to add that “with respect to the human exemplar the cited authors do not speak in such an explicit way” (DM 25.1.7). This hesitation is quite understandable because many sixteenth-­century authors spoke about things with objective existence but committed themselves to different metaphysical theories about the status of these things. For an overview, see Marco Forlivesi, “La distinction entre concept formel et concept objectif: Suárez, Pasqualigo, Mastri,” Les Études Philosophiques 60 (2002), 3–30. 15  See Peter Aureol, Scriptum super primum Sententiarum, dist. 3, sect. 14, ed. Buytaert (vol. II, 696–8 and 712–15); Scriptum dist. 27, pars 2, art. 2, ed. Friedman, 10–12. For an analysis, see

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

254  Dominik Perler house builder has a mental plan, she immediately apprehends a house in her mind—a special entity that exists before the house is built. Since it is a model for the material house, that house will be similar to it when it gets built. Is this a convincing theory? One immediate objection that could be made is that as long as it remains unclear how the special thing with objective existence comes into existence, this theory is hardly a compel­ ling one. But this is not the problem that Suárez points out, presumably because he takes it for granted that God can create something with objective existence out of his own essence, and that a human being can produce it by using and combining elements of previously perceived things (e.g. by using elements of houses one saw earlier and stored in the mind). The real problem Suárez sees does not concern the production of that thing, but rather its epistemic function. For as soon as one claims that it is this thing that is immediately apprehended, it is assigned the function of the first cognitive object. The external thing will then be only the secondary object, that is, an object that is cognized through or even after the first object. This is exactly the point that Suárez takes to be unacceptable: But a created craftsman cognizes directly and immediately the art­ work, and he produces it by means of this cognition. Therefore, there is no third thing that could have the function (ratio) of an exemplar. (DM 25.1.18)

Let me flesh out this critique using the example of the plan for a house.16 Suppose you think very hard about all the rooms, the windows, and Perler, Theorien der Intentionalität, 283–94, and Adriaenssen, Representation and Scepticism, 82–99. One might also mention John Duns Scotus, who in his Ordinatio referred to things with objective existence that are produced by the mind and exist in it. However, in his later theory he defended a reductionist account and identified these things with mental acts; see Richard Cross, Duns Scotus’s Theory of Cognition (Oxford: Oxford University Press, 2014), 189–95. Since Suárez has a non-­reductionist account in mind, Aureol’s theory is a better ex­ample. Note that it is controversial among recent commentators how Aureol’s theory is to be understood. In presenting it as a theory that invokes a special internal thing, I follow an interpretation that was already adopted by medieval commentators (e.g. Ockham) but that is not uncontested. For a concise discussion of various interpretations, see Robert Pasnau, After Certainty: A History of Our Epistemic Ideals and Illusions (Oxford: Oxford University Press, 2017), 273–7. 16  When introducing his critique in DM 25.1.18, Suárez presents two ways of understanding the “third thing,” namely as an external thing (e.g. a material house in imitation of which

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  255 many other details of the house. When you do that, you are not thinking about a mental house with mental rooms, mental windows, etc. Rather, your thought is directed at the real house that is intended to have rooms built with bricks and windows made of glass. That is, you do not turn towards an internal object. Instead, you are directed at an external thing, even if this thing does not yet exist. Your thought is like an arrow targeting a goal in the material world, regardless of whether it will ever reach this goal. It would therefore be erroneous to posit an internal thing that serves as a “third thing,” that is, something that stands between your mind and the external thing. There is no such thing; indeed, if there were such a thing it would prevent you from targeting the house to be built. The material house would then be hidden, as it were, behind the mental house, and you could apprehend it only by thinking that there will eventually be something corresponding to the mental house that is present to you. In mentioning a suspicious “third thing,” Suárez presents the classic objection to a representationalist theory that posits an internal repre­ sentation as the first cognitive object and thereby denies that we have immediate access to an external thing. This objection is not new. It can be found in Ockham, for example, who presented it against Aureol’s theory of things with objective existence, and in other fourteenth-­ century critics of this theory.17 What is remarkable here is that Suárez does not only raise this objection when discussing cases of perception.18 Of course, in his theory of perception he also rejects the thesis that when seeing something we just have direct access to a special thing with ob­ ject­ ive existence; he thinks that an act of seeing is immediately

another house is built) or an internal thing. But he quickly dismisses the first option, noting that in many cases there simply is no external thing. I therefore focus on the second option. 17  See William of Ockham, Ordinatio I, dist. 27, q. 3, Opera philosophica et theologica (= OPh and OTh), OTh IV, 238–58. For other critics, see Katherine  H.  Tachau, Vision and Certitude in the Age of Ockham: Optics, Epistemology and the Foundations of Semantics 1250–1345 (Leiden: Brill, 1987), and Pasnau, After Certainty, 256–64. 18  This is the context in which Aureol introduces objects with objective existence, primarily in order to explain cases of sensory illusion, for in these cases the thing we immediately per­ ceive and think about does not match the thing that exists in the material world. See Peter Aureol, Scriptum, dist. 3, sect. 14, ed. Buytaert (vol. II, 696–7), and an analysis in Dominik Perler, “Can We Trust Our Senses? Fourteenth-­Century Debates on Sensory Illusions,” in D. G. Denery II, K. Ghosh, and N. Zeeman (eds.), Uncertain Knowledge: Scepticism, Relativism, and Doubt in the Middle Ages (Turnhout: Brepols, 2014), 63–90.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

256  Dominik Perler directed at an external thing.19 But the important point is that he defends direct realism not just with respect to perception. On his view, this p ­ os­ition is also to be defended with respect to ideas. Even in a case where a ma­ter­ial thing is yet to produced and eventually perceived, it is wrong to posit a special thing in the mind that serves as the immediate cognitive object, for even in this case the mind can reach out to something outside itself and make this thing its immediate object. Metaphorically speak­ ing, it can send off an arrow and target something in the material world. A representationalist theory that denies this must be rejected, for it ­limits the scope of our thoughts to things inside the mind. Moreover, it merely mentions a similarity relation between these things and the material things that will eventually be produced, but neglects to explain what this similarity amounts to and how we could ever notice that there is indeed a similarity. How is that possible if we lack direct access to material things and if we cannot compare them with the mental things that are present to us? We are, as it were, imprisoned in the mind with its own things and cannot look to external things. Suárez does not confine himself to presenting this well-­known argu­ ment against representationalism. He adduces still another argument that emphasizes the dynamic character of our thoughts about future things. This is in fact a distinctive feature that sets them apart from many other thoughts: Second, I argue that a human craftsman never conceives beforehand in its completeness and as an individual the thing that he will produce, and later does produce, because he can never know with certainty what the future thing to be made by his art will be and how it will be as an individual, for the termination of an effect in an individual depends on various circumstances or on a superior agent.  (DM 25.1.21)

The crucial point is that the initial thought does not match the future thing although it is really directed at it. It rather undergoes many 19 See De anima, disp. 8, q. 2 (vol. III, 46–8). For a detailed analysis of his perceptual real­ ism, see James B. South, “Suárez and the Problem of External Sensation,” Medieval Philosophy and Theology 10 (2001), 217–40.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  257 changes until it matches it. Let me use again the house example to illus­ trate this point. Suppose you have a detailed plan of the house you want to build. Unfortunately, you are running out of money when you start building it and therefore need to make changes. The house you end up building is much smaller and more modest than what you originally planned. This means that you cannot simply use the plan to construct a house that perfectly matches the one you thought of. The same is true for many other things we produce: the end product often differs from the original plan due to external circumstances or other people who intervene and prevent us from doing what we originally wanted to do.20 How is this relevant for evaluating the theory that appeals to special internal things? This theory assumes that we first apprehend an internal thing and then produce an external copy of it, but in making this assumption it neglects the fact that the external thing quite often fails to be a true copy. When working on something, we often make adjust­ ments and do not simply create a replica of something in our mind. Hence, there is no strict similarity relation between an internal and an external thing. In fact, we do not even aim at establishing such a rela­ tion. We are rather from the beginning directed at an external thing and try to produce it by taking into account various constraints. But no mat­ ter how many changes and adjustments we make, we remain focused on an external thing. To use the arrow metaphor again, one could say that we often redefine the goal of the arrow, but we always target an exter­ nal goal. Would proponents of the theory that posits internal things accept this objection? Probably not. They could concede that there is a dynamic process in which we change our plan and redefine the thing we want to produce, but this change, they could say, simply amounts to a new pro­ duction of internal things: every time we change our plans due to exter­ nal circumstances, we produce a new internal thing. The material thing 20  Of course, one might wonder if the initial and the final thought are always about the same object, despite all the changes that occur. Suppose you first think about a luxurious villa but then end up building a small cottage. Is your first thought then really about the thing you finally produce? Suárez is not explicit, but he would presumably concede that the objects of the initial and the final thought can differ. However, this is only the case if a new decision is made, for instance if you decide to build a cottage instead of a villa. Without a new decision, the ini­ tial object is adjusted but not radically changed.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

258  Dominik Perler we eventually produce will be in accordance with the last in a long series of internal things. No doubt, this would be a possible defense. However, it would require that there be not just one “third thing,” but a series of things, each of them standing in a similarity relation to a possible future thing. It is exactly this kind of ontological proliferation that Suárez intends to avoid. On his view, it is superfluous to posit a series of internal things as long as there is a mental act that is directed at an external thing. The person producing this act needs to be flexible, like the archer who needs to be flexible when the wind changes. Nevertheless, it is an act that this person produces, not an internal thing.

