Oxford Studies in Medieval Philosophy Volume 8 9780198865728, 9780198865742, 0198865724

Table of contents :
Cover
Oxford Studies in Medieval Philosophy Volume 8
Copyright
Contents
What Does the Happy Life Require?: Augustine on What the Summum Bonum Includes
1. Introduction
2. Augustine’s Conditions for True Happiness
3. Augustine’s Initial Agreement with the Stoics
4. The Mature Augustine: Happiness Includes the Body and Externals and Cannot Be Currently Achieved
5. Augustine against the Stoics on the Value of Natural Goods
6. Virtue and Loving Goods Now
7. The Emotions and Caring for Others’ Happiness in the Good Life
8. Augustine on Happiness and Security
9. The Future Life and Augustine’s Eudaimonism
10. Conclusion
Bibliography
Perception in Augustine’s De Trinitate 11: A Non-Trinitarian Analysis
1. Introduction
2. Consciousness and Levels of Awareness in Augustine
3. Outer Vision and Inner Vision in De Trin. 11
a. Outer Vision
b. Inner Vision
4. Limits of the Trinitarian Analysis: Locating Perception in De Trin. 11
a. Negative Thesis: Outer Vision Is Not Perception
b. Positive Thesis: Perception Is a Hybrid State
5. Memory and the Trinitarian Mind
Bibliography
The Justice and Mercy of God: Seneca in the Eleventh Century
1. Seneca on Imperial Clemency
2. Anselm: The Justice and the Mercy of God
3. A Common Template?
4. Influence or Reinvention: Seneca in the Eleventh Century?
5. Conclusion
Bibliography
Meditating on the Meditations: Al-Ghazālī, Teresa of Ávila, Descartes
1. Descartes, Teresa, al-Ghazālī
a. Descartes
b. Teresa
c. The Youthful al-Ghazālī—The Cognitive Problem
d. Al-Ghazālī’s Second Crisis—The Sufi Challenge
e. Further Lessons
2. Cognitive and Moral Certitude
3. Towards an Unsettled Historiography
Bibliography
Thomas Aquinas on How the Soul Moves the Body
1. Self-Motion and the Grades of Life
a. Action among Inanimate Bodies
b. The Self-Motion of Plants
c. The Self-Motion of Animals and Humans
2. The Diversity of Parts and the Grades of Physical Life
a. The Diversity Proper to Living Beings
b. An Objection from Divine Simplicity
c. How the Soul Moves the Body
3. Conclusion: Situating the Soul Relative to other Principles of Action
Bibliography
Thomas of Sutton’s Intellectualist Doctrine of the Will’s Self-Motion
1. Introduction
2. Thomas Aquinas on Specification, Exercise, and the Will’s Self-Motion
3. Thomas of Sutton on Specification and Exercise
4. Thomas of Sutton on the Will’s Self-Motion
5. A Development of Aquinas’s View?
6. Conclusion
Bibliography
Walter Burley on Co-Signification in Opaque Contexts
1. The Challenge of Opacity
2. Direct Signification
3. Real Propositions and Mental Language
4. The Semantics of Propositional Attitude Reports
5. Conclusion
Bibliography
Social Powers and Mental Relations: William Ockham on the Semantics and Ontology of Lordship and Ownership
1. The Origin and Nature of Lordship and Ownership
2. The Ontological Status of Ownership
a. Reducing Ownership I: Mental Relations
b. Reducing Ownership II: ‘Owner’—‘Ownership’ and the Semantics of Abstract and Concrete Terms
3. Is Ownership Real?
4. Conclusion
Bibliography
Briefly Noted
Notes for Contributors
Index of Names

Citation preview

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

Oxford Studies in Medieval Philosophy Volume 8

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

A DV IS O R Y B O A R D 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 Mark Boespflug, University of Colorado

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

Oxford Studies in Medieval Philosophy Volume 8 Edited by ROBERT PASNAU

1

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

3

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 2020 The moral rights of the authors have been asserted First Edition published in 2020 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: 2014254512 ISBN 978–0–19–886572–8 (hbk.) 978–0–19–886574–2 (pbk.) 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, 17/8/2020, SPi

Contents Articles

What Does the Happy Life Require? Augustine on What the Summum Bonum Includes

1

Caleb Cohoe

Perception in Augustine’s De Trinitate 11: A Non-Trinitarian Analysis

41

Susan Brower-Toland

The Justice and Mercy of God: Seneca in the Eleventh Century

79

Mary Sirridge

Meditating on the Meditations: Al-Ghazālī, Teresa of Ávila, Descartes

111

Taneli Kukkonen

Thomas Aquinas on How the Soul Moves the Body

144

David Cory

Thomas of Sutton’s Intellectualist Doctrine of the Will’s Self-Motion

187

Michael Szlachta

Walter Burley on Co-Signification in Opaque Contexts

221

Nathaniel Bulthuis

Social Powers and Mental Relations: William Ockham on the Semantics and Ontology of Lordship and Ownership

248

Jenny Pelletier

Briefly Noted Schuessler—Vasalou—Faucher—Aquinas’s Quodlibeta

277

Notes for Contributors Index of Names

281 283

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

What Does the Happy Life Require? Augustine on What the Summum Bonum Includes Caleb Cohoe

1. Introduction Many critics of religion insist that believing in a future life makes us less able to value our present activities and distracts us from accomplishing good in this world.¹ Augustine is a frequent target, given his insistence that this life is a vale of tears, where we must wait in hope for a better life, while only loving things insofar as they relate to God. Martha Nussbaum insists that “Augustinian love is committed to denying the importance of worldly losses and injustices.”² She claims that an orientation towards God and towards a future life leads one to neglect the importance of justice and charity in this life. “Death is irrelevant, real suffering in this world is irrelevant, all that is relevant is coming into God’s presence.”³ Instead of looking to the next life, we should be “directing compassion altogether toward the theater of history and not at all toward the shadowy and uncertain realm that may or may not lie outside it.”⁴ Augustine’s focus on connection to God leaves him without a strong foundation for caring “when people are hungry, when they mourn, when they are persecuted.”⁵

¹ For example, Zarathustra speaks of the “soporific virtues” that religion engenders, passively waiting for a future life: “blessed are the sleepy ones because they will soon drop off ” (Friedrich Nietzsche, Thus Spake Zarathustra, Part 1, “On the Virtuous”). ² Martha C. Nussbaum, Upheavals of Thought: The Intelligence of Emotions (Cambridge: Cambridge University Press, 2001), 552. ³ Ibid. ⁴ Ibid., 553. ⁵ Ibid., 552. Caleb Cohoe, What Does the Happy Life Require? Augustine on What the Summum Bonum Includes In: Oxford Studies in Medieval Philosophy Volume 8. Edited by: Robert Pasnau, Oxford University Press (2020). © Caleb Cohoe. DOI: 10.1093/oso/9780198865728.003.0001

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

2

 

In this paper, I show that Nussbaum has things backwards. It is while Augustine is trying to achieve happiness in this life that he denies “the importance of worldly losses and injustices,” treats suffering as “irrelevant,” and focuses on cultivating his own divine contemplation, neglecting the material world. His attitudes change once he comes to believe that true happiness can only be achieved in the next life. In fact, this belief is what enables Augustine to compassionately engage with present suffering and allows him to see material and social goods as part of the summum bonum.⁶ The future life plays two key roles in Augustine’s ethics. Looking to a future life allows Augustine to affirm the value of compassion (misericordia) in this life. As long as Augustine thinks happiness is attainable in this life, he insists that we should aim to avoid any negative emotions. By postponing complete happiness to the next life, Augustine can continue to agree with the Stoics and other ancient philosophers that the blessed life must be tranquil, while taking a very different position on what a good human life in our present condition looks like. We should experience sorrow and fear, for ourselves and others, when such emotions help us to better direct our steps. Christians mourn with those who mourn and seek to relieve their distress, looking forward to a future state in which there will be no tears. Secondly, the hope of a perfect future life enables Augustine to develop an inclusive and expansive view of which goods we should seek as part of our happiness. The early Augustine denies that the state of our bodies or the well-being of our friends and loved ones affect our happiness. While the mature Augustine continues to hold that happiness is centered on ⁶ Translations throughout are mine unless noted. Throughout this paper, my analysis uses both ‘happiness’ and ‘blessedness’ to speak of the highest good which Augustine thinks all rational beings desire. ‘Happiness’ picks up on Augustine’s insistence that everyone is seeking beatitudo and that we cannot fail to will it. However, as many have noted, beatitudo (the word Augustine uses as the Latin equivalent of eudaimonia) refers primarily to the overall status of one’s life, not to a momentary feeling or psychological state, and so has similarities with contemporary notions of well-being. ‘Blessedness’ picks up on this, as well as on the idea that the good life should be secure and untroubled. In translating Augustine, I consistently render beatus and beatitudo as “blessed” and “blessedness” and felix and felicitas as “happy” and “happiness.” As we shall see, Augustine usually equates beatitudo with the summum bonum and takes it to involve not just delighting in the good, but securely possessing it (De Civitate Dei [civ. Dei] XI 13; cf. civ. Dei XIX 3–4), whereas felicitas requires only that one be enjoying good things, even if one’s state is not secure (e.g. Adam and Eve before the fall (civ. Dei XIV 10).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

3

enjoying God, he now insists that we should also take the well-being of our minds and bodies and our friendships with others to be part of our blessedness. For Augustine, the awaited triumph of the whole city of God allows us to expand our circle of care outwards to love all humans and seek a common happiness with all who share our nature. Christians look forward to the entire city of God living together in perfect peace and everlasting love.

2. Augustine’s Conditions for True Happiness From De Beata Vita, his first extant work, through to the end of his writings, Augustine’s inquiries are driven by his search for the blessed life. As Étienne Gilson puts it “Augustine desires truth for the sake of happiness.”⁷ For Augustine, pursuit of intellectual questions should never be idle or disconnected from life, it should always be tied to our desires for fulfillment. Augustine returns again and again to the question of what the ultimate good, the summum bonum, is and how to obtain it. To see why he initially takes a more restrictive view of happiness, we need to start with the conditions Augustine thinks the blessed life must meet. Throughout his writings, he consistently insists that to be blessed one must meet two key conditions, both of which are inspired by the Stoics.⁸ We see the first condition come out in an early definition of happiness: What else is it to live blessedly but to possess an eternal object through knowing it (aliquid cognoscendo habere)? For the eternal is that in which alone one can rightly place his confidence, it is that which cannot

⁷ The Christian Philosophy of Saint Augustine (New York: Random House, 1960), 5. Cf. Robert O’Connell, Augustine’s Early Theory of Man,  386–91 (Cambridge, MA: Harvard University Press, 1968), 205: “if there is one constant running through all of Augustine’s thinking, not only in [his] early period but throughout his career, it is his preoccupation with the question of happiness.” ⁸ For further discussion of the ancient philosophical background to Augustine’s views on happiness see Knut Ragnar Holte, Béatitude et sagesse: Saint Augustin et le problème de la fin de l’homme dans la philosophie ancienne (Paris: Études Augustiniennnes, 1962).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

4

  be taken away from the one who loves it (amanti auferri non potest), and it is that very thing which one possesses solely by knowing it.⁹

Augustine here emphasizes the security of the object enjoyed. Unlike wealth or a delicious meal, the eternal object of knowledge is not liable to be taken away against one’s will. The one who truly loves the eternal must also truly know it, in order to love it. Since knowledge and love always go together, you cannot truly love the eternal without also knowing it and thus possessing it. There is no risk of disappointment or loss in loving something eternal, as it does not change. It will always be accessible and it will always be worth loving. Augustine’s reasoning here is based around what I will call his Security Condition: We only properly count as beati when we possess and enjoy the summum bonum in such a way that we cannot lose it unwillingly.

If, against our will, we can lose a good we are enjoying, then this possibility of loss either directly ruins our enjoyment or makes our state less than optimal. Hence we will be experiencing something less than the summum bonum.¹⁰ As we will see, the need for security motivates both the Stoics and the early Augustine to exclude bodily and social goods from the summum bonum. Despite Augustine’s framing of security in relation to an eternal object, ancient philosophers endorsed this condition without holding that the supreme good is God or that we can live forever. Indeed, Augustine is following the Stoics’ insistence that a truly happy life requires securely enjoying the things in which your good consists: Who is able to be blessed (beatus) when he is unsure of the good things he has? . . . Who will be able to be assured of either strength of body or stability of fortune? But no one is able to be blessed unless his good is stable, fixed, and enduring (stabili et fixo et permanente).¹¹

For the Stoics, only virtue cannot be taken away from you against your will, so only virtue is good and all the virtuous are happy. The Security ⁹ de diversis quaestionibus octoginta tribus, q. 35. ¹⁰ Cf. civ. Dei XI 13; XIV 25; XIX 10; De Trinitate (trin.) XIII 5–6. ¹¹ Cicero, Tusc. Disp. V 13.40–14.40.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

5

Condition also explains why Augustine agrees with the Stoics in thinking that tranquility is required for blessedness. If we possess goods in such a way that they cannot be taken away from us unwillingly, we will not worry about losing them and so we will not experience fear.¹² By contrast, if we enjoy having certain goods but do not stably possess them, we will fear losing them. This negative emotion will then impair our current enjoyment and our happiness. Lack of security undermines our experience of good things. We also need to have all the goods that we desire. Following Cicero, whom he consistently refers to throughout his career, Augustine insists that “someone is only blessed if he has everything which he wills and wills nothing wrongly (habet omnia quae vult, et nihil vult male).”¹³ I will call this Augustine’s Complete Satisfaction Condition: Beatitudo both requires and consists in having all that one wills.

For Augustine, blessedness requires having all that we desire. If we fail to obtain something that we will, we are not truly blessed. If I love something and take that thing to be necessary to my good but have not obtained it, I cannot yet count as happy. The summum bonum must be sufficient on its own to make my life worth choosing and lacking nothing.¹⁴ I cannot have achieved the summum bonum if, by my own lights, I am lacking some important part of my good.¹⁵ As in the case of the Security Condition, meeting this condition leads to tranquility. If I am completely satisfied, I will be experiencing the positive feeling of joy or contentment.¹⁶ I will, of course, not have any unmet desire for future goods to pain me and I will not be experiencing pain or sorrow about goods I currently lack or evils that are present.¹⁷ The Complete Satisfaction Condition guarantees psychological well-being.¹⁸ ¹² Metus, translating the Stoic phobos. De Libero Arbitrio (lib. arb.), I 4–5, 11–12; civ. Dei XIV 8; cf. civ. Dei XI 13; XIV 25; XIX 10; De Trinitate (trin.), XIII 5–6. ¹³ trin. XIII 5.8; cf. de Beata Vita (b. vita) 4.25; for discussion of this definition in Cicero see Alberto Grilli in Cicero, Ortensio, testo critico (Bologna: Pàtron Editore, 2010), fragm. 59, pp. 66–9. ¹⁴ Cf. Aristotle, EN I 7, 1097b15–18. ¹⁵ Cf. civ. Dei XIV 25, trin. XIII 5–6. ¹⁶ Gaudium, translating the Stoic chara. ¹⁷ Cupiditas, translating the Stoic epithumia, and tristitia, translating the Stoic lupē. ¹⁸ civ. Dei XIX 8; cf. A. A. Long and D. N. Sedley, The Hellenistic Philosophers, 2 vols. (Cambridge: Cambridge University Press, 1987), fragments 65A–E.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

6

 

Augustine consistently takes these two conditions to be necessary and jointly sufficient for achieving happiness. People who meet both conditions will enjoy the summum bonum in undisturbed peace. Since they meet the Complete Satisfaction Condition, they will delight in what they have without any unfulfilled desires or any pain or sorrow. Since they meet the Security Condition, they will not fear losing what they love. While Augustine’s conditions are inspired by the Stoics, he parts ways with them from the beginning when it comes to the question of how to achieve the ultimate good. He follows Platonists such as Plotinus in his contemplative understanding of virtue.¹⁹ Instead of the Stoic conception of attaining blessedness through making excellent choices about our external actions, Augustine, throughout his writings, holds that happiness should be pursued through directing our understanding and our love upwards. Augustine consistently holds that happiness comes from the soul clinging to God. God is to be enjoyed, frui, and other rational creatures are to be enjoyed in God, frui in Deo.²⁰ This, however, leaves open the question of how we should value our bodies and external goods. What role, if any, do our bodies, our friendships, and the state of our society play in our happiness? It is here that Augustine radically changes his views.

3. Augustine’s Initial Agreement with the Stoics Before seeing how Augustine’s thought is transformed by looking to a future life for happiness, we need to understand his early views on happiness, when he still thought it was attainable in this life. As we will see, the disconnected and solely spiritual view of happiness that Nussbaum critiques is precisely the view Augustine adopts when he thinks that happiness is possible in this life and attempts to achieve it. The early Augustine lacks feeling for the suffering of others, while it is the mature Augustine, waiting for future blessedness, who shares in their griefs and pains and seeks to relieve them. ¹⁹ E.g. Soliloquia I.6.13. ²⁰ de doctrina Christiana I.22.20–I.34.38; cf. civ. Dei XIX 13.1; trin IX 8.13.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

7

Once Augustine converts to Christianity and abandons worldly pursuits to dedicate himself to philosophy, Augustine believes that blessedness is within his grasp.²¹ The early Augustine follows Cicero (influenced by Aristotle, among others) in suggesting that excellence for human beings is found in using our best part well: “what else do you think a blessed life is, except to live in accordance with that which is the best element in the human being? . . . That part . . . may be called mind or reason.”²² This means that if we achieve wisdom and use our reason well, we also achieve blessedness. At this point in his life, Augustine agrees with the Stoics that bodily and external things cannot meet the Security Condition. Since they can be lost, they cannot be part of the summum bonum. Further, if we were to intrinsically value them, our desires would often be unfulfilled and we would fail to meet the Complete Satisfaction Condition. This is why the newly converted Augustine insists that changeable goods cannot contribute to happiness. Indeed, at this point, Augustine, with the Stoics, insists that a good will is sufficient for virtue and happiness.²³ In lib. arb., he argues that the life of those with a good will is praiseworthy and not to be avoided. But the wretched life is to be avoided. Since lives must be either wretched or blessed, the life of those with a good will must be blessed.²⁴ Now, like the

²¹ In the introduction to his b. vita, Augustine states that if Theodore, the dedicant, helps him, “with a little effort I will come easily to that blessed life to which I assume you already hold fast” (b. vita 5). While we should read this claim within the context of dedicatory rhetoric, the language is still striking. ²² Contra Academicos (c. Acad.) I 2.5; cf. b. vita 2.7–8; Cicero, de Finibus [fin.] 5.13.36; 14.38. Given that we either are our minds and souls, or are them most of all, it is their excellences that make us good and blessed: “If you ask for my opinion, I judge that the supreme good of a human is in the mind” (c. Acad. III 12.27). The later Augustine will include more than the excellence of the mind in the human good, but will continue to take reason or mind to be superior to the body (Retractationes [retr.] 1.1.2; civ. Dei XII 21–3, 26; XIX 3; XX 29–30). ²³ In I 13, Augustine argues that those with a good will possess all four of the cardinal virtues, due to their love of things within their power and lack of attachment to things outside of their power. ²⁴ I 13. As Michael Frede rightly notes, Augustine operates with the “stock Stoic contrast of wisdom and folly” throughout lib. arb. (A Free Will: Origins of the Notion in Ancient Thought (Berkeley: University of California Press, 2011), 166). Here Augustine is also using the disjunctive view on happiness endorsed by the Stoics: our lives are either blessed or wretched, with no intermediate states. Augustine and Evodius both insist that attaining a good will provides us with joy which “calmly, quietly, and steadily bears up the soul” (I.13.29, tr. Williams). This joyful possession of such a will “is the blessed life. For doesn’t a blessed life consist precisely in the enjoyment of true and unshakeable goods?” (ibid.). Cf. John Bussanich, “Happiness, Eudaimonism,” in A. D. Fitzgerald (ed.), Augustine Through the Ages: An Encyclopedia

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

8

 

Stoics, Augustine insists that his view is compatible with being careful about one’s body, wealth, and other externals, as this displays one’s wisdom and virtue. [The wise person] will avoid death and pain, insofar as he is able and insofar as it is fitting to do so. If he did not avoid these things at all, he would be wretched, not because they happened to him, but because when he was able to avoid them, he willed not to (noluit), which is a clear indication of foolishness. In failing to avoid such things, then, he will be wretched not through enduring them, but through his foolishness.²⁵

If you are enduring death and pain because of poor and foolish choices, then you are wretched precisely because you made such choices. The suffering itself is not bad and is not what makes you miserable. Monica observes, with Augustine’s approval, that the person who has all material goods “is not blessed through the possession of these things, but through the moderation of his mind.”²⁶ Using such goods well is evidence of happiness, but one’s blessedness comes from the exercise of virtue, not from the externals themselves. Indeed, Augustine insists that the wise person will be happy no matter the circumstances: If, however, he does not prevail in avoiding such happenings [i.e. death, pain, etc.], although he has attended to this carefully and appropriately, they will not make him wretched . . . .“Since what you will cannot be, will what can.” [Terence, Andria, II 1, 305–306] In what way will someone be wretched to whom nothing happens against his will? Because he is unable to will anything which it seems to him he cannot bring about. He sets his will on the most certain things, that is, in whatever he pursues, he pursues it only according to virtuous duty and the divine law of wisdom, and these things can in no way be taken from him.²⁷

(Grand Rapids, MI: Eerdmans, 1999), 413–14: “Like the Stoics, [Augustine] insists that beatitude must be immune from fortune and independent of external goods.” ²⁵ b. vita 4.25.

²⁶ b. vita 2.11.

²⁷ b. vita 4.25.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

9

Augustine insists that we should not love anything external. We should not unconditionally endorse desires to avoid pain, injury, and death. If you suffer such things despite acting wisely, then you remain happy, because you continue to act from duty and follow divine wisdom. This is sufficient for blessedness. True blessedness also requires tranquility, so we cannot care about externals. Worrying about what happens to us and others and experiencing sorrow and fear over circumstances would destroy the happiness at which we are aiming. The qualities of the soul alone are what determine our blessedness or wretchedness. Our souls maintain their perfect and self-sufficient condition through avoiding fear or anger about external states. The early Augustine closely mirrors Stoic teaching, advising us to avoid any negative emotions and to take externals as irrelevant to achieving the human good.²⁸

²⁸ Cf. lib. arb., where Evodius claims to understand “the nature of that blameworthy cupidity that is called inordinate desire. Obviously it is the love of those things that one can lose against one’s will” (I.4, tr. Williams). He is implying, like the Stoics, that any attachment to such things would be excessive. Indeed, Evodius is quite clear that externals should not be intrinsically valued at all: we should always be willing to abandon them and even give up our life rather than injuring someone else (I.5.12). Augustine does not go quite so far as to say that these externals are never worth pursuing, but he is clear that valuing them is a distraction and that the good will is on a whole other scale of worth (I.12). Augustine’s description of the person lacking a good will and ordered desires closely resembles Epictetus’s description of the person who values externals: Fear attacks from one side and desire from the other; from one side, anxiety; from the other, an empty and deceptive happiness; from one side, the agony of losing what one loved; from the other, the passion to acquire what one did not have. (lib. arb I.11, tr. Williams) Whoever longs for things that are not within his power, or seeks to avoid them, can neither be trustworthy nor free, but must necessarily be subject to change, and be tossed in all directions along with those things, and is inevitably placing himself under the domination of other people, namely, those who can secure or prevent such things. (Disc. I.5.19, tr. Hard) Even the early Augustine, however, does not deny that we can call excellent states of the body or external things good. He never endorses the Stoic claim that virtue is the only good. He is already committed to the Platonic position on the convertibility of being and goodness: whatever is, is good, especially insofar as it participates in order, number, and measure. Nevertheless, calling these things good in some way is compatible with denying that they partially constitute the supreme good for humans. My clothing can be good (by fitting well, providing appropriate protection from the environment, etc.) or bad (being torn or broken, fitting poorly, etc.) without its good or bad condition affecting whether my life is good or bad. Just so, states of the body, relationships, or reputation and wealth may be acknowledged as good or bad without thinking that they themselves constitute (even partially) the human good.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

10   Indeed, at this point, Augustine thinks we should put no value on temporal relationships, on ties of family or earthly loves. We should only care for each other’s souls: If we are on fire with love for eternity we will hate temporal bonds (temporales necessitudines) . . . .Whoever loves another as himself ought to love that in the other which he himself is. Our bodies are not truly what we are, so we should not place our wishes and desires in the human body.²⁹

The young Augustine insists that such an attitude is not inhuman, but the proper way to love a rational soul. Although we should use temporal things well and show concern for others, what happens to them does not affect our well-being. [The just person] is not made sorrowful by the death of anyone . . . . He is not made wretched by the wretchedness of another, any more than he is made unjust by the injustice of another. Just as no one can take from him justice and God, so no one can take from him his blessedness. If at some time he is perhaps moved with feeling by another’s danger or error or grief, he lets it go so far as to help or correct or console that other, but not so far as to overthrow his own wellbeing.³⁰

Like the Stoics, the early Augustine thinks we should help others, but without truly feeling their pain.³¹ Indeed, Augustine thinks we progress by entirely leaving temporal and perceptible things behind. At this point in his life, Augustine holds that happiness is attainable in this life through directing the mind’s eye to God. In his other early works De Ordine (ord.) and De Animae Quantitatae (quant.), he sets forth a program of education and spiritual ascent that is to lead to an unshakeable wisdom. This ascent “is done rather quickly, in some cases; in others, slowly,” but attaining contemplative knowledge of God is possible in this life, at least for the wise and dedicated.³² All seven steps of the ascent are activities of the soul that it can perform by itself allowing for the possibility of tranquility, through ²⁹ De Vera Religione [vera rel.] 46.89. ³⁰ vera rel. 47.91. ³¹ Cf. Seneca, De Clementia, II.5–6; Epictetus, Enchiridion, 16.

³² quant. 36.80.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

11

separating ourselves from the cares and worries of life.³³ When such heights are reached, “death, the flight and escape from this body in every way, which previously was dreaded, is now yearned for as the ultimate gift.”³⁴ The soul, having attained its end, no longer values or cares for anything temporal. As Robert O’Connell summarizes Augustine’s early views: “Heaven can be dwelt in even now; even in the body, the soul can achieve effective disincarnation, unabating contemplation, unbroken happiness.”³⁵ Augustine does affirm the resurrection of the body, but its purpose remains obscure.³⁶ In order to achieve happiness in this life, the early Augustine advocates for a disconnected and solely spiritual view of happiness. If we are to be secure and completely satisfied during our mortal lives, we cannot value anything external. We cannot wholeheartedly endorse desires about the state of our bodies or what happens to us or our friends, because such desires might fail to be satisfied, robbing us of blessedness. We must turn away from the body and emotionally withdraw from others, avoiding grief even at the death of a loved one. While Augustine always holds that we should help others, at this point he insists that we must avoid being burdened by their sufferings if we are to be at peace ourselves.

4. The Mature Augustine: Happiness Includes the Body and Externals and Cannot Be Currently Achieved Does Augustine hold on to this solely spiritual view of happiness throughout his life or does he change his mind? Some scholars insist that Augustine’s overall views on happiness are fundamentally the same throughout his post-conversion life. Étienne Gilson claims that

³³ quant. 35.79. ³⁴ quant. 33.76. Augustine notes that the seventh and last step is “not really a step, but a dwelling place . . . What shall I say are the delights, what the enjoyment of the supreme and true Good; what breath of serenity and eternity?” (ibid.). This is still a difficult goal, “to this knowledge few are able to arrive in this life, but no one is able to progress beyond it even after this life” (ord. II 9.26; Ad quam cognitionem in hac vita pervenire pauci, ultra quam vero etiam post hanc vitam nemo progredi potest). ³⁵ Augustine’s Early Theory of Man, 253; cf. 270. ³⁶ quant. 33.76.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

12   Augustine is consistent throughout his life in thinking that, “happiness is a good of the spirit, and cannot be endangered by the loss of any material good. Besides, a wise man desires only what is possible, precisely in order that he may not see his desires frustrated.”³⁷ John Bussanich similarly insists that, for Augustine, “happiness in reality is a transcendent and purely spiritual state.”³⁸ Gilson and Bussanich, however, fail to properly acknowledge that the views endorsed in Augustine’s early works are ones that he ultimately rejects. While Augustine does always think blessedness consists primarily in the knowledge and love of God achieved through our souls, his views on how we should value our bodies and the lives of others change radically. After Augustine reflects further on the Christian hope of the resurrection, he comes to think that what happens to our body and what happens to our friends and broader human society is crucial to our happiness. These interpreters have too exclusive a reading of the goods Augustine loves. Due to his changing views on happiness, he alters his views on how to live well in this life, putting much greater emphasis on helping others and sharing in their emotions and experiences, and insisting that tranquility and blessedness are not suitable goals in our present situation, where we should feel sorrow and fear and aim to purify ourselves. In his Reconsiderations, Augustine notes that his early claims about happiness were wrong: I said that, during this life, the blessed life dwells only in the soul of a wise person regardless of the condition of his body; yet the Apostle hopes for a perfect knowledge of God, the greatest that a human can have, in the life to come [cf. 1 Cor 13.12–13], for that alone should be called a blessed life where the incorruptible and immortal body will be subject to its spirit without any vexation or resistance.³⁹

Augustine is here making it clear that human wisdom is not enough for blessedness. First of all, true happiness requires a perfect knowledge of God, which we cannot yet have in this life. Secondly, blessedness requires an incorruptible and immortal body. Such a body cannot be harmed or ³⁷ The Christian Philosophy of St. Augustine, 5; cf. 249–50. ³⁹ retr. I 2.2; tr. Mary Inez Bogan with modifications.