3.  The Act Theory It has become clear so far that Suárez rejects any appeal to an internal thing, be it identical to or distinct from the external thing, in order to explain what an idea is. How then are ideas to be understood? Suárez presents his own view by calling the idea a “formal concept,” or some­ thing that “exists subjectively and formally in the intellect” (DM 25.1.9 and 25.1.26). This means that the idea exists in the mind as in a subject; hence it has subjective existence. And what has this kind of existence is an act, for an act is a quality of the mind that exists in it as in a subject. When introducing the technical term ‘formal concept,’ Suárez explicitly states: “The formal concept is said to be the act or, what is the same, the word (verbum) by which the intellect conceives of a thing or a common aspect” (DM 2.1.1).21 An idea is thus nothing but the act by which some­ thing is conceived, and this act is immediately directed at an object. The basic thesis Suárez defends can be summarized as follows: 21  See also DM 25.1.37 and the characterization of the mental word as “the act by which the thing is conceived and through which the craftsman operates” (De anima, disp. 5, q. 5, n. 23; vol. II, 396). Suárez insists that the mental word is not a special internal object, and thereby confirms his rejection of the objective existence theory. Metaphysically speaking, there is noth­ ing but an act, which is a quality existing in the intellect. For details, see Daniel Heider, “Suárez’s Metaphysics of Cognitive Acts,” in R. A. Maryks and J. A. Senent de Frutos (eds.), Francisco Suárez (1548–1617): Jesuits and the Complexities of Modernity (Leiden: Brill, 2019), 23–45.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  259 (3) An idea in the mind is about x because it is an act that is directed at x. At first sight, this thesis does not look very informative, since it says nothing about the specific features that make an act be directed at a thing; the “aboutness” is simply rephrased as “directedness.” It will therefore be important to look at the way Suárez explains the act’s directedness. But the salient point here is that he mentions nothing but the act that is directed at a thing. He thereby subscribes to what could be called an act theory.22 This becomes clear if one looks more closely at the way he contrasts the act, that is, the formal concept, with the objective concept. On his view, there is indeed an objective concept, but it is not the kind of con­ cept that proponents of the objective existence theory postulate; that is, it is not a special internal thing that exists in addition to the act in the mind.23 Suárez emphasizes that the objective concept is “the thing or aspect (ratio) that is properly and immediately cognized or represented by the formal concept” (DM 2.1.1). This thing, he hastens to add, is not inside the mind, but outside. He adduces a telling example to illustrate the difference between the formal and the objective concept. When someone thinks about a human being, the act of thinking is the formal concept. Metaphysically speaking, it is a quality that inheres in the mind of the thinker. The objective concept, by contrast, is the human being who is thought of. To be sure, it is not the human being as such, but the human being insofar as he or she is the object of an act of thinking. This is why there needs to be an “extrinsic denomination” (DM 2.1.1) or as we would say today, an external relation that links the human being to 22  I borrow this label from Marilyn McCord Adams, William Ockham (Notre Dame, IN: Notre Dame University Press, 1987), 75, who introduced it to characterize Ockham’s final intellectio theory. However, Suárez’s act theory differs in some important points from Ockham’s theory, as will become clear in this section. 23  It is therefore important to distinguish Suárez’s own understanding of the objective con­ cept from that of his opponents. Renemann, “The Mind’s Focus,” 707, neglects to make this distinction when he accuses Suárez of an equivocation. Suárez does not use the expressions ‘objective concept’ and ‘objective existence’ equivocally. He simply presents two different uses of these expressions: in DM 25 he reports his opponents’ use, and in DM 2 (as well as DM 54) he spells out his own use. For a concise presentation of the different uses, see Norman J. Wells, “Objective Reality of Ideas in Descartes, Caterus, and Suárez,” Journal of the History of Philosophy 28 (1990), 33–61 (especially 39–43).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

260  Dominik Perler the act of thinking. Nevertheless, it is the human being made of flesh and blood that is thought of, and not a mental copy. Hence it is this external thing and not an internal one that is the objective concept. This example makes clear that the objective concept can be a material thing that really exists. But Suárez is fully aware that other things, for instance products of the imagination, can also be thought of. He therefore emphasizes that an objective concept “is not always a true and positive thing” (DM 2.1.1). All that matters is that it is a thing that is thought of, whatever its ontological status may be.24 This contrast between formal and objective concepts shows that Suárez strictly follows his own rule that one should avoid introducing a “third thing” that stands between the thinking mind and the external world. All one needs to accept is (i) an act of thinking and (ii) an external target for this act, whatever that target may be.25 The act is immediately directed at this target, thus making it the object that is thought of. Now, it seems as if Suárez is in full agreement with Ockham, who also defends an act theory and rejects every kind of intermediary between act and external object. Ockham insists that there is no need for any mental entity in addition to the act. It is not only things with objective existence that are superfluous, but also intelligible species, since an act is by itself directed at an external object.26 In making this claim, Ockham applies his famous principle of parsimony, holding that one should not posit such intermediaries because they play no explanatory role: they do not help to make it intelligible why an act is about an object. 24  The thing thought of can be an impossible object as well as a possible one. Suárez is an ontological pluralist, as argued convincingly by Brian Embry, “Francisco Suárez on Beings of Reason and Non-­Strict Ontological Pluralism,” Philosophers’ Imprint 19:27 (2019), 1–15. In fact, Suárez accepts different types of things as objective concepts. What is common to all of them is that they are related to an act (or “grounded in it,” as Embry puts it), although not all of them have real existence. 25  In certain special cases, namely, when one thinks about one’s own act of thinking or about one’s own soul, the target is of course an internal thing. But Suárez emphasizes that reflection is required for these cases. See De anima, dist. 9, q. 5 (vol. III, 168–78), and for an analysis Dominik Perler, “Suárez on Consciousness,” Vivarium 52 (2014), 261–86. 26  This is the main thesis of his later act theory; see Summa Logicae I, cap. 12 (OPh I, 42–3), and Quodl. IV, q. 35 (OTh IX, 474). On his rejection of intelligible species, see Reportatio II, qq. 12–13 (OTh V, 268–76). For detailed discussion, see Perler, Theorien der Intentionalität, 322–42, and Claude Panaccio, Ockham on Concepts (Aldershot: Ashgate, 2004), 23–31.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  261 Suárez defends a less radical view than Ockham. He thinks that in­tel­ li­gible species are required in addition to the act.27 These entities play an important role, because human beings cannot bring about an act out of nothing. Unlike God, who is self-­sufficient and produces every act out of his own essence, we are not self-­sufficient beings. We always need some working material out of which we form an act and give it a ­dis­tinct­ive content. Intelligible species, which are cognitive devices in the mind, have exactly this function: they make the fixing of an act’s content possible. However, Suárez points out that “an intelligible species is not an exemplar in the strict sense” (DM 25.1.30). It is only that which makes it possible to produce an act that will be directed at a well-­defined object. It does the preparatory work as it were, and thus presents “only the seed of an object” (DM 25.1.27). But as long as one does not reflect upon the way one produces an act, the species is not the object that is targeted by the act. Let me spell out this important point with the house example. Suppose you start thinking about the house you want to build. You can­ not do so unless you know in principle what a house looks like and what its basic features are. To gain this knowledge, you need to see a number of houses so that you have a number of house descriptions at your dis­ posal. These descriptions are a prerequisite for thinking now about the future house, but they are not what you are thinking about. After all, you are not thinking about the house descriptions you have acquired in the past; rather, you are using them as the raw material for thinking about your own house. This material must not be conflated with the object of your act of thinking. This way of defending intelligible species shows that Suárez does not endorse Ockham’s radical act theory, for he thinks that it fails to explain on what basis an act can get a well-­defined content. On his view, it is not enough simply to appeal to visual images that cause an act and thereby serve as the basis. After all, sensory images are material items, whereas a mental act is immaterial. How should it be possible for a material item to immediately cause an immaterial item and to give it a content? An 27  Suárez defends the existence of intelligible species at great length in De anima, disp. 5, qq. 3–4 (vol. II, 340–68) and disp. 9, q. 3 (vol. III, 106–52).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

262  Dominik Perler immaterial act cannot arise unless there is some immaterial ground for it, and the function of intelligible species is precisely to provide this ground. They somehow translate the content of the material images into an immaterial format and thereby provide the basis on which an act with a well-­defined content can be produced.28 This is why they are ne­ces­sary for forming an act, but in a normal situation they never serve as its object. The object is rather the external thing. But how can an act target an external thing? Since that thing does not yet exist, there can be no direct causal relation that makes the targeting possible. Suárez is fully aware of this problem and therefore refrains from appealing to a causal relation. He speaks rather about something internal to the act that makes it intentional. When introducing the tech­ nical term ‘formal object,’ he states that an act is a formal object because “it is the ultimate form of the mind, or because it formally represents a known thing to the mind” (DM 2.1.1). And when defending the claim that no internal thing in addition to the formal concept (i.e. the exem­ plar or the act) is required, he says: “An exemplar is not constituted as something that is formally and precisely an exemplar because it is known as such, but because of its proper and, so to speak, intrinsic being” (DM 25.1.32, similarly 25.1.37). Admittedly, these technical statements are not easy to understand, but at least one point is clear: it is because of its form or its inner constitution (its “intrinsic being”) that an exemplar presents something to the mind. Using modern terminology, one could say that it is the inner structure that matters.29 It is precisely this structure that gives the act a distinctive content and thereby makes it directed at an object, whether or not that object ever exists.

28  In fact, intelligible species make the content of the sensory images available “in some spiritual way,” as Suárez claims; see De anima 9.2.17 (vol. III, 104). Using modern terminology, one could say that they serve as the transducers that map the content of material images onto an immaterial level. On Suárez’s account of the transduction, see Perler, “Suárez on Intellectual Cognition,” 23–31. 29  Suárez compares the exemplar in the mind to a material model, saying about this model that “its being painted or figured in a particular way is the proper being of such an exemplar” (DM 25.1.32). Thus, it is its specific configuration or structure that makes a material model the very model it is. Similarly, it is its specific structure that makes an exemplar in the mind the very exemplar it is.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  263 This appeal to an inner structure makes clear why Suárez takes an act to be intentional. The main thesis of his act theory, which was first stated in a rudimentary form, can now be expanded and rephrased as follows: (3*) An idea in the mind is about x because it is an act that is directed at x in virtue of its inner structure. This way of accounting for the intentionality of an idea may still look quite vague. Let me spell it out with the house example. Suppose you think very hard about how your future house should look and decide that it should have five rooms (feature A), a pointed roof (feature B), and a large terrace (feature C). When you do this you assign three character­ istic features to your house and produce an act of thinking that has exactly these features as its content. That is, you form an act with the set {A, B, C} as its content. Should you think about a different house, say one that also has five rooms but a flat roof (feature D) and a balcony instead of a terrace (feature E), you would produce an act that has a dif­ ferent set of features as its content, namely, {A, D, E}. Each of these acts would have its own content, and the two would be clearly dis­tin­guish­ able from each other, whether or not they are ever related to a ma­ter­ ial house. I hope this example makes clear in what sense an act can have an inner structure.30 I also hope it illustrates why Suárez does not refer to a similarity relation when explaining the intentionality of an act. Admittedly, he occasionally speaks about similarity and concedes that one can call an act the thing “in whose similarity the effect is made and which the craftsman imitates in his production” (DM 25.1.41). But the crucial point is that it is not the similarity relation that makes an act intentional. The act is already intentional before the craftsman produces a thing that is similar to it, simply because it has an inner structure. It 30  An appeal to an inner structure presupposes, of course, that an act can be complex. But complexity does not compromise its status as a single quality in the mind. What exists as one thing can have a complex form and hence a structure displaying many features. When explain­ ing Ockham’s act theory, Adriaenssen, Representation and Scepticism, 115, rightly emphasizes that an act can have “certain internal properties” that make it a distinctive act, distinguishable from other acts. Suárez seems to follow Ockham on this point by claiming that an act has an inner structure.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