³⁸ “Happiness,” 414.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

13

damaged and is also entirely in harmony with and under the control of the spirit and of reason (which in turn is entirely subservient to God). As he puts it in one of his sermons, “The Stoic . . . is quite mistaken (fallitur); I mean it’s simply untrue, it’s absolutely incorrect that a person who has the enjoyment of the virtue of his soul (in fructu virtus animae eius) is happy.”⁴⁰ Augustine no longer holds that human wisdom is enough on its own for blessedness. He is also now clearly taking a more inclusive view on what happiness consists in: it involves enjoying the good of mind and body as well as knowing God. How and why does Augustine change his mind? There are, no doubt, several factors. Both the difficulties of achieving perfection and his developing theology of grace and its relation to our wills play a role.⁴¹ Once Augustine changes his anthropology and comes to believe that our bodies are part of what we are, he can no longer hold that the condition of our bodies is irrelevant to the human good. At the same time, as we will see, his reflections on the resurrection and the triumphant city of God offer him a way to include more within the summum bonum without giving up on the Security and Complete Satisfaction Conditions. His personal experience of the loss of his mother and many friends also deepened his understanding of the value of social goods.⁴² His words in vera rel. perhaps ring true in a different way than he expected. When we have [changeable goods] we imagine that we do not love them, but when they begin to leave us we discover what kind of person we are. We have a thing without loving it when we can let it go without grieving (sine dolore discedit).⁴³

Augustine is also shaped by his pastoral experience consoling and guiding Christians in the midst of the collapse of the Western half of the Roman Empire. He has to consider the plight of those raped by soldiers ⁴⁰ Sermo 156.7, tr. Sarah Byers. ⁴¹ E.g. de spiritu et littera, de Natura et Gratia. ⁴² In this respect, the Confessions is an intermediate work. He is not as detached and scornful of external goods as in the dialogues of Cassiciacum but is still confused about how our experience of sadness fits with our desire for happiness (IX.12.29–13.34) and unsure of the appropriate degree to which we should suffer with others and feel sorrow and anger at our own condition and the conditions of others (IV.7.11–12; IV.12.18). ⁴³ vera rel. 47.92.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

14   and determine when clergy are allowed to flee danger and when they should stay with their flock.⁴⁴ His life ends with Hippo, the town where he has served as bishop for decades, besieged by the Vandals and sheltering hundreds of defeated soldiers and refugees.⁴⁵ Instead of tracing all the contours of his thought’s development, in the remainder of this paper, I lay out and examine the positions on happiness Augustine takes in his later works, especially in civ. Dei and trin. Focusing on his ultimate views will allow us to more effectively examine the effects that postponing happiness to the next life has for our present conduct. As we will see, Augustine takes the excellent condition of both our minds and bodies and our membership within a well-ordered and secure society to be part of the summum bonum precisely insofar as he thinks that they can be achieved securely and completely at the eschaton. Does Augustine fit within the classical eudaimonist tradition or is appealing to a future life a deformation of eudaimonism? Augustine is certainly transformed by his reflections on grace, human achievement, and the Christian doctrines of the resurrection of the body and the new heavens and earth. However, Augustine also changes his position to solve some of the inherent tensions in ancient thought on eudaimonia. He comes to reject the Stoic view that external goods are irrelevant to happiness, but he is also unconvinced by the Peripatetic view that happiness comes in degrees and only requires a moderate amount of such goods.⁴⁶ Augustine is drawn to a maximalist position, on which blessedness consists in securely possessing all the goods of body, soul, and society. From this perspective, looking to the hope of a future life is the only plausible way to seek such a summum bonum. Augustine changes his thought on what happiness involves not because he increasingly devalues the material world but because he comes to more fully recognize its value. Hope in the blessedness of the future life enables him to acknowledge the goodness and desirability of the world around us. ⁴⁴ civ. Dei I 16–18; ep. 228.2. ⁴⁵ See Serge Lancel, Saint Augustine (London: SCM Press, 2002), 473–5 for discussion of the end of Augustine’s life. For discussion of Augustine’s advice to his correspondents and how he practically applies his order of goods see Joseph Clair, Discerning the Good in the Letters and Sermons of Augustine (Oxford: Oxford University Press, 2016). ⁴⁶ For discussion of the debate see Julia Annas, The Morality of Happiness (Oxford: Oxford University Press, 1993), chs. 18–21.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

15

5. Augustine against the Stoics on the Value of Natural Goods In book XIX of the City of God Augustine critiques the attempts of the philosophers to achieve the summum bonum in this life. Augustine explicitly critiques the philosophers for putting the highest good in some condition of the human body or soul instead of looking to a higher reality, the supreme Good that is God. [The city of God] will reply [to the philosophers] that life eternal is the supreme good, death eternal the supreme evil, and that to obtain the one and escape the other we must live rightly . . . . As for those who have supposed that the supreme good and evil are to be found in this life, and have placed it either in the soul or the body, or in both . . . all these have, with a marvelous shallowness, sought to find their blessedness in this life and in themselves . . . . For what flood of eloquence can suffice to detail the miseries of this life? . . . For when, where, how, in this life can these primary objects of nature be possessed so that they may not be assailed by unforeseen calamities?⁴⁷

Augustine’s framing initially suggests that he is simply rejecting the views of the philosophers on theological grounds. On my reading, however, Augustine is using their own philosophical principles against them. Admittedly, Augustine’s initial critique in this passage is consistent with his early view that full human fulfillment requires our mind to be informed and governed by the best thing, God. This is why he always preferred the Platonists to the Stoics.⁴⁸ Just reading this, one might be tempted to Gilson’s reading on which beatitudo is wholly a matter of our spiritual condition. Indeed, Christian Tornau claims that Augustine continues to side with Cicero and the Stoics in “denying external and bodily things all relevance for happiness . . . whatever is relevant for

⁴⁷ civ. Dei XIX 4, tr. Marcus Dods with modifications. ⁴⁸ And why, perhaps, the Platonists largely drop out of the discussion of the philosophers in book XIX, despite his extensive engagement with them in earlier books (e.g. VIII–X, XIV).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

16   happiness is virtue.”⁴⁹ But this does not make sense of Augustine’s procedure in the rest of civ. Dei XIX. Why go on to catalogue all the ailments of the body, failures of friendship and family, and war and civil strife, if all these things are irrelevant to happiness? Such an interpretation does not explain Augustine’s extended catalogue of the miseries of this life. These, after all, were the sorts of things that the Augustine of b. vita insisted were irrelevant to happiness. Augustine critiques the current state of our minds, bodies, and relationships precisely because he has come to think that they are important to our blessedness. If our mental acuity, beauty, health, or friendships were irrelevant, then complaining about our deficiencies in these areas would be inappropriate, especially when addressing philosophers such as the Stoics, who insist that these are not real goods. Augustine’s catalogue of miseries shows how expansive his vision of the summum bonum is. The mature Augustine thinks that we can have rightly ordered desires for health, friendship, and social harmony. Augustine no longer holds that we need only perfect our minds to achieve happiness. Augustine has turned away from the Stoic view for philosophical as well as theological reasons, as we see in his use of Peripatetic dialectical arguments against them. Augustine’s discussion of the way illnesses and harms to our bodies and minds affect our happiness draws on Cicero’s critique of the Stoics in his de Finibus. There Cicero’s Piso insists that completing the human work and pursuing what is good by nature requires succeeding in obtaining the things at which one is aiming, not just, as the Stoics suggest, pursuing them: We love ourselves and want every aspect of mind and body to be perfect. This shows that we love all these aspects on their own account, and that they are of the greatest importance in determining whether we live well.⁵⁰

⁴⁹ Christian Tornau, “Happiness in this Life? Augustine on the Principle that Virtue Is SelfSufficient for Happiness”, in Ø. Rabbås, E. K. Emilsson, H. Fossheim, and M. Tuominen (eds.), The Quest for the Good Life: Ancient Philosophers on Happiness (Oxford: Oxford University Press, 2015), 272. ⁵⁰ fin. 5.13.37, tr. Raphael Woolf.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

17

The mature Augustine wholeheartedly agrees with this claim.⁵¹ Our good requires actually possessing full and stable excellences of both body and mind, states that we often lack and are fragile even when present.⁵² The rest of book XIX shows Augustine using the principles of the philosophers against them, pulling on tensions inherent to eudaimonism to argue that happiness cannot currently be achieved. Augustine continually appeals to the Security and Complete Satisfaction Conditions, but he now does so in order to show that our current lives cannot be truly blessed, not to restrict which things we should include in the summum bonum. The Stoics were right in thinking that bodily health, as well as friendships and social goods, cannot be securely achieved in this life, but they were wrong to conclude that these things are not part of the human good. We see Augustine’s explicit inclusion of externals within the sphere of goods sought for their own sake in his ep. 130 to Honora, where he considers which things Christians can appropriately request. Augustine tells Honora that Christians should wish and pray for mental and physical wholeness and friendship for both themselves and others. These have intrinsic value, unlike riches and earthly power, which Augustine continues to insist should only be valued as instruments: The health and friendship of a human being are sought for their own sake, but a sufficient amount of necessary goods is generally sought not for their own sake but for the sake of the two previous things, when they are sought in a proper fashion. But this health consists of life itself and of the wholeness of mind and body (incolumitas porro in ipsa vita, ac salute, atque integritate animi et corporis constituta est). Likewise, ⁵¹ Piso, following on this earlier passage, goes on to claim that “in fully realizing [the mind], [human nature] will attain its end, its supreme good. After all, understanding and reason are by far the outstanding elements” (fin. 5.14.40). He endorses the Aristotelian view that the best activity of our best part is sufficient for happiness. Augustine himself is sympathetic to this view in his own b. vita. However, in his Retractationes, he finds his earlier view insufficient insofar as it neglects the need for a transcendent connection to being itself and the divine life. “Insofar as the nature of man is concerned, there is nothing better in him than mind and reason—and yet the man who wills to live blessedly should not live according to this, for then he lives as man lives although, in order to be able to attain blessedness, ‘he should live as God lives [cf. 1 Peter 4:6].’ To attain this, our mind should not be self-contented, but should be subject to God” (Retractationes 1.4.3, tr. Bogan with modifications). ⁵² civ. Dei XIX 4.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

18   friendship should not be bounded by narrow limits, for it embraces all to whom we owe affection and love though it is inclined more eagerly toward some and more hesitantly toward others . . . .We ought to pray that, when we have these goods, we may retain them and that, when we do not have them, we may acquire them.⁵³

Here we see Augustine including the excellences of mind and body, whose fragility he bemoans in civ. Dei XIX, within the scope of the intrinsic goods for which the Christian should pray. It is proper, Augustine insists, to will that you and those you love “live safely and soundly.”⁵⁴ Tornau is thus wrong to say that “Augustine fully endorses the traditional ideal of the sage’s—or the saint’s—independence from everything external and bodily.”⁵⁵ Augustine now includes our friendships and the state of our bodies within the scope of our care. How can Augustine include these states of mind and body within the summum bonum, given their fragility? Augustine has not given up on the Security and Complete Satisfaction Conditions, conditions he appeals to repeatedly within civ. Dei.⁵⁶ He does so precisely because he is looking to a future happiness. These conditions are fragile in this life, but, Augustine insists, things will change after the resurrection of the dead. I may not have the health and mental acuity I will right now, but in the eschaton I will have mental and bodily perfections that I cannot lose and whose abilities satisfy all my desires. Augustine insists that external goods are included in the heavenly city, “where peace is entirely complete and entirely certain (ubi pax plenissima atque certissima est).”⁵⁷ [In that city] there will be gifts of nature (naturae munera), that is, those things of our nature (naturae nostrae) that are given by the Creator of all natures, gifts not only good, but also truly everlasting; gifts not only of the spirit, healed now by wisdom, but also truly of the body, revived by the resurrection.⁵⁸

While Augustine rejects the idea that our highest good could be found through excellences of body or soul on their own apart from God, he ⁵³ ep. 130.6.13, tr. Roland Teske with modifications. ⁵⁴ ep. 130.5.11. ⁵⁵ “Happiness in This Life?” 279. ⁵⁶ E.g. XI 12, XIX 4. ⁵⁷ civ. Dei XIX 10. ⁵⁸ civ. Dei XIX 10.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

19

strongly affirms the inclusion of all human goods, both bodily and mental, within the ultimate human good. As we saw when introducing the Security Condition, Cicero and the Stoics thought that no features of your body, relationships, or social standing could meet this condition, asking who can be “assured of either strength of body or stability of fortune?”⁵⁹ The early Augustine feels the force of this objection but also deeply loves his friends, seeking their wellbeing and sorrowing at their losses, even when he thinks he should not. He is torn. The mature Augustine has an answer to the Stoics. After the resurrection, our bodies can be in a state worthy of being called good, where they will comply perfectly with the soul, where they cannot be harmed, and where they do make a positive contribution to blessedness. For Augustine, it is because the goods of body and soul will ultimately meet the Security and Complete Satisfaction Conditions, that it is appropriate to will and value them. Christians should seek and pray for the goods of temporal health and friendship, looking forward to a future in which they will securely enjoy all the goods of human nature. Another key feature of Augustine’s argumentation in civ. Dei XIX is the idea that the supreme good must be unsurpassable. This, again, is not primarily a Christian stipulation. This characterization goes back to Aristotle’s insistence that the best good must be teleion, final or complete, and self-sufficient, autarkēs, with nothing further required.⁶⁰ The idea of the summum bonum, a highest and unsurpassable good, plays a key role in both Hellenistic and Imperial philosophy, with Epicureans, Stoics, Platonists, and others putting forward their candidates. Seneca argues strongly that happiness, precisely because it is the best good, is possessed completely or not at all. Indeed, Seneca uses this idea against the later Peripatetics and others who defended a view on which virtue was sufficient for the happy life, but external goods such as beauty, health, and wealth could make a life even happier. If a life has the highest good it is supremely happy. Just as the highest good does not admit of an addition (for what is above the highest?) then neither does the blessed life, which cannot exist without the ⁵⁹ Tusc. Disp. V 14.40.

⁶⁰ EN I 7.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

20   highest good. But if you introduce someone who is “more” blessed, then you can also introduce someone who is “much more” blessed. You will generate countless distinctions within the highest good, when on my understanding the highest good is that which has no level above it. If one person is less blessed than another, it follows that he will have a stronger desire for this life of the other person than for his own; but a blessed person prefers nothing to his own life. Either of these two propositions is unbelievable: (a) that there is something left for the blessed person to prefer to be the case than is already the case; or (b) that he does not want what is better than he is. For certainly the more prudent a person is the more he will strive towards what is best and desire to achieve it in any way possible. But how can someone be blessed if he can—indeed, should—desire something even now?⁶¹

This argument explains why the Stoics do not recognize external goods as goods. If they were goods, then they would be included in the summum bonum. The virtuous person would desire more of them and one virtuous person would be more blessed than another by having better health or greater riches. In civ. Dei XIX, Augustine now reverses the argument. The fact that we do appropriately will the goods of mind and body, and social goods, the goods of friendship and peace, but cannot securely hold on to them shows that whatever well-being we have now can be surpassed and hence is not the summum bonum. Augustine agrees with the Stoics that blessedness must be complete and unsurpassable. Contra the Peripatetics, I can only be satisfied if I am lacking in nothing that would make my life better. This, though, means that we must wait in hope for a condition in which we can enjoy all good things to the greatest extent possible, instead of trying to convince ourselves that we have already achieved the highest good.

6. Virtue and Loving Goods Now While Augustine does come to believe that goods such as friendship, bodily health, and mental agility will be present in our summum bonum, ⁶¹ ep. 85.20–1, tr. Brad Inwood with alterations; cf. Cicero, Tusc. Disp. V 8.23–9.25.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

21

is this actually grounds for desiring temporal goods now? The sort of friendships, bodily health, and mental agility that we have in this life seem to be easily lost, even against our will. Perhaps the sorts of external goods we can presently obtain are not worth valuing, even if there will be external goods worth having in the next life. To see why Augustine rejects such reasoning, we need to look at his views on virtue. Although he thinks that true virtue is unattainable in this life, there is continuity between the moral character that we develop now and our final state of virtue or vice. Further, Christians develop virtue through valuing goods as they will be valued in the eschaton, practicing the right ordering of their loves. As we saw, the early Augustine had thought virtue is sufficient for happiness and sometimes spoke as though we ourselves can achieve its firm excellence.⁶² Augustine now thinks the level of virtue we can attain in this life is not stable and enjoyable enough to constitute a blessed life.⁶³ Indeed, virtue itself, which is not among the primary objects of nature (prima naturae) . . . though it holds the highest place among human goods (bonorum humanorum), what is its activity in this world save to wage perpetual war with vices—not those that are outside of us, but within; not those of others, but our own . . . . But what is it we will to do when we seek to attain the supreme good? Surely it can only be that the flesh cease to desire against the spirit, and that there be no vice in us against which the spirit desires? Since, however ardently we will this, we cannot attain to it in this life, let us by God’s help accomplish at

⁶² See fns. 21 and 23–7 for discussion and citations. ⁶³ The mature doctor of grace is convinced that virtue cannot be achieved by ourselves, but depends on God’s gift. Both our acquiring this virtue or love and our continuing in it come by the grace of God (e.g. De spiritu et littera 3.5). Cf. Augustine’s doubts about whether any righteous person, apart from Jesus himself, has lived without sin (De Natura et Gratia 36.42). We are not sufficient on our own to achieve virtue. Augustine is now firmly opposed to the insistence of the philosophers that a person cannot be wise and happy “unless he judges that all goods for him are placed in himself (nisi omnia [bona] sibi in se posita censebit)”, Cicero, Tusc. Disp. V 14.42; cf. Plato, Menexenus, 247e–248. Augustine sharply criticizes the philosophers’ attempts to secure happiness for themselves in trin. XIII 7.10; civ. Dei XIV 14.25, breaking sharply with the Hellenistic philosophers’ emphasis on happiness as being up to us. Further, due to the role externals play in blessedness, virtue is no longer, on its own, sufficient for happiness. Instead, it is sufficient for being “blessed in hope (spe beatus)” (civ. Dei XIX 20; trin. XIII 7.10).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

22   least this, to preserve the soul from succumbing and yielding to the flesh that lusts against it, and to refuse our consent to carrying out sin. Far be it from us, then, to believe that while we are still engaged in this internal war, we have already found the blessedness which we seek to reach by victory. And who is there so wise that he has no conflict at all to maintain against his vices? . . . So long, therefore, as this weakness, this plague, this disease, is within us, how shall we dare to say that we are safe (salvos)? And if not safe (salvos), then how can we dare to claim that we are already blessed with that complete blessedness (finali beatitudine)?⁶⁴

The Christian’s virtue falls short in two ways. First, Augustine emphasizes that the virtues possessed in this life are painful to exercise. They are struggles against sin and disordered desires. Exercising courage and prudence is hard and unpleasant work, it is not the joyful self-satisfied excellent activity of Aristotle’s virtuous gentleman. Even more importantly, however, our selves are still divided. We continue to be liable to the temptation to misorder our loves. As long as that temptation remains, our virtue, which consists in the order of our loves, is not secure. No one is safe yet, as Augustine insists that all Christians need to pray for steadfastness and perseverance in avoiding sin and overcoming temptation.⁶⁵ Like bodily health or temporal friendships, even the best virtues of this life do not meet the Security Condition, so even the wisest and most moderate philosophers should admit that they cannot achieve the summum bonum. Augustine is not just giving an ad hominem critique of pagan failures. For Augustine, all Christians have remnants of sin to overcome even as they progress towards holiness.⁶⁶ This does not, however, mean that we should not value and cultivate virtue. Instead, our goal in life should be to love rightly and overcome our remaining disordered desires as far as we can. If we do this, there will be a continuity between the virtues we develop in this life and their

⁶⁴ civ. Dei XIX 4.3–4, tr. Dods with modifications. ⁶⁵ ep. 130.11.21; cf. Augustine’s De dono perseverantiae (persev.). ⁶⁶ Augustine’s influence led the Council of Carthage to insist in 418 that even the saints and the just in this life continue to pray for their sins to be forgiven; see Canons VI–VIII.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

23

culmination in the next life, as Augustine insists in his letter to Macedonius: By the help of him by whose bounty they were given, [your virtues] will grow and become perfect so that they will without any doubt bring you to the truly blessed life, which is none other than eternal life. In it, prudence will not distinguish evil, which will not exist, from what is good, nor will courage endure troubles (adversa), because we will find there only what we love, not what we endure, nor will temperance bridle desire (libido) where we will not feel its enticements. Nor will justice aid the needy with help where we will have no one poor and needy. In that life there will be only one virtue, and it will be both virtue and the reward of virtue . . . . There this will be complete and everlasting wisdom, and this same wisdom will also be the truly blessed life (veraciter vita beata). It is, of course, the attainment of the eternal and highest good, and to cling to it for eternity is the goal that holds all our good. This might be called prudence, because it will with perfect foresight cling to the good that will not be lost. It might be called courage because it will most firmly cling to the good that will not be torn away. It might be called temperance because it will most chastely cling to the good by which it will not be corrupted. And it might be called justice because it will with full righteousness cling to the good to which it is rightly subject.⁶⁷

The virtues are developed now in the struggle against disordered desires but will be exercised in a new way after the resurrection. Their purgative and conflictual aspect will be taken away, but the ordering of loves in which they consist will be made stable, so as to secure the blessedness of the righteous.⁶⁸ To get to this stable condition, however, we must train ourselves by rightly loving things now. We should love God above all, but we also need to love bodily integrity and friendship more than merely instrumental goods such as riches and reputation. For Augustine, we use our knowledge of what is included in the summum bonum to direct our

⁶⁷ ep. 155 to Macedonius, 3.12, tr. Teske with modifications; cf. trin. XIV 9.12. ⁶⁸ Cf. civ. Dei XIX 10. For more discussion of the continuity of ante and post-mortem virtue see Tornau, “Happiness in This Life?”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

24   loves now, appreciating what is good and chosen for its own sake while avoiding attachment to what is merely useful.⁶⁹

7. The Emotions and Caring for Others’ Happiness in the Good Life Contra Nussbaum, Augustine’s postponement of happiness also allows for greater emotional engagement and expands our circle of care. By acknowledging that we cannot achieve happiness in this life, Augustine finds latitude to both acknowledge brokenness and faults in himself and others and long for transformation for us all. While Augustine’s criticisms of the Stoics on the emotions are well known, examining his mature views in light of his development on the question of happiness clarifies several things. To begin with, as we have seen in section 3, Augustine did not always reject the Stoic view on emotions. He initially endorses the Stoic claim that we should not feel for others, even when helping them, and he did so precisely because this is the only way one could conceivably hold on to tranquility and remain undisturbed in the face of suffering. Augustine never rejects the claim that blessedness requires tranquility. In this respect, he is clearly within the main Greco-Roman philosophical tradition, which insisted that only someone undisturbed and at peace can be living the happy life. Augustine only affirms the value of negative emotions in this life once he comes to believe that we cannot yet achieve blessedness and so should not yet aim at tranquility. For Augustine, virtuous people in this life feel sorrow and anger as well as joy and act based on these emotions. Their feelings are really affected by what happens to those around them. And not only on their own account do [those who love God] experience these emotions, but also on account of those whose liberation they desire and whose loss they fear, and they sorrow if those perish and rejoice if they are liberated (et dolent si pereunt et gaudent si liberantur)

⁶⁹ Cf. civ. Dei XIX 10; ep. 130.6.13.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

25

. . . . Very joyfully do we with the eyes of faith behold [Paul] rejoicing with those that rejoice, and weeping with those that weep [Romans 12.15] . . . because these emotions, when they are exercised appropriately, follow the guidance of right reason, who will dare to say that they are diseases or vicious passions? . . . For as there was in [the Lord Himself] a true human body and a true human soul, so was there also a true human emotion (humanus affectus).⁷⁰

Throughout his later works, Augustine appeals to the apostle Paul’s advice to weep with those who weep and rejoice with those who rejoice. For Augustine, the examples of the apostle and of the suffering Christ illustrate how to live virtuously in this life. Living well as a wayfarer does not consist in tranquility and aloofness, but rather in struggle, pain, and sorrow. Augustine acknowledges that the negative emotions we feel for ourselves and others prevent us from being at peace. But this is good given our role in the world. Avoiding appropriate feelings makes us inhuman, not happy: Since we must live rightly in order to attain to a blessed life (recta vita ducenda est, qua perveniendum sit ad beatam), a right life has all these emotions right, a wrong life has them wrong. However, the blessed life eternal will have love and joy, not only right, but true and assured (verum etiam certum); while it will have no fear and grief . . . those moved or bent by no emotion do not achieve true tranquility (veram tranquillitatem) but instead lose all humanity (humanitatem totam). For something is not right, because it is harsh, nor is something healthy (sanum) because it lacks feeling (stupidum).⁷¹

Augustine has turned decisively against the Stoic strategy of avoiding negative emotions. He now sees this detached approach, which he had earlier endorsed, as a misguided attempt to manufacture a counterfeit happiness through suppressing the feelings we should feel in response to the evils of this life.⁷²

⁷⁰ civ. Dei XIV 9, tr. Dods with modifications. ⁷¹ civ. Dei XIV 9. ⁷² civ. Dei XIV 25; trin. XIII 7.10; see vera rel. 47.91, quoted in section 3, for his earlier view.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

26   Augustine is not, however, rejecting what Nicholas Wolterstorff calls the Stoic principle of emotional detachment: In the truly estimable life there would be no negative emotions—no fear, no grief, no regret, no remorse, nothing of this sort.⁷³

Instead, Augustine is separating the virtuous or estimable way of living in this life, which Jesus Christ and Paul embodied in their earthly journeys, from the ultimate blessed life, which does still satisfy this principle. The lives of Jesus and Paul are exemplary, but the suffering and death of Christ and the many trials and torments of Paul gain their value from the way in which they contribute to ultimate happiness, both for themselves and others. Augustine still holds that the truly blessed life is incompatible with negative emotions. Christians can sorrow and feel compassion now, because they are looking forward to a future life in which only positive emotions will be present. These themes are developed in his letters, where Augustine urges his correspondents to extend their circle of care outwards, beyond close friends and family to the whole body of Christ. That is what living well and choosing the good looks like for human life under present sinful and broken conditions. Augustine follows the Stoics in calling for us to show concern for all of humanity. Augustine emphasizes that the commandment to love our neighbor means that we need to love all other humans. We should, of course, judge who our neighbor is in this passage not on the basis of blood relationship but on the basis of our sharing in the community of reason (rationis societate), a community to which all humans belong.⁷⁴

This line of reasoning, of course, goes directly back to the Stoics. Augustine takes from them the emphasis on valuing every human and on promoting the welfare of all. Both Augustine and the Stoics make more universal claims than traditional Roman values, which restricted one’s concern to either the citizens of your particular city or to Romans ⁷³ Nicholas Wolterstorff, “Augustine’s Rejection of Eudaimonism,” in J. Wetzel (ed.), Augustine’s City of God: A Critical Guide (Cambridge: Cambridge University Press, 2012), 156. ⁷⁴ ep. 155.4.14; cf. ep. 130.6.13.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

27

more generally. As Peter Brown notes of Roman giving, “the benefactors of cities gave to their ‘fellow citizens’ and never to the poor. Some of these citizens might well be poor, but their poverty in itself entitled them to nothing . . . to love the city was also to love its citizens with ‘unique affection’—and to love no one else.”⁷⁵ Augustine insists that the city of God is the universal city, to which all owe their first allegiance. We should hope for all our fellow humans to become citizens of this city through loving God. This insistence on the universality of Christianity and the hope of including all the world within it is something new. Even a generation earlier, Latin Christians of the early fourth century could, Claire Sotinel says, imagine “Christianity as present in all parts of a [social] universe, but not a social universe that was entirely Christian.”⁷⁶ Augustine shares the Stoics’ universal concern for rational creatures, but he goes beyond the Stoics in not just acting for the benefit of others but in sharing their feelings, both positive and negative. For the Stoics, we should benefit others, but without sharing in their emotions and without regarding their conditions as intrinsically good or bad. My global concern only affects which actions to pursue and how to promote things that are in themselves indifferent. While Augustine draws on the Stoics’ cosmopolitan ideals and their expanding circles of concern, his views on the future life allow him to care about and be emotionally affected by what happens to others now.⁷⁷ For the Stoics, my blessedness is unaffected by what happens to my friends, family, or the rest of the cosmos. The excellence of the ruling part of my mind is the one and only thing that matters for my happiness. Thinking otherwise ruins our life. As Epictetus puts it, Although we have it in our power to apply ourselves to one thing alone, and devote ourselves to that, we choose instead to apply ourselves to ⁷⁵ Peter Brown, Through the Eye of a Needle: Wealth, the Fall of Rome, and the Making of Christianity in the West, 350–550  (Princeton, NJ: Princeton University Press, 2012), 68. ⁷⁶ Claire Sotinel, “La sphère profane dans l’espace urbain,” in É. Rebillard and C. Sotinel (eds.), Les frontiers du profane dans l’antiquité tardive (Rome: École française de Rome, 2010), 344 (tr. Brown). Cf. Brown, Through the Eye of a Needle, 34: “[Christians in the earlier age of Constantine] were proud of the fact that Christians could be found everywhere. But this did not mean that they expected that everyone everywhere would become a Christian.” ⁷⁷ For discussion of the influence of the Stoic notion of expanding circles of concern on Augustine’s thought and his practical advice, see Clair, Discerning the Good, ch. 1 and interlude.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

28   many things, and attach ourselves to many, to our body, and our possessions, and our brother, and friend, and child, and slave. And so, being attached in this way to any number of things, we’re weighed down by them and dragged down.⁷⁸

If I think my happiness depends on anything outside my own power of choice, I will never be happy. This means that I cannot place value on any external occurrences and I cannot have any negative emotions such as fear or anger or compassion, either for what happens to me or to my family or city.⁷⁹ Such feelings would show that I have not achieved wisdom and tranquility. Seneca, for example, insists that we need to act to benefit others without feeling their pain. All good men (boni viri) will show mercy and mildness (clementiam mansuetudinemque), but will avoid compassion [misericordiam], which is a vice of weak spirits (pusilli animi) which cannot endure the sight of another’s sufferings . . . .Compassion is a disorder of the soul caused by the sight of others’ miseries (aegritudo animi ob alienarum miseriarum speciem) or it is a sadness caused by the evils with which it believes others to be undeservedly afflicted (tristitia ex alienis malis contracta, quae accidere immerentibus credit): but the wise man cannot be affected by any disorder: his mind is calm (serena eius mens est), and nothing can possibly happen to cloud it. Moreover, nothing becomes a human more than greatness of soul (magnus animus): but a human cannot be great and sad . . . the wise person will not feel compassion, but will help others and be of service, since he is born for the common aid and the public good (in commune auxilium natus ac bonum publicum), and from them he will give to each a share.⁸⁰

The Stoic sage helps others, but without being affected by their pain and wretchedness. He remains at peace, undisturbed in spirit. By contrast, Augustine insists that actually feeling compassion is appropriate and will help us to better love others:

⁷⁸ Discourses I 1.14–15, tr. Hard. ⁷⁹ Cicero, Tus. Disp. V 6.15–17; 10. 28–30; 13.40–14.42.

⁸⁰ De Clementia II 5–6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

29

In our teaching, we do not so much inquire whether a pious soul is angry, as why he is angry; not whether he is sad, but what is the cause of his sadness; not whether he fears, but what he fears. For I do not know anyone of sound mind (sana consideratione) who would blame someone for being angry at wrongdoers in order that they be set right or someone who sorrows for the suffering in order that they be liberated or someone who fears for those in danger in order that they do not perish. The Stoics, indeed, are accustomed to condemn compassion . . . but what is compassion except a fellow-feeling in our heart for another’s wretchedness, by which we are compelled to help if we are at all able (Quid est autem misericordia nisi alienae miseriae quaedam in nostro corde compassio, qua utique si possumus subvenire compellimur)? And this passion is obedient to reason, when compassion is shown while preserving justice, as when the poor are provided for or the penitent forgiven.⁸¹

Augustine thinks that our feelings have a role to play in guiding us to help others, both spiritually and materially. Just as sorrow and fear help motivate us to flee evils and embrace the good when we feel them about ourselves, so these feelings motivate us to aid others.⁸² Another key difference between Augustine and previous GrecoRoman thinkers is that he seeks happiness not only for himself, but for the whole city of God. For the Stoic, I only control myself, so I can only seek my own happiness. I would value appreciating the cosmos together with my fellow citizens, but I cannot place my hope in their joining me. For the Aristotelian, I seek happiness for myself and for those close to me, my family and fellow citizens. To achieve happiness, my love needs

⁸¹ civ. Dei IX 5, tr. Dods with modifications. ⁸² While still insisting on our obligation to love all and help all, the later Augustine is also more flexible on the role of existing relationships in how we love others than his earlier self: We ought to pray for these things for ourselves and for ours, for strangers and even for enemies, although for different persons a different affection arises in and lifts up the heart of the person in prayer in accord with the closeness and the distance of their relationships (propinquitates vel longinquitates necessitudinum). (ep. 130.12.23, tr. Teske) Augustine now recognizes that circumstances and earthly ties can appropriately affect our loves. Christians need to have affection for all, but their prayers and actions can be informed by the relationships and responsibilities they have based on their earthly situations.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

30   to be restricted to a tight circle: those whose well-being I can effectively influence. That a man’s happiness should lie with himself seems to have been a prominent idea in classical Greek thought, one we find expressed in the funeral oration of Plato’s Menexenus: For that man’s life is best arranged for whom all, or nearly all, the things that promote happiness depend on himself. Such a man does not hang from other men and necessarily rise or fall in fortune as they fare well or badly . . . he above all, when wealth and children come and when they go, will pay heed to the adage: because he relies on himself, he will be seen neither to rejoice nor to grieve too much.⁸³

The virtuous person of this oration is not completely unaffected by what happens to his family or city, but he does place his happiness primarily in himself, in what he does and how he reacts to the chances of fortune. Indeed, Cicero cites this speech as a precursor of the Stoic view.⁸⁴ Requiring others’ happiness in order for you to be happy is dangerous. It seems in tension with both the Security Condition, because you cannot guarantee or preserve the happiness of others, and the Complete Satisfaction Condition, because you will not be happy if anyone whose happiness you desire fails to obtain it. For Epictetus, you should always play your role in a relationship well and do your part. If, however, the relationship suffers due to the other’s behavior, it will not affect your happiness or well-being.⁸⁵ Augustine, in contrast to these earlier thinkers, emphatically wills happiness not just for himself but for the whole city of God. We see this in the advice he gives about what Macedonius, as a Christian and a public official, should be seeking: Let us ask the Lord our God, who made us, for the virtue to conquer the evils of this life and for the blessed life that we may enjoy after this life ⁸³ 247e–248a, tr. Paul Ryan. In this dialogue, Socrates recites a funeral oration for the fallen Athenians to Menexenus, a speech he claims is actually composed by Aspasia, his rhetoric teacher (and the lover and partner of Pericles) (236a–c). The speech’s relationship to the views of Aspasia, Pericles, Socrates, and Plato is contested and complex. For discussion, see J. M. Cooper, in Plato, Complete Works (Indianapolis: Hackett, 1997), 950–1. On any reasonable construal, however, it can fairly be taken as a reputable ancient Athenian view on how to achieve happiness. ⁸⁴ Tusc. Disp. V 12.36. ⁸⁵ Disc. I.15, Ench. 30.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

31

in his eternity . . . let us desire this for ourselves; let us desire this for the city of which we are citizens. For a city’s blessedness comes from the same source as a human’s, since a city is nothing but a multitude of humans with a common goal.⁸⁶

Augustine and Macedonius are to desire virtue and the blessed life, not just for themselves, but for all those they love and for the whole city of God. Augustine does not just outwardly help others, he expands his love towards all his brothers and sisters in Christ. He can weep when they weep, because he looks forward to the “perfectly ordered and perfectly harmonious community of enjoying God and one another in God (ordinatissima et concordissima societas fruendi Deo et invicem in Deo).”⁸⁷ In the end, all members of the city of God find their happiness in a shared vision of God and of one another in God. Augustine insists that only this order can be called the “peace of a rational creature,” because it is the only stable condition in which all the included rational creatures are successfully ordered to both their own happiness and the happiness of everyone else in the group.⁸⁸ Each rational member of the city of God, each human or angel, will be sharing in blessedness with and loving every other member. In this perfect heavenly society, there is no danger that anyone the righteous love will fail to be happy.⁸⁹ Since no citizen of the divine city will ultimately be sad or afraid, Augustine can share in their negative feelings now. Christian friendships in this life both prepare for and anticipate the union of loves and wills that Augustine thinks members of the city of God will ultimately achieve. His monastic way of life and the rule that Augustine himself develops are meant to serve as the first manifestations of that ultimate peaceful society. As Peter Brown puts it, “Augustine was confident that the love generated in a monastery pointed the way to an unimaginable future at the end of time. Then the firestorm of love would engulf the entire company of the blessed: ‘Out of many souls ⁸⁶ ep. 155.3.9, tr. Teske. ⁸⁷ civ. Dei XIX 13.1; cf. XIX 18. ⁸⁸ civ. Dei XIX 18. ⁸⁹ Keeping in mind, however, that Augustine’s prayers and his desire for happiness for others do not extend to the fallen angels or to those who have irrevocably excluded themselves from the city of God (civ. Dei XXI 24; cf. XX 22): his common society extends only to those who are ultimately saved and made good. Contra Origen, he insists that not all will ultimately be redeemed (civ. Dei XXI 17).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

32   there will arise a City and a People with a single soul and heart, turned towards God.’ ”⁹⁰

8. Augustine on Happiness and Security If we can only be happy in the triumphant city of God, does Augustine consign us to misery in this life? The mature Augustine clearly thinks that no one obtains the summum bonum in this life. His reasons for holding this, however, are compatible with experiencing enjoyment and delight in this life. As I noted in section 2, Augustine consistently holds on to the Security Condition for happiness: happiness must be stably possessed and, having attained it, you cannot lose it. It is this condition that rules out true happiness in this life. The Security Condition is central to Augustine throughout the civ. Dei. His criticism of Platonist views on reincarnation is based around it, as reincarnation would make the supposed joys of the separated soul false and impermanent, given its eventual return to the body.⁹¹ We see the need for security most clearly in his insistence that even the righteous Christian has not yet achieved blessedness: However, if we look at this a little more closely, we see that no one lives as he wills but the blessed, and that no one is blessed but the righteous. But even the righteous person does not live as he wills, until he has arrived where he cannot die, be deceived, or injured, and until he is assured that this shall be his eternal condition. For this nature demands; and nature is not fully and perfectly blessed till it attains what it seeks. But what person is at present able to live as he wills, when it is not in his power so much as to live?⁹²

⁹⁰ de bono coniugali 18.21, tr. Brown, Through the Eye of a Needle, 177. For discussion of the monastic life as a “kind of divine and heavenly republic” see Through the Eye of a Needle, chapter 12 and Lancel, Saint Augustine, chs. XXI–XXIII. ⁹¹ civ. Dei XII 20. It is also striking that in his discussion of Adam and Eve before the fall he repeatedly describes the first humans as felices, but not as beati, the word he generally reserves for the summum bonum and takes to involve secure possession of the best good as well as enjoyment (civ. Dei XIV 10). ⁹² civ. Dei XIV xiv.25, tr. Dods with modifications.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

33

Augustine is not saying that the righteous are entirely wretched in the present life, just that they are not yet secure, both because they can lose some of the goods which they will (violating the Complete Satisfaction Condition) and because they could still fall from grace. The failure to completely fulfill these conditions means that even the truly righteous cannot count as fully happy. He clarifies this in book XIX: Here, indeed, we are called blessed when we have peace, to whatever minute extent this can be had in a good life (quantulacumque hic haberi potest in vita bona); but this blessedness is found to be absolute wretchedness (prorsus miseria) compared to that blessedness, which we call ultimate.⁹³

For Augustine, we can live a good life now, but the peace and enjoyment possible are so limited in comparison to the ultimate goal—a peace “than which a better and greater cannot be (qua melior et maior esse non possit)”—that this lack of security and completion make blessedness in this life unworthy of the name.⁹⁴ Augustine allows for some enjoyment and satisfaction in the goods available to us now, including the knowledge and love of God available to Christians. Yet the possibility of suffering and evil and the uncertainties of our lives mean that we cannot yet live the best possible life in perfect harmony. Augustine’s discussion of the angels before the fall shows again how highly he values security as a condition of the highest good. The angels all knew God, but were not truly blessed until they were secure in their positions. This leads him to contrast the condition of Adam with that of the saints striving in the church militant: The first man was more blessed in paradise than any righteous person in this state of mortal frailty, with respect to enjoyment of the present good. But as for the hope of the future, any just person in the extreme of bodily suffering is happier than the first-created. For to the just person it has been shown with the certainty of truth—not mere opinion—that, free from all distresses, he will share with the angels that endless enjoyment of God Most High, whereas that first man, even ⁹³ civ. Dei XIX 10.

⁹⁴ civ. Dei XIX 10.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

34   in the great happiness of paradise (in magna illa felicitate paradisi), was uncertain about his fate. Anyone can now easily gather that the blessedness which the understanding being desires with unswerving resolution is the product of two causes working in conjunction, the untroubled enjoyment (sine ulla molestia perfruatur) of the changeless Good, which is God, together with the certainty of remaining in him for eternity, a certainty that admits of no doubt or hesitation, no mistake or disappointment (nec ulla dubitatione cunctetur nec ullo errore fallatur).⁹⁵

Augustine is glad to say that in an important way the martyr being tortured to death is more blessed than Adam, insofar as the martyr has a sure and certain hope of attaining the final good and enjoying it forever. Adam enjoyed the full range of goods and Augustine describes him as having “great happiness in paradise.”⁹⁶ However, Adam fails to meet the Security Condition, because he is unsure about what will happen to him and this means that he was not truly blessed. The just person’s certainty about his future condition allows for a kind of happiness now, as the just can look forward to certain enjoyment.⁹⁷ More generally, Augustine allows for enjoying goods in this life in ways that might qualify as happiness in the weaker sense often employed in colloquial English. Augustine denies that we can presently achieve beatitudo, not because he thinks our lives must be invariably sorrowful and miserable but because even the best human life under current circumstances will fail to meet the Security Condition and the Complete Satisfaction Condition.

9. The Future Life and Augustine’s Eudaimonism It is the hoped-for future triumph of the city of God that allows Christians to count as pursuing blessedness in their current actions, ⁹⁵ civ. Dei XI 13, tr. Bettenson with modifications. ⁹⁶ Felicitas, not beatitudo. As I mentioned in fn. 6, Augustine uses felicitas more freely and without the same Security Condition that he applies to beatitudo. Cf. civ. Dei XIV 10 on Adam and Eve in paradise, which also describes their condition in terms of felicitas, but not beatitudo. ⁹⁷ At least for those who know their salvific status; cf. the complications of persev.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

35

even if they cannot yet achieve it. Augustine’s ethics continues to be eudaimonist, contra Nicholas Wolterstorff, who insists that the mature Augustine breaks with eudaimonism: I see nothing in Augustine’s thought that would explain his affirmation of compassion other than that he saw this as an implication of obedience to Christ’s command to love one’s neighbor as one loves oneself . . . [Augustine] has to be interpreted as parting ways with eudaimonism and entering territory untracked by the ancient philosophers.⁹⁸

On this interpretation, Augustine’s love and care for his neighbors breaks with what Wolterstorff calls the agent-wellbeing proviso: “one may seek to promote the wellbeing of anyone whatsoever as an end in itself— provided that doing so promises to enhance one’s own overall wellbeing.”⁹⁹ Ancient eudaimonists do endorse some version of this proviso, but Augustine does as well. Indeed, the shared blessedness of the triumphant city of God is precisely what allows him to hold on to it. As we saw, for Augustine, we will find happiness in enjoying God and others in God. In this way, each Christian’s happy life involves loving every other Christian and loving the love every other Christian has for God. This mutual love of God and our appreciation for God is part of the well-being of the blessed. Their well-being is enhanced by the other saints and so their love of each other still fits with the agent-wellbeing proviso.

⁹⁸ Nicholas Wolterstorff, “Augustine’s Break with Eudaimonism,” in Nicholas Wolterstorff, Justice: Rights and Wrongs (Princeton, NJ: Princeton University Press, 2010), 203–4. ⁹⁹ Wolterstorff, “Augustine’s Rejection of Eudaimonism,” 155. Wolterstorff also holds that ancient eudaimonists all shared “the activity thesis concerning the estimable life. They all held that the estimability of a person’s life is determined solely by the actions and activities to be found in that person’s life; the state and events to be found in his or her life play no role in determining its estimability” (ibid., 152). He then uses this to distinguish Augustine from his ancient predecessors. This, in fact, is a distinctively Aristotelian thesis, not one shared by all eudaimonists. Aristotle has to extensively defend it, precisely because it is not a commonly acknowledged truth. Indeed, traditional Greek thought sees blessedness as largely found in the states and events in that person’s life, as, for example, we see in Solon’s conversation with Croesus (Herodotus, Historia, I 29–33 and 86–90). Indeed, it is not clear that Aristotle himself would want to defend the stark formulation Wolterstorff offers (“no role” vs. EN I 9–10). The Epicureans, Cyrenaics, and Pyrrhonian Skeptics, among others, would reject this thesis. Augustine himself is quite aware that the role of states and events play in a person’s happiness is disputed. In civ. Dei, when he goes through views on the summum bonum, many of the philosophical options Augustine and Varro present involve intrinsically valuing things that are not themselves activities, either the primary blessings of nature, pleasure, or the absence of pain.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

36   This proviso is vital for Augustine’s psychology. Augustine never gives up on the claim that my happiness is my summum bonum. The will for blessedness is, he says, the only voluntas common to all human beings, something that even the Academic skeptics “did not doubt (although Academics doubt everything).”¹⁰⁰ For Augustine, happiness turns out to consist in enjoying God, but he never countenances the possibility that my happiness and my love of God could come apart. Throughout all his writings, he remains committed to psychological eudaimonism, to the idea that we love everything and pursue it so as to further our happiness. He never gives up on happiness. Augustine does not emphasize compassion and values externals more because he becomes a disinterested advocate of duty. Instead, once he sees how, in the end, his good involves the mutual love and excellent conditions of all members of the city of God, he sees how he can contribute to happiness by loving them now and feeling sorrow and anger alongside them. In the Latin West, we will have to wait for Anselm or John Duns Scotus to see a true rejection of eudaimonism.¹⁰¹

10. Conclusion Augustine is the first in the Greco-Roman philosophical tradition to explicitly put forward a view of my happiness that goes beyond myself and my close connections to encompass the whole order of the universe. Without the hope of ultimate peace to which he looks forward, there are strong reasons to restrict the scope of who can affect your happiness. This is why Aristotle limits the bounds of care to one’s family and polis. While contemporary secular eudaimonistic theories can expand the

¹⁰⁰ trin. XIII 3.6; trin. XIII 4.7. ¹⁰¹ I am agreeing here with Jeff Brower, “Anselm on Ethics,” in B. Davies and B. Leftow (eds.), The Cambridge Companion to Anselm (Cambridge: Cambridge University Press, 2004), 251 note 5, that Augustine is not the source or beginning of medieval movements away from eudaimonism, contra John Hare, “Scotus on Morality and Nature,” Medieval Philosophy and Theology 9 (2000), 15–17. For the turn away from eudaimonism in Anselm and Duns Scotus see Brower, “Anselm on Ethics,” 223–4; 240–9; Thomas Williams, “How Scotus Separates Morality from Happiness,” American Catholic Philosophical Quarterly 69 (1995), 425–45, and Hare, “Scotus on Morality and Nature.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

37

circle of care, the more such theories do so, the more impossible it becomes for me to aim at or achieve happiness. If I really care equally about the welfare of all sentient beings, then I cannot be happy until all are happy and there is no practical way for me to achieve this. Without any justified expectation for a perfect culmination or any reasonable hope of attaining such a state, we are tempted either to despair or to restrict our care to those close to us. As long as Augustine thinks happiness is attainable in this life, he insists that we should aim to avoid any negative emotions. Once he postpones complete happiness to the next life, Augustine can continue to agree with the Stoics and other ancient philosophers that the blessed life must be tranquil, while taking a very different position on what a good human life looks like now, in our condition as wayfarers. Those still running the race should experience sorrow and fear, for themselves and others, whenever such emotions help them to better direct their steps and those of others. Augustine’s vision of blessedness also becomes more encompassing. The young Augustine denies that the state of our bodies or the well-being of our friends and loved ones makes a difference to our happiness. The good we are aiming at must be entirely spiritual. While he always thinks that blessedness consists in enjoying God, once Augustine holds that happiness is only to be found in the eschaton he is able to include the well-being of our minds and bodies and our friendships with others within the summum bonum. Augustine looks forward in hope to the point at which the wise and blessed resurrected Christians will be at peace, achieving what the philosophers were seeking. After the last judgment, every member of the city of God will rejoice together, with weeping, anger, and fear no longer having any place, since all their causes have been defeated. Because of the divine assurance that things will go well—for the whole community of God, not just for the individual—the Christian can, unlike the Stoic, desire and act for the happiness of every member of the kingdom of God.¹⁰² Augustine’s belief in a future life motivates action in this life by ¹⁰² Cf. Immanuel Kant’s claim, in his Critique of Practical Reason II.5, pars. 125–31, that we must hope for a future life so that happiness and good will might be harmonized into the best possible good.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

38   allowing for greater emotional range in the virtuous. It grounds his appreciation of the value of mind and body and it provides hope that present evils will be defeated.¹⁰³ Metropolitan State University of Denver

Bibliography Adams, Marilyn McCord. Horrendous Evils and the Goodness of God (Ithaca, NY: Cornell University Press, 1999). Annas, Julia. The Morality of Happiness (Oxford: Oxford University Press, 1993). Augustine. The City of God, tr. M. Dods (Buffalo, NY: Christian Literature Publishing Co., 1887). Augustine. The Retractations, tr. M. I. Bogan (Washington, DC: Catholic University of America Press, 1968). Augustine. The City of God, tr. H. Bettenson (New York: Penguin, 1995). Augustine. Letters: Volume 2, ed. and tr. R. J. Teske (Hyde Park, NY: New City Press, 2002). Brower, Jeffrey. “Anselm on Ethics,” in B. Davies and B. Leftow (eds.), The Cambridge Companion to Anselm (Cambridge: Cambridge University Pres, 2004), 222–56. Brown, Peter. Through the Eye of a Needle: Wealth, the Fall of Rome, and the Making of Christianity in the West, 350–550  (Princeton, NJ: Princeton University Press, 2012). Bussanich, John. “Happiness, Eudaimonism,” in A. D. Fitzgerald (ed.), Augustine Through the Ages: An Encyclopedia (Grand Rapids, MI: Eerdmans, 1999), 413–14. Byers, Sarah. Perception, Sensibility, and Moral Motivation in Augustine: A Stoic-Platonic Synthesis (Cambridge: Cambridge University Press, 2012).

¹⁰³ Cf. Marilyn McCord Adams, Horrendous Evils and the Goodness of God (Ithaca, NY: Cornell University Press, 1999). I would like to thank the participants in the 2019 Cornell Summer Colloquium in Medieval Philosophy for their insightful and helpful questions on an earlier version of this paper. I am also grateful to Joseph Clair for advice on Augustine, Matt Morginsky for inspiration and insight in thinking about the future life, and an anonymous referee for pushing me to clarify several aspects of my argument.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

      

39

Cicero. On Moral Ends [De Finibus], tr. R. Woolf (Cambridge: Cambridge University Press, 2001). Cicero. Ortensio, testo critico, ed. A. Grilli (Bologna: Pàtron Editore, 2010). Clair, Joseph. Discerning the Good in the Letters and Sermons of Augustine (Oxford: Oxford University Press, 2016). Council of Carthage. “The Canons of the Council of Carthage (417 or 418) on Sin and Grace,” Early Church Texts . Retrieved June 29, 2018. Epictetus. Discourses, Fragments, Handbook, tr. R. Hard (New York: Oxford University Press, 2014). Frede, Michael. A Free Will: Origins of the Notion in Ancient Thought (Berkeley: University of California Press, 2011). Gilson, Étienne. The Christian Philosophy of Saint Augustine (New York: Random House, 1960). Hare, John. “Scotus on Morality and Nature,” Medieval Philosophy and Theology 9 (2000), 15–38. Holte, Knut Ragnar. Béatitude et sagesse: Saint Augustin et le problème de la fin de l’homme dans la philosophie ancienne (Paris: Études Augustiniennnes, 1962). Kant, Immanuel. Critique of Practical Reason, tr. M. Gregor, rev. ed. (Cambridge: Cambridge University Press, 2015). Lancel, Serge. Saint Augustine (London: SCM Press, 2002). Long, A. A. and D. N. Sedley (eds.). The Hellenistic Philosophers, 2 vols. (Cambridge: Cambridge University Press, 1987). Nietzsche, Friedrich. Thus Spoke Zarathustra, tr. W. Kaufmann (New York: Viking, 1954). Nussbaum, Martha C. Upheavals of Thought: The Intelligence of Emotions (Cambridge: Cambridge University Press, 2001). O’Connell, Robert J. Augustine’s Early Theory of Man,  386–391 (Cambridge, MA: Harvard University Press, 1968). Plato. Complete Works, ed. J. M. Cooper (Indianapolis: Hackett, 1997). Seneca. Selected Philosophical Letters, tr. B. Inwood (Oxford: Oxford University Press, 2007).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

40   Sotinel, Claire. “La sphère profane dans l’espace urbain,” in É. Rebillard and C. Sotinel (eds.), Les frontiers du profane dans l’antiquité tardive (Rome: École française de Rome, 2010), 319–49. Tornau, Christian. “Happiness in this Life? Augustine on the Principle that Virtue Is Self-Sufficient for Happiness,” in Ø. Rabbås, E. K. Emilsson, H. Fossheim, and M. Tuominen (eds.), The Quest for the Good Life: Ancient Philosophers on Happiness (Oxford: Oxford University Press, 2015), 265–80. Williams, Thomas. “How Scotus Separates Morality from Happiness,” American Catholic Philosophical Quarterly 69 (1995), 425–45. Wolterstorff, Nicholas. “Augustine’s Break with Eudaimonism,” in Nicholas Wolterstorff, Justice: Rights and Wrongs (Princeton, NJ: Princeton University Press, 2010), 180–207. Wolterstorff, Nicholas. “Augustine’s Rejection of Eudaimonism,” in J. Wetzel (ed.), Augustine’s City of God: A Critical Guide (Cambridge: Cambridge University Press, 2012), 149–66.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

Perception in Augustine’s De Trinitate 11 A Non-Trinitarian Analysis Susan Brower-Toland

1. Introduction The latter half of Augustine’s De Trinitate is as much a treatise in human cognitive psychology as it is a work in trinitarian theology. Augustine holds that our best (and perhaps only) prospect for understanding something of God’s trinitarian nature is to study its reflection in the human mind.¹ For, on his view, the mind at its deepest (or highest) level possesses a trinitarian structure—one that is likewise manifested in its various episodic cognitive acts. And, while the most adequate representation of the divine trinity has its locus in the higher reaches of human rational activity, traces or quasi-images of the trinity can be found even in the soul’s non-rational, sense-based activities. In this paper, I explore Augustine’s account of sense cognition in book 11 of De Trinitate. His discussion in this context focuses on two basic types of sensory state—what he himself refers to as “outer vision” and “inner vision,” respectively. His analysis of both types of state is designed to show that our lowest, outermost cognitive acts—i.e., those involving external and internal sense faculties—are susceptible of a

¹ In what follows, translations of De Trinitate are adapted (with occasional, slight modification) from the version of Edmund Hill. The editions and translations on which I rely for Augustine’s other works are listed in the bibliography. Susan Brower-Toland, Perception in Augustine’s De Trinitate 11: A Non-Trinitarian Analysis In: Oxford Studies in Medieval Philosophy Volume 8. Edited by: Robert Pasnau, Oxford University Press (2020). © Susan Brower-Toland. DOI: 10.1093/oso/9780198865728.003.0002

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

42  - trinitarian analysis. In this respect, the theological significance of his discussion is perfectly straightforward. As a project in philosophical psychology, however, its significance is rather less obvious. Indeed, it is far from clear what sense-cognitive phenomena correspond to Augustine’s notion of outer and inner vision, respectively. One very natural (and fairly common) way to read De Trin. 11, however, is to interpret Augustine’s analysis of “outer” vision as his account of sense perception (i.e., of perceptual acts of seeing, hearing, touching, etc.) and the analysis of “inner” vision as an account of occurrent acts of sensory memory and imagination.² On this reading, the division between outer and inner vision yields a mutually exclusive and exhaustive division of the main types of sensory state that occur in human beings such that all sense-based states turn out to be trinitarian in structure. Although this reading has both an appealing tidiness as well as some prima facie textual support, it cannot, I claim, be correct. In what follows, therefore, I argue for an alternative reading of book 11. In particular, I advance two theses: one negative, one positive. The negative thesis is that what Augustine calls “outer vision” is not, in fact, sense perception— at least not in the ordinary or paradigmatic sense. In making this claim, I take for granted that, paradigmatically, sense perception is a kind of conscious perceptual awareness. To be sure, there are various forms of non-conscious sensory processing (blindsight; or other kinds of nonconscious, or sub-personal sensory events), but these are not paradigmatic cases of sense perception. Hence, insofar as we take perception to

² Although I know of no paper which advances such an interpretation as part of a focused, exegetical study of perception in De Trin. 11 (partly because there are very few such studies), nevertheless, there are treatments of book 11 that clearly do presuppose some such interpretation. Recent examples include: Scott MacDonald, “Augustine’s Cognitive Voluntarism in De Trinitate 11,” in E. Bermon and G. O’Daly (eds.), Le De Trinitate de saint Augustin: exégèse, logique et noétique (Paris: Vrin, 2012), 322–39 and Mark Kalderon, “Trinitarian Perception,” Proceedings of the Aristotelian Society, supp. vol. 91 (2017), 21–41. This reading also seems to be taken for granted by Peter King, “Augustine on Knowledge,” in D. Meconi and E. Stump (eds.), The Cambridge Companion to Augustine, 2nd ed. (Cambridge: Cambridge University Press, 2014), 142–65, Johannes Brachtendorf, The Structure of the Human Mind according to Augustine: Self-Reflection and Knowledge of God in De Trinitate, tr. A. Looney (Hamburg: Felix Meiner Verlag, 2000), 193–8, and Luigi Gioia, The Theological Epistemology of Augustine’s De Trinitate (Oxford: Oxford University Press, 2008), 190–8.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

43

be a phenomenally conscious mode of sensory awareness, my negative claim is that Augustine’s analysis of outer vision is not an analysis of sense perception. Rather, I claim, in order for the deliverances of outer vision to reach the threshold of consciousness, outer vision must occur in conjunction with the activation of the faculties involved in inner vision. Hence my positive thesis: for Augustine, sense perception is a complex, hybrid state—one that involves the tandem operation of both outer and inner vision.³ If I am right, the end result is a rather less tidy analysis of perception since, on my reading, acts of sense perception turn out not to be susceptible to trinitarian analysis. Even so, the account is interesting and nuanced for all that. As will be clear, my argument presupposes not only that Augustine recognizes distinctions between different levels or types of awareness, but also that such distinctions are plausibly framed in terms of some notion of ‘consciousness.’ In order to prepare the way for my reading of book 11, therefore, I begin (§2), first, by clarifying what I mean in characterizing a given state as ‘conscious’ and, second, arguing for the claim that Augustine has a systematic way of marking the distinction between states that are conscious and non-conscious in this sense. In so doing, I also introduce the basic cognitive mechanisms Augustine relies on to explain such a distinction. Next (§3), I turn to Augustine’s discussion of outer and inner vision in book 11 and, thereafter (§4), to the defense of my two central interpretive theses. I conclude (§5) with some brief observations about how my interpretation finds further support and motivation when situated vis-à-vis some of Augustine’s broader philosophical and theological commitments.