264  Dominik Perler points by itself to an external thing, whether or not this thing ever exists. Thus, the act with the content {A, B, C} points to a house with five rooms, a pointed roof, and a large terrace, whether or not such a house will ever be built. Of course, once it is built with all three features it will perfectly match the idea; then there will indeed be a similarity relation. But similarity is not constitutive of intentionality: it comes into play after intentionality has been established.31 It is a mere consequence of the fact that whatever displays the same set of features as the act’s content must correspond to it and fully match it. In short, it is a consequence of structural isomorphism. The appeal to an inner structure has a further advantage. It enables Suárez to provide an elegant solution to the problem of the missing ­correspondence between the initial idea and the final product (see sec­ tion  2). How can it be that the house one ends up building does not match the original plan? This is possible because one can change the content of the original plan. Thus, one can start with the content {A, B, C}, then change it to {A, D, C} and eventually change it again so that one ends up with {A, D, E}. The material house that is finally built will have  this last set of features so that there will be a similarity relation with the last plan in the series of plans. Nevertheless, the very first plan was already intentional because it was from the beginning an act with a dis­tinct­ive content. Changing the original plan simply amounted to making adjustments in the configuration of the features that fixed the content. It is precisely this reconfiguration that is responsible for the dynamic character of the process of planning. Since the single features can constantly be recombined or replaced with others, it is always pos­ sible to adapt the plan to new circumstances (e.g. if one begins to run out of money while building the house) or to external interventions (e.g. if the city rejects the initial house plan). In any case, it is always a change of the inner structure and hence of something intrinsic that accounts for adjustments to the plan. Since the fixing of the content is the relevant issue here, it is quite understandable why Suárez takes intelligible species to be indispensable. 31  Similarity is a real relation that requires the existence of two relata, as Suárez emphasizes (DM 47.4.2). Therefore, it cannot exist if there is just the act as one relatum.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  265 As noted above, they prepare the ground for an act, and do so by provid­ ing features that can be combined in different ways. What is important here is that they precede the act and make the fixing of its content ­pos­sible. However, they do not play the role of an internal object; this is why introducing intelligible species does not pave the way for represen­ tationalism, as Ockham thinks.32 On the contrary, it explains why an act can target an external object. If there were no species, it would hardly be intelligible how an act can ever get a well-­defined content.33 Now, it seems as if Suárez is defending an act theory that differs from Ockham’s version of that theory but nevertheless agrees with it on an important point: once the act has been formed, it is by itself directed at an object. No special relation linking the act to an object is required. This is exactly the thesis Ockham defends, thereby subscribing to a non-­ relational theory of intentionality. He holds that an act is an “absolute thing,” i.e. a thing that is distinct from all other things and therefore not dependent on the existence of some other thing.34 Given this status, an act can be intentional even if there is no external relatum. For instance, when a young person thinks about her future grandchild (a possible object) or about a golden mountain (an impossible object), she has an intentional act in her mind even though there is no real thing to which her act could be related. This is the case because her act, an absolute quality, is intrinsically intentional. 32  Nor does it contradict our experience, as Ockham assumes when he says that “nobody intuitively sees a species; hence experience does not lead to this [sc. the introduction of spe­ cies]” (Reportatio II, qq. 12–13; OTh V, 268). Species are not internal objects we see or other­ wise apprehend (at least not as long as we do not reflect upon the production of a mental act). They play an indispensable role before we form an act, for they make possible its content and hence its directedness at an object. 33  This is the main reason why Suárez holds that the principle of cognition is not the bare intellectual power, but “the power informed by the species” (De anima, disp. 5, q. 4, n. 15; vol. II, 364). The bare power could not produce an act because it would not have anything at its disposal that would enable it to fix the act’s content. 34 See Ordinatio I, prol., q. 1, art 1 (OTh I, 39). Since an act is an absolute thing and not a relational one, God can create it even if no external thing exists. Thus, God can create an act presenting a star even if there is no star, as Ockham explicitly holds. The act thus presents the star in virtue of its own inner structure, not in virtue of some relation to a star. For an analysis of this case, which is a test case for a non-­relational theory, see Dominik Perler, Zweifel und Gewissheit: Skeptische Debatten im Mittelalter (Frankfurt am Main: Klostermann, 2006), 253–66, and Susan Brower-­Toland, “Causation and Mental Content: Against the Externalist Reading of Ockham,” in J.  Pelletier and M.  Roques (eds.), The Language of Thought in Late Medieval Philosophy (Cham: Springer, 2017), 59–80.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

266  Dominik Perler It is tempting to ascribe the same view to Suárez, since he insists that there is no special relation that makes an idea intentional. However, it would be incorrect to conclude from this that he rejects every relation. This becomes clear if one looks at his classification of relations.35 Like most scholastic authors, he first distinguishes real relations, which obtain between two really distinct relata (e.g. the relation between father and son), from mere conceptual relations, which obtain between two relata that are not really distinct (e.g. the relation of self-­identity). But he then introduces a further distinction within the realm of real relations, which is less common and deserves closer examination (DM 47.3.10–13). He claims that there are categorical relations, which are so called because they belong to the category of relation. These are the real relations one usually talks about, for instance the relations of similarity or equality. But in addition, there are also transcendental relations, which are so called because they go beyond (or literally transcend) the cat­egor­ic­al classification and can be found in entities that belong to other cat­egor­ies than that of relation. Thus, they can be found in the categories of sub­ stance and quality. What is characteristic of them, no matter where they are found, is that they are essential to the things in which they are present (DM 47.3.13); that is, they do not belong to these things because of their combination with something else, but simply because of their inner structure. Transcendental relations thus do not require the existence of two relata. A single relatum is sufficient: as soon as a thing with an inner structure comes into existence, a transcendental relation exists as well. At first sight, this kind of relation looks quite enigmatic. Does not every relation require at least two relata? Is it not like a bridge that needs to have a starting point and an end point? Or speaking technically, does not every relation require a foundation and a terminus? Suárez agrees that two relata are indeed required for a categorical relation (DM 47.8.4), but he insists that not every real relation needs to fulfill this require­ ment. To support his thesis, he presents the case of a power that is striv­ ing to produce something (DM 47.8.5). In this case, there is a relation 35  For a helpful reconstruction of Suárez’s classificatory scheme, see Jorge Secada, “Suárez on the Ontology of Relations,” in D. Schwartz (ed.), Interpreting Suárez: Critical Essays (Cambridge: Cambridge University Press, 2012), 62–88; Sydney Penner, “Suárez on the Reduction of Categorical Relations,” Philosophers’ Imprint 13:2 (2013), 1–24 (especially 3–7).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  267 between the power and its future product. But of course, the product does not yet exist; this is why there is no second thing that actually exists in addition to the power. There is nothing but the power, which, in vir­ tue of its inner structure, is directed at a future product. Using modern terminology, one could say that the power has a natural teleology that makes it goal-­directed.36 For instance, heat is a power that is directed at something hot it intends (in a non-­mental sense) to produce. What makes it a specific power, distinguishable from all other powers, is ­precisely this inner goal-­directedness, which exists whether or not there will ever be a hot thing. This is why there can be a transcendental rela­ tion without the existence of a second relatum.37 The same point applies to mental acts. When they come into exist­ ence, they have an inner structure and hence a content that makes them be directed at an object. Like heat or other natural powers they have a  goal-­directedness, whether or not the goal is ever reached. Suárez explicitly draws this parallel. Just after mentioning powers that have a transcendental relation in virtue of their inner structure, he says: For a similar reason one can say that an act of the intellect has a tran­ scendental relation to some being of reason, because this being can certainly be a sufficient object of such an act. For that reason it is not a problem for this relation that a being of reason is something fashioned by the intellect; the transcendental relation is in fact founded in it. This argument correctly proves that no other things besides the acts of the mind can have transcendental relations to beings of reason. It is by these acts that beings of reason are thought of or fashioned, and I include among them certain acts of the imagination insofar as im­agined and impossible things can be fashioned and represented by them. (DM 47.8.5–6)

36 Some contemporary metaphysicians would even call it “physical intentionality”; see George Molnar, Power: A Study in Metaphysics (Oxford: Oxford University Press, 2003), 60–5. 37  Secada, “Suárez on the Ontology of Relations,” 80, aptly remarks that the foundation has “an in-­built reference to the term, which it brings with it and possesses, so to speak, in virtue of its absolute character.” The crucial point is that this in-­built reference to the term is given even if the term does not exist. This is why the power of heat is directed at something hot even if no hot thing exists.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

268  Dominik Perler In this passage Suárez clearly states that a transcendental relation between the intellect and a being of reason is grounded in an act of the intellect and therefore has one relatum (the foundation) but lacks a sec­ ond relatum (the terminus). The intellect strives, as it were, for the sec­ ond relatum, although no such relatum really exists.38 The intellect can do so simply because it has an act with a distinctive content. The act alone suffices for the striving, or speaking technically, the act alone suf­ fices for a transcendental relation. It is important to note that this relation is not some special entity added to the act. Here again, Suárez draws a clear line between cat­egor­ ic­al and transcendental relations. In the case of categorical relations there is always “an accidental form that comes to (adveniens) a founda­ tion that is fully constituted in its essential and absolute being” (DM 47.4.3). Thus, fatherhood is an accidental form that comes to exist in a man who is fully constituted. It is added to that man as soon as his child is born, and it would be removed from him if his child were to die.39 By contrast, a transcendental relation is not an accidental form that is added at a certain moment and removed at some later moment. It is simply an “essential mode” (DM 47.4.15), that is, a way of being that a thing has in virtue of its inner structure. That thing always preserves this way of being as long as it preserves its structure, no matter what else exists. In mentioning a mode, Suárez appeals to an ontological resource that was not available to Ockham. Ockham accepted just substances and qualities as really existing entities. For Suárez, by contrast, there are also modes that really exist. However, he insists that modes are not add­ ition­al entities or things. They are rather inner features of the basic en­tities—features that always stick, as it were, to them and naturally