³ Since beginning work on this paper, I discovered José Silva’s excellent 2014 article on active perception in Augustine (“Augustine on Active Perception,” in J. Silva and M Yrjönsuuri (eds.), Active Perception in the History of Philosophy (Dordrecht: Springer, 2014), 79–98). While Silva’s focus is not on questions about consciousness, nor specifically on De Trin. 11, nevertheless, in the course of defending a thesis about the soul’s activity in perception, he offers an account of perception similar to the one I defend here. Since its publication, there has been some criticism of Silva’s interpretation—especially on the issue of whether it is consistent with what Augustine says in De Trin. 11. See Kalderon, “Trinitarian Perception” and Silva’s reply in José Silva, “Perceptiveness,” Proceedings of the Aristotelian Society, supp. vol. 91 (2017), 43–61. It will be clear, however, that my own account of De Trin. 11, in many ways, supports Silva’s position.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

44  -

2. Consciousness and Levels of Awareness in Augustine Since there is both a wide range of psychological states that might be characterized as ‘conscious’ and no standard use of terminology for classifying various kinds of conscious phenomena, I need to say something about how I am using the relevant terminology, and how it applies to the various psychological phenomena Augustine recognizes and discusses. For my purposes, talk of ‘consciousness’ is simply a way of signaling phenomenality. Conscious states are those that register at the level of phenomenal awareness. To say that a given psychological act or state of awareness is conscious, then, is just to say that it is experienced by its subject; there is, as the familiar refrain goes, ‘something it is like’ for the subject to have or be in that state. Thus, conscious states, in the sense in which I am interested, are states a subject is aware of or experiences herself as being in.⁴ For clarity in what follows, it will be important to distinguish consciousness in the foregoing sense not only from cases of complete absence of awareness, but also from two other modes of psychological awareness: namely, what we might think of as non- or sub-conscious awareness, on the one hand, and introspective awareness on the other. In speaking of non-conscious awareness, I have in mind psychological states that we do not experience, but which nevertheless causally impact our conscious thinking and/or our behavior and actions. Thus, while the content of sub-conscious states does not feature in conscious experience, we do have a kind of unconscious, or functional access to it. Similarly,

⁴ Nowadays, treatments of phenomenal consciousness are very often focused around qualia—that is, the purely qualitative features associated with phenomenally conscious states. Thus, philosophers often take what-it’s-like talk to refer to its qualitative character—say, the particular greenish way the tree appears in awareness. That said, it is also widely acknowledged that there may be more to phenomenal consciousness, more to our experience of our mental states, than qualitative character. Thus, philosophers from Franz Brentano to Uriah Kriegel have called attention to the self-conscious or subjective character of phenomenally conscious states. The core idea here is just that phenomenally conscious states are states one is aware of oneself as being in. For recent treatments of this aspect of phenomenal consciousness see: Uriah Kriegel, Subjective Consciousness (Oxford: Oxford University Press, 2009) and “Consciousness and SelfConsciousness,” The Monist 87 (2004), 185–209; Colin McGinn, “Consciousness and Content,” in N. Block, O. Flanagan, and G. Güzeldere (eds.), The Nature of Consciousness: Philosophical Debates (Cambridge, MA: MIT Press, 1997), 295–308; and Joseph Levine, Purple Haze: The Puzzle of Consciousness (Oxford: Oxford University Press, 2001).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

45

consciousness as I am understanding it, is to be distinguished from introspective awareness. Introspection is a kind of deliberate, focal, attentive awareness of one’s own occurrent conscious states. When we introspect, we not only experience our mental states, but we explicitly attend to that very experience. Such attention to our own states is, however, relatively rare; in the ordinary course of things, our conscious experience occurs without our being, at the same time, introspectively aware of it. I call attention to these various modes of awareness as a way of introducing some basic distinctions Augustine himself draws in De Trin. between different levels or types of awareness.⁵ I should say, however, that Augustine’s clearest presentation of these distinctions, and of the cognitive mechanisms associated with them, occurs in contexts in which he is focusing on the mind’s higher, and purely rational (i.e., intellective) activities. And so it is on these texts that I shall, at least initially, be focusing. As will become clear, however, Augustine recognizes the same kinds of distinctions (and appeals to the same basic mechanisms) in book 11 when discussing inner and outer vision. To get a start at some of the distinctions Augustine draws between levels of awareness, consider his observations in the following passage: . . . in the recesses of the mind (abdito mentis), there are various awarenesses of various things (quarundam rerum quasdam notitias), and they come out somehow into the open and are set, as it were, more clearly in the mind’s view (in conspectu mentis) when they are thought about (cogitantur); it is then that the mind discovers and remembers and understands and loves something which it was not thinking about while it was thinking about something else. And if it is something that we have not thought about for a long time and are unable to think about unless we are reminded of it, then in heaven knows what curious way it is something that we do not know that we know. (De Trin. 14.9) ⁵ Since, for Augustine, we come to knowledge of the divine only by first turning inward and coming to know the nature of our own minds and thoughts, it is no surprise that he gives careful attention to both the nature and the phenomenology of our various mental activities and states.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

46  - Here, the main distinction on which Augustine focuses is between what I am calling conscious and non-conscious modes of awareness. In his own terminology, the distinction is between cases in which a given content occurs within one’s “mental field of view” (in conspectus mentis) and cases in which such content is present to (or present in) the mind, but not experienced as such. It is typical for Augustine to signal the former, conscious, mode of awareness by using variants of the verb ‘cogitare.’⁶ While this expression is almost always rendered into English as ‘to think,’ a translation that might suggest a specifically intellective activity, it is worth noting at the outset that Augustine uses the expression more broadly to refer to acts of occurrent awareness involving either rational or non-rational contents.⁷ Thus, for him, cogitatio involves the conscious processing of a given content (whether intellective or sensory) in such a way that it is, as it were, before the mind’s eye. Indeed, Augustine habitually appeals to visual metaphors not only to characterize the nature of acts of cogitatio, but also to describe the psychological mechanism responsible for conscious awareness. Thus, says Augustine, “thought is a kind of sight of the soul” (De Trin. 15.16) wherein “the mind’s gaze” (acies mentis, mentis aspectum) is directed upon a given content in such a way that one experiences one’s awareness of it.⁸ It is worth noting, moreover, that conscious awareness in this sense is to be distinguished from introspection; for Augustine, introspection is a matter of having these very acts of thinking (or knowing or understanding)

⁶ Augustine takes the verb, cogito (“I think”) to be an iterative form of cogo (“I collect”) and so an appropriate label for the soul’s activity drawing or bringing together any range of content (whether sensory, or purely rational) before the mind’s conscious awareness. See De Trin. 11.6 and Confessions 10.11.18. ⁷ This will become clear when we turn, in section 3, to Augustine’s discussion of sense cognition in book 11, where he speaks of conscious sensory awareness as ‘cogitatio.’ This broad usage is also made explicit at De Trin. 15.16, where Augustine points out that “thought is a kind of gaze of the soul, whether things are present that are seen with bodily eyes or sensed with other senses . . . or whether they are things only taught by the disciplines of a liberal education, or whether the higher causes and ideas of all these things are being thought about in their unchanging nature.” When Augustine is speaking of conscious thought involving purely intelligible contents, he will use ‘understanding’ (intellegere) to characterize the specific type of ‘thinking’ (cogitare) in question. In keeping with standard usage, I will continue to translate the Latin expressions ‘cogitare’/‘cogitatio’ in the customary way (namely, as ‘thinking’/ ‘thought’), but the reader must bear in mind the broad meaning of these terms. ⁸ As we will see, when referring to this gaze or view in connection with perceptual awareness Augustine speaks of it as the acies animi (soul’s gaze) rather than acies mentis (mind’s gaze).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

47

in view. In his terms, then, introspective awareness involves seeing one’s thinking (videt . . . cogitationem).⁹ Now, as Augustine frequently reminds us, the contents of the mind far outstrip what we presently have in view. In this way, we can be said to know (nosse) or have awareness (notitia) of far more than what we are presently (consciously) thinking of.¹⁰ Augustine gives numerous examples, but one such illustration runs as follows: It is like one who is expert in many disciplines: the things which he knows are contained in his memory, but nothing thereof is within the view of his mind except that which he thinks about. All the rest is stored up in a kind of secret awareness (arcana quadam notitia), which is called memory . . . . Thus, we say the man knows (nosse) letters even when he is thinking (cogitat) about other things, not letters. (De Trin. 14.8)¹¹

Here as elsewhere, the mechanism to which Augustine appeals in speaking of any kind of non- or sub-conscious awareness is that of

⁹ See, e.g., De Trin. 15.16 where Augustine puzzles over how it is that we do not, in fact, see all of our acts of thinking, while at the same time apparently recognizing that we do not: “Who fails to see (videt) his own thoughts (cogitationem)? And on the other hand, who does see his own thoughts? . . . Who fails to see them, and who does see them? After all, thought is a kind of sight of the soul.” Another context in which Augustine treats introspective awareness is his discussion at De Trin. 15.21 of whether we can know that we know, and know that we know that we know—and so on. ¹⁰ While ‘notitia’ is the nominal form of the verb ‘nosse’ (meaning ‘to know’), it would be a mistake to think that Augustine uses “notitia” in a technical sense for contents that are known in any strict, epistemic sense of that term. Indeed, in introducing book 11—which is narrowly concerned with non-rational awareness—he characterizes his subject matter as “notitia of bodily things” (rerum corporearum notitia) insofar as it is acquired through bodily senses (sensu corporis). See De Trin. 10.19. For this reason, I often translate ‘notitia’ as “awareness” in order to capture this more neutral sense of the expression. Like the expression ‘knowledge,’ ‘notitia’ can be used to refer as much to the contents of a cognitive act as to the act itself. ¹¹ Cf. De Trin. 14.9: “This person whom you perceive disputing about geometry is also a perfect musician, for he both remembers music, and understands, and loves it; but although he both knows and loves it, he is not now thinking of it, since he is thinking of geometry, of which he is disputing. And hence we are warned that we have a kind of knowledge of certain things stored up in the recesses of the mind, and that this, when it is thought of, as it were, steps forth in public, and is placed as if openly in the sight of the mind; . . . But in the case of that of which we have not thought for a long time, and cannot think of it unless reminded; that, if the phrase is allowable, in some wonderful way (I know not how) we do not know that we know. In short, it is rightly said by him who reminds, to him whom he reminds, ‘You know this, but you do not know that you know it; I will remind you, and you will find that you know what you had thought you did not know.’ . . . “But he who cannot contemplate these things, even when reminded, is too deeply buried in the darkness of ignorance, through great blindness of heart and too wonderfully needs divine help, to be able to attain to true wisdom.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

48  - ‘memory’ (memoria). As this passage makes clear, moreover, memory extends to any content which is retained in the mind, but is not within one’s current, conscious field of view. Indeed, for Augustine, memory functions broadly as the source or principle of all that we can access in occurrent, conscious awareness.¹² As he puts it elsewhere: memory constitutes “all that we know, even if we are not thinking of it” (De Trin. 15.40). Thus, the contents of memory are not restricted to past events or experiences, but rather include “the whole of a person’s knowledge together” (De Trin. 15.17). What exists in memory is present to the soul in such a way that it not only remains accessible for conscious consideration (in some cases, only with great effort), but may also actively structure conscious awareness and action.¹³ Now, as we have already noted, conscious awareness, on Augustine’s model, is a matter of directing the soul’s gaze so as to bring some content into view. Owing to our noetic limitations, however, the scope of our capacity for conscious awareness is highly restricted. Thus, of all that is psychologically present, known to us, and, thus, held in memory, what appears in conscious awareness does so only “bit by bit and one by one” and, as Augustine explains, only as a result of one’s turning the [mind’s] view (conspectu) from here to there and there to here, and again from there or there to this and that, as though one could not see something unless one stopped seeing other things. (De Trin. 15.23)¹⁴ ¹² Thus, importantly, for Augustine, memory is not a distinct faculty or power of the mind, but rather the mind itself insofar as it functions in different capacities. This extremely broad notion of memory characterizes Augustine’s account of memory not only in De Trin., but in the Confessions (where Augustine claims that God is among the things in memoria) and in De Genesi ad Litteram (where, in book 12, memory plays an important role in occurrent sense perception). The literature on Augustine’s theory of memory is vast. A very brief, and helpful overview can be found in Roland Teske, “Augustine’s Philosophy of Memory,” in E. Stump and N. Kretzmann (eds.), The Cambridge Companion to Augustine (Cambridge: Cambridge University Press, 2001), 148–58. See also John Mourant, Saint Augustine on Memory (Villanova, PA: Villanova University Press, 1980), and Gerard O’Daly, Augustine’s Philosophy of Mind (Berkeley: University of California Press, 1987), chapter 5. Of particular relevance to memory in De Trinitate is Paige Hochschild, Memory in Augustine’s Theological Anthropology (Oxford: Oxford University Press, 2012). ¹³ Cf. the case discussed in n. 11 above. This is also the case, as we will see, for non-conscious sense-based awareness. ¹⁴ Cf. De Trin. 11.12: “But the gaze of the soul cannot look at everything contained in memory at one glance, and so trinities of thoughts follow one another in succession, and one gets this innumerably numerous trinity.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

49

What we consciously experience is, therefore, to some degree, up to us. Indeed, Augustine thinks that, in many cases, what we know or become aware of is to be explained by appeal to the contribution (or failure thereof) of our will.¹⁵ Indeed, it is this feature of Augustine’s account of consciousness that secures his trinitarian analysis of it. For, as we shall see, on Augustine’s analysis, every occurrent act of cognition involves not only an object, on the one hand, and an act of awareness of it, on the other, but also an action or volition on the part of the subject in directing her attention to the object thus cognized. Hence consciousness is itself a trinitarian structure: acts of conscious awareness involve (i) the soul’s gaze (acies) being (ii) intentionally directed via the will (voluntas) to (iii) some object or content present to it. As there will be occasion to return—at some length—to Augustine’s trinitarian analysis of conscious awareness in connection with his account of sense-based cognition, the foregoing will suffice for the discussion to come. I now want to turn to book 11 and to Augustine’s account of outer and inner vision.

3. Outer Vision and Inner Vision in De Trin. 11 As I have indicated already, the overarching thesis that drives Augustine’s project in De Trinitate is the idea that the human mind is itself a likeness and image of the divine Trinity.¹⁶ While, strictly speaking, this trinitarian image lies solely in the intellect, Augustine holds that traces of it redound even to the lowest—i.e., non-rational—operations of the soul. Thus, as a way of elucidating the image of the Trinity as it exists in the ¹⁵ See De Trin. 9.18 and 10.4 for a discussion of the role of will in knowing. The role of will in sense cognizing will be explored in more detail below. ¹⁶ Drawing on Genesis 1:26, Augustine takes it as given that humans are the imago Dei; we are, in other words, created in God’s image and likeness. Given God’s Trinitarian nature, moreover, Augustine likewise takes for granted that humans bear the image and likeness of the divine Trinity. Insofar as the locus of this image of the Trinity is the human mind, it is possible, Augustine thinks, to come to some understanding of the Trinity by turning inward and studying the nature of our own mind and thoughts. In this way, he thinks we come “somehow to see him by whom we were made by means of this image which we ourselves are, as through a mirror” (De Trin. 15.14). For an exploration of the question of how the discovery of such images aids knowledge of the divine see Peter King, “Augustine’s Trinitarian Examples,” Medioevo 37 (2012), 83–106.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

50  - soul at the highest level, Augustine proposes to begin by exploring models of the trinity as they are found in these lower—more familiar, more accessible—cognitive activities.¹⁷ In this, Augustine pursues a characteristic ascent from exterior to interior and from inferior to superior. In book 11, therefore, he begins by examining the sensory-functions of “the outer human being”—that is, the human person with regard to its capacity for sense-based cognition of material bodies.¹⁸ The project, as Augustine characterizes it, is “to look for some model of the trinity in this human who is decaying; for even if it is not a more accurate model, it may perhaps be easier to distinguish.” In this regard, book 11 is a remedial exercise: a way to begin preparing us for the ideas to come. In subsequent books, Augustine gradually ascends to images of the Trinity as they exist in ‘inner human being’—that is, in the human person with respect to its capacity for various kinds of rational activity.¹⁹ What all this means, then, is that, in book 11, Augustine is not merely focusing on sense-based cognitive states, but focusing on them in such a way as to wholly exclude the contributions of higher, rational or intellective faculties in judging or interpreting sensory content. For, on Augustine’s view, what pertains to the outer human being is precisely those features of our mind (animus) that are shared in common with non-rational animals.²⁰ Accordingly, the discussion in book 11 covers ¹⁷ For reasons we will consider presently, the trinitarian structures involved in sense cognition so understood do not in fact constitute an imago Dei. Insofar as vestigial traces of the Trinity can be found in all created things, it is, nonetheless, possible to see a trinitarian likeness in the structure of perception. ¹⁸ Augustine cites Paul (2 Corinthians 4:16) for the distinction between the inner and outer man, but the distinction has Platonic sources as well. For more on this see Kaldreron, “Trinitarian Perception,” 25. ¹⁹ Augustine’s trinitarian analysis of thought as well as the methodology of inward ascent have precedent in the Neoplatonic tradition. For discussion of such influences, see Christopher Tournau, “The Background of Augustine’s Triadic Epistemology in De Trinitate 11–15,” in E. Bermon and G. O’Daly (eds.), Le De Trinitate de saint Augustin: exégèse, logique et noétique (Paris: Vrin, 2012), 251–66. ²⁰ Augustine opens book 12 by explicitly delineating the boundary between the outer human being, with which he was concerned in book 11, and the inner human being to which he is turning in book 12. The dividing line, he tells us, is rationality. “Well now, let us see where we are to locate what you might call the border between the outer and the inner human being. Anything in our mind (in animo) that we have in common with animals is rightly said to be still part of the outer human being. It is not just the body alone that is to be reckoned as the outer human, but the body with its own kind of life attached, which quickens the body’s structure and all the senses it is equipped with in order to sense things outside. And when the images of things sensed that are fixed in the memory are looked over again in recollection, it is still something

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

51

both acts of cognizing external, material bodies, and acts of memory and imagination, all of which involve sense-based representations. Insofar as Augustine’s project is merely to uncover the trinitarian structure of our non-rational, sense-based cognitive acts, his analysis is not (and is not intended as) a full-blown or exhaustive treatment of sense-based cognition. Rather, he focuses selectively on just those aspects of sense cognition relevant to his broader, theological pursuits. For the same reason, however, the precise cognitive application and implication of his analysis can be difficult to pin down. I begin, therefore, by simply tracing the details of his presentation of each of the two trinities of book 11, namely, that of outer and inner vision respectively. Not surprisingly, each type of vision shares the same basic structure. Both outer and inner vision consist in a kind cognitive union between (i) the cognized object, on the one hand, and (ii) the relevant sensory power on the other. This cognitive contact is facilitated by (iii) the will. Indeed, the act of ‘seeing’ or ‘sight’ (visio) (where this is just the actualization of the relevant cognitive power in receiving the form or species of the object cognized) is itself the culmination of the will’s activity in directing awareness.

a. Outer Vision In presenting the details of this picture, Augustine begins by focusing on outer sense cognition—that is, on cognition of external, material objects. In developing his account, he uses vision as the representative sense power (though the basic analysis is intended to apply to any sense modality).²¹ Accordingly, the first trinitarian structure is a trinity of outer vision. Augustine presents it this way:

belonging to the outer human being that is being done. In all these things, the only way that we differ from animals is that we are upright, not horizontal, in posture” (De Trin. 12.1). ²¹ As Augustine explains, “So then, the outer person is endowed with sensation, and with it senses bodies; and this sensation, as can be readily verified, is divided into five parts, seeing, hearing, smelling, tasting, touching. But it would be too much, and quite unnecessary to ask all these five senses about what we are looking for. What one tells us will go for the others. So let us use for preference the evidence of the eyes; this is the most excellent of the body’s senses, and for all its difference in kind has the greatest affinity to mental vision” (De Trin. 11.1).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

52  - When we see some particular body, there are three things which we can very easily remark and distinguish from each other. First of all there is (i) the thing (res) we see, a stone or a flame or anything else that the eye can see, which of course could exist even before it was seen. Next there is (ii) the actual sight or vision (visio), which did not exist before we sensed the object presented to the sense. Thirdly, there is (iii) what holds the sense of the eyes on the thing being seen as long as it is being seen, namely the soul’s intentional aim (animi intentio). These three are not only manifestly distinct, but also different natures. (De Trin. 11.2)

Here, Augustine opens by simply calling our attention to the three constitutive elements of sensory seeing, namely: (i) the object or thing (res) that is seen, (ii) the seeing (visio) itself, and (iii) the soul’s intentional aim (animi intentio) upon the object. As Augustine will go on to make clear, each of the three elements are not only distinct from one another, but also jointly necessary for the occurrence of outer, bodily vision. The mutual distinctness of each item from the other can be shown, Augustine thinks, from the fact that (i) each has a different nature and (ii) each can exist without the other. On Augustine’s view, “the visible body is of quite a different nature than the sense of the eyes.” For, the visible object is entirely corporeal, whereas the body’s sensory organs— insofar as they are animated by the soul—are an “admixture” of body and soul.²² But the soul’s intentional aim differs from both the visible body and the sensory organs since it is a power that “belongs just to the soul”—a fact which Augustine takes as evident given its active, volitional nature.²³ Again, each constituent can exist without the others. Not only can a visible object exist independently of being seen, but insofar as the form received in the eyes is distinct from the body itself, it is also possible ²² On Augustine’s view, it is only because the soul is somehow present at or in the physical organs that such organs are powers for sensory cognition. As he says: absent the soul (exanimae) there is no capacity for bodily sensing. See De Trin. 11.2. ²³ In fact, at points, he refers to this third element of sensory cognition as a “volition” of the soul. Cf. De Trin. 11.5. Although soul’s “intentio” functions as the third member both in the trinities of outer and inner vision, Augustine tends to place more emphasis on the role of will (voluntas) as the source of the soul’s intentional uniting with the objects stored in memory. See MacDonald, “Augustine’s Cognitive Voluntarism,” for a discussion and limited defense of Augustine’s conception of sense perception as involving volition.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

53

for the eyes to be informed (at least for a time) even when the body itself is no longer present.²⁴ Finally, the soul’s intentional aiming persists even in the absence of a functioning sense organ or visible objects. After all, Augustine claims, the soul’s “appetite for seeing (videndi appetitus) remains intact whether this happens to be possible or not.”²⁵ While the main point of all of this is to establish the distinctness of each of the members of this trinity from the others, Augustine’s discussion also makes clear why he thinks all three components are jointly necessary for outer vision. Without the first member of the trinity, namely, (i) the visible object, nothing would be seen. In such a case, there would be no uniting of the sensory organ of vision to an external visible object. Again, if a visible object is present, but without the eye being actually informed from it, there is no (ii) sight. After all, vision requires that the sense faculty actually take in, or receive, information from the surrounding environment. But notice that this very event, namely, the eyes receiving such information, requires (iii), that the eyes are actually being directed at the relevant visible object. Seeing requires looking, as it were. In this regard, the cognitive contact that constitutes outer vision is, on Augustine’s model, the consummation of a prior inclination or will for seeing (voluntas videndi).²⁶ Together these three things—the object, the aiming to see, and the sight—yield what Augustine characterizes as “a kind of unity,” namely, the cognitive “joining” that constitutes bodily vision.²⁷ According to Augustine, this analysis of outer vision yields a trinitarian image not merely because we have a case in which three distinct things yield a kind of unity (a single, unified cognitive event), but because the relations among the three reflect those that hold among the three persons of the Trinity. For example, the external object as the source of ²⁴ “It often happens that when we look at some lights for a little while and then close our eyes, certain luminous colors continue to revolve in our vision, changing their hues and gradually becoming less brilliant until they cease altogether. We can understand them as being the remnants of that form which was produced in the sense while we were looking at the luminous body . . . ” (De Trin. 11.4). ²⁵ De Trin. 11.2. ²⁶ See De Trin. 11.10. ²⁷ Here’s Augustine: “Although these three differ in nature they are compounded in a kind of unity, that is to say, the form of the body which is seen, and its image imprinted on the sense which is sight, or formed sense, and the volition of the soul (voluntas animi) which applies the sense to the sensible thing and holds sight on it” (De Trin. 11.5).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

54  - bodily sight serves as the analogue to the first person of the Trinity, namely, God the Father. And because sight itself involves (i) a species—or likeness—of that object that is (ii) produced (or “begotten”) by that object, sight itself functions as the second person of the Trinity, namely, as God the Son. Augustine spells it out this way: So it is that sight, that is the form which is produced in the sense of the beholder, has its quasi-parent (quasi parens) in the form of the body from which it is produced. But this is not a true parent, and so the former is not a true offspring; it is not wholly begotten by it since something else is presented to the visible body for sight to be formed out of it, namely the sense of the one who is seeing . . . So the will which joins them both together as quasi-parent and quasi-offspring is more spiritual than either of them . . . . It does not proceed from that quasiparent, or for that matter from this quasi-offspring, [rather] . . . the will was already there before sight occurred and it applied the sense of the body to be formed from it by observing it. However it was not yet pleased; how could it be with something not yet seen? (De Trin. 11.9)

In Augustine’s terminology, the external object is “quasi-parent” (i.e., analogue to the Father), and the informed eye its image and, so, “quasi-offspring” (i.e., analogue to the Son). The will for seeing (voluntas videndi), being neither parent nor offspring (and, hence, analogue to the Holy Spirit), completes the trinity in its role as what “joins them both together.” This trinity is not, of course, anything like a proper image of God given that, among other things, it is comprised of entities of wholly different natures (body, soul, and an admixture thereof) and so lacks the requisite unity. It is merely a start. From here we are ready to move to more inward operations of the soul—albeit still just insofar as the soul is occupied with sensory representations. Hence, Augustine turns now to his second sensory trinity: namely, that of inner vision.

b. Inner Vision Augustine begins his discussion of this second trinity by pointing out that when the species of an external object is received in a corporeal sense

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

55

organ, it is (typically) also transmitted to memory. Thus, even when the object itself is no longer present it is possible, nevertheless, to recall it in an occurrent act of awareness. Even when the species of a body sensed by corporeal senses is taken away, there remains a likeness of it in memory to which the will can again turn its gaze to be formed by it from within just as the sense was formed from without by the sensible body presented to it. (De Trin. 11.6)

In this way, when stored sensory forms or species are utilized in occurrent acts of sensory awareness, we get a second sensory trinity: a trinity of ‘inner vision.’ Thus, Augustine continues: And so one gets another trinity out of the memory and internal sight and the will which couples them together; and when these three are coagitated (coguntur) into a unity the result is called cogitation or thought (cogitare), from the very act of co-agitation (coactu). Nor among these is there a difference in substance . . . . Instead of the species of the body which was sensed outside, there now appears memory retaining (i) that species which the soul (anima) drank in through the sense, and instead of that external sight of the sense being formed from the sensible body, we now have a similar (ii) internal sight when the soul’s gaze (acies animi) is formed from what the memory retains, and absent bodies are thought about (cogitantur); and (iii) the same will that, in the first case, applied the sense for formation to body presented to it outside and kept it joined to it once formed, now turns the gaze of the soul of the one recollecting to the memory in an act of recollection for it to be formed from what the memory has retained, and there is produced in thought something like sight. (De Trin. 11.6)