38  Suárez clearly denies that beings of reason have real existence. They are mere “shadows of beings” (DM 54, preface 1) and exist only insofar as they are targeted by an act of the mind. For an exposition of this thesis, see Christopher Shields, “Shadows of Being: Francisco Suárez’s Entia Rationis,” in B. Hill and H. Lagerlund (eds.), The Philosophy of Francisco Suárez (Oxford: Oxford University Press, 2012), 57–74. 39  On the status of categorical relations, see DM 47.2 and the discussion in Penner, “Suárez on the Reduction of Categorical Relations.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  269 belong to them.40 It is therefore not surprising that he takes the modes to be features that cannot be separated from the basic entities: they come into existence as soon as these entities come into existence, simply in virtue of their inner structure. This has a consequence for our understanding of the transcendental relation as a mode in the case of mental acts. Suppose you have an act with the content {A, B, C}. As soon as this act comes into existence, a transcendental relation will also exist as an essential mode of the act. The act will then be like an arrow pointing at an object with the features A, B, and C, even if this object does not exist: {A, B, C}→. The crucial point is that the transcendental relation is not some kind of extra entity that is added to your act and that could be removed at any time. Rather, the relation immediately emerges when the act with its distinctive con­ tent emerges, and remains in existence as long as the act exists. Given this insistence on a special relation, the core thesis Suárez defends needs to be further specified. It can now be put as follows: (3**) An idea in the mind is about x because it is an act that is directed at x in virtue of its inner structure, thereby grounding a transcen­ dental relation to x. Does that mean that Suárez endorses a relational theory of intentionality?41 It all depends on how one understands such a theory. If one takes it to be a theory according to which an intentional act is a rela­ tion, Suárez does not endorse a relational theory. On his view, an act is clearly a quality, not a relation. However, if one takes it to be a theory according to which an act is always accompanied by a transcendental relation, he does indeed endorse a relational theory. Unlike Ockham, who eliminated relations, Suárez does not think that there is nothing but an act as an “absolute thing”; on his view, the act always comes along with a relation. In fact, it must come along with a relation: the relation 40  They are real, as Suárez emphasizes, but do not constitute a separate category of things (DM 7.1.16–18). On their ontological status, see Tad  M.  Schmaltz, The Metaphysics of the Material World: Suárez, Descartes, Spinoza (Oxford: Oxford University Press, 2020), 41–63. 41  This is how Hamid Taieb, Relational Intentionality: Brentano and the Aristotelian Tradition (Cham: Springer, 2018), 100–2, interprets him.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

270  Dominik Perler follows the act like a shadow. But this relation does not constitute the act’s intentionality. It is rather the content that makes it intentional, for if the act displays an inner structure and hence a well-­defined content, it is directed at an object. This is true not just for ideas, but for all acts of thinking. All of them come along with a transcendental relation that follows them like a shadow. Ideas bring to the fore what is distinctive of all mental acts. They are all intentional because of their inner structure, not because of something external that is added to them. Thus, acts of perceiving are also directed at particular objects because of their inner structure, and this structure also gives rise to a transcendental relation. Of course, in that case the structure is determined by a causal relation to a material thing; hence the transcendental relation is ultimately based on a causal relation. But there can be an inner structure even if there is no causal relation, as the case of ideas shows. Consequently, there can be a tran­ scendental relation without a causal relation.42 All that matters is that the act has an inner structure, no matter how it comes about.

4.  Acts and Their Causal Function It should be clear by now that ideas are intentional acts. But why do they cause something? Why, for instance, does the idea of a house in your mind make you build a house? At first sight, the answer to this question seems obvious. An idea always results from a process of deliberation about what one should do. Once one has completed this deliberation, one has fixed the goal of an action, and it is precisely this goal that an idea indicates. This is why an idea has right from the start an action-­ guiding function. Suárez emphasizes this point by saying that “the func­ tion (munus) of the proper and inner exemplar which we are discussing is to direct the action of an agent in its specification” (DM 25.1.25). 42  To be precise, there need not be a direct causal relation. Since intelligible species are required for the fixing of an idea’s content, there must be an indirect relation. After all, no in­tel­ li­gible species can be in the mind without earlier perceptions (and hence causal relations) to material things. However, once the species are available, the content of an idea can be fixed inside the mind without there being an additional causal relation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  271 Thus, the function of the house idea in your mind is to specify the action you want to bring about: it should make you produce your desired house and nothing else. However, this general statement leaves open how an idea can make it that the house really comes into existence. Is the mere presence of the idea enough, or is something more required? Suárez acknowledges that this problem needs to be addressed, for unlike God, who is always fully active and hence fully productive, we human beings need to become active at a given moment. It is therefore important to give an account of why and how we become active. Otherwise the transition from merely thinking that a certain object should be produced to really producing it remains a mystery. One way of explaining this transition would be to appeal to a reflexive act. Suárez attributes such an explanation to Aquinas (DM 25.1.34–5), claiming that according to him an idea does not cause anything unless one reflects upon it and thereby makes it an action-­causing idea.43 Thus, the house builder does not start building a house unless he or she has reflected upon the house idea. “So he cognizes the idea,” Suárez writes, “and through this cognition he applies, as it were, the exemplar so that he causes something through it” (DM 25.1.35). This means that the house builder produces a higher-­order idea to make the first-­order idea, which is directed at the house, a causing idea. The main idea of this reflexivity model can be summarized as follows: (4)  It is by reflecting upon an idea that one makes it a causing idea. At first sight, this explanation looks quite plausible. After all, when you start building a house you do not contact a construction company on the basis of simply thinking about a lovely home with a terrace. Rather, you need to say to yourself something like the following: “Yes,

43  Suárez refers to STh I, q. 15, art. 2, ad 2, where Aquinas does indeed say that the house builder “understands that he understands” when he is building the house. Note, however, that it is not so clear that Aquinas has explicit reflection in mind here, that is, reflection that involves a higher-­order idea. One can also understand him as saying that the house builder is aware of his own thinking. For this deflationary reading, see Therese Scarpelli Cory, Aquinas on Human Self-­Knowledge (Cambridge: Cambridge University Press, 2014), 142–53.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

272  Dominik Perler this is exactly the house that should be built!” And to do that, you need to focus on your house idea and deliberately use it as the plan for your future actions. That is, you need to form a higher-­order idea that picks out the first-­order idea as the action-­guiding idea. This is exactly what you do when you reflect upon your idea. Without such a reflection, you would have a nice plan for what you could (or even should) do, but you would never become active. However, Suárez does not accept this reflexivity model. He objects that it is against our experience, for “someone who is writing or building does not always engage in an actual reflection when he acts” (DM 25.1.36).44 Thus, you do not reflect upon your house idea when you actually build the house, not even when you start building it. Otherwise you would be so much focused on your own idea that you would neglect the house itself. Of course, there may be moments in which you do in fact focus on your idea, such as when you think about what you really think about the house. But when you talk to the construction company, you are focused on the house itself, not on your thought of it. In fact, it is because you think about the rooms, the terrace, and other parts of the house that you engage in an action that makes all these parts become reality. In technical terms, this means that there is no ascent to a higher-­ order idea: merely having a first-­order idea is sufficient for becom­ ing active. This objection shows that Suárez rejects what might be called an over-­ intellectualization of the productive process. No explicit and deliberate reflection upon one’s own thoughts is necessary for becoming active. Indeed, if it were required that there be a higher-­order idea, an infinite regress could result. For how could the higher-­order idea cause an action? After all, it would also be just a mental act, and it would there­ fore not be better suited to cause an action than the first-­order idea. Would one then need still another idea to become active? If this were 44  Note that Suárez criticizes the strong thesis that a person always engages in an actual reflection. A proponent of the reflexivity model could also defend the weaker thesis that such a person sometimes reflects, namely, when she initiates an action, but once the action has started, reflection is no longer necessary. Suárez does not consider this possibility, presumably because he assumes that the reflexivity model requires that the person stay focused on the idea. If she were to cease to reflect, she could be distracted by other thoughts and her idea would lose its causal force.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  273 the case, one would have to posit a higher-­order idea for every idea, and the regress would never stop. Given this problem, it is better not to appeal to a higher-­order idea in the first place. It must be then that the mere presence of the first-­order idea can cause an action. Does this mean that no reflection at all is required? Suárez does not draw this radical conclusion. All he rejects is the thesis that explicit reflection, which involves the production of a higher-­order idea, is required. But he concedes that another form of reflection is necessary. He describes it as follows: It is therefore not necessary that the exemplar be known as the object, but it suffices that it be known implicitly and in some way through its tendency to an object. This occurs by a merely implicit and virtual reflection, which any act of the mind includes in itself. For this reason, it is said that the act is known by itself as that by which something is known, although it is not known as that which is known. (DM 25.1.39)

Here Suárez clearly holds that implicit reflection is required, thereby subscribing to the following modified thesis: (4*) It is by implicitly reflecting upon an idea that one makes it a caus­ ing idea. This kind of reflection is built into every act that is actively and atten­ tively directed at an object. Using modern terminology, one could speak of same-­order consciousness, that is, consciousness that is present at the same level as the act of thinking.45 What this amounts to can be explained by means of the house example. Suppose you think very hard about the house you want to build. In that case you do have an act of thinking and you are immediately conscious of it because the act imme­ diately presents itself. Should someone ask you if you are really thinking, 45 I borrow this terminology from Uriah Kriegel, Subjective Consciousness: A Self-­ Representational Theory (Oxford: Oxford University Press, 2011), who also speaks of a self-­ representational act, that is, a first-­order act that makes itself present to the mind.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