As in the case of outer vision, Augustine characterizes inner vision as a cognitive event in which a cognitive power is joined or united to its object. In this case, however, the object in question is not an external body, which, Augustine stipulates, is no longer present. Rather, the object is a sensory representation of an external body—a species in memory; and what is informed from it is not a corporeal eye, but the soul’s inner, incorporeal gaze (acies animi). Nevertheless, it is “the same will that in

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

56  - the first case applied the sense for formation to the body presented to it outside and kept it joined to it once formed, [that] now turns the soul’s gaze to the memory.”²⁸ The uniting of the soul’s gaze (acies animi) to the species in memory yields an occurrent, conscious act of sense-based awareness—or what Augustine, in his usual way, refers to as ‘thought’ (cogitatio). Although the etymological connection between the Latin verb ‘cogere’ (to collect together or join) and the verb ‘cogitare’ (to think or cogitate) does not translate in English, it is precisely this connection that Augustine means to signal in using the term ‘thought’ (cogitatio) in this context. For, the actualization of the acies consists in its being conjoined to, and so informed by, a sense-representation present in memory—a conjunction that yields an act of conscious remembering. In cases where retained images are recombined in various ways (often in ways that do not correspond to reality), the sense-based thought in question constitutes an act of imagination.²⁹ Either way, each type of act is an instance of inner vision: that is, a kind of interior awareness which, while not involving a bodily eye, nevertheless, yields “something like sight”— namely, an inward seeing and, hence awareness, of a given sensory content.³⁰ Now, as Augustine’s discussion makes clear, acts of inner, sense-based thinking, namely, acts of remembering or imagination, are structurally isomorphic to acts of outer vision. In this case, the three members of the trinity are: (i) the inner object, namely, the species in the memory, (ii) the inner seeing—i.e., the soul’s gaze (acies animi) being informed by the species from memory; and (iii) the intentional directedness of the soul or will (intentio voluntas) upon the inner object. Having identified the three constituents of this second trinity, Augustine goes on, as he did ²⁸ Cf. De Trin. 11.7. ²⁹ E.g., at De Trin. 11.17, Augustine notes that the will can act not only to unify conscious attention and the contents of memory, but also to creatively recombine such contents present in memory. ³⁰ It is only when inner vision occurs in the absence of (or absent attention to) corresponding external objects that it constitutes an act of memory or imagination. In such cases, what one is aware of is not anything in external reality, but something in memory or imagination. As will be clear, however, on my view, when inner vision occurs (i) as part of a unified process that includes outer vision, and (ii) has external bodies (not inner images) as object, the awareness is perceptual. (See section 4 below.)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

57

in the case of outer vision, to argue for the distinctness of each member; and he likewise calls attention to their unity in the production of inner vision. There is no need to survey the details of his discussion here except to note that, in this case, we get a unity that is much greater than that among the three constituents of outer vision. As Augustine observes: the three members of the trinity of inner vision “are not now differentiated by a diversity of nature but are of one and the same substance, since all of this is inside and it is all one soul” (De Trin. 11.7). For the same reason, the members of this trinity are likewise a better (though still far from proper) image of the relations among the first, second, and third persons of the divine Trinity. Toward the very end of his discussion in book 11, Augustine sums up his results by considering the triune components of both outer and inner vision in the context of a broader, ordered series of sense-cognitive events. As this passage will be of particular significance for the discussion to come, it is worth quoting at length: So it is that in this series which begins with the species of the body and ends with the species which is produced in the thinking gaze (contuitu cogitantis), four species are brought to light, born as it were step by step one from the other; the second from the first, the third from the second, the fourth from the third. From (i) the species of the body that is seen arises (ii) the species that is produced in the sense of one seeing, and from this (iii) the one produced in the memory, and from this (iv) the one that is produced in the gaze of the one thinking (acie cogitantis). So the will couples quasi-parent with its offspring three times: first the species of the body with the one it begets in the sense of the body; next, this one with the one that is produced from it in the memory; and then a third time this with the one that is brought forth from it in the gaze of thought (cogitantis intuitu) . . . .And, in this series, there are two visions: one of sensation, the other of thought. It is to make possible the sight of thought that there is produced from the sight of sensation something similar in the memory, which the gaze of the soul can turn to in thought just as the gaze of the eyes turns to a body in observing it. That is why I have wished to propose two trinities of this kind, one when the sensation of sight is formed from the external body,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

58  - the other when the vision of thought is formed from internal memory. But I did not wish to propose a middle trinity in between because it is not usually called a vision when the form that is produced in the sense of the observer is committed to memory. In every instance, however, the will only appears as coupling quasi-parent with its offspring. And for this reason, wherever it proceeds from, it cannot itself be called parent or offspring. (De Trin. 11.16)

Here, Augustine shows how the trinities of outer and inner vision form part of a series involving the transmission of sensory information from the external world all the way to inner awareness. Indeed, because the process starts with the external visible object, the sensible form of that object is counted among the ‘species’ included in the series itself. Considered in this way, the process as a whole includes: “four species . . . born, as it were, step by step”; a three-fold coupling of “quasi-parent with its quasi-offspring” by the will; and “two visions—one of sensation, one of thought.” Represented in diagram fashion, we get something like Figure 1. Interestingly, the chain of transmission Augustine presents here includes an intermediate (and hitherto unmentioned) stage, namely, one at which the species in the animated sense organ is transmitted (via yet another conjunctive application of will) to memory. We will have reason to return to this part of his discussion presently. For the moment, it suffices just to call attention to the fact that, in his account of sensory cognition, Augustine includes not merely an analysis of inner and outer vision respectively, but also of their serial occurrence. And, he does so Outer Vision

External object

species1

Inner Vision

Sensory organs

species3

species2 application 1

Memory

application 2

Figure 1 Trinities of outer vision

Acies

species4 application 3

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

59

even though the middle stage in the series presents something of a complication for him. For in this middle stage, there is an object, an informing of a cognitive power, and a conjunctive application of will uniting them, yet the occurrence of the three do not constitute a trinity. (And, apparently, this is because the information received in memory does not register in any distinct act of occurrent awareness and, so, does not constitute “vision of any sort.”)

4. Limits of the Trinitarian Analysis: Locating Perception in De Trin. 11 Having established the main elements of Augustine’s account of both outer and inner vision, the natural question to consider is this: to which—and what range of—sense-cognitive states does the trinitarian analysis apply? One might very naturally suppose that Augustine intends his analysis as having fairly wide application; that is, we might think that he intends it to apply in a general way to all sense-based cognition. After all, Augustine tells us that his account of outer vision applies equally to any of the five outer, sensory faculties; and he is also explicit about the fact that his analysis of inner vision is applied both to acts of memory and imagination. Hence, on the face of it, it could appear that the analysis covers not only all modes of sense perception, but also occurrent states of sensory memory and imagination. Indeed, we could even take the unheralded, middle, pseudo-trinity of informed memory, as giving us an account of dispositional sensory memory. This is not, however, what I think Augustine has in mind. Rather, as I read book 11, the trinitarian analyses Augustine develops have a rather narrower range of application. In particular, the two trinitarian images he develops do not apply to what we might think of as the central case of sensory cognition: namely, ordinary sense perception. In order to establish this, I begin with my negative thesis: namely, that what Augustine calls ‘outer vision’ is not plausibly identified with what we (or he) would think of as ordinary, paradigmatic cases of sense perception. And this is because, unlike ordinary cases of perceptual experience, the sensory states that Augustine classifies as instances of ‘outer vision’ are not

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

60  - phenomenally conscious states. From here, it is a relatively short step to my positive thesis: namely, that insofar as perceptual awareness involves both the deliverances of the “outer” sense faculties, as well as the involvement of “inner” faculties of both memory and the acies animi, sense perception turns out to be a kind of hybrid state—one involving the joint occurrence of both outer bodily sensing and inner conscious awareness.

a. Negative Thesis: Outer Vision Is Not Perception In support of my negative thesis, I offer two kinds of evidence. The first comes from a close reading of a single, telling passage in book 11—one that I think sheds a good deal of light on the phenomenology Augustine associates with outer and inner vision respectively. The second sort of evidence comes from systematic considerations, having to do with Augustine’s broader views about conscious awareness. I begin with the latter, systematic considerations. It will be useful to have these considerations in mind before we turn to the key text. For starters, recall Augustine’s systematic identification of conscious awareness—at both the intellective and sensory level—with acts of cogitation or thought (cogitatio). As we have now seen, thought consists in the activation of the acies animi—i.e., the activation of the soul’s power for occurrent, conscious awareness. Thus, when the soul’s gaze is informed by a given content, that content is present within one’s conscious field of view (contuitu cogitantis).³¹ In this way, the acies animi serves consistently, for Augustine, as the psychological mechanism for conscious awareness. Add to this the fact that, as Augustine points out in the following passage, the acies animi is always informed from memory. We very frequently believe people when they tell us true things which they have themselves perceived with their senses . . . . It does not seem in this case as if our attention is being bent back to our memory to ³¹ Indeed, it would appear that, for Augustine, the activation of the soul’s gaze in an act of cogitatio is not only sufficient for conscious awareness, but also necessary. Cf. De Trin. 14.8: “Nothing can be in the mind’s view except what is being thought about.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

61

produce sights; after all, we are not thinking about them because we are recalling them, but because someone else is telling us them . . . . However, if we look at the matter a little more closely, not even in this case do we depart from the limits set by memory. I could not even begin to understand what he was telling me if I was hearing all the things he said and what they added up to for the first time, and did not have a general memory of each of them . . . . So the limits of thinking are set by the memory just as the limits of sensing are set by bodies. The senses receive the species of a thing from the body we sense, the memory receives it from the senses, and the gaze of the one thinking from the memory. (De Trin. 11.14)

As Augustine puts it here: “memory is the limit of thought.” You cannot, in other words, have an act of cogitatio involving some content if that content is not first in your memory. In this particular context, Augustine is considering an example in which you are picturing an event narrated to you by your friend. In the circumstance, it would seem that you are thinking about something that is not, in fact, in your memory since the story involves an event that did not happen to you. Thus, on the face of it, the case appears to be a counterexample to the principle that the gaze of thought is always informed by a content derived from one’s own memory. But, in response, Augustine simply denies that this is the case, insisting instead that “if we look at the matter a little more closely, not even in this case do we depart from the limits set by memory.” The point, of course, is that our ability to understand the import of our friend’s story owes to the fact we have concepts and experiences within our own mind—memoria—to draw on. Thus, even in such a case, our conscious thinking about the events in our friend’s story is informed by what is present in our own memory. Augustine reinforces the point by reiterating, at the end of the passage, the process by which sensory information reaches conscious awareness: “The senses receive the species of a thing from the body we sense, the memory receives it from the senses, and the gaze of the one thinking from the memory.” The significance of all of this, for our purposes, is that this discussion signals a general principle about the connection between conscious awareness and memory: namely, the acies animi is always informed

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

62  - from, and hence posterior to, memory. In other words, conscious sensory awareness, insofar as it requires the activation of the acies animi also necessarily requires the involvement of memory. Of course, insofar as conscious attention is selective and active, such awareness also involves a contribution from the will. Thus, the will’s contribution, as Augustine describes it, involves its activity in making some (but not others) of the representations that are present in memory salient in one’s conscious awareness. As Augustine says: “the will turns the gaze (acies) here and there and back again to be formed and once formed it keeps it joined to the image in memory” (De Trin. 11.7). If all this is right—that is, if the acies is always informed from memory, and is directed by the will—we can expect that where there is both a failure of memory (i.e., some content fails to be taken up in memory) and a failure of will (i.e., absence of attention) there will be no conscious awareness. And, this, in fact, is precisely what Augustine says in a key passage. The passage I have in mind comes toward the end of book 11. In it, Augustine is exploring the nature of will’s contribution to sense cognition. He pursues the matter, however, by considering what happens, cognitively speaking, when will is absent from one or more of the various stages of sense cognition. While Augustine typically lays emphasis on the will’s positive, unitive contribution to the formation and constitution of sensory states, in this text he calls attention to the fact that will’s activity can also be cited in explaining the non-occurrence of such states. Thus, in the opening lines, he points out that the will may act not only to apply one’s attention to some object, but also to avert it. He says: “Just as it is the will that fastens sense to body, so it is the will that fastens memory to sense and the gaze of the one thinking to the memory. And what fastens them together and assembles them also unfastens and separates them, namely the will again” (De Trin. 11.15). Augustine immediately goes on to specify particular ways in which the will can separate (rather than unify) a cognitive power and its object, and then to explain the cognitive outcome of such cases. What he has to say is illuminating. Here’s the text in its entirety: [STAGE-1 FAILURE:] It is by movements of the body that it separates the senses of the body from the bodies to be sensed, either to avoid

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

63

sensing or to stop sensing something; as when we shut our eyes or turn them away from something we do not want to see . . . .Thus it is by moving the body that the will avoids coupling the senses of the body to sensible things. And it does it as far as it can. When it suffers difficulty in this respect because of our condition of servile mortality, the result is torment, and nothing is left to the will but endurance. [STAGE-2 FAILURE:] Memory is averted from sense by the will when, intent on something else, it does not allow present things to adhere to it. This is easy to observe when, as often happens, we are with someone talking to us and appear not to have heard what they are saying, because we are thinking of something else. It is not true, though; we have heard, but we do not remember the sounds slipping that very instant through our ears, because the will has been disinclined to give the permission which is needed as a rule to fix them in the memory. So it would be truer to say “We do not remember” than “We did not hear” when something like that happens. It also happens to people reading, extremely often at any rate to me, that I find I have read a page or a letter and have not the slightest idea what I have read, and have to repeat it. The will’s interest has been intent on something else and so the memory has not been applied to the sense of the body as that sense has been applied to the letters. So too, you go for a walk with your will intent on something else and so you do not know what path you have taken. If you actually had not seen, you would either not have gone for the walk or you would have walked by feeling your way with great attention, especially if you were going along a way you did not know. But you walked quite easily, so of course you did see . . . . [STAGE-3 FAILURE:] Finally, the way the will averts the gaze of the soul from what is in the memory is simply by not thinking about it. (De Trin. 11.15)

The cases Augustine identifies here fall into three groups, each corresponding, respectively, to a failure at one of the three stages of will’s application depicted in Figure 1, which, for convenience, I reproduce here. Before considering the precise significance of this passage vis-à-vis my negative thesis, it will be useful to begin by first reviewing the basic details of each of the three types of case Augustine considers.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

64  - Outer Vision

External object

species1

Inner Vision

Sensory organs

species3

species2 application 1

Memory

application 2

Acies

species4 application 3

Figure 1 Trinities of outer vision

The first set of cases Augustine considers (those labeled as ‘stage-1 failures’ in the text) are cases in which there is a failure of the will’s application at the very first stage of sense cognition. Here, the failure of application of the sense organs to the external object consists in the will positively acting so as to separate the “senses of the body” from “bodies that are to be sensed.” It does so, for example, by simply averting the relevant sensory organ—say, by willing to shut or avert the eyes, cover one’s ears or nose, etc. In these cases, will’s role is characterized in terms of resistance. As the examples make clear, where will succeeds in separating, it prevents or terminates outer vision.³² And where there is no outer vision, there is, of course, no awareness whatsoever of the surrounding physical environment.³³ Stage-1 failures thus represent absence of any outer, bodily sensing.

³² Augustine does allow, however, that even if one wills against the application of senses to corporeal objects, outer vision may, nevertheless, persist. In such cases, he describes the subject as “unwillingly enduring” an act of outer vision. Here the famous scene from Clockwork Orange comes to mind in which the central character, Alex, has his eyes wired open and is forced to watch violent films. But perhaps Augustine himself had Plato’s case of Leontius and the corpses in mind (Republic IV, 439e–40a). ³³ Outer vision fails not only in cases where the will actively resists application of the senses to external bodies, but also when the senses of body are inert for some reason (as in sleep, or as in cases of physical defect say, blindness/deafness), or, more rarely, in cases in which the soul’s “intentional aim” (animi intentio) is “fixed by a kind of necessity on some inner image” (as in cases of madness, divination, or spiritual ecstasy). Augustine offers a fuller description of cases of the latter sort elsewhere in De Trin. 11 and there he characterizes them as cases in which the will “concentrates its whole energy on the inner image (interiorem phantasiam), and withdraws the gaze of the soul altogether from the presence of the bodies that surround the senses, and from the senses of the body themselves, and directs it utterly on the image that is seen within” (De Trin. 11.7). Thus, while the eyes might even be open (or other senses equally at the ready),

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

65

In the second set of cases (namely, those labeled ‘stage-2 failures’), there is a failure in the application of memory to information present in sense organs. Outer vision exists, but its input is not recorded in memory. In these cases, the failure owes not to active resistance on the part of the will, but to distraction. When one’s attention is sufficiently focused elsewhere (because, say, will is “intent on something else”), the data received by the outer senses is neither registered nor processed in memory. Where there is a stage-2 failure, information in bodily senses simply does not penetrate further than the outer senses. The final type of case (namely, ‘stage-3 failures’) represents the selective nature of occurrent, sense-based thought. Since there is more present in memory than is (or can be) entertained in any occurrent act of awareness (be it an act of memory, imagination, etc.), the will functions to separate the soul’s gaze from any range of sensory contents by simply failing to bring or “apply” the acies to them (or, presumably, by actively resisting doing so).³⁴ What stage-3 failures show, then, is just that consciousness of a given (non-rational) content does not occur without the soul’s gaze being applied-to or informed-from memory. This, of course, fits the general model outlined above. With these details in mind, we are now better positioned to consider the import of the passage for the thesis that outer vision is not—by itself— sufficient for perception. Indeed, what is particularly valuable about Augustine’s presentation of these cases—especially the cases of failure at stage 2—is the way in which it allows us to isolate the occurrence of outer, physiological sensing from more inward psychological processing. Doing so allows us to identify not only what sort of state, and what kind of phenomenal experience, outer vision yields when it occurs on its own, but also what states correspond to its conjunction and coordination with inner awareness. For starters, what Augustine’s discussion shows is that outer vision can, in principle, occur without the informing of memory, or the the subject, nonetheless, senses nothing—there is no awareness of perceptible objects. Augustine considers similar cases in De Genesi ad litteram; see for example Gen. Litt. 12.25 and 12.27. ³⁴ Likely, Augustine would allow, as he does for outer vision, that inner seeing can occur against (or without) one’s willing it. Consider, e.g., nightmares, or cases in which you are unable to get a song out of your mind, or in which someone is haunted by the vivid memory of a traumatic event.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

66  - subsequent informing of the soul’s gaze. This alone—given what we have already seen of Augustine’s systematic account of the psychological mechanisms associated with consciousness—suggests that outer vision is a non-conscious or sub-conscious mode of awareness. Not only that, but the three examples Augustine offers to illustrate the nature of isolated outer vision support precisely this conclusion. Each involves some failure of awareness. His first example is a case in which someone in your vicinity is talking—say, giving a lecture in your department—but, having gotten distracted by thoughts about an upcoming appointment, you discover that you have not been hearing anything. In the second case you are reading (perhaps this very paper) but then you realize that, although your eyes have been scanning the words and your hands turning the pages, you “have not the slightest idea what [you] have just read.” The final example is Augustine’s analogue to David Armstrong’s famous example of the long-distance truck-driver.³⁵ In Augustine’s version, however, “you go for a walk with your will intent on something else and so you do not know what path you have taken.” Unlike cases in which outer vision fails, in each of these examples there is clearly some level of sensory awareness. Indeed, as Augustine points out in the walking case, you manage to reach your destination safely and without feeling your way, so, “you did see.” On the other hand, he is also keen to emphasize that in each case you also fail to experience any sensory awareness. Thus, it appears to you that you have not heard, not read, nor observed the path by which you walked. ³⁵ See David Armstrong, “The Nature of Mind,” in The Nature of Mind and Other Essays (Ithaca, NY: Cornell University Press, 1981), 1–15. Armstrong’s case, which has become quite familiar in the current literature on consciousness, runs as follows: “If you have driven for a very long distance without a break, you may have had experience of a curious state of automatism, which can occur in these conditions. One can suddenly ‘come to’ and realize that one has driven for long distances without being aware of what one was doing, or, indeed, without being aware of anything. One has kept the car on the road, used the brake and the clutch perhaps, yet all without any awareness of what one was doing” (“The Nature of Mind,” 12). Armstrong claims that in such a case “something mental is lacking” and he goes on to argue that what’s lacking on the part of the driver is awareness of his perception of the road. “The driver in a state of automatism perceives, or is aware of, the road. If he was not, the car would be in a ditch. But he is not currently aware of his awareness of the road. He perceives the road, but he does not perceive his perceiving, or anything else that is going on in his mind. He is not, as we normally are, conscious of what is going on in his mind . . . . The driver in the automatic state is one whose ‘inner eye’ is shut: who is not currently aware of what is going on in his own mind” (“The Nature of Mind,” 14).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

67

The reason you seem not to have heard or seen anything is that, in such cases, you do not have any awareness or experience of outer acts of bodily sensing. For when outer vision occurs without simultaneously engaging memory and the soul’s gaze, its deliverances fail to register at the level of conscious awareness. What stage-2 failures show, then, is that outer vision alone is not sufficient for perceptual awareness. It is, rather, a sub- or non-conscious mode of awareness. Admittedly, in describing such cases, Augustine places particular emphasis on them being instances of a failure of memory. For example, in the case where we are inattentive to someone who is speaking, he says: “we have heard, but we do not remember the sounds . . . . So, it would be truer to say ‘we do not remember’ than ‘we did not hear.’” Likewise, in the case of inattentive walking. As I have just noted, Augustine insists: “you walked quite easily, so you did see.” One might think, therefore, that what stage-2 failures represent is not a failure of conscious perceptual awareness, but merely a failure to recall its (perhaps fleeting) occurrence.³⁶ Although this alternative reading of stage-2 failure has some initial plausibility, it is here that reflection on the broader, systematic considerations just canvassed can serve to guide our interpretation. It goes without saying that stage-2 failures are rightly characterized chiefly as a failure of memory: after all, they are cases in which the will has not applied the memory to the external sense organs. Thus, information present in the senses has not been taken up or processed by memory. What this means, however, is that such information is also not received in the acies animi. For, what we have now seen is that, on Augustine’s view, the acies is always informed from memory. Thus, where there is no uptake of sensory content in memory there is no possibility for its uptake in one’s conscious field of view. Hence such cases cannot represent failure to recall conscious perceptual experiences. No such experience has (yet) occurred. When, therefore, Augustine insists that “you did see” or “you did hear,” he is merely highlighting the occurrence of outer, bodily vision and outer, bodily hearing. After all, it is precisely the fact that outer seeing (and hearing) does occur in such cases that distinguishes them ³⁶ I am grateful to Scott MacDonald for calling my attention to (and pressing) this alternative reading of the passage.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

68  - from instances of stage-1 failure—namely, cases where outer vision also fails. But such seeing and hearing is nothing more than the actualization of the external sense organs in receiving the species of external bodies. And this, as I have been arguing, is not sufficient for perceptual experience. For that we need the informing of the soul’s gaze. So much for the defense of my negative thesis: namely, that outer vision is not the equivalent of sense perception. It is, I think, a fairly short step from the negative thesis to my second, positive thesis—namely, that perception is a hybrid state requiring not only the occurrence of outer vision, but the simultaneous contributions of memory and inner vision. It is, therefore, to the defense of this thesis that I now turn.

b. Positive Thesis: Perception Is a Hybrid State If outer vision (taken by itself ) is a non-conscious mode of sensory processing, and inner vision (again, taken by itself ), while conscious, is not a form of perceptual awareness, we might be tempted to conclude that the trinities of sense cognition Augustine identifies in book 11 are not relevant to or even intended as part of an analysis of ordinary conscious perception. If Augustine’s project in book 11 is simply to focus on those sensory states susceptible of a trinitarian analysis, perhaps it is a mistake to expect an analysis that applies to all or even the paradigmatic cases of sense-based cognition. Indeed, it may be that Augustine’s discussion in book 11 omits conscious perceptual states altogether in order to focus singularly on sensory states that fit the trinitarian model he seeks to illuminate. In fact, however, I think this is not the case. As we have seen, Augustine’s discussion in De Trin. 11 includes not only an individual analysis of outer and inner vision, respectively, but also consideration of them together as an ordered series. Although no single stage in this series—taken by itself—constitutes ordinary conscious perception, it is plausible to think that Augustine intends the whole series together to provide us with an account of just this phenomenon. Indeed, it seems no accident that, immediately after discussing the separation cases, Augustine turns directly to consider “the series which begins with the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

69

species of a body and ends with the species which is produced in the thinking gaze.” I have already made the case for thinking that outer vision, by itself, does not constitute perception. It is quite clear, moreover, that just the first two stages alone (i.e., outer vision, plus the informing of memory) are likewise insufficient for conscious perceptual awareness. As Augustine himself points out, there isn’t anything rightly “called vision when the form that is produced in the sense of the observer is committed to memory.” Rather, conscious awareness requires the activation of the soul’s gaze. It is likewise clear, moreover, that the third stage (i.e., the activation of inner vision) depends on the second as it is not possible for the soul’s gaze to be informed directly from the outer senses: acies is always informed from memoria. Hence, consciousness of a given content does not occur without the soul’s gaze being applied to and, thereby, informed from memory. Hence, conscious sense perception appears to require all three members of the series: namely, (i) actualization of the sense organs by species received from external objects, (ii) memory’s reception of this sensory information, and (iii) the immediate activation of inner gaze by the sensory information thus received by memory from the senses. Without the first component, namely, the actualization of the sense organs, there is no outer vision—no awareness of external bodies. Without the second component, namely, memory’s reception of information from the senses, there is no content available to inform the soul’s conscious gaze. And lacking the third component, there is no conscious awareness of information present to the senses. Hence, my positive thesis: conscious perception is a hybrid state requiring the tandem activation of the mechanisms involved in both outer and inner vision. If this is right, however, it turns out that perception is not itself any kind of trinity. After all, the ordered series that constitutes an occurrent perceptual state includes four species, two visions, and three applications of the will. Still, perceptual awareness is comprised from such trinities: namely, trinities of outer and inner vision. Indeed, taken together (or, rather, in various combinations), the mechanisms required for both outer vision, inner vision, and memory appear to supply the basic cognitive ingredients (or sub-structure) for a broad (perhaps even

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

70  - exhaustive) range of sensory cognitive states. In this regard, then, whatever the limits of the range of states included under inner and outer vision taken singly, the analysis of sense cognition offered in De Trin. 11, can be seen as having a fairly broad application.

5. Memory and the Trinitarian Mind I have been arguing that, taken singly, neither of two sensory trinities Augustine discusses in De Trin. 11 yields an analysis of sense perception. My argument rests both on the assumption that sense-perceptual awareness is, paradigmatically, conscious as well as on the contention that conscious awareness requires the activation of the acies animi. That such awareness requires the informing of the soul’s gaze is, I believe, beyond dispute: not only does Augustine explicitly identify conscious awareness with episodic acts of thinking (cogitatio), but also explicitly analyzes the latter in terms of the actualization of the soul’s acies. This is not surprising given Augustine’s commitment to the view that it is the soul (and not the body) that is the subject of conscious experience.³⁷ I have also argued that careful attention to the details of Augustine’s discussion in book 11 shows that the acies animi is activated upon receiving sensory information present in memoria. Thus, on my reading of Augustine, it turns out that perceptual awareness requires that sensory information be received in memory prior to its informing conscious

³⁷ Cf. De Trin. 11.2: “Nevertheless, it is the soul, conjoined to the body, that senses through a bodily instrument.” Augustine lays emphasis on the soul as the subject of sensory or bodily awareness in other contexts as well. For example, in Gen. Litt., he claims that it is “not the body that is the subject of sensation, but the soul through the body” (Gen. Litt. 3.7) and argues that “even bodily pain in any animate creature is itself a great and wonderful power of the soul” (Gen. Litt. 3.25). Part of what motivates Augustine’s insistence on this point is his commitment to the view that the soul exercises at least some agency in directing its awareness toward one thing rather than another. But if it is the soul that is the source of such agency, it must also be the subject of the awareness thus achieved. To be clear, however, to say that the soul is the subject of conscious experience does not entail any kind of transparency thesis. For example, just because the soul animates the body, it does not follow that it is thereby conscious of all bodily operations. Indeed, it is precisely Augustine’s rejection of such a transparency thesis that motivates his insistence on the limits of the soul’s gaze. Conscious awareness requires that a given content is present to the acies, not just that it is present somehow in the soul (e.g., in the animated sense organs or in memory).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

71

awareness. On the face of it, this result is perhaps somewhat surprising. For, while it is natural to suppose that memory is needed to recall a perceptual experience, it is far less obvious that memory should be a prerequisite for such experience. Thus, even if my reading is wellmotivated in the context of the details of Augustine’s discussion in book 11, it is harder to see what broader motivation there could be for assigning memory such a role in sense perception. Indeed, absent such motivation, one might be inclined to resist my interpretation on grounds of charity if nothing else. As it turns out, however, such motivation is not far to seek. I want to conclude, therefore, by briefly situating my reading of De Trin. 11 vis-à-vis some of Augustine’s wider philosophical and theological commitments. Doing so has the benefit of both highlighting the broader plausibility of Augustine’s views as well as providing further corroboration and support for my interpretation of book 11. To begin, it is worth bearing in mind that, for Augustine, the notion of memory is extremely broad. Memory is not strictly identified with a single psychological faculty or power, and its function goes well beyond the basic role of retaining and recollecting past experiences. In general, memory functions, for Augustine, as a kind of principle or source of human knowing and thinking. Thus, in some sense, memory encompasses everything we can mentally access or think about (or love, or seek)—even when we are not currently doing so.³⁸ That said, it remains true that memory’s function in sense-based cognition is largely that of gathering and holding information received in the external sense organs. In this capacity, however, it plays a key role in structuring conscious perceptual awareness.³⁹ In this respect, memory’s function in perception is continuous with the role it plays in mental activity in general: namely, as a principle or source of occurrent awareness. ³⁸ E.g., at De Trin. 15.40, Augustine sums up his account this way: “I have been attributing to memory all that we know even though we are not thinking about it . . . . But there are more hidden depths in our memory, where we found this thing even when we thought about it for the first time.” Cf. section 2 above. ³⁹ I do not claim to be particularly original in calling attention to memory’s role in perception. O’Daly, for example, has a lot to say on this score (Augustine’s Philosophy of Mind, chapter 3). Still, much more attention has been given to the role played by will in Augustine’s account of perception than the role played by memory.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

72  - To see what I have in mind, let us take a particular example—one of Augustine’s own—namely, one involving memory’s role in our (auditory) perception of language. Because spoken words are comprised of syllables consecutively articulated, it is clear, Augustine thinks, that our ability to perceive even a single word requires the retention of sensory input in memory. As Augustine says, in De Genesi ad litteram: “unless the spirit immediately formed in itself an image of the voice heard by the ears and stored it in memory, you would not know that the second syllable was second since the first, having vanished after striking the ear, no longer exists.” Thus, insofar as the sounding of a word is a temporally extended event, there needs to be some way to retain and structure sensory information in order to generate a unified perceptual experience. This is the role memory performs. Augustine makes this explicit when discussing the same example in De musica: “unless memory helps us when we hear even the shortest syllable . . . we cannot say that we have heard anything.”⁴⁰ The same will hold true, Augustine thinks, for other sense modalities as well. For each sense is directed upon complex, extended objects—the parts of which cannot be grasped all at once. Augustine points out, for example, that memory is likewise required for perception of shape: the sense of touch and vision, say, cannot always take in the whole of a three-dimensional object simultaneously (e.g., side to side, front and back). Hence, here too memory receives, retains and integrates visual representations of the shape (or feel) of corporeal bodies.⁴¹ Not only that, but insofar as perceptual awareness requires the integration of sensory input both within a single sense faculty, and also across distinct modalities—a point to which Augustine calls attention in book 2 of De libero arbitrio—it is natural to suppose that memory plays a role here as well.⁴² ⁴⁰ De musica 6.21. ⁴¹ Ibid. ⁴² De libero arbitrio 2.7. In this context, however, he speaks of the relevant faculty not as memory, but only as “a kind of inner sense to which everything is conveyed from the five familiar senses” and which is “neither sight nor hearing nor smell nor taste nor touch, but some other thing that presides over all of them.” Interestingly, Augustine assigns to this faculty the same sort of role he assigns to memory in other contexts, including that of facilitating conscious awareness of sensible objects. For more discussion of inner sense in De libero arbitrio and the way it fits within Augustine’s account of perception in other works see Charles Brittain, “NonRational Perception in the Stoics and Augustine,” Oxford Studies in Ancient Philosophy 22 (2002), 288ff.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

73

In general, then, Augustine is both explicit and consistent in defending—on philosophical grounds—the view that perception constitutively involves not merely the activation of the external sensory organs in receiving species of external bodies, but also the production, retention, and integration of sensory representations in memory.⁴³ He is equally explicit, moreover, that without this, perceptual awareness is not possible. Where there is a failure of uptake in memory, there is nothing to structure or inform conscious awareness—in which case, Augustine will say, there is no perception. This is the reason why it often appears to us that we have not heard people speaking in front of us when we were occupied by some other thought . . . .The sound undoubtedly reaches the ears . . . but . . . the impetus of that [sound’s] motion is immediately extinguished because of the soul’s interest being on something else. If the impetus remained, it would remain in the memory so that we would both find it and perceive that we had heard. (De musica 6.21)⁴⁴

What is required for auditory perception, thus, is that sensory representations are held in memory in order that the soul may “find” them there

⁴³ Because Augustine wants to defend a view according to which perception is efficiently caused by the soul (rather than by external bodies) he is wont to emphasize the soul’s active, ongoing role in generating sensory representations from information present in the senses. For example, in Gen. Litt. 12.33 he tells us that “although we first see a body which we had not seen before, and from that moment its image begins to be in our spirit, by which we can remember it when it is no longer there in front of us, still it is not the body that makes its own image in the spirit, but the spirit itself which makes it in itself with a wonderful swiftness that is infinitely removed from the sluggishness of the body. Thus, no sooner is the body seen by the eyes than its image is formed without the slightest interval of time in the spirit of the person seeing.” What Augustine’s discussion in the De musica passage quoted just above seems to suggest, however, is that the mere production of such images does not ensure their reception or retention in memory. ⁴⁴ The full text runs as follows: “Thus, unless memory helps us when we hear even the shortest syllable, so that the motion, which was created when the beginning sounded, remains in our soul during that moment of time, when no longer the beginning but the end of the syllable is sounding, we cannot say that we have heard anything. This is the reason why it often appears to us that we have not heard people speaking in front of us when we were occupied by some other thought, and this happens not because the soul does not produce those occurring rhythms at that moment—since the sound undoubtedly reaches the ears, and since the soul cannot be inactive during this reaction of its body, nor can it be moved in a different way than if this reaction did not take place—but because the impetus of that motion is immediately extinguished because of the soul’s interest being on something else. If the impetus remained, it would remain in the memory so that we would both find it and perceive that we had heard” (De musica 6.21).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

74  - and, thus, “perceive that we have heard.” It is in this sense, then, that memory functions as a kind of antecedent principle or source of perceptual awareness. To the extent that memory is needed to inform and structure the soul’s awareness of external, corporeal reality, there is good reason for Augustine’s insisting on memory’s functional priority— sensory content must be received in memory in order that we may be conscious of it.⁴⁵ There are, finally, significant theological considerations motivating Augustine’s view about both the priority and necessity of memory’s role in perceptual awareness. As I noted at the outset, on Augustine’s view, occurrent conscious awareness (be it an act of thinking involving non-rational, sensory content, or one involving purely intelligible content) manifests a trinitarian structure. The best way to see Augustine’s commitment to a trinitarian analysis of thought is to note that while he habitually relies on metaphors involving vision or sight to characterize acts of thinking, he also introduces and develops a second metaphor: namely, thought as inner word (verbum).⁴⁶ It turns out that conscious thinking can be regarded not only as a kind of inner vision or seeing, but also and equally as an inner word or kind of speaking. As Augustine explains: Thoughts are a kind of utterance of the heart . . . though just because we say thoughts are utterances of the heart does not mean they are not also acts of seeing (visiones), arising when they are true from vision of awareness (de visionibus notitiae). When these things happen

⁴⁵ It is worth noting that the claim that inner representations are involved in structuring perceptual awareness need not entail that such representations are the object of perceptual experience. Thus, I do not mean to be saddling Augustine with the sort of view that O’Daly attributes to him, namely, that, for Augustine, “sense perception is perception of incorporeal images of the objects perceived” (O’Daly, Augustine’s Philosophy of Mind, 106). It is perfectly compatible with the view I am advancing that the external body—not an inner likeness of it— serves as the object of perceptual awareness. For a defense of the view that external things are objects of perception, see Brittain, “Non-Rational Perception,” Silva, “Augustine on Active Perception,” and Gareth Matthews, “Knowledge and Illumination,” in D. Meconi and E. Stump (eds.), The Cambridge Companion to Augustine, 2nd ed. (Cambridge: Cambridge University Press, 2014), 171–85. ⁴⁶ The image of the verbum is first introduced in book 9 (at 9.12ff.) but Augustine discusses and develops it further in book 15 (at 15.17–26).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

75

outwardly through the body, speech is one thing, sight another; but when we think inwardly they are both one and the same. (De Trin. 15.18)

While Augustine is often willing to move freely between metaphors of inner seeing and speaking, the image of thought as the soul’s word (verbum) carries important theological resonance, especially in connection with the broader trinitarian aims that guide his project. For, understood as word, conscious thought serves as mind’s analogue to the second person of the Trinity—namely, God the Son, the divine Word spoken by God the Father (the first person of the Trinity).⁴⁷ Note, however, that in the human mind, it is memory that always functions as the analogue to the first person of the Trinity.⁴⁸ Thus, it is memory that must beget—i.e., be the source or principle of—acts of conscious thought. As Augustine tells us: All these things then that the human soul knows by perceiving them through itself or though the senses of the body or through the testimony of others, it holds onto where they are stacked away in the treasury of memory (thesauro memoriae condita). From them is begotten (gignitur) a true word when we utter what we know . . . .For it is then that the word (verbum) is most like the thing known, and most its image because the sight that belongs to thinking (visio cogitationis) springs direct from the sight that belongs to knowing (visione scientiae) . . . . (De Trin. 15.22)

It is precisely the fact that the soul’s gaze is formed from memory that secures the trinitarian image in general, and conscious thought as begotten word in particular. As Augustine says, “our word, the one that has ⁴⁷ Here, of course, Augustine is drawing on the Gospel of John (John 1:1) where Jesus is introduced as the Word of God and as the Word made flesh (John 1:14). See De Trin. 15.19–20. ⁴⁸ See, for example, De Trin. 15.40: “Certainly I have been trying as best I could to delineate God the Father and God the Son—that is, God the begetter, who in his Word co-eternal with himself somehow or other uttered all that he has substantially, and God this Word of his, who also substantially has neither more nor less than what is in him who begot him as a true and not a false Word—I have been trying to delineate all this, not as it might already be seen face-to-face, but as it might be seen by whatever kind of limited inference from this likeness in a puzzle, which we find in the memory and understanding of our mind. In this likeness I have been attributing to memory all that we know even if we are not thinking about it, and to intelligence in the proper sense a kind of formation of thought.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

76  - neither sound nor thought of sound . . . is at least something like that Word of God which is also God since this one [in us] is born of our knowledge just as that one was born of the Father’s.”⁴⁹ Here, then we have further—and, in this case, theological—motivation for the priority of memory in perceptual awareness. After all, if, as the foregoing passages suggest, the informing of the acies animi in perceptual awareness is itself a kind of inner word, then such awareness (qua analogue of God the Son) must have memory (qua analogue of God the Father) as its immediate source and principle. Awareness, in other words, must be produced by or “begotten from” memory. Not only that, but where memory serves as the source of conscious awareness, there is the requisite equality and substantial unity between that which begets (memoria) and what is begotten (informed acies): both are the soul itself.⁵⁰ In light of this, it is difficult to see how perceptual awareness could qualify as an image of the Word if it is not formed from memory. And yet it seems quite clear that Augustine means for such awareness to count as an image (or, at least a quasi-image) of the divine Verbum.⁵¹ Saint Louis University

Bibliography Armstrong, David. “The Nature of Mind,” in D. M. Armstrong (ed.), The Nature of Mind and Other Essays (Ithaca, NY: Cornell University Press, 1981), 1–15. ⁴⁹ De Trin. 15.24. ⁵⁰ This equality and unity is at the very heart of the image of the verbum. See, for example, De Trin. 15.19: “For when we utter something true, that is when we utter what we know, a word is necessarily born from the knowledge we hold in memory, a word which is absolutely the same kind of thing as the knowledge it was born from.” Indeed, it is for this reason (i.e., failure of substantial unity) that outer vision fails to generate a proper image of the trinity. ⁵¹ I am deeply grateful to Scott MacDonald and Charles Brittain, who generously included me in a series of workshops and seminars they organized around the De Trinitate starting in the early 2000s. It is thanks to them that I began thinking about De Trinitate as a whole and, later, about these issues in book 11 more specifically. I have learned an enormous amount from both of them over the years. Back in 2012, I gave a very early version of this paper at a SMRP memorial session honoring Gareth Matthews (to whom I am also indebted for much of what I know about Augustine) and, more recently, I presented it at the 2019 Cornell Summer Colloquium in Medieval Philosophy. I’m grateful to audiences on both occasions for their valuable feedback. Finally, thanks go to Therese Scarpelli Cory, Mary Sirridge, Jeff Brower, Brent Brower-Toland, and an anonymous referee for helpful comments on previous drafts of the paper.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  ’   

77

Aurelius Augustine. De Genesi ad litteram libri duodecim [= Gen. litt.], ed. J. Zycha (Vienna: F. Tempksy, 1894). Aurelius Augustine. De libero arbitrio, ed. W. M. Green (Turnhout: Brepols, 1956). Aurelius Augustine. De Trinitate libri quindecim [= De Trin.], ed. J. W. Mountain (Turnhout: Brepols, 1960). Aurelius Augustine. The Trinity, ed. J. E. Rotelle, tr. E. Hill (Hyde Park, NY: New City Press, 1991). Aurelius Augustine. On Free Choice of the Will, tr. T. Williams (Indianapolis: Hackett, 1993). Aurelius Augustine. De musica liber VI, ed. and tr. M. Jacobsson (Stockholm: Almqvist and Wiksell, 2002). Aurelius Augustine. The Literal Meaning of Genesis, ed. J. E. Rotelle, tr. E. Hill (Hyde Park, NY: New City Press, 2002). Brachtendorf, Johannes. The Structure of the Human Mind according to Augustine: Self-Reflection and Knowledge of God in De Trinitate, tr. A. Looney (Hamburg: Felix Meiner Verlag, 2000). Brittain, Charles. “Non-Rational Perception in the Stoics and Augustine,” Oxford Studies in Ancient Philosophy 22 (2002), 253–308. Clark, Mary. “De Trinitate,” in E. Stump and N. Kretzmann (eds.), The Cambridge Companion to Augustine (Cambridge: Cambridge University Press, 2001), 91–102. Gioia, Luigi. The Theological Epistemology of Augustine’s De Trinitate (Oxford: Oxford University Press, 2008). Hochschild, Paige E. Memory in Augustine’s Theological Anthropology (Oxford: Oxford University Press, 2012). Kalderon, Mark. “Trinitarian Perception,” Proceedings of the Aristotelian Society, supplementary volume 91 (2017), 21–41. King, Peter. “Augustine’s Trinitarian Examples,” Medioevo 37 (2012), 83–106. King, Peter. “Augustine on Knowledge,” in D. Meconi and E. Stump (eds.), The Cambridge Companion to Augustine, 2nd ed. (Cambridge: Cambridge University Press, 2014), 142–65. Kriegel, Uriah. “Consciousness and Self-Consciousness,” The Monist 87 (2004), 185–209.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

78  - Kriegel, Uriah. Subjective Consciousness (Oxford: Oxford University Press, 2009). Levine, Joseph. Purple Haze: The Puzzle of Consciousness (Oxford: Oxford University Press, 2001). MacDonald, Scott. “Augustine’s Cognitive Voluntarism in De Trinitate 11,” in E. Bermon and G. O’Daly (eds.), Le De Trinitate de saint Augustin: exégèse, logique et noétique (Paris: Vrin, 2012), 322–39. McGinn, Colin. “Consciousness and Content,” in N. Block, O. Flanagan, and G. Güzeldere (eds.), The Nature of Consciousness: Philosophical Debates (Cambridge, MA: MIT Press, 1997), 295–308. [Reprint of “Consciousness and Content,” Proceedings of the British Academy 76 (1988), 219–39.] Matthews, Gareth. “Knowledge and Illumination,” in D. Meconi and E. Stump (eds.), The Cambridge Companion to Augustine, 2nd ed. (Cambridge: Cambridge University Press, 2014), 171–85. Mourant, John. Saint Augustine on Memory (Villanova, PA: Villanova University Press, 1980). O’Daly, Gerard. Augustine’s Philosophy of Mind (Berkeley: University of California Press, 1987). Silva, José Filipe. “Augustine on Active Perception,” in J. Silva and M. Yrjönsuuri (eds.), Active Perception in the History of Philosophy (Dordrecht: Springer, 2017), 79–98. Silva, José Filipe. “Perceptiveness,” Proceedings of the Aristotelian Society, supplementary volume 91 (2017), 43–61. Teske, Roland. “Augustine’s Philosophy of Memory,” in E. Stump and N. Kretzmann (eds.), The Cambridge Companion to Augustine (Cambridge: Cambridge University Press, 2001), 148–58. Tournau, Christopher. “The Background of Augustine’s Triadic Epistemology in De Trinitate 11–15,” in E. Bermon and G. O’Daly (eds.), Le De Trinitate de saint Augustin: exégèse, logique et noétique (Paris: Vrin, 2012), 251–66.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

The Justice and Mercy of God Seneca in the Eleventh Century Mary Sirridge

In chapter 9 of his Proslogion,¹ written at the monastery of Bec in 1077–8, Anselm asks of God: But how do you spare the wicked if you are totally just and supremely just? For does the one who is totally and completely just do something that is not just? And what sort of justice is it to give everlasting life to someone who merits eternal death? (Pros. ch. 9: I: 108, 86)

Anselm’s preliminary answer in this chapter is that God is merciful to the wicked and spares some of them the loss of salvation—“eternal death”— because he is just (Pros. ch. 9: I: 108, 86). Nevertheless, he continues, it is also just for God to punish the wicked (Pros. ch. 10: I: 108, 87). Anselm concludes his discussion with an admission that when faced with a final question reason must admit defeat: But even if one can somehow grasp why you can will to save the wicked, certainly no reasoning can comprehend why, from those who are similarly wicked, you save some rather than others through your supreme goodness and condemn some rather than others through your supreme justice. (Pros. ch. 11: I: 109, 88)

¹ Anselm of Canterbury, Proslogion in Sancti Anselmi Catuarensis Archiepiscopi Opera Omnia, 6 vols., ed. F. S. Schmitt (Edinburgh: Nelson, 1946; repr. Stuttgart: Frommann Verlag, 1968), vol. 1, 93–122. References to Anselm’s works use the following abbreviations: Proslogion, Pros.; De Concordia Praescientiae et Praedestinationis et Gratiae Dei, De Concordia. English translations are those of Thomas Williams, in Anselm, Basic Writings (Indianapolis: Hackett Publishing, 2007), adapted. References to Anselm’s works are followed by an indication of volume and page in Schmitt’s edition and the page in Williams’s translation. Mary Sirridge, The Justice and Mercy of God: Seneca in the Eleventh Century In: Oxford Studies in Medieval Philosophy Volume 8. Edited by: Robert Pasnau, Oxford University Press (2020). © Mary Sirridge. DOI: 10.1093/oso/9780198865728.003.0003

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

80   It is tempting to think that Anselm ought to have admitted the defeat of reason far earlier in his investigation, that his position comes down to the view that both divine mercy extended to the wicked and divine punishment imposed upon the wicked are of necessity just simply because whatever God wills is just. Indeed, in his appeal to the Psalmist he seems to say just this: Truly therefore all the ways of the Lord are mercy and truth,² and the Lord is just in all his ways.³ And indeed there is no inconsistency here, for it is not just for those to be saved whom you will to punish and it is not just for those to be damned whom you wish to save. For only what you will is just, and only what you do not will is not just. (Pros. ch. 11: I: 109, 88)

Anselm’s very compressed account of the justice and mercy of God is, I think, more complex than this; and the pattern of his reasoning becomes clearer if we look to Seneca’s De Clementia in which the firstcentury Roman Stoic philosopher argues for a policy of imperial clemency.⁴ Anselm is likely to have been familiar with Seneca’s De Clementia, and there are significant similarities between Anselm’s reasoning about divine justice and mercy and Seneca’s arguments in favor of imperial clemency. The most striking of these is that both Anselm and Seneca present the remission of penalties as a prerogative of someone with great power and authority that constitutes the highest form of justice.

² Psalm 25:10: “All the ways of the Lord are mercy and truth.” ³ Psalm 145:17: “The Lord is just in all his ways.” ⁴ L. Annaeus Seneca, De Clementia, translated by Susanna Braund with Latin and English text on facing pages, accompanied by an introduction and commentary (Oxford: Oxford University Press, 2009, 2011). Braund’s Latin text is taken from Ermanno Malaspina, L. Annaei Senecae: “De Clementia” Libri Duo (Turin: Edizione del Orso, 2001, 2004). Malaspina’s edition of the text is preceded by introductory material in which he presents a reconstruction of the history of textual transmission. The text of De Clementia is cited as DC. References to De Clementia are by book, chapter, and section. Translations from DC are those of Braund, adapted. Material from Braund’s introduction and commentary is identified by her name and location in the edition. I will consistently use “clemency” to translate the Latin “clementia”.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

81

1. Seneca on Imperial Clemency Seneca would have found the vocabulary he needed and the basis for arguments in favor of clemency in very well-known works by Cicero, who had presented clemency as a personal virtue of “someone with the highest power.”⁵ In his Caesarian Orations, particularly in Pro Marcello, and Pro Ligario,⁶ Cicero attached the expression “clemency” to Caesar’s established policy, once victorious, of not taking revenge on those who opposed him, fought against him, or insulted him—ostensibly on the grounds that his former opponents were honorable and had been loyal to the cause for which they fought or to the leader whom they followed. In Pro Marcello, Cicero, sincerely or not, praises Caesar’s clemency as a particular virtue of his character: It is impossible to pass over in silence such humanity, such exceptional, indeed unheard-of clemency, such invariable moderation, exhibited by one who has attained supreme power, such incredible and almost divine wisdom. (Pro Marcello I. 1–2)

The opposite of clemency is identified in Pro Ligario as “cruelty” (Pro Ligario III. 9–100). And in his De Officiis Cicero advises rulers that popular support depends upon conspicuous “beneficence,” which is even more important than perceived justice and prudence; political power, and the ruler’s very safety, are undermined by harshness and cruelty (De Officiis II. vii. 24; ix. 31–4).⁷ Seneca moves well beyond Cicero, however. Unlike Cicero, Seneca will carefully distinguish clemency from pity and compassion. And conspicuously absent from Cicero’s works is the characterization of clemency as

⁵ Although the term “clemency” was not new as a designation of actions of leniency or mercy to conquered populations, Cicero’s innovation was to establish the official ideology which attached clementia to Caesar as a particular virtue of character. See Stefan Weinstock, Divus Julius (Oxford: Oxford University Press, 1972), esp. 237–9; Braund’s introduction (Braund, introduction, 27–44); David Konstan, “Clemency as a Virtue,” Classical Philology 100 (2005), 337–46; David Konstan, Pity Transformed (London: Duckworth, 2001), 95–104. ⁶ Caesarian Orations, ed. and tr. N. H. Watts (Cambridge, MA: Harvard University Press, rev, and repr. 1953). Pro Marcello, 422–51; Pro Ligario, 452–93. The translations are those of Watts, adapted. ⁷ Cicero, De Officiis, ed. and tr. W. Miller (London: Harvard University Press, 1913).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

82   the highest form of justice, which is the most distinctive element of Seneca’s approach. The emphasis is instead on Caesar’s generosity and leniency.⁸ Seneca’s own presentation of his theory of clemency is complicated by the fact that De Clementia is officially devoted to urging the young Nero to manifest and persevere in his natural clemency. Particularly in Book I, Seneca’s strategy is to motivate Nero by “holding a mirror up to him,” so that he will be moved to the further development of his virtuous clemency by admiration of his own natural tendency to exercise this virtue,⁹ in contrast to the awful spectacle presented by Gaius Caesar,¹⁰ and even the younger Augustus. Thus the bulk of Seneca’s arguments for the indispensability of clemency are framed in terms of its role in fostering the ideal relationship between an absolute ruler and the citizenry, and a number of them have to do with what is politically expedient. It is not until Book II that Seneca focuses on giving a precise philosophical account of the virtue of clemency and distinguishes clemency from the closely related phenomena that are in fact vices: pity or compassion and pardoning.¹¹

⁸ Cicero’s understanding of clemency seems substantially to fit Braund’s account (Braund, introduction, 40–3), which is that in the transition from republic to empire, imperial decisionmaking power came to be understood to be above or beyond the law when it came into play and substantially displaced or superseded pre-imperial considerations of justice. ⁹ Seneca thus imputes to Nero a tendency to exercise clemency that he already did not conspicuously possess. Tacitus, Annales, ed. and tr. J. Jackson (Cambridge, MA: Harvard University Press, 1937), 13. 2. 1, describes an escalating climate of violence and treachery in the social milieu that revolved around Nero: “Things were headed toward slaughter, had not Seneca and Burrus stood in the way.” The preceding paragraphs describe a trail of murder and treachery, as various people jockey to secure Nero’s favor by removing those who pose obstacles to their plans, or simply exploit the situation by seizing the opportunity to pay off old scores. ¹⁰ The cruelty and savagery which are described in general terms in DC are specifically attributed to Gaius Caesar, among others in Seneca’s De Ira, Dialogorum Libri Duodecim, ed. L. D. Reynolds (Oxford: Oxford University Press, 1977), 39–128; tr. J. Basore, Moral Essays (London: Heinemann, 1928), vol. 1, 106–355. See especially De Ira II.33.3–6; III.18.2–19.5. Braund (introduction, 70–3) argues that De Clementia and De Ira are complementary, since De Ira, which is devoted to warning about the bad effects of anger, hypothesizes that the unchecked anger of rulers escalates to savagery and cruelty, opposites of clemency. The brief discussion in De Clementia of anger, cruelty, savagery, and brutality parallels Seneca’s more extensive accounts of these phenomena in De Ira. ¹¹ For a detailed analysis of Seneca’s rhetorical strategy in DC I. 1.1–I. 1.9, see Braund’s commentary (153–82); Miriam Griffin, “Seneca’s Pedagogic Strategy: Letters and De Beneficiis,” in R. Sorabji and R. W. Sharples (eds.), Greek and Roman Philosophy 100 BC–200 AD (London: Institute of Classical Studies, 2007), 89–113.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

83

Since the first book of De Clementia is focused on imperial power, and since instances of clemency would be a matter of common knowledge, Seneca can afford to give a very brief preliminary account of clemency before moving on to his main objective, arguing for the importance of clemency in establishing and sustaining ideal imperial power: What is needed, consequently, is an exercise of a moderation that knows how to distinguish the curable from the hopeless characters . . . We should hold to the measured response—but because a perfect balance is difficult, if there is to be anything beyond parity, it should verge towards the more humane. (DC I. 2. 2)

This very compressed account presents several challenges. The initial characterization must, it seems, mean that extending clemency is to be understood as a regulation of disciplinary responses that is guided by a reasoned assessment of the prospects for improvement. Seneca’s “beyond parity” is slightly puzzling; but given that clemency is a virtue and aims at a “perfect balance,”¹² it cannot be read as a recommendation to depart from justice. Seneca must mean, it seems, that especially with respect to the particular circumstances of the case, the ideal for clemency is to strike the finest balance between offense and penalty,¹³ which includes a tendency toward the “more humane” of a range of legitimate responses, i.e., toward a response that manifests as positive an orientation toward the good of human beings as is consistent with the situation:

¹² For the connection between the Stoic “perfect balance” and virtue, see. Letter 85. 10, L. Annaeus Seneca, Ad Lucilium Epistulae Morales, vol. 2, ed. and tr. Richard Gummere (London: Heinemann, 1920), 290. Brad Inwood, in Seneca, Selected Philosophical Letters (Oxford: Oxford University Press, 2007), observes in connection with Letter 85 that this use of “perfect balance” (temperamentum) in moral psychology comes from medicine, where it is used to refer to a proper balance of bodily humors; its meaning then spreads to a broader use: “an optimal balance of consistent and harmonious beliefs which prevents outbreaks of the passions,” 226–7. ¹³ See Sarah Byers, “The Psychology of Compassion: A Reading of City of God 9.5,” in J. Wetzel (ed.), The Cambridge Critical Guide to The City of God (Cambridge: Cambridge University Press, 2012), 130–48, for a similar interpretation of Seneca’s notion of clemency. Seneca seems to have very little interest in locating the virtue of clemency as a mean between extremes, though he will eventually identify cruelty as its opposite; his interest is fixed instead on the proportionality between offense and penalty.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

84   “No virtue suits a human being more, since no virtue is more humane” (DC I. 3. 2).¹⁴ It follows immediately from the fact that clemency is a matter of striking the proper adjustment of penalty to offense that it would not in principle be fitting to grant clemency indiscriminately—and certainly not as a commonplace. An additional reason against indiscriminate or universal pardon, Seneca says, is that “when the distinction between the bad and the good is removed, the result is confusion and an outbreak of vicious behavior.” To pardon everyone, he says is as cruel as to pardon no one (DC I. 2. 2).¹⁵ Seneca does not offer a precise account of the virtue of clemency until the beginning of Book II. Like all the virtues, clemency is for Seneca a quality of the mind that inclines someone who possesses it to a particular sort of action. Clemency is defined as “restraint of the mind when it has the power to take revenge,” or “leniency of someone in a higher position to someone lower down with respect to imposing penalties,” “a drawing back from what could deservedly be imposed” (DC II. 3. 1). More important than a definition for a clear understanding of clemency is delineating its relationship to other moral phenomena. The opposite of clemency is not, Seneca says, “severity,” which is itself a virtue, but cruelty. Seneca does not further define severity; but it must, it seems, be the virtue of moderating disciplinary responses by a discernment that leads to the imposing of the heavier of legitimate penalties, again on the basis of a reasoned assessment of the facts about the person being punished or the situation. Cruelty, the true opposite of clemency, is