274  Dominik Perler you could spontaneously respond: “Of course I am thinking. In fact, I am thinking so vividly and intensely that I can describe not only the object of my act of thinking, but also the act itself. I can characterize it as a focused act that absorbs my intellectual energy.” If you were to say that, you would make clear that you have what Suárez calls “implicit and virtual reflection,” that is, reflection that comes along with your act and makes it a conscious act. This would clearly differ from the case in which you are lying on the couch dozing. In that case you would perhaps have a loose series of acts because you would not be fully asleep, but these acts would not be present to you. You could neither say what exactly you are thinking nor confirm that you are thinking, simply because you would not be conscious of your acts. In that case you would lack implicit and virtual reflection. I hope this example illustrates what Suárez has in mind here.46 I also hope it makes clear that he does not accept the simple view that the mere presence of an idea is sufficient for becoming active. It is only the presence of a conscious idea that causes an action, for only such an idea has an action-­guiding function. This is a significant thesis, for it shows that Suárez does not look at an act as an isolated item that causes an action by itself. It does so only when it is conscious, which presupposes its existence in a mind. If God were to intervene in the world and mi­racu­lous­ly extract an act from a given mind, this act could not cause anything because it would not be conscious: to be conscious it needs to be present in and to a mind. This is why it is only an embedded act, that is, an act in an active and attentive mind, that can cause an action. To avoid misunderstandings, it is important to note that conscious­ ness alone does not make an idea productive. For instance, consciously thinking about a fancy castle does not make you build a castle. Consciousness is at most a necessary, but not a sufficient condition for activity. You also need to decide that you really want to build a castle; only then will you become active. Typically, you first go through a pro­ cess of deliberation, then decide that you want to produce a certain 46  It is not only in this context that Suárez refers to first-­order consciousness. When analyz­ ing perceptual acts, he also claims that we are immediately aware of them without having higher-­order acts. There he speaks of an experience in the mere having or exercising of the act (in actu exercito); see De anima, disp. 6, q. 4, n. 7 (vol. II, 510).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  275 object, and finally form an idea of that object. If this idea is then con­ sciously present, it gives rise to an action. So, to make clear that it is not just any conscious idea that is productive, Suárez’s thesis needs to be specified as follows: (4**) It is by implicitly reflecting upon an idea as the outcome of a deci­ sion that one makes it a causing idea. But how can an idea cause an action once it is present in this way? So far, it has only been shown that an idea is an act that has an inner struc­ ture and hence a content. But why should that give it some causal power? To answer this question, it is important to take a closer look at Suárez’s characterization of cognitive acts. These acts belong, meta­phys­ic­al­ly speaking, to the category of quality.47 As noted above (see section  3), each of them is a quality that exists in the mind as in a subject; hence it has subjective existence in the mind. Like many other qualities, an idea is part of a causal chain; that is, it is caused by the intellectual power, and in turn causes other entities. First of all, it causes other acts.48 This is possible because each act has two aspects (DM 44.8.13). Considered under the “aspect of an action” (ratio actionis), it is an item that presents an object. This is the intentional aspect that has already been described in detail. Suárez speaks of an action here not because he assumes that the act produces the object it presents, but because he wants to empha­ size that the act is directed at the object and somehow reaches out for it. It goes for that object just as an arrow goes for the object at which it is aimed. But an act can also be looked at under another aspect, namely, the “aspect of a quality” (ratio qualitatis). When considered in this way,

47  Suárez emphasizes that “the act of knowing is a special quality that is really distinct from the power” (De anima, disp. 5, q. 3, n. 2; vol. II, 342). It is therefore not just an actualization of the intellectual power, which is also a quality, but a distinct quality. 48  A cognitive act can also cause habits, as Suárez hastens to add, for if it repeatedly occurs it brings about a stable disposition, which in turn produces new acts of the same type. This dis­ position is thus a new quality that comes into existence. See DM 44.8–9 and a discussion in Dominik Perler, “Suárez on the Metaphysics of Habits,” in N. Faucher and M. Roques (eds.), The Ontology, Psychology and Axiology of Habits (Habitus) in Medieval Philosophy (Cham: Springer, 2018), 365–84.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

276  Dominik Perler an act is a power that can produce something: it can give rise to other acts.49 The crucial point is that a mental act is not an inert thing that has a distinctive content but remains passive. If this were the case, an act would be like a book that includes many sentences but cannot do any­ thing. An act is rather an active thing; that is, it is a power that can bring about other things. To avoid misunderstandings, one should note that an act cannot be productive as an isolated item. Since it is an embedded act, it becomes active in cooperation with other powers. In particular, it becomes active together with the underlying intellectual power. For if there were no power that causes the act and maintains it in existence, the act itself could not become active either; the intellectual power in turn could not become active if it were not rooted in the soul, which is a basic power.50 Metaphysically speaking, this means that the act is a dependent power that cannot act on its own. It is part of a whole system of powers—the system of the soul—and can exercise its causal function only when it is integrated into that system. Nevertheless, it is a real power that produces a real effect. This thesis gives rise to the question of how an idea can produce a real effect. What kind of causality is at stake here? It is in his answer to this question that Suárez’s commitment to the Aristotelian framework becomes visible, for he holds that the act must belong to one of the four Aristotelian kinds of cause. It would be pointless to appeal to a special type of cause that has its own way of acting. Such an appeal would be misleading, for it would suggest that an idea is a special thing that has its own, very special force. But an idea is not a special thing, as has already become clear (see section  2). In particular, it is not an internal thing with “objective existence” that could somehow act upon the mind or the

49  As far as I can see, Suárez does not explain why an idea is a power. Presumably, he assumes that it belongs to the second type of quality, which comprises capacities and powers (Cat. 5; 9a9–27). In any case, in his general classification of qualities he points out that “all the qualities that are principles of acting belong to the second species of quality” (DM 42.3.10). 50 In DM 18.5 Suárez even claims that the underlying soul has a coordinating function. It guarantees that the intellect, the will, and the sensory powers form a coherent whole and become active in a well-­ordered way. On this function, see Marleen Rozemond, “Unity in the Multiplicity of Suárez’s Soul,” in B. Hill and H. Lagerlund (eds.), The Philosophy of Francisco Suárez (Oxford: Oxford University Press, 2012), 154–72 (especially 161–4).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  277 body and produce an effect in its own way. There is no such thing, and hence an idea does not have some special way of acting (DM 25.2.9). As an act, an idea is part of a natural causal chain and must therefore be accounted for like any other cause. But what kind of cause is it? It is clear that an act cannot be a material cause since such a cause is something that must be united with a form (DM 25.2.2). But an act is already a form, or more precisely, it is an accidental form that is present in the mind. If it were not a form, it could not be an active thing that produces something. Is it therefore a formal cause? No, because such a cause simply has the function of structuring a given thing. But to do that, the thing to be structured needs to exist, and the formal cause as such does not make it exist. What is required here is something that pre­ cedes the formal cause.51 So, both material and formal cause, the first two Aristotelian candidates for a cause, are ruled out. The final cause seems to be a more promising candidate. A final cause is always the goal of an action, and it acts insofar as it attracts the will and makes it go for what is present as a good and desirable object. Suárez characterizes the activity of this kind of cause by speaking of a “meta­ phorical motion” (DM 25.2.3). In using this expression he wants to stress that the goal moves the will in such a way that the will really wants to go for it. This is not motion in the physical sense; after all, the will is an immaterial power and cannot be moved like a billiard ball. Nevertheless, there is some kind of motion insofar as the will is impelled or incited to go for the goal.52 Thus, when you are presented with a delicious-­looking apple, your will is somehow impelled by the apple to go for it. It is exactly this kind of motion that is responsible for the fact that you really want to eat the apple and that you reach out for it. If there were no motion or attraction in this non-­physical way, you would never act. Or to put it in 51  Suárez is aware that one could call the formal cause at stake an “extrinsic form” (DM 25.2.7), that is, a form that exists only in the mind and precedes the intrinsic form that will exist in the material thing. But he objects that this does not solve the causation problem (DM 25.2.9). For how could an extrinsic form produce an intrinsic form? It would have to be appre­ hended and used to produce an intrinsic form. In that case it would not be the extrinsic form but the act of using it that would have a causal function. 52 In DM 23.4.8 Suárez explicitly says that the metaphorical motion stems “from an object that allures and pulls the will toward itself.” For a detailed analysis of this characterization, see Stephan Schmid, Finalursachen in der frühen Neuzeit: Eine Untersuchung der Transformation teleologischer Erklärungen (Berlin: De Gruyter, 2011), 121–31.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

278  Dominik Perler modern terms, if the object presented to you did not exercise a norma­ tive power over you, you would never respond to it.53 Now one might assume that it is exactly this kind of motion that is at stake here, and hence that an idea acts as a final cause. Tempting as this account might be at first sight, it is not the account Suárez chooses. On his view, an idea is not a final cause but must be an efficient cause (DM 25.2.8–10). To be sure, he does not deny that a final cause is required, but he insists that final and efficient causes have two different functions (DM 25.2.3). The final cause has the function of attracting the will so that a person chooses the goal of an action. But once she has made a choice, she needs to carry out the action and really go for the thing she has chosen, and for that she needs an efficient cause.54 Let me illustrate this point by adding some details to the house example. Suppose you have saved some money but are not so sure what to do with it. Should you use it to build a house? Should you spend it on luxurious trips? Or should you do something else with it? When you evaluate different options, you consider various goals. Each of them has some advantages, but in the end it is the house option that seems best because it has long-­ term benefits. Since it is present to you as the best option, it attracts you so that you decide to choose it and to discard the other options. To put it in Suárez’s terminology, your will is then metaphorically moved by this option. However, this kind of motion is not sufficient for the production of the house, for once you have chosen the best option you need to become active. This is why you need a detailed plan, which must be ­vividly present to you. It must even incite or somehow push you so that you contact a construction company, give them an order, and supervise their work. This is exactly the function of the idea as an efficient cause: it makes you go for what you have chosen. Or as Suárez says, it “deter­ mines the action of the agent” (DM 25.2.3) and thereby makes the agent produce what she has decided to produce.

53 For an insightful interpretation of final causation in terms of normative power, see Thomas Pink, “Agents, Objects, and Their Powers in Suarez and Hobbes,” Philosophical Explorations 21 (2018), 3–24. 54 Schmid, Finalursachen in der frühen Neuzeit, 144, nicely spells out this difference by not­ ing that the final cause explains why a person chooses a certain action, while the efficient cause explains what she really does once she has made a choice.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  279 Given this insistence on the presence of an efficient cause, it is im­port­ ant to distinguish two stages in the process that leads to the production of a thing. In the first stage, a person needs to imagine a thing so that it can attract the will and trigger a decision in favor of producing it. It is in this stage that the final cause plays a crucial role. In fact, it is in this stage that the imagined object plays the role of the final cause. But when the imagined thing has been chosen as the desired object, the process is not yet finished. After all, merely choosing a thing is not the same as really going for it. Hence, there needs to be a second stage in which an act is formed that causes an action. This is exactly the stage in which the idea plays a crucial role, for it is in this stage that the idea presents the chosen thing and causes other acts that lead to the production of that thing. To do that, the idea needs to be an efficient cause that can really bring about something. The crucial point is that an idea is able to act as an efficient cause simply by being present in a person. For if it is conscious, it is an active thing that can produce something. Here again, it is important to note that an idea is not an inert thing that displays a distinctive content but remains passive. Its status as a power guarantees that it can be pro­duct­ ive, and as an efficient cause it can bring about something that is distinct from it and depends on it, namely, a bodily activity that leads to the pro­ duction of the chosen thing.55 To avoid misunderstandings, it should be noted that the fact that an idea is active does not entail that it always successfully produces some­ thing. After all, not every efficient cause is a successful cause. It may well be that internal or external factors prevent a person from bringing about the product she intends to bring about with her idea. Suárez is fully aware of the possibility of failure, and therefore cautiously remarks:

55  According to Suárez’s general definition, an efficient cause “gives being to something else” so that this thing “really flows forth and emanates from the cause” (DM 17.1.6). The crucial point is that cause and effect are really distinct and that there is a one-­sided existential depend­ ence relation between them. For a detailed analysis, see Stephan Schmid, “Efficient Causality: The Metaphysics of Production,” in J. L. Fink (ed.), Suárez on Aristotelian Causality (Leiden: Brill, 2015), 85–121 (especially 91–4), and Tuttle, “Suárez’s Non-­Reductive Theory of Efficient Causation,” 128–32.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

280  Dominik Perler The craftsman by himself only intends to bring about such a form [sc. the form of the house] in his effect, and for that he needs only a proper conception of this form. This form is then extrinsically assimilated to something else, and this is accidental to the causality of the craft as such; therefore it is also accidental to the causality of the exemplar. (DM 25.2.10)

The crucial point is that the craftsman only intends to produce a thing that has a certain form. He does so by coming up with an idea of that thing, because it is in this idea that he conceives of the form of that thing. But merely intending to produce a thing is not the same as really producing it. Many things can happen that will make the process of ­production fail. For example, while building the house you may all of a sudden have a strong desire to do something else; this desire will then overpower your idea and prevent it from producing an effect. Or your partner may think that building the house will lead to a financial disas­ ter and therefore prevent you from actually building it, or something else may stop you. Whether or not you really build the house does not depend just on your initial idea. This is why Suárez emphasizes in the passage just quoted that the assimilation of the form to something else, that is, its becoming present in a real house, is “accidental to the causal­ ity.” It is not essential to the idea as a cause that it really bring about an effect. What is essential to it is that it is a power that strives to produce an effect, and it always remains such a power as long as it is present. It is therefore with good reason that an idea can be called a cause, whether or not it ever produces an effect: being active as a cause is one thing, bring­ ing about an effect is quite another.

5. Conclusion Where does that leave us? I opened this paper by saying that the seemingly innocuous case of house planning poses at least two deep problems: the intentionality problem and the efficacy problem. I hope it has become clear that Suárez offers a solution to both problems.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  281 He attempts to solve the intentionality problem by presenting an act theory: making a plan amounts to forming a mental act that is directed at something in virtue of its content. The act targets an external thing like an arrow that targets a goal, whether or not it ever reaches the goal. The crucial point about this theory is that the act directly targets an  external thing. Its targeting is not mediated by something else; in particular, it is not mediated by an internal object that serves as the immediate goal. As has become clear (see section 2), Suárez rejects the existence of mediating entities. Admittedly, he accepts the existence of intelligible species, but this does not commit him to introducing in­tern­al objects that prevent a person from reaching out for external things. Intelligible species are nothing but devices that make the fixing of the act’s content possible. In arguing for this thesis, Suárez defends direct realism and rejects any attempt to explain cognition by reference to rep­ resentations that function as internal objects. To be sure, he repeatedly speaks of representations, but he takes them to be nothing but mental acts that are directed at external things. The case of ideas brings to the fore what is crucial for all cases of cognition: talk about representations does not compromise direct realism. It is therefore not surprising that Suárez argues in favor of direct realism in many areas of his cognitive theory.56 No matter whether we perceive things, think or judge about them, we are always immediately directed at the things themselves, not at some inner proxies. Having representations as mental acts does not pre­ vent us from targeting external things. On the contrary, it is precisely the presence of representations that makes this targeting possible. It is also of crucial importance that acts are intrinsically intentional once their content has been fixed. Of course, they presuppose in­tel­li­ gible species, but once these cognitive devices are present nothing else is required. In particular, no causal relation or similarity relation is required. This thesis sheds light on Suárez’s entire cognitive theory. For if it is the content that matters, it is important to look at the content in 56  This is most visible in his theory of perception, where he also speaks about representa­ tions and also accepts species as cognitive devices, but insists that we directly see or otherwise perceive external things. For a detailed account, see South, “Suárez and the Problem of External Sensation.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

282  Dominik Perler each and every case—not just in the case of ideas, but also in cases of perception, imagination, etc. This is why Suárez always attempts to spell out what the content consists in and what devices (sensible or in­tel­li­ gible species) are required for determining it.57 The case of ideas is of special significance, for in this case there clearly is no external thing and hence no causal relation or similarity relation to such a thing. When studying ideas one sees as under a magnifying glass what is at the core of every intentional act. It is their inner structure and hence their content that needs to be examined. But why is it that an idea can be intentional? Why can it point at something in virtue of its content? As far as I can see, Suárez never addresses this problem. What he repeatedly discusses is the question of how an idea can be directed at a thing, but he does not offer an answer to the question of why it can be directed at all. Does this mean that he neglects the most obvious question that cries for an answer? It is tempt­ ing to make this reproach, but one should resist the temptation. Suárez takes it be a fundamental fact that a mental act is always directed at a thing. This becomes clear if one looks again at his claim that every ­mental act comes along with a transcendental relation (see section 3). Whenever there is an act with a well-­defined content, he claims, there is also a pointing towards something and hence a transcendental relation. This relation is an “essential mode” (DM 47.4.15) that always ac­com­pan­ ies the act, or is even built into it. As has already been mentioned, he compares the essential mode of a mental act to that of a natural power: as a natural power (e.g. heat) is essentially goal-­directed and points at something (e.g. a hot thing), so also a mental act is essentially goal-­ directed and points at something. The only difference between the two cases lies in the fact that a natural power produces the thing at which it points (provided there are no hindering factors), whereas a mental act simply presents that thing. Whether or not it also produces something depends on what kind of act it is; only an idea is a productive act that 57  Thus, when discussing external and internal sensation Suárez spells out the acts that are at stake; that is, he examines their specific content and the powers that are required for deter­ mining it. For details, see Daniel Heider, “Late Scholastic Debates about External and Internal Senses: In the Direction of Francisco Suárez (1548–1617),” in S.  Schmid (ed.), Philosophy of Mind in the Late Middle Ages and Renaissance (London: Routledge, 2019), 165–84.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  283 both presents and produces a thing. But what is common to all mental acts is that they are essentially directed at something in virtue of their content. It would therefore be pointless to ask why they are intentional: being intentional is simply part of their essence.58 What about the efficacy problem? Suárez attempts to solve this prob­ lem as well, and here again his discussion of ideas as productive thoughts sheds light on his broader theory. As has become clear, he holds that ideas are efficient causes that immediately bring about an action (see section  4), or to be more precise, they immediately initiate an action. Whether or not the action will be accomplished depends on additional factors, but the salient point is that an idea is in and by itself a cause: it is not simply used as an instrument by some other cause. Since it is a power, it can give rise to mental and bodily activities that will eventually lead to the production of a material thing. This is possible because an idea is from the start integrated into a whole system of powers. To be sure, this does not mean that an idea can somehow intervene in the lower sensory powers and produce in them activities like moving one’s hands or calling someone. All it can immediately produce is another thought, that is, another act in the rational power. Thus, when you have an idea of the house, this idea can directly produce only a thought about, for example, the construction company that should be contacted. Once this thought is present, the sensory power will become active and produce the activities that are required—for instance, the activities of picking up the phone and talking to construction workers. Suárez insists that there is a perfect coordination between the rational and sensory powers so that the sensory power is triggered once a certain thought is present in the rational power. There is, as he says, a perfect “harmony” or “sympathy” between the powers.59 In making this claim, he assumes that there is a network of powers and that the activity of one

58  Note, however, that this applies not just to mental acts. Since natural powers are also essentially directed at something, they also have some kind of intentionality, which could be called “physical intentionality” (see note 36). It would therefore be inappropriate to say that intentionality is the mark of the mental. It is the mark of all natural powers that are directed towards a goal. 59  See, for instance, DM 18.5.3, and De anima, disp. 2, q. 5, n. 5 (vol. I, 328).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

284  Dominik Perler power has an immediate consequence for all other powers.60 The exist­ ence of a network of powers is indeed the main reason why the mere having of a thought can be productive: in thinking something, one trig­ gers other powers, and the whole chain of activities will then lead to the production of a material thing. It is striking that Suárez takes the entire network of powers to be a network of efficient causes. He spells out an action by referring to a number of efficient causes that interact. To be sure, he does not reduce the whole action to a chain of efficient causes. As has become clear (see section 4), no action is initiated unless there is a final cause that moves the will and makes it choose a certain goal. But once the goal has been fixed, there is a long chain of efficient causes that makes it possible to aim for that goal and to produce it. The function of an idea is precisely to act as an efficient cause and to activate other causes. It is therefore not surprising that Suárez looks at a person as a system of efficient causes. Of course, he still follows the Aristotelian tradition in claiming that it is the person as a whole that acts, but when he then analyzes an action he goes beyond that tradition by looking at the efficient causes inside the person and by examining the process that leads from an idea as the ini­ tial efficient cause to the production of a material thing. In short, he goes from the personal level to the sub-­personal level of efficient causes. He thereby paves the way for non-­Aristotelian theories that attempt to explain an action in terms of efficient causation. Admittedly, Suárez himself never describes his project as an endeavor to revise or even abandon the Aristotelian theory. But in referring to a network of effi­ cient causes, he contributes to a deep transformation of that theory.61 Humboldt-­Universität zu Berlin

60  Suárez emphasizes that the powers do not literally act upon each other. They are merely well coordinated so that the activity in one power goes along with the activity in another power. For details, see Josef Ludwig, Das akausale Zusammenwirken (sympathia) der Seelenvermögen in der Erkenntnislehre des Suarez (Munich: Karl Ludwig Verlag, 1929), and more recently Simo Knuuttila, “The Connexions between Vital Acts in Suárez’s Psychology,” in L.  Novak (ed.), Suárez’s Metaphysics in Its Historical and Systematic Context (Berlin: De Gruyter, 2014), 259–74. 61  I am very grateful to Sebastian Bender, Ian Drummond, Can Laurens Löwe, Stephan Schmid, Hamid Taieb, and an anonymous referee for detailed comments on earlier versions of this paper.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  285

Bibliography Primary Sources Aristotle. Categoriae, ed. L. Minio-Paluello (Oxford: Clarendon Press, 1949). Aristotle. De anima, ed. W. D. Ross (Oxford: Clarendon Press, 1956). Augustine. De diversis quaestionibus octoginta tribus, ed. A. Mutzenbecher, Corpus Christianorum Series Latina XLIV (Turnhout: Brepols, 1975). Francisco Suárez. Disputationes Metaphysicae, in C.  Berton (ed.), Opera omnia, vols. 25 and 26 (Paris: Vivès, 1861). Francisco Suárez. De anima, ed. S. Castellote, 3 vols. (Madrid: Sociedad de estudios y publicaciones, 1978–91). Peter Aureol. Scriptum super primum Sententiarum, ed. E.  M.  Buytaert (St. Bonaventure, NY: Franciscan Institute, 1956). Peter Aureol. Scriptum super primum Sententiarum: distinctio 27, ed. R. L. Friedman, http://www.peterauriol.net/editions/electronicscriptum. Thomas Aquinas. Summa theologiae, ed. P. Caramello (Rome: Marietti, 1952). Thomas Aquinas. Quaestiones disputatae de veritate: quaestiones 1–7, editio Leonina XXII/I/2 (Rome: Sancta Sabina, 1970). Thomas Aquinas. Sentencia libri de anima, editio Leonina XLV/1 (Rome: Commissio Leonina, 1984). William of Ockham. Opera philosophica et theologica (= OPh and OTh), ed. G. Gál et al. (St. Bonaventure, NY: Franciscan Institute, 1967–88).