¹⁴ Seneca goes on to reason that clemency will be understood to be the most humane virtue both by Stoics, on the grounds that it manifests and furthers the right sort of positive social orientation toward other human beings, and by Epicureans, since this virtue is the most conducive to peace and leisure. Thus “humane” here has the force of “proper to human nature” or “in line with the highest human good.” ¹⁵ “Cruelty” will receive a technical definition in Book II; DC II. 4. 1: “barbarousness of the mind in exacting punishment.” It cannot, it seems, be the case that pardoning everyone counts as cruelty in this technical sense. Possibly Seneca is sacrificing precision for effect; or, as Braund (commentary, 190) proposes, he means to call attention to the need for a “visible” scale in the dispensation of clemency, so that blanket decisions in either direction are equally useless. Seneca may also be blurring distinctions that his philosophically unsophisticated addressee cannot yet appreciate, as he appears to do in other works. See Griffin, “Seneca’s Pedagogic Strategy,” 112. In Book II “pardon” has a technical meaning, the remission of due penalties, and thus does not count as clemency.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

85

clearly defined in De Clementia II as “the mind’s savagery in exacting penalties” (DC II. 4. 1–3). In this sense the cruel are those who have “an inclination of the mind to excessive harshness,” who “have reason to punish, but do not do so in a measured way” (DC II. 4. 2).¹⁶ Clemency is not to be confused with pity or compassion, which is a vice so close to clemency that it is easy to confuse the two, just as it is easy to confuse cruelty and severity. Pity or compassion is a vice. It is a “sickness of mind which succumbs to the sight of others’ misery” (DC II. 5. 4). No such sickness can or should affect the wise man, who will not feel pity or compassion, but will nonetheless help even those who merit disapproval and correction, and still more readily those who are truly afflicted and struggling mightily (DC II. 6. 3). The sage will willingly and in a generous spirit go to the assistance of other people who sorrow, but without commiserating with them. Clemency is also explicitly to be distinguished from pardoning, which is a matter of not punishing someone, despite one’s reasoned judgment that the person ought to be punished (DC II. 7. 3). Pardoning in this sense is unjust, which clemency never is. Letting someone go free because he was led astray or drunk, letting enemies go unpunished if they fought from honorable motives, and the like, are exercises of clemency, not pardon. According to Seneca, extending clemency is an exercise of “freedom of judgment.” Someone who exercises this freedom, an “arbiter,” is not understood by Seneca as someone upon whose decisions there is no constraint, but as a particular kind of judge who has the skill and the authority to form judgments and reach settlements on the basis of what is right and also good, in a way that goes beyond the letter of the law. In De Beneficiis¹⁷ Seneca once again refers some very important decisions to an arbiter, this time judgments about whether an appropriate response

¹⁶ There are other behaviors that we consider cruel, he says, but we are concerned here with the context of imposing punishments. Seneca’s example of cruelty in this sense is Phalaris, whose savagery was directed toward the guilty, not the innocent, but “went beyond the bounds of humanity and reasonable measure” (DC II. 4. 2). ¹⁷ Seneca, De Beneficiis, in Moral Essays, vol. 3, and in M. Griffin and B. Inwood, On Benefits (Chicago: University of Chicago Press, 2011), hereafter Ben.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

86   has been made to receiving a benefit. Such decisions should not be a matter for courts to decide, he says, but not because there is no offense to judge in cases of ingratitude. Rather the calculations are so complex and subtle that no ordinary judge can make such a decision, since no law or combination of laws is adequately precise; and applying the laws is the office of the ordinary judge (Ben. III. 7. 7–8). In cases involving gratitude for benefits received and judgments about exercise of clemency—or severity—the decision of an arbiter is called for: “Clemency does not hand out less than justice demands, but what is most just” (DC II. 7. 3).¹⁸ Much of De Clementia is devoted to showing that imperial clemency, and avoidance of its opposite, imperial cruelty, are essential to sustaining the ideal relationship between the emperor and the citizenry. Clemency is universally prized and admired, Seneca says. This is because everyone can have hopes for clemency, which is not, like other blessings, dispensed in a way proportionate to one’s desert, since like receiving a benefit, clemency is undeserved (DC I. 1. 9). Moreover no one can be so confident of his own guiltlessness as not to rejoice at the general prospect of clemency for human errors (DC I. 1. 9); and no one can be sure of never needing clemency because a twist of fate is sometimes as ruinous as guilt and because something praiseworthy can well end up being subject to punishment (DC I. 2. 1). Many of Seneca’s arguments for the importance of imperial clemency are based on political expediency. The conspicuous exercise of imperial clemency supports the popular perception of a relationship of mutual support and dependence between the emperor and his subjects by generating a popular perception of the emperor as ceaselessly acting with an eye to the safety and wellbeing of each and every one of his subjects (DC I. 3. 3). It supports the perception of the emperor as possessed of the greatness of spirit that befits the greatness of his position, which is what gives him the ability to look with equanimity upon insults and injuries to himself (DC I. 5. 5–6). And his subjects’ loyalty and commitment to the emperor contribute in turn to his security: ¹⁸ On “freedom of judgment” see Braund (commentary, 420); Brad Inwood, Reading Seneca (Oxford: Oxford University Press, 2005), 212–17.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

87

“Clemency renders rulers not just more noble, but safer. Clemency is not only the most splendid ornament of imperial rule, but at the same time its greatest source of security” (DC I. 11. 4).¹⁹ De Clementia also offers arguments against imposing frequent, harsh punishments. The emperor has standing too great to require compensation for injuries to himself, and his power is too great for him to need to look for confirmation of its extent by imposing harsh punishments (DC I. 11. 4). As for injuries to others, the important consideration in imposing penalties is the prospect of improvement: improvement in the person to be punished if possible, and if not, betterment of the rest of the community by punishing the guilty—or in extreme cases improving things for the rest of the community by removing the guilty person. The same considerations apply when extending clemency. Clemency for the very wicked is sometimes advisable, Seneca says, because people are more likely to improve if they still have something left to lose (DC I. 22. 1). By allowing himself to be drawn into harshness, the emperor in fact lessens his power. Whereas clemency attests to the emperor’s greatness of soul, forbearance, and good judgment, from which everyone draws confidence and hope, in contrast the failure to extend clemency makes people reckless and desperate, even if the punishments are deserved. “Frequent punishment crushes the hatred of a few, but it rouses the hatred of everyone” (DC I. 8. 6). The worst consequence of imperial cruelty is that an emperor who is harsh and cruel effectively reduces his own freedom and puts himself in the power of others: “The worst thing of all about cruelty is that it has to be continued, and there is no way to return to a better course of action; crimes have to be protected by more crimes” (DC I. 13. 2). There is also a strain of argumentation in De Clementia that is interwoven with such pragmatic political concerns, but is substantially independent of them. Seneca stresses repeatedly the degree to which clemency befits a king or emperor. “The fact is that great strength is

¹⁹ For Seneca ‘nobility’ and the human good are coextensive, though not identical, since it is by achieving nobility of character that someone attains the human good. See Inwood, Reading Seneca, 283–4.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

88   admirable and glorious only if its power is beneficent” (DC I. 3. 3). “Greatness of spirit befits greatness of standing” (DC I. 5. 5). To refrain from punishing to the extent to which punishment could be imposed is a prerogative reserved solely to someone with great power and great majesty (DC I. 5. 6).²⁰ The emperor’s power and his elevated position impose a constraint which the emperor shares with the gods: It is a slavery of the greatest magnitude, not to be able to become less. The emperor and the gods alike are bound by this necessity. It is impossible for the gods to sink below their great position, and it is as unfitting as it is unsafe for the emperor to do so. (DC I. 5. 6)

2. Anselm: The Justice and the Mercy of God In Proslogion, Anselm has a definite overall agenda: to deduce the existence and attributes of the deity from the formula: God is “something than which nothing greater can be thought” (Pros. ch. 1: I: 101, 81). From this formula Anselm first demonstrates the existence of God and his supreme goodness, from which it follows that God must be just, truthful, and blessed. In subsequent chapters Anselm further argues that God must be percipient, omnipotent, merciful,²¹ and impassible because if he were not, he would be less than something than which nothing greater can be thought. In Proslogion 5–11 Anselm sets out to deal with challenges to his agenda. In some cases, e.g., divine omnipotence and divine mercy, it is difficult to explain how a divine being of the sort Anselm describes can possess the attribute at all, whereas in the case of divine mercy and divine justice the problem is that the two attributes seem inconsistent with each other: If, as Anselm seems to assume, justice is a matter of giving people their due, then where is the space for mercy, which seems precisely to be a matter of not giving someone what is due?²² ²⁰ See also DC I. 19. 1. ²¹ In the discussion of Anselm the Latin ‘misericors/misericordia’ will be translated ‘merciful/ mercy.’ ²² Anselm is interested here specifically in divine justice and divine mercy. By the early twelfth century, however, the relationship between mercy and the virtue of justice and the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

89

Divine omnipotence appears to be incompatible with statements that are undoubtedly true, e.g., “God cannot lie” and “God cannot make something true be false.” Anselm’s answer to this objection is that divine omnipotence is completely compatible with the truth of such statements. It is precisely from God’s omnipotence that these “limitations” flow, since someone’s being able to do something that is not beneficial to himself and that he ought not to do comes not from power, but from weakness; and to the extent that he can do them, misfortune and wickedness (literally “adversity and perversity”) have power over him, and he lacks power over them. Someone’s “having the power” to do such things in fact constitutes not power, but a lack of power. He is not said to be able to do such things because he has the power to do them, but because his lack of power gives something else power over him. (Pros. ch. 7: I: 105, 84)

We often speak loosely in this way, Anselm adds, for example when we say that someone who is being silent or sitting is doing something, when the person in fact is not doing something, i.e., talking or moving about.²³ The mercy of God raises the question of how God can be merciful if he is at the same time impassible, which he must be, since it is better not to be subject to feelings than to be susceptible to them. But if God is impassible, he cannot feel sorrow, hence cannot have compassion for suffering, which is precisely what it is to be merciful (Pros. ch. 8: I: 106, 85). Anselm’s answer is that God is “merciful with respect to us” in the sense that he rescues the wretched and spares those who sin against him, so that we feel the effect of someone being merciful. But as he is in

proper understanding of justice itself has a well-established broader context in moral philosophy, pastoral direction, and scriptural exegesis; and it is possible that scholastic reflections on the problem carry over into jurisprudential thinking. See Philippa Byrne, Justice and Mercy: Moral Theology and the Exercise of Law in Twelfth-Century England (Manchester: Manchester University Press, 2019), 16–86. ²³ Anselm’s main point does not, however, seem to be that this is a simple case of misleading surface grammar. Choosing to lie, like choosing to remain sitting or silent, is a definite choice, not a non-choice; and as a choice of something other than the good, it is not open to God. Being susceptible to making such choices would in fact make God less than omnipotent and less than the being than which no greater can be thought.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

90   himself, “with respect to yourself,” Anselm says, God is not merciful because for God there is no affect, no feeling—God feels no compassion (Pros. ch. 8: I: 106, 85). God is thus both merciful and not merciful. Setting aside for the moment the question of whether Anselm’s proposed solution to the problem of divine mercy is not unnecessarily paradoxical, this sparing of sinners by God in which the mercy of God is supposed to consist does raise the larger question with which we began: How does God in his mercy spare the wicked and those who sin against him if he is completely and supremely just, which Anselm seems to understand as giving people their due?²⁴ It would seem that sparing the wicked, giving everlasting life to someone who deserves eternal death, is doing something unjust (Pros. ch. 9: I: 106–7, 85). Anselm’s initial answer does not get us very far: God is beneficent, even to the wicked, because he is totally and supremely good: After all, you would be less good if you were not beneficent to any of the wicked. For someone who is good both to the good and to the wicked is better than someone who is good only to the good, and one who is good to the wicked both by punishing and sparing them is better than one who is good only in punishing them. Therefore you are merciful because you are totally and supremely good. (Pros. ch. 9: I: 107, 86)

This answer just seems to make the problem more obvious. Admittedly, handing out good to the good and to the wicked alike will result in more good being handed out. But this just pushes the difficulty back a step: in what non-trivial sense is God supremely good on account of being good to some of the wicked instead of punishing them? Doing good by doing good where good should not be done seems not really to be doing good at all and seems therefore not to flow from God’s supreme goodness. As Anselm himself says, it remains thoroughly mysterious at this point why God, “who is wholly just, and in need of nothing, gives good things to the wicked and guilty” (Pros. ch. 9: I: 107, 86).

²⁴ Pros. ch. 9: “True, out of goodness you repay the good with good and the evil with evil, but the very nature of justice seems to demand this” (I. 107: 86).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

91

The subsequent prayer makes matters worse: You save the just in accord with justice, you free those whom justice condemns. The just you save with the help of their merits, sinners despite their merits, recognizing in the just the good things you have given them, in the wicked overlooking the evils you hate. O immeasurable Goodness that thus surpasses all understanding.²⁵ Let the mercy that proceeds from your great riches come upon me. Let that which flows forth from you flow over me. Spare by your clemency, lest you exact retribution by your justice. (Pros. ch. 9: I: 107–8, 86)²⁶

In this picture, divine justice and divine mercy seem to be operating independently—almost to be in competition. Apparently God produces a certain amount of good by justly saving the just (at this point the distinction between the good and the wicked seems to be clear-cut), and then sacrifices some justice in order to produce more good by sparing some of the wicked. Anselm needs to show not just that the mercy of God can co-exist with his justice, but that God is merciful precisely because he is supremely just. As Anselm himself puts it: Help me, just and merciful God, whose light I seek. Help me to understand what I am saying. You are indeed merciful because you are just. (Pros. ch. 9: I: 108, 86)

Anselm’s solution involves a reversal of the strategy he used in his earlier explanation of divine mercy. There he argued that there is divine mercy “with respect to us,” since we feel the effects of God’s activity of sparing and preserving us; as God is in himself, however, there is, if we are being precise, no mercy, since there is no feeling of compassion for human sorrow and sin. With respect to justice, he now distinguishes between divine justice with respect to human merit and God’s justice with respect

²⁵ Philippians 4:7: “And the peace of God that surpasses all understanding will guard your hearts and your minds in Christ Jesus.” ²⁶ This is the only occurrence of ‘clemency’ in Pros. 5–11.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

92   to himself. God is just with respect to us and gives us our due when he exacts retribution and punishes and spares according to our merit. But God does what befits himself as the being than which no better and more powerful can be thought when he spares the wicked. Divine mercy flows from God’s justice—as it applies to himself—since it befits a being than which no greater can be thought to give good things in cases in which it might justly condemn and punish: . . . because it is just for you to be so good that you cannot be understood to be better and to act so powerfully that you cannot be thought to act more powerfully. For what could be more just than this? This would not be true if you were good only by exacting retribution and not sparing, or if you made only those who are not-good to be good, and not the wicked as well. (Pros. ch. 9: I: 108, 87)

There are several developments in this answer. Anselm now makes it fully explicit that the good God does for human beings is to make them good, i.e., to free them from sin and save them. Moreover, in this response Anselm has gone from speaking about the good versus the wicked to speaking about the not (completely) good versus the wicked. From God’s exalted point of view, none of us deserves saving or being made good on our own merits; God could in justice save no one. On the basis of our human desert we might all justly be abandoned. Still it seems that there is a real difference with respect to desert between the wicked and the merely not-good, who are apparently those Anselm referred to earlier as “already having been given good things by God,” having been offered the grace of salvation and accepted it. The not-good are to be justly spared and made good; and divine mercy, which flows from God’s justice in relation to himself, demands that some of the wicked be made good as well, and not punished as they deserve. It looks problematic, though, that this solution seems to separate the punishment of the remaining wicked from what befits God, from what God owes himself as the being than which no better and more powerful can be thought. If God renders to himself what is his due, when he spares not only the not-good, but some of the wicked as well, then why not spare everyone, the not-good and all the wicked alike, thereby doing even

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

93

more good? Anselm’s answer is that God punishes some of the wicked in accord with their desert—which is just—and saves the rest of the wicked out of his great goodness and power and thus out of what befits him as supremely good and supremely powerful; it does not befit the divine omnipotence and divine freedom to be constrained by human desert. Thus due and deserved punishments are sometimes imposed, but also sometimes tempered or remitted on account of a mercy that flows from justice—because it does justice to and befits God as supremely good and supremely powerful to save some who do not deserve to be saved. By sparing the wicked, you are just with respect to yourself, and not with respect to us, just as you are merciful with respect to us, and not with respect to yourself. For just as you are merciful by saving us, whom you might justly abandon, not because you experience any affect, but because we experience the effect, so you are just, not because you give us our due, but because you do what is fitting for you who are supremely good. In this way, then, you justly punish and justly spare without any inconsistency. (Pros. ch. 10: I: 109, 87)

Thus “in some sort of way,” Anselm says, we can “perhaps” understand how God as supremely just can will good for the wicked and spare them: God is always, even when he is merciful to the wicked, giving someone or other his due, in the case of sparing the wicked, Himself. There is a remaining unclarity, however, which perhaps explains why Anselm is reserved about his achievement: What determines which of the wicked God in his justice to himself will save and which of them will be condemned as a just response to their wickedness? One possibility is that although there is now no absolute contrast between the good and the wicked, and even if no saving grace is deserved, Anselm is thinking that there is ordinarily a connection between human choices and divine assistance and response. He seems to recognize a real difference between the not-good and the wicked, since he never seems to doubt that, liable to abandonment as all human beings are without the support of divine aid, the not-good will be spared and saved. Anselm clearly thinks that it is in accord with their desert that God justly does not save some of the wicked. Perhaps, then, there are degrees of wickedness

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

94   from God’s point of view, some of the wicked so entrenched in their resistance to divine grace that God does not owe it to his magnitude and power to save them—though he could in his great goodness do so.²⁷ Surely by the time of Anselm’s later On the Harmony of God’s Foreknowledge, Predestination, and Grace with Free Choice, in which he readily refers his audience to his early work, he is clearly committed to a correspondence between human choices and divine response. Even if human rectitude is a result of receiving divine grace to start with, and cannot be acquired or preserved, save by the support of divine grace and mercy, no one acquires rectitude except by choosing to accept divine grace or preserves rectitude except by willing it (De Concordia III. 4: II: 267–8, 378). This choice to preserve rectitude in turn requires the continued support of divine grace, but that divine gift will not be withdrawn unless free choice abandons its rectitude and wills something else (De Concordia III. 4: II: 268, 378). Moreover, temptation will never be allowed to overwhelm the human will, which if it wills rectitude will receive the grace to preserve it (De Concordia III. 4: II: 268, 378). Thus the normal pattern of divine response to human merit is the further support by divine grace to the extent that human beings choose to accept saving grace and struggle to maintain rectitude. Always keeping in mind that no one deserves God’s initial offer of rectitude and that preserving rectitude requires divine grace, God in his justice gives the not-good their due when he assists and saves them. The final difficulty posed by Anselm, which he has pronounced incomprehensible for reason, shows that even in the case of the wicked who are abandoned according to their desert he is thinking, at least generally, of degrees of moral rectitude and the lack of it: And perhaps this is why the one who is supremely just can will good things for the wicked. But even if one can somehow grasp why you can

²⁷ If, as seems probable, Anselm is familiar with Seneca’s De Beneficiis, he may well be thinking about divine assistance along the lines of Seneca’s “benefits,” gifts and good offices, which, though not owed to anyone, are nonetheless normally to be apportioned very precisely according to complex guidelines. One relevant consideration is whether the intended recipient can be expected to recognize having received the benefit, accept it and reciprocate.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

95

will to save the wicked, certainly no reasoning can comprehend why from those who are alike in wickedness, you save some rather than others through your supreme goodness and condemn some rather than others through your supreme justice. (Pros. ch. 11: I: 109, 88)

This is a question that should not arise if degrees of human wickedness were wholly irrelevant to the dispensation of divine mercy, since otherwise there is a simple solution to the difficulty: God saves whomever he wills to save, and everything God wills is just, and so he can justly will to abandon and punish someone who deserves to be abandoned and punished—or not—regardless of whether he saves someone else equally wicked.

3. A Common Template? Anselm’s usual practice is to adapt his source material to the problem at hand and integrate it into his own agenda sometimes even subverting the intent of his sources.²⁸ Given the fact that Anselm’s and Seneca’s agendas are very different, we can expect certain differences between Seneca’s reasoning in De Clementia and Anselm’s approach to divine power, justice, and mercy. The will of Anselm’s God is immutable and is fixedly attached to the good, unlike the choices of a worldly emperor who can make less than optimal choices and be urged to follow this or that policy. Another very important difference between the Roman emperor and Anselm’s God is that, unlike Anselm’s God, the emperor is not omnipotent, even if he is the supreme political power. Thus Seneca’s arguments in favor of imperial clemency that are based on political expediency, being rooted in the fact that the emperor’s power is not unlimited, will

²⁸ See Thomas Williams, “Anselm’s Quiet Radicalism,” British Journal for the History of Philosophy 24 (2016), 3–32: “It is characteristic of Anselm to adopt the formulations of his authorities while giving them meanings of his own, hiding conceptual disagreement by means of verbal echoes” (3). We perhaps find a “verbal echo” in Pros. ch. 9: “Spare me through your clemency, lest you exact retribution through your justice” (I: 108, 86), which resembles Seneca’s first definition of clemency; DC II. 3. 1: “clemency is moderation of the mind in the case of someone with the power to exact retribution.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

96   have no direct relevance for Anselm’s project.²⁹ Still, even if some of Seneca’s arguments have no direct application to Anselm’s God, there are nonetheless striking similarities between Seneca’s and Anselm’s reasoning. In addition, at least one peculiarity of Anselm’s presentation is plausibly explained as the result of borrowing from Seneca and adapting his arguments. If Anselm had access to De Clementia (the issue of accessibility will be discussed in detail in section 4 below) it would have been difficult for him to overlook the fact that De Clementia repeatedly and explicitly applies its understanding of the emperor’s high position to the exalted greatness, goodness, and power of the gods (DC I. 8. 3). The emperor is urged to imitate them: The state looks at its governor with the same feeling with which we would look at the gods if they made it possible for us to see them—with veneration and respect. Does he not occupy the position nearest to the gods if he conforms to their nature, if he is beneficent and generous and wields his power for the better? It is fitting to make this your aspiration and your model: so to be thought the Greatest that at the same time you will be held to be the Best. (DC I. 19. 8–9)

The emperor should adopt for himself the spirit of the gods; he should delight in those of his citizens who are good and put up with the rest (DC I. 5. 7). “Real happiness consists of offering deliverance to many people and calling them back to life from the point of death . . . the power to preserve lives in droves and the community as a whole is divine” (DC I. 26. 5). There is some overlap in terminology between De Clementia and Anselm’s reasoning, and the several obvious differences turn out to be superficial. For example, Anselm only once uses “clemency” and usually uses “mercy,” which Seneca very carefully distinguishes from “clemency.”³⁰ But beyond the terminological difference there is considerable agreement

²⁹ I am indebted to Marcia Colish for stressing this point in correspondence. ³⁰ For Seneca’s probable philosophical background motivations in distinguishing clemency from compassion and pardoning, see Braund’s introduction (66–8).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

97

in substance. For Anselm divine impassibility follows from being something than which no greater can be thought; thus Anselm very clearly separates divine mercy from any feeling or emotion on God’s part. From our point of view, God is merciful because he saves the miserable and pardons sinners, though in himself God is not merciful, properly speaking, because he feels no compassion (Pros. ch. 8: I. 106, 85). This is very similar to what Seneca says about clemency; it is precisely because susceptibility to this sort of feeling is a defect that pity or compassion is inconsistent with supreme goodness and power—the person who exercises the virtue of clemency acts from virtue and right reason and will do dispassionately what someone moved by compassion would do inconsistently, undependably, and for the wrong reason. Thus with respect to the negative value of feeling or emotion, Anselm’s notion of divine mercy very much resembles Seneca’s understanding of imperial clemency. Similarly, Anselm very rarely uses the expressions like “freedom of judgment”³¹ to describe the divine will, whose operation is not a matter of choosing between alternatives, but is constituted by its immutable attachment to the good.³² But in attributing “freedom of judgment” to the emperor in his exercise of clemency, Seneca means to attribute to the emperor the great discretionary power he attributes to the “arbiter,” who is a particular kind of judge endowed with exceptional insight.³³ In his later work, De Concordia Praescientiae et Praedestinationis et Gratiae Dei, Anselm describes the great precision of divine justice, which goes beyond the abilities of any ordinary judge:

³¹ The issue is complicated by the fact that Seneca uses the same expression for the emperor’s discretionary power as Anselm, following a tradition that goes back at least to Augustine, used to designate freedom of the will or the exercise of freedom of choice. ³² One exception is Anselm, De Libertate Arbitrii, in Opera Omnia I, 201–26 (tr. Williams, 145–65), in which Anselm argues that the closely related “freedom of choice” cannot be defined as “the power to sin or not to sin,” since then this definition would not apply to God or the good angels, who have no power to sin. Instead “freedom of choice” has to be defined as “the power to preserve rectitude of will for the sake of rectitude itself,” I: 207–12, 147–51. ³³ Cf. Ben. VI. 6. 1–2: “One law does not combine with another; each pursues its own course . . . a benefit is subject to no law; it makes me the arbiter. I am free to make a comparison between how much he helped me and how much he harmed me, and then to pronounce whether I am owed more than I owe or the reverse. In the cases you adduced [cases in which two laws yield incompatible results] nothing is in our power and we must follow where we are led; in dealing with a benefit, all the power rests with me and I pass judgment.” See also Ben. III. 7. 7–8.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

98   I do not hold that eternal life is promised to all the just; it is promised only to those who are just without any injustice . . . Some people, after all, are just in one respect and unjust in another: for example someone who is chaste, but envious. The blessedness of the just is not promised to such people, for just as true blessedness is without any shortcomings, so too it is granted only to those who are just without any injustice. For since the blessedness that is promised to the just will be a likeness to the angels of God, it follows that just as there is no injustice in the good angels, so too no one with any injustice will have fellowship with them. (De Concordia III. 4; II: 268, 379)

Both Seneca and Anselm stress the limits that come with the highest power. It is a repeated theme in Seneca’s work that greatness brings with it its own constraints; like the gods, who cannot become less, the emperor is constrained not to become less (DC I. 8. 3). Thus certain things that should not be done are proscribed for the emperor both because they are unfitting for someone with the highest station and power and because they would reduce his power; having aroused hatred and resistance, he would lose the ability to change, effectively placing him in the control of others (DC I. 13. 2). For Anselm divine omnipotence makes it impossible for God to do what he should not do; for someone to have the “power” to do so is in fact to have a weakness, to be subject to misfortune and wickedness (“adversity and perversity”); for God to have such an ability would thus amount to being less than the being than which no greater and more powerful can be thought. The most striking similarity between De Clementia and Anselm’s theory about divine mercy is that in both cases the apparent tension between justice and mercy is resolved by identifying mercy as the highest justice. Granted, there is no precise parallel in Seneca’s theory to Anselm’s explanation that divine mercy is the highest justice because in exercising mercy, and not being bound by human desert, God does what is owing to himself as the being than which no better or more powerful can be thought.³⁴ We do, however, find in De Clementia many ³⁴ At one point, Seneca argues that in sparing others the emperor spares himself; the context, however, is an elaborate metaphor in which the emperor is compared to the mind which directs

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

99

variants of the idea that extending clemency is particularly appropriate to the person with the highest power. Extending clemency bespeaks the greatness of spirit that befits the highest power, which cannot fall below its great standing without dragging down that standing as well (DC I. 5. 5). Preserving and saving is the prerogative of the most exalted rank (DC I. 5. 7). For both Seneca and Anselm, clemency/mercy is undeserved. As Anselm sees it, everyone deserves to be abandoned; divine mercy is not earned. And for Seneca, clemency is not a matter of desert; there is no one so innocent as to be certain of never needing clemency (DC I. 1. 9): We have all committed offenses, some of us serious ones, some of us more trivial ones, some on purpose, some by chance impulse or carried away by the wickedness of others. Some of us have not stood firm enough in our good intentions and have lost our guiltlessness unwillingly and while trying to hold on to it. And it is not only that we have been at fault, but we shall continue to be at fault to the very end of life. Even if a person has already purified his mind so thoroughly that nothing can disturb him or mislead him anymore, all the same he reached his state of guiltlessness by committing offenses. (DC I. 6. 3–4)³⁵

This pronouncement is easily read as an anticipation or endorsement of the view that unaided human nature cannot attain or preserve rectitude. In the end it is clear that, for Anselm, what the good God does to the not-good and the wicked alike is a matter of saving them and making them good. For Seneca, the important determinant in deciding to whom and to what extent clemency should be extended is the prospect of

the citizenry as a body. He concludes: “For if, as the argument so far shows, you are the mind of the state and it is your body, you see, then, how necessary clemency is: you are showing mercy to yourself when you seem to be showing it to others” (DC 1. 5. 1). ³⁵ Cf. Seneca’s De Ira II. 28. 1–2: “If we want to be just judges in all things, let us first persuade ourselves of this, that none of us is without guilt. For our greatest source of indignation is our saying: “I am not to blame” and “I have done nothing wrong.” Say instead that you admit to no wrongdoing . . . What man is there who can claim to be innocent with respect to all the laws? And even if that should be so, how limited an innocence it is to be good in the eyes of the law! How much more comprehensive are the rules of duty than the rules of law!”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

100   improvement, of “preserving” or “saving” the person granted clemency. In some cases that call for clemency, there are mitigating circumstances that lessen responsibility, making improvement more likely; the malefactor was young, led astray, a victim of misplaced loyalty. In other cases the deciding factor is simply the prospect of improvement. For Seneca, as for Anselm, though for different reasons, clemency should sometimes be extended to the very wicked; but the intended result is the same, i.e., to make them better than they otherwise would be. In fact, one peculiarity of Anselm’s reasoning is more easily explained if he is using a Senecan template. According to Anselm, God is both merciful and not merciful because we feel God’s mercy as the effect of his action, although in God as he really is there is no mercy because God, being impassible, feels no affect or emotion. This solution to the puzzle of divine mercy seems unnecessarily paradoxical. It would have been more plausible, it seems, for Anselm to imitate the strategy he adopted with respect to divine omnipotence, where he argued in effect that “being able to lie” never is really a description of a power, but always indicates a weakness, so that the absolute lack of such a “power” is a necessary constituent of omnipotence. In the case of mercy, he might, it seems, have argued that “mercy” always essentially designates the action of rescuing the suffering and sparing sinners and that in the case of human beings, though not for God, such behavior springs from feelings of compassion and pity. Possibly Anselm is simply responding to the etymology of the Latin “misericordia,” (“sorrowful heart”),³⁶ terminology to which Anselm is bound by scripture and theological tradition. But another possible reason for Anselm’s claim that from God’s side there is no mercy because, given the divine impassibility, God feels no compassion, is that he is reluctant to give up the close connection between mercy and feelings of pity and compassion that is plainly asserted in De Clementia. Despite many interesting similarities, there is an important disanalogy between Seneca’s and Anselm’s theories. For Anselm divine mercy is the highest justice, it seems, because it befits the being than which no greater and more powerful can be thought not to be constrained by human lack ³⁶ This is suggested in Williams’s note to Pros. ch. 8 (Basic Writings, 85, n. 22).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

101

of merit. Thus divine freedom of judgment allows God to save some and condemn others among those who are equally wicked. For Seneca, however, the emperor enjoys “freedom of judgment” only in the sense that he is free to distinguish circumstances and prospects more finely than the provisions of the law. It is for this reason that exercising clemency is the highest form of justice—because it is the most exact form of justice. Seneca has got a very good explanation for why clemency should not be extended to everyone: a policy of universal clemency would mean extending it in cases in which there was no mitigating factor and no prospect of improvement, and would thus be contrary to striking the most just balance. Anselm in the end has not got a very good reason for why God should not save everyone. To do so would not be just according to human desert, but it is hard to see why that consideration should have any weight, since this is already true of all divine mercy, and especially of divine mercy to some of the wicked. There is another important consequence of this difference between the two theories. Anselm has to admit at the end of his discussion that it cannot be comprehended by human reason how it can be that among those similarly wicked some are saved and some are abandoned. But for Seneca it does not at all follow from the fact that no one is wholly innocent and that the emperor has “freedom of judgment” that punishment and the remission of punishment is not straightforwardly understandable. Thus, given Seneca’s idea of clemency it seems that Anselm’s question will not arise; the emperor will never impose different degrees of punishment upon those equally guilty and identically circumstanced. Clemency as an exercise of moderation that requires discernment, which is the judgment of right reason, will in the case of those equally guilty and identically circumstanced impose and remit exactly the same degree of punishment.

4. Influence or Reinvention: Seneca in the Eleventh Century? Is it plausible that Anselm did in fact look to Seneca’s De Clementia as a model for his theory of divine mercy and justice? Anselm himself gives

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

102   us no definite evidence. He rarely identifies his sources, even when quoting from scripture, but ordinarily weaves his source material seamlessly into his own argumentation.³⁷ In the absence of any decisive positive evidence, we are faced with two questions. Keeping in mind that “access to a work” need not mean acquaintance with the entire work, or with an uncontaminated version, and still less an understanding of its context, could Anselm have had sufficient access to De Clementia to understand its relevance to his project? Second, is there any other material that is likely to have been available to Anselm and which could itself have served as the model for his approach to divine mercy and justice or could have served as a conduit for the ideas in De Clementia? It is probable that Anselm did have access to De Clementia and De Beneficiis, with which it circulated frequently, though not always.³⁸ These two works were part of the popular proliferation of Senecan and pseudoSenecan materials at the turn of the century in Northern France and Norman England.³⁹ In addition, collections of Seneca’s Ad Lucilium Epistulae Morales seem to have circulated, though not always widely, from the Carolingian period on;⁴⁰ it has been argued convincingly that the letters were a principal source for Anselm’s ideas about free will and

³⁷ For Anselm’s method of dealing with his scriptural sources, see R. W. Southern, Saint Anselm: A Portrait in a Landscape (Cambridge: Cambridge University Press, 1990), 69–73. For an example, see notes 3 and 2 above, for texts from Psalms that Anselm weaves into his own argumentation. With respect to Anselm’s method of dealing with scriptural and other source material, I am indebted to Marcia Colish in correspondence. ³⁸ See Malaspina, in De Clementia 81–2, for a comparison of manuscripts of De Clementia circulating with De Beneficiis and those in which it circulated alone. ³⁹ Anthony Hayward, “The Earls of Leicester, Sygerius Lucanus, and the Death of Seneca: Some Neglected Evidence for the Cultural Agency of the Norman Aristocracy,” Speculum 91 (2016), 328–55, describes one late eleventh- or early twelfth-century collection in which Seneca’s De Clementia and Apocolocyntosis are bound together with the pseudo-Senecan Epitaphium, widely supposed to have been written by Seneca on his deathbed, which circulated in numerous collections of Seneca’s ethical works (at 341–3). Hayward argues that it is this short poem about which it is said in another early collection of manuscripts in which the Epitaphium occurs (British Library Ms. Burney 357), that Robert, Earl of Leicester, “was used to recite it from memory,” and that the anecdote attests to the likelihood of a considerable acquaintance on the part of the Earl, probably Robert II (d. 1168), with “the Roman philosopher’s life and ethos” (345). More generally such collections are evidence of the familiarity and popularity of Senecan materials in the Norman culture of Northern France and Norman England at the turn of the century. ⁴⁰ L. D. Reynolds, Texts and Transmission: A Survey of the Latin Classics (Oxford: Oxford University Press, 1983), 370–1.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

103

weakness of will.⁴¹ There was already a copy of Seneca’s Naturales Quaestiones in the library at the monastery of Bec from the time of Lanfranc,⁴² and it seems very probable that Anselm took the famous Proslogion description of God, “something than which nothing greater can be thought,” from the preface to Book I of this Senecan work, though the passage in question might just as well been taken from a florilegium to which Anselm very probably also had access.⁴³ Thus there is considerable evidence for general familiarity with Senecan ideas and works on Anselm’s part. With respect to the accessibility of De Clementia itself to Anselm, Malaspina’s recent critical edition finds no extant manuscripts that are demonstrably from the period between the ninth century, when the archetype of all subsequent manuscripts of De Clementia was produced, and the many manuscripts of the twelfth century. Still, Reynolds speculated already in 1983 that De Clementia and De Beneficiis were on the move in the late eleventh century,⁴⁴ spreading through the monasteries of the Loire valley to England. Both works are listed in the section of a twelfth-century catalogue of the holdings of the library at Bec that seems to pre-date the donation of a large number of books to the Abbey by Philip of Harcourt, ca. 1163; De Beneficiis is listed as having been in Philip’s donation as well.⁴⁵ And a close examination of the relationships between the extant manuscripts of De Clementia has led Malaspina to the conclusion that the twelfth-century manuscript tradition is “already so

⁴¹ Riccardo Federiga and Roberto Limonta, “Le debolezza di voluntà in Anselmo e le sui fonti,” Rivista di storia della filosofia 71 (2016), 357–86. The majority of Anselm’s works cited by the authors were written during his time at Bec. ⁴² See Jacqueline Hamesse, “Les florilèges à l’époque de saint Anselm,” Rivista di storia della filosofia 48 (1993), 477–95, esp. 492–3. ⁴³ Hamesse, “Les florilèges,” 477–8. K. D. Nothdurft, Studien zum Einfluss Senecas auf die Philosophie und Theologie des zwölften Jahrhunderts (Leiden: Brill, 1963), 192–5, suggested already in 1963 that the source for Anselm’s famous formula is Seneca’s Naturales Quaestiones I, Pref., 13: “What is god? The mind of the universe. What is god? All that you see, all that you do not see. In short, only if he alone is all things and if he maintains his own work both from within and without, is he rendered what is due to his magnitude, than which nothing greater can be thought,” 10. The translation is that of Corcoran, adapted. ⁴⁴ Reynolds, Texts and Transmissions, 365. ⁴⁵ Laura Cleaver, “The Monastic Library at Bec,” in B. Pohl and L. Gathagan (eds.), A Companion to the Abbey of Le Bec in the Central Middle Ages (11th–13th Centuries) (Leiden: Brill, 2018), 171–205, esp. 204.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

104   advanced and well developed that it cannot plausibly be attributed just to the exceptional Senecan enthusiasm of the 12th century,” but argues for a tradition of copying and emendation that extends back probably through the whole eleventh century, and surely to well before the first appearances of the item in medieval library catalogues.⁴⁶ In short, in addition to general familiarity with Seneca’s ideas and Senecan works, much of which could have come from exposure to florilegia,⁴⁷ Anselm could easily have had access to De Clementia as a whole or to significant portions of it. As to possible intermediaries through which Senecan ideas might have reached the eleventh century, starting in late antiquity, there is no shortage of relevant Senecan material to be found, often excerpted, quoted, abbreviated, or expropriated.⁴⁸ There are numerous quotations from Seneca’s works and descriptions of his ideas among Church Fathers.⁴⁹ Lactantius (240–325 ) deals with the relationship between justice and mercy in his Divine Institutes and gives a summary of Stoic views on the emotions in his De Ira Dei; both of Lactantius’s works contain a good deal of material from Seneca’s De Beneficiis and De Ira, including fragments of Senecan works no longer extant.⁵⁰ Martin of Braga (d. 579 ), who had available a number of Senecan works,

⁴⁶ Malaspina, in Seneca, De clementia, 77. Medieval catalogues do not show how long an item has been in the collection, and possibly not all that is in the collection. ⁴⁷ See Michael Lapige, “The Stoic Inheritance,” in P. Dronke (ed.), A History of TwelfthCentury Western Philosophy (Cambridge: Cambridge University Press, 1988), 81–112, esp. 94–6. ⁴⁸ Reynolds, Texts and Transmission, 357–65. ⁴⁹ Reynolds, Texts and Transmission, 358. Marcia Colish, The Stoic Tradition from Antiquity to the Early Middle Ages, vol. 2: Stoicism in Christian Latin Thought through the Sixth Century (Leiden: Brill, 1990), gives an analysis of Stoic influences in the early church (9–70), and the relationship between Stoicism and the thought of Augustine, to whom we owe the preservation of a number of Senecan texts (153–252). See also Nadia Bray, La tradizione filosofica stoica nel Medioevo: un approccio dossografico (Rome: Edizioni di storia e letteratura, 2018), 1–33, for an examination of texts from the Church Fathers that show identifiable influence from specific Stoic sources, though she concentrates principally on passages from Cicero’s works. ⁵⁰ Lactantius’s De Ira Dei and Divinae Institutiones supply some material that is missing from Seneca’s De Ira on account of a lacuna in the preserved text of Seneca’s work. For an examination of Lactantius’s quotations from Senecan works, including some works that have not survived, see Marion Lausberg, Untersuchungen zu Senecas Fragmenten (Berlin: De Gruyter, 1970).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

105

including some that are no longer extant, wrote a De Ira that was based on Seneca’s De Ira.⁵¹ The language of Martin of Braga’s short De Ira is Senecan, and the short argument that no one is so sinless as not to need pardon somewhat resembles the reasoning of De Clementia I. 6. 3–4; but Martin’s argument is demonstrably best understood as a patchwork of snatches taken verbatim from Seneca’s De Ira (De Ira III. 24–6). Indeed, Martin’s De Ira as a whole is a pastiche of dramatic passages from Seneca’s De Ira. Moreover this work seems not to have circulated very far beyond its place of origin until quite a bit later than the eleventh century.⁵² More promising as a source are Lactantius’s De Ira Dei and some sections of Institutiones Divinae,⁵³ where he incorporates a mix of ideas from De Ira, De Clementia, and De Beneficiis⁵⁴ into his own theories about divine action. Lactantius’s main argument in De Ira Dei runs counter to Seneca’s views about mercy; contrary to the Stoic view that emotions like anger and pity are vices, and therefore not fitting for someone of the highest greatness and the highest rank, Lactantius sees anger, solicitude, and pity as emotions that are proper to the divine nature (De Ira Dei 17. 17–20). Indeed he sees divine anger, defined as “a movement of the mind rising up to suppress offenses,” as necessary precisely because it motivates divine just and providential governance of

⁵¹ Martin of Braga, De Ira (Opera Omnia, ed. C. W. Barlow (New Haven: Yale University Press, 1950), 145–58). Barlow gives a full comparison of Martin’s De Ira and Seneca’s De Ira in his apparatus to the text. See particularly the apparatus on pages 153–4. ⁵² Barlow, in Opera Omnia, 146, suggests that the distribution of manuscripts of Martin’s works is key to determining where the works circulated; in this case, the distribution would seem to indicate that Martin’s De Ira did not circulate beyond the Iberian Peninsula until later than Anselm’s time, at least not under his own name. ⁵³ These two works by Lactantius are not found in the twelfth-century catalogue of the library of Le Bec. For De Ira Dei, there are not many manuscripts extant for the period between the seventh- and ninth-century archetypes and the proliferation of fourteenth- and fifteenthcentury manuscripts, but there are numerous fragments and excerpts. For Divinae Institutiones there are more manuscripts, more widely distributed. See Lactantius, Opera Omnia, introduction, xiii–xlv. ⁵⁴ In De Ira Dei 18, Lactantius argues that someone who when moved to anger is nonetheless immoderately lenient or pardons more often than necessary, or even always, fosters greater crime in those he pardons and provides himself with perpetual trouble. See also De Ira Dei, 16. 1–5, where Lactantius argues that a God who did not reward the good would be guilty of ingratitude, reasoning which seems to be based on Seneca’s De Beneficiis, though the additional argument that rewarding the just requires also punishing the wicked seems to be Lactantius’s own addition.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

106   the world by moving or inciting God to hate and punish the wicked and love and reward the good (De Ira Dei 17. 12).⁵⁵ And in his Divine Institutes Lactantius brings “humanity” and “compassion” under “perfect justice.” The first obligation that justice imposes on us, without which nothing is really virtue, he says, is to conjoin ourselves with God; “our every deed, every thought, every effort must be aimed at earning his approval.” But the second obligation that justice imposes is to conjoin ourselves to our fellow human beings, since they are images of God; this is called “humanity” or mercy (Divinae Institutiones VI 9.24–10.2). Anselm, had he had access to Lactantius’s works, would surely have agreed that God’s mercy is present and manifest in the world by his aid and governance, but he would surely have rejected Lactantius’s view that such divine action requires inciting or moving by emotions of anger, solicitude, and pity and that divine mercy is a feeling of pity or compassion on God’s part for the sufferings and struggles of human beings. Moreover, for Lactantius, divine mercy and compassion are just divine responses to the tribulations of the good, not a dispensation that no one deserves; and Lactantius is unrelievedly retributionist in his understanding of divine action against the wicked (De Ira Dei 17. 4–6). Finally, for Lactantius it is not God’s exercise of mercy that is identified with the highest justice, but the mercy that human beings extend to other human beings. Lactantius would thus be a very indirect and not very likely source for Anselm’s theory, though Seneca’s reasoning is a probable common source for both Lactantius and Anselm. It is not possible to say with certainty that Anselm adapted the reasoning of De Clementia to his approach to divine justice and mercy. The parallels in strategy are striking, however. And in the absence of plausible intermediary sources, given the probability that Anselm had De Clementia and De Beneficiis available and that he seems to have used Senecan materials on other occasions, it seems probable that De ⁵⁵ De Ira Dei 17. 12. Lactantius claims that anger and compassion have “matter” in God (108), which seems here to reflect his general view that God has a material nature, and can thus be moved and be informed by emotions that motivate divine actions. See Colish, Stoic Tradition, 40; Gérard Verbeke, L’Évolution de la doctrine du pneuma du Stoïcisme à S. Augustin (Louvain: Desclée de Brouwer, 1945), 69–82.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

107

Clementia, adapted and woven into a new agenda, was source material for the reasoning of Proslogion 6–11.

5. Conclusion Neither Seneca nor Anselm purports to have resolved the tension between justice and mercy in a general way. Anselm is specifically interested in divine justice and divine mercy, which is the exclusive prerogative of a supreme, omnipotent being. Seneca’s arguments have to do with a prerogative explicitly limited to the emperor and others who have sovereign power in some situation. In the end, however, it is not clear that either Seneca or Anselm succeeds in answering satisfactorily the more limited question he addresses, and for similar reasons. Neither Seneca nor Anselm gives an indication of intending to revise radically the conception of justice as giving someone his due, but each in his own way stretches that conception in a philosophically problematic way. There seems to be a serious tension between Seneca’s view that no one deserves clemency and his understanding of clemency as “the highest justice,” demanding the judgment of an “arbiter,” who is the most exact sort of moral judge. The call for an arbiter points to an understanding of decisions about clemency as a very precise and comprehensive calculus of desert. The problem is not solved by shifting to the prospect of improvement as the basis for decisions about clemency, since arbitration on the basis of prospects for improvement is no longer based on a calculus of desert, even if a very large number of cases will be resolved in the same way on either criterion.⁵⁶ Anselm strays less from a consistent notion of justice. In all justice, human and divine alike, justice is a matter of giving someone his due. Still, in putting forward his account of how divine mercy flows from divine justice, Anselm moves divine justice and mercy away from their counterparts in the human context, where the recipient of justice and mercy is the subject whose desert demands it.

⁵⁶ Not all cases will be resolved in the same way, however. The very wicked are sometimes to be granted clemency because they are then unlikely to put at risk what they still have by further offenses.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

108   Divine justice to human beings that manifests itself as divine mercy is divine justice because in extending it to human beings God gives himself what he is due.⁵⁷ Louisiana State University

Bibliography Anselm of Canterbury. Sancti Anselmi Cantuarensis Archiepiscopi Opera Omnia, 6 vols., ed. F. S. Schmitt (Edinburgh: Nelson, 1946; repr. Stuttgart: Frommann Verlag, 1968). Anselm of Canterbury. Basic Writings, tr. T. Williams (Indianapolis: Hackett Publishing, 2007). Bray, Nadia. La tradizione filosofica stoica nel Medioevo: un approccio dossografico (Rome: Edizioni di storia e letteratura, 2018). Byers, Sarah. “The Psychology of Compassion: A Reading of City of God 9.5,” in J. Wetzel (ed.), Augustine’s City of God: A Critical Guide (Cambridge: Cambridge University Press, 2012), 130–48. Byrne, Philippa. Justice and Mercy: Moral Theology and the Exercise of Law in Twelfth-Century England (Manchester: Manchester University Press, 2019). Cicero, Marcus Tullius. De Officiis, ed. and tr. W. Miller (London: Harvard University Press, 1913). Cicero, Marcus Tullius. Caesarian Orations, ed. and tr. N. H. Watts, rev. ed. (Cambridge, MA: Harvard University Press, 1953). Cleaver, Laura. “The Monastic Library at Bec,” in B. Pohl and L. Gathagan (eds.), A Companion to the Abbey of Le Bec in the Central Middle Ages (11th–13th Centuries) (Leiden: Brill, 2018), 171–205.

⁵⁷ “Thus in saving us whom you might justly destroy . . . you are just, not because you give us our due, but because you do what is fitting for you who are supremely good” (Pros. ch. 10: I: 109, 87). Earlier versions of this paper were delivered to the Humanities Forum at Louisiana State University sponsored by the Voegelin Institute and to the Cornell Summer Colloquium in Medieval Philosophy, and I am grateful for profitable discussions. I also wish to thank Marcia Colish, who read the paper and made helpful comments and suggestions.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

    

109

Colish, Marcia. The Stoic Tradition from Antiquity to the Middle Ages, vol. 2: Stoicism in Christian Latin Thought through the Sixth Century (Leiden: Brill, 1990). Federiga, Riccardo and Roberto Limonta. “Le debolezza di voluntà in Anselmo e le sui fonti,” Rivista di storia della filosofia 71 (2016), 357–86. Griffin, Miriam. “Seneca’s Pedagogic Strategy: Letters and De Beneficiis,” in R. Sorabji and R. W. Sharples (eds.), Greek and Roman Philosophy 100 –200  (London: Institute of Classical Studies, 2007), 89–113. Griffin, Miriam. Seneca on Society: A Guide to De “Beneficiis” (Oxford: Oxford University Press, 2013). Hamesse, Jacqueline. “Les florilèges à l’époque de saint Anselm,” Rivista di storia della filosofia 48 (1993), 477–95. Hayward, Anthony. “The Earls of Leicester, Sygerius Lucanus, and the Death of Seneca: Some Neglected Evidence for the Cultural Agency of the Norman Aristocracy,” Speculum 91 (2016), 328–55. Inwood, Brad. Reading Seneca (Oxford: Oxford University Press, 2005). Inwood, Brad. Seneca: Selected Philosophical Letters (Oxford: Oxford University Press, 2007). Konstan, David. Pity Transformed (London: Duckworth, 2001). Konstan, David. “Clemency as a Virtue,” Classical Philology 100 (2005), 337–46. Lactantius. Divinae Institutiones, in C. Brandt and G. Laubmann (eds.), L. Caeli Firmiani Lactanti Opera Omnia (Corpus Scriptorum Ecclesiasticorum Latinorum 19, 1–672) (Vienna: F. Tempsky, 1890). Lactantius. De Ira Dei, in C. Brandt and G. Laubmann (eds.), L. Caeli Firmiani Lactanti Opera Omnia (Corpus Scriptorum Ecclesiasticorum Latinorum 27, 67–132) (Vienna: F. Tempsky, 1897). Lapige, Michael. “The Stoic Inheritance,” in P. Dronke (ed.), A History of Twelfth-Century Western Philosophy (Cambridge: Cambridge University Press, 1988), 81–112. Lausberg, Marion. Untersuchungen zu Senecas Fragmenten (Berlin: De Gruyter, 1970). Martin of Braga. De Ira, in Martini Episcopi Bracarensis Opera Omnia, ed. C. W. Barlow (New Haven: Yale University Press, 1950), 145–58. Nothdurft, Klaus-Dieter. Studien zum Einfluss Senecas auf die Philosophie und Theologie des zwöften Jahrhunderts (Leiden: Brill, 1963).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

110   Reynolds, L. D. The Medieval Tradition of Seneca’s Letters (Oxford: Oxford University Press, 1961). Reynolds, L. D. “The Medieval Tradition of Seneca’s Dialogues,” The Classical Quarterly 18 (1968), 355–72. Reynolds, L. D. Texts and Transmission: A Survey of the Latin Classics (Oxford: Oxford University Press, 1983), 370–86. Seneca, L. Annaeus. Epistulae Morales ad Lucilium, ed. and tr. R. Gummere (London: Heinemann, 1920). Seneca, L. Annaeus. Moral Essays, ed. and tr. J. Basore (London: Heinemann, 1935). Seneca, L. Annaeus. Naturales Quaestiones, ed. and tr. T. Corcoran, 2 vols. (Cambridge, MA: Harvard University Press, 1971). Seneca, L. Annaeus. De Ira, Dialogorum Libri Duodecim, ed. L. D. Reynolds (Oxford: Oxford University Press, 1977), 39–128. Seneca, L. Annaeus. De Clementia Libri Duo, ed. E. Malaspina (Turin: Edizione del Orso, 2001, 2004). Seneca, L. Annaeus. Selected Philosophical Letters, tr. B. Inwood (New York: Oxford University Press, 2007). Seneca, L. Annaeus. De Clementia, ed. and tr. S. Braund (Oxford: Oxford University Press, 2009). Seneca, L. Annaeus. On Benefits, tr. M. Griffin and B. Inwood (Chicago: University of Chicago Press, 2011). Southern, R. W. Saint Anselm: A Portrait in a Landscape (Cambridge: Cambridge University Press, 1990). Tacitus. Annales, ed. and tr. J. Jackson (Cambridge, MA: Harvard University Press, 1937). Verbeke, Gérard. L”Évolution de la doctrine du pneuma du Stoïcisme à S. Augustin (Louvain: Desclée de Brouwer, 1945). Weinstock, Stefan. Divus Julius (Oxford: Oxford University Press, 1972). Williams, Thomas. “Anselm’s Quiet Radicalism,” British Journal for the History of Philosophy 24 (2016), 3–32.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

Meditating on the Meditations Al-Ghazālī, Teresa of Ávila, Descartes Taneli Kukkonen

In a recent article, Christia Mercer has argued that the key moves made by René Descartes in the Meditations find precedent in Teresa of Ávila’s (1515–1582) spiritual manual The Interior Castle, published sixty-odd years earlier and enormously popular throughout Roman Catholic Europe in the late sixteenth and early seventeenth centuries. Most prominently, what Mercer calls the evil deceiver argument—a fictional inquisitor whose job it is to instill radical doubt—is pressed to serve an important epistemological role in both Teresa and Descartes. Mercer maintains that although the evidence is necessarily circumstantial, seeing as how Descartes was not given to acknowledging his sources, “in the end, there can be little doubt that Descartes was familiar with Teresa of Ávila’s ideas when he composed his own Meditations in the 1630s.”¹ If Mercer’s verdict is correct then this raises fresh questions, e.g., about Descartes’ reasons for casting his philosophical masterwork in the genre of spiritual meditations.² But what if Teresa’s and Descartes’ moves alike were shown to be anticipated by a Muslim writer of the early twelfth century? Would that further change the story or how we tell it? ¹ Christia Mercer, “Descartes’ Debt to Teresa of Ávila, or Why We Should Work on Women in the History of Philosophy,” Philosophical Studies 174 (2017), 2547. ² Mercer has previously contributed to our understanding of how Descartes’ Meditations reflect trends in Christian meditative literature prevalent in the late Middle Ages and the Reformation period: see Christia Mercer, “The Methodology of the Meditations: Tradition and Innovation,” in D. Cunning (ed.), The Cambridge Companion to Descartes’ Meditations (Cambridge: Cambridge University Press, 2014), 23–47. For further examples of the kind of contextualizing work now being done see Roger Ariew, John Cottingham, and Tom Sorell (eds.), Descartes’ Meditations: Background Source Materials (Cambridge: Cambridge University Press, 1998). Taneli Kukkonen, Meditating on the Meditations: Al-Ghazālī, Teresa of Ávila, Descartes In: Oxford Studies in Medieval Philosophy Volume 8. Edited by: Robert Pasnau, Oxford University Press (2020). © Taneli Kukkonen. DOI: 10.1093/oso/9780198865728.003.0004

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

112   Point for point, as I will demonstrate in the first part of this study, the flow of Descartes’ and Teresa’s argument, as per Mercer’s construal, finds reflection in Abū Hāmid Muhammad al-Ghazālī’s (1056–1111) intellec

tual autobiography The Deliverer from Error, published and released in the eastern reaches of the Islamicate world around 1106 . What is more, in describing two separate perspective-altering bouts with radical doubt—two “crises of knowledge,” as Carol Bargeron labels them— al-Ghazālī effectively manages to presage both Descartes’ and Teresa’s meditations.³ Al-Ghazālī anticipates Descartes’ version of the skeptical crisis, which has a cognitive character and is directed at the claims made by the senses and the intellect, in the first part of the Deliverer. This story of a youthful wrestling with skepticism has earned al-Ghazālī comparisons with Descartes, sometimes with attendant intimations that the former could or should have influenced the latter somehow.⁴ Al-Ghazālī’s second crisis, by contrast, has to do with translating beliefs into practice and is closely reminiscent of Teresa. It is characterized by a lack of existential or ethical conviction in propositions that supposedly have been accepted and affirmed before (e.g., that final judgment is imminent and that therefore devotion to God is imperative, or that knowledge of God should naturally translate into a passionate love for Him and a desire to behold His countenance). It is pegged to the historical events of 488 AH/1095 , when al-Ghazālī experienced some sort of breakdown and departed his professorial post in Baghdad. ³ See Carol L. Bargeron, “Sufism’s Role in al-Ghazālī’s First Crisis of Knowledge,” Medieval Encounters 9 (2003), 32–78; also Eric L. Ormsby, “The Taste of Truth: The Structure of Experience in al-Ghazālī’s Al-munqidh min al-dalāl,” in W. B. Hallaq and D. P. Little (eds.), Islamic Studies Presented to Charles J. Adams (Leiden: Brill, 1991), 133–52. ⁴ Claims about al-Ghazālī influencing Descartes keep cropping up, but concrete evidence is hard to come by. Mahmud H. Zakzuk in Al-Ghazālīs Grundlegung der Philosophie: mit einer Erörterung seines philosophischen Grundansatzes im Vergleich mit Descartes, 2nd ed. (Munich: Ludwig