Secondary Sources Adams, Marilyn McCord. William Ockham (Notre Dame, IN: Notre Dame University Press, 1987). Adriaenssen, Han Thomas. Representation and Scepticism from Aquinas to Descartes (Cambridge: Cambridge University Press, 2017). Biard, Joël (ed.). Le langage mental du moyen âge à l’âge classique (Louvain: Éditions de l’institut supérieur de philosophie, 2009). Boulnois, Olivier, Jacob Schmutz, and Jean-Luc Solère (eds.). Le contemplateur et les idées: modèles de la science divine, du néoplatonisme au XVIIe siècle (Paris: Vrin, 2002).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

286  Dominik Perler Brower, Jeffrey  E. and Susan Brower-Toland. “Aquinas on Mental Representation: Concepts and Intentionality,” Philosophical Review 117 (2008), 193–243. Brower-Toland, Susan. “Causation and Mental Content: Against the Externalist Reading of Ockham,” in J. Pelletier and M. Roques (eds.), The Language of Thought in Late Medieval Philosophy (Cham: Springer, 2017), 59–80. Carraud, Vincent. Causa sive ratio: la raison de la cause, de Suarez à Leibniz (Paris: Presses Universitaires de France, 2003). Cory, Therese Scarpelli. Aquinas on Human Self-Knowledge (Cambridge: Cambridge University Press, 2014). Cory, Therese Scarpelli. “Knowing as Being? A Metaphysical Reading of the Identity of Intellect and Intelligible in Aquinas,” American Catholic Philosophical Quarterly 91 (2017), 333–51. Cross, Richard. Duns Scotus’s Theory of Cognition (Oxford: Oxford University Press, 2014). De Rijk, Lambert  M. “Quaestio de ideis: Some Notes on an Important Chapter of Platonism,” in J. Mansfeld and L. M. de Rijk (eds.), Kephalaion: Studies in Greek Philosophy and Its Continuation (Assen: Van Gorcum, 1975), 204–13. Des Chene, Dennis. “Suárez on Propinquity and the Efficient Cause,” in B.  Hill and H.  Lagerlund (eds.), The Philosophy of Francisco Suárez (Oxford: Oxford University Press, 2012), 89–100. Embry, Brian. “Francisco Suárez on Beings of Reason and Non-Strict Ontological Pluralism,” Philosophers’ Imprint 19:27 (2019), 1–15. Falà, Jacopo and Irene Zavattero (eds.). Divine Ideas in Franciscan Thought (XIII–XIVth Century) (Canterano: Aracne Editrice, 2019). Fink, Jakob Leth (ed.). Suárez on Aristotelian Causality (Leiden: Brill, 2015). Forlivesi, Marco. “La distinction entre concept formel et concept objectif: Suárez, Pasqualigo, Mastri,” Les Études Philosophiques 60 (2002), 3–30. Heider, Daniel. “Late Scholastic Debates about External and Internal Senses: In the Direction of Francisco Suárez (1548–1617),” in S.  Schmid (ed.), Philosophy of Mind in the Late Middle Ages and Renaissance (London: Routledge, 2019), 165–84. Heider, Daniel. “Suárez’s Metaphysics of Cognitive Acts,” in R. A. Maryks and J.  A.  Senent de Frutos (eds.), Francisco Suárez (1548–1617): Jesuits and the Complexities of Modernity (Leiden: Brill, 2019), 23–45.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  287 Hoffmann, Tobias. Creatura intellecta: Die Ideen und Possibilien bei Duns Scotus mit Ausblick auf Franz von Mayronis, Poncius und Mastrius (Münster: Aschendorff, 2002). Knuuttila, Simo. “The Connexions between Vital Acts in Suárez’s Psychology,” in L. Novak (ed.), Suárez’s Metaphysics in Its Historical and Systematic Context (Berlin: De Gruyter, 2014), 259–74. Kriegel, Uriah. Subjective Consciousness: A Self-Representational Theory (Oxford: Oxford University Press, 2011). Ludwig, Josef. Das akausale Zusammenwirken (sympathia) der Seelenvermögen in der Erkenntnislehre des Suarez (Munich: Karl Ludwig Verlag, 1929). Molnar, George. Power: A Study in Metaphysics (Oxford: Oxford University Press, 2003). Panaccio, Claude. Le discours intérieur: de Platon à Guillaume d’Ockham (Paris: Seuil, 1999). Panaccio, Claude. Ockham on Concepts (Aldershot: Ashgate, 2004). Pasnau, Robert. Theories of Cognition in the Later Middle Ages (Cambridge: Cambridge University Press, 1997). Pasnau, Robert. After Certainty: A History of Our Epistemic Ideals and Illusions (Oxford: Oxford University Press, 2017). Penner, Sydney. “Suárez on the Reduction of Categorical Relations,” Philosophers’ Imprint 13:2 (2013), 1–24. Perler, Dominik. Theorien der Intentionalität im Mittelalter (Frankfurt am Main: Klostermann, 2002). Perler, Dominik. Zweifel und Gewissheit: Skeptische Debatten im Mittelalter (Frankfurt am Main: Klostermann, 2006). Perler, Dominik. “Can We Trust Our Senses? Fourteenth-Century Debates on Sensory Illusions,” in D.  G.  Denery II, K.  Ghosh, and N.  Zeeman (eds.), Uncertain Knowledge: Scepticism, Relativism, and Doubt in the Middle Ages (Turnhout: Brepols, 2014), 63–90. Perler, Dominik. “Suárez on Consciousness,” Vivarium 52 (2014), 261–86. Perler, Dominik. “Suárez on the Metaphysics of Habits,” in N. Faucher and M.  Roques (eds.), The Ontology, Psychology and Axiology of Habits (Habitus) in Medieval Philosophy (Cham: Springer, 2018), 365–84. Perler, Dominik. “Súarez on Intellectual Cognition and Occasional Causation,” in D. Perler and S. Bender (eds.), Causation and Cognition in Early Modern Philosophy (London: Routledge, 2020), 18–38.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

288  Dominik Perler Pink, Thomas. “Agents, Objects, and Their Powers in Suarez and Hobbes,” Philosophical Explorations 21 (2018), 3–24. Renemann, Michael. Gedanken als Wirkursachen: Francisco Suárez zur geistigen Hervorbringung (Amsterdam: Grüner, 2010). Renemann, Michael. “The Mind’s Focus as an Efficient Cause: Francisco Suárez’s Re-interpretation of the Traditional Understanding of the Idea,” American Catholic Philosophical Quarterly 84 (2010), 693–710. Rozemond, Marleen. “Unity in the Multiplicity of Suárez’s Soul,” in B. Hill and H.  Lagerlund (eds.), The Philosophy of Francisco Suárez (Oxford: Oxford University Press, 2012), 154–72. Schmaltz, Tad  M. “Efficient Causation: From Suárez to Descartes,” in T.  M.  Schmaltz (ed.), Efficient Causation: A History (Oxford: Oxford University Press, 2014), 139–64. Schmaltz, Tad M. The Metaphysics of the Material World: Suárez, Descartes, Spinoza (Oxford: Oxford University Press, 2020). Schmid, Stephan. Finalursachen in der frühen Neuzeit: Eine Untersuchung der Transformation teleologischer Erklärungen (Berlin: De Gruyter, 2011). Schmid, Stephan. “Efficient Causality: The Metaphysics of Production,” in J.  L.  Fink (ed.), Suárez on Aristotelian Causality (Leiden: Brill, 2015), 85–121. Secada, Jorge. “Suárez on the Ontology of Relations,” in D. Schwartz (ed.), Interpreting Suárez: Critical Essays (Cambridge: Cambridge University Press, 2012), 62–88. Shields, Christopher. “Shadows of Being: Francisco Suárez’s Entia Rationis,” in B.  Hill and H.  Lagerlund (eds.), The Philosophy of Francisco Suárez (Oxford: Oxford University Press, 2012), 57–74. Silva, José Filipe and Christina Thomsen Thörnqvist (eds.). Assimilation and Representation in Medieval Theories of Cognition, Vivarium 57:3–4 (2019). South, James B. “Suárez and the Problem of External Sensation,” Medieval Philosophy and Theology 10 (2001), 217–40. Tachau, Katherine  H. Vision and Certitude in the Age of Ockham: Optics, Epistemology and the Foundations of Semantics 1250–1345 (Leiden: Brill, 1987). Taieb, Hamid. Relational Intentionality: Brentano and the Aristotelian Tradition (Cham: Springer, 2018).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Productive Thoughts  289 Tuttle, Jacob. “Suárez’s Non-Reductive Theory of Efficient Causation,” Oxford Studies in Medieval Philosophy 4 (2016), 125–58. Tuttle, Jacob. “Suárez on Creation and Intrinsic Change,” American Catholic Philosophical Quarterly 93 (2019), 29–51. Wells, Norman  J. “Objective Reality of Ideas in Descartes, Caterus, and Suárez,” Journal of the History of Philosophy 28 (1990), 33–61.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Briefly Noted Peter Adamson, Medieval Philosophy: A History of Philosophy without any Gaps, volume 4 (Oxford University Press, 2019), xxii + 637 pp. History, like time itself, displays sufficient continuity that any attempt at a summary representation of it, no matter how large, will inevitably contain some omissions. Each generation chooses the periods it is willing to pass over in silence. To write about medieval philosophy at all is to make some contribution to the gap-­filling project, but this volume in Adamson’s steadily growing series argues that previous histories of medieval phil­ oso­phy have themselves contained far too many gaps. Over 78 chapters brief chapters, in addition to the usual territory, we get discussions of various female voices: Heloise, Hildegard, Hadewijch and Mechthild, Marguerite Porete, Julian of Norwich, Catherine of Siena, and Christine de Pizan. There is literature: Jean de Meun, Dante, Chaucer, Langland; and there is law and politics: canon law, Franciscan poverty, just war, Church versus State, economic theory, sexuality. Of course, making room for all of this means introducing new gaps elsewhere: Aquinas gets a mere 16 pages. Still, for breadth and concision, Adamson’s treatment is unmatched. He has a fantastic capacity for extracting the essential elements of each of these debates, and for each topic there is a brief but expert bibliography. One might suppose that Adamson here distills the scholarly work of a lifetime, and yet he somehow is able to produce a new volume in this series every year. Chris Schabel, Pierre Ceffons et le déterminisme radical au temps de la peste noire (Vrin, 2019), 244 pp. This study, based on Schabel’s 2016 Pierre Abélard lectures in Paris, offers a path into the underappreciated work of Pierre Ceffons, the Cistercian theologian who lectured on the Sentences in Paris in 1348–49. Schabel updates and deepens our understanding of Ceffons’s career, and offers a preliminary account of the unique manuscript in which his work survives, Troyes 62, “un manuscript splendide” (24), but complex in its details. Turning to Ceffons’s distinctive ideas, Schabel devotes successive chapters to the necessity of the past, the contingency of the future, divine foreknowledge, and predestination. Each chapter offers a masterful précis of scholastic views, and shows where Ceffons makes a distinctive contribution. Particularly interesting are his “radical” ideas regarding future contingents, inasmuch as he holds, in persona philosophi, that such contingency should be denied. In contrast to Peter Auriol, who had argued that bivalence entails determinism, and thus concluded that bivalence is false, Ceffons maintains bivalence and embraces determinism (120). At the end of the day, in propria persona, as a Christian, he endorses contingency. So why, he asks himself, would he bother with reciting philosophical views that he does not judge to be ultimately true? To test

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Briefly Noted  291 these ideas, he explains, and to see “whether it is purely to be believed that not all things occur of necessity” (126). The implication of his work is that here, and in many other domains, the Christian faith is purely to be believed, in the sense that such belief cannot be founded on philosophical reason. Daniel Schwartz, The Political Morality of the Late Scholastics: Civic Life, War and Conscience (Cambridge University Press, 2019), 234 pp. The flourishing field of late scholastic studies has not only deepened our understanding of familiar scholastic questions but also widened our sense of the sorts of questions with which scholastic authors engaged. This book is an excellent example of the latter project. Schwartz shows us that the sixteenth and seventeenth centuries contain fascinating debates on a range of topics that have scarcely received any scholarly notice. The book’s subtitle describes the rough contours of the discussion, but cannot do justice to the rich interest of the specific issues canvassed. Part One alone ranges over the ethics of election campaigning, tax evasion, immigration, privacy, and artistic in­decency, finding not just isolated mentions of these issues but extensive debates among a large network of authors. Although the names discussed are often obscure, and even relatively well-­known figures like Luis de Molina are presented in unfamiliar contexts, Schwartz takes no time over biographical introductions or other historical niceties. Instead, he dives straight into the content of these debates, treating the material as if it were continuous with twenty-­first century discussions. Readers looking for an overarching historical perspective might supplement this book with Rudolf Schuessler’s excellent Debate on Probable Opinions in the Scholastic Tradition (Brill, 2019), but the advantage of Schwartz is that he takes us directly to the philosophical payoff, leaving others to supply the background. Claude Panaccio, Récit et reconstruction: Les fondements de la méthode en histoire de la philosophie (Vrin, 2019), 230 pp. Having retired from his professorship in Montreal, Claude Panaccio here steps back from his long career of research into nominalism to reflect on the larger question of method in the history of philosophy. The initial questions are the ones we all find fascinating at the start of our career and then at some point start to find tedious: to what extent is the history of philosophy continuous with philosophy today? how do we engage with that history while avoiding anachronism? But Panaccio is too much the philosopher, and too creative a thinker, to rest content with these old chestnuts. The book in fact comes to devote itself principally to thinking at a much more abstract level about what exactly it is that historians of philosophy study, and how that subject can be continuously studied over centuries. What gives these less familiar questions their bite, for Panaccio, is that, even when thinking about historical methodology, he remains a nominalist. Thus he cannot simply and vaguely say that, in studying (e.g.) Ockham’s thought, we are interested in the ideas that his words express. We have to explain, in concrete nominalist fashion, exactly which concrete words we are talking about and which concrete ideas. To this end, Panaccio must tell a story about how the ink on the page of a book in his library provides information about the ur-­text (l’exemplaire étalon) that is the ultimate concrete subject of inquiry, and how that singular text relates to specific ideas in the author’s mind. Then he needs to explain how there can be

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

292  Briefly Noted con­tinu­ity across time and across languages between those fourteenth-­century res and the things we prod­uce when we think, talk, and write about Ockham. To explain this, Panaccio embarks on a nominalist semantics for the transmission of ideas through linguistic and theoretical change. By the end, the familiar questions of methodology with which he begins take on a whole new interest.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Notes for Contributors OSMP welcomes submissions in all areas of medieval philosophy. Papers received will be evaluated in a timely manner and an effort to provide significant feedback will be made in every case. To the fullest extent possible, papers will be refereed according to a triple-­blind process, so that the author will not know the identity of the referee, the referee will not know the identity of the author, and the editor will learn the identity of the author only after a first-­round decision about the paper has been reached. In addition to articles, we welcome editions of texts and brief critical discussions of recently published articles (both in OSMP and elsewhere). Book reviews, however, will be published only when solicited. Submissions should be in English, without the author’s name or any other information that would impede blind refereeing. Papers may be of any length, although contributors should bear in mind that longer submissions will be judged by higher standard for acceptance. Papers should be submitted as either.pdf or Word-­compatible files. The formatting of the initial submission is immaterial, but accepted papers will ultimately need to adhere to OSMP style, as on display in this present volume. All submissions, as well as queries, should be addressed to [email protected]. Please do not send queries directly to the editor, since this compromises the process of blind review.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

Index of Names For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. ʿAbd al-Jabbār  56–8, 65–6 Adams, Marilyn McCord  225, 242–3 Adamson, Peter  290 Albert the Great  150–1, 163–72, 193–4 Alexander of Aphrodisias  3, 6–7, 80, 87, 90–2, 107, 169n.68 ʿAlī Ibn al-Samḥ  41–5, 58–64, 68, 70–1 Aristotle  7, 32, 80–95, 99n.46, 100–1, 102n.54, 104–5, 107, 112, 114–15, 123, 131–2, 134–6, 147–9, 150n.14, 156–7, 162, 167–8, 180, 187–95, 197–9, 202–9, 211–12, 214–18, 252n.11 Augustine  1–34, 41n.3, 61–2, 187–8, 196n.18, 198–9, 215n.82, 218–19, 247 Averroes  44–5, 191 Avicenna  52, 67–8, 77–84, 93–139, 163–5, 175–8 Boethius  147–8, 161, 168, 173–4, 178, 214 Boethius of Dacia  171–2 Bonaventure 187–219 Bourke, Vernon J.  14–15 Brittain, Charles  15–17 Clanton, J. Caleb  226 Cleomedes 20–1 Cottrell, Jane E.  26–7 Descartes, René  6–7 Dutton, Blake D.  20

Empedocles 4 Epicurus 20–1 Euclid  9, 22–4 al-Fārābī  46, 67–8 Francisco Suárez  24–5, 30–1, 246–84 Freppert, Lucan  241n.45 Galen  6–7, 22, 24–5, 27–8, 32 al-Ghazālī  44–5, 60n.60, 63, 67–8, 70–1, 79 Gilson, Étienne  15, 187–90, 195 Hume, David  40–1, 61–2 Ḥ unayn Ibn Isḥāq al-‘Ibādī  48, 64–5, 90–1 Iamblichus 4 Ibn al-Khammār  68n.80 Ibn Rushd, see Averroes Ibn Sīnā, see Avicenna Ibn al-Ṭayyib  41–5, 64–8, 71 al-ʿIrāqī, Muḥammad ʿĀt ̣if 106–7, 131, 133–4 Irwin, Terence  223, 239–41 ʿĪsā Ibn Zurʿa  41–5, 51–9, 64–71 Isḥāq Ibn Ḥ unayn  85n.19, 91–2 John Duns Scotus  189n.3 al-Kindī, Abū Yūsuf  22–4, 106n.63, 158n.36

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 23/06/21, SPi

296  Index of Names Longino, Helen  151–5 MacFarlane, John  149–50 Marius Victorinus  11–12 Martin, Kraig  226 Mayer, Toby  98, 108 McGinnis, Jon  80, 98n.45, 107–8, 124–5 McGrade, A. S.  223–4 Merleau-Ponty, Maurice  4 Nemesius of Emesa  9, 29–30 Nicholas of Paris  150–1, 157–63 Noë, Alva  24 Panaccio, Claude  291–2 Peter Auriol  253–6 Peter of John Olivi  1, 4 Peter of Spain  147–8 Philoponus  32, 80, 87–8, 90–2, 101n.51, 107, 114n.87, 128–30 Pierre Ceffons  290–1 Plato  3–4, 6–7, 10–12, 14–15, 24–5, 27–9, 89n.22, 96n.41, 107, 188–9, 191, 195–6, 198–9, 208n.60 Plotinus  3–6, 11–12, 15, 20–1, 212n.74 Porphyry  4–6, 11–12, 20–1, 147–8, 212n.74 Priscian  4, 18–19

Proclus  4, 33–4, 114n.87 Ptolemy  2, 22–4, 27–8 Radulphus Brito  150–1, 172–80 al-Rāzī, Abū Bakr  70 al-Rāzī, Fakhr al-Dīn  95n.40, 96n.41 Roger Bacon  1, 157n.31 Schabel, Chris  290–1 Schwartz, Daniel  291 Shayegan, Yegane  107–8, 121n.100, 124–5 Simon of Faversham  147–8, 171–2 Simplicius  87–8, 90–2, 163 Sutton, Peter  1 Themistius 91–2 Theodore Abū Qurrah  48 Thomas Aquinas  250–2, 271 van Steenberghen, Fernand  187–8, 190–1, 195 Ward, Thomas  224, 229n.15 William of Ockham  222–44, 255–6, 260–2, 265 Winer, Gerald A.  26–7 Yaḥyā Ibn ʿAdī  41–52, 54–8, 62, 64–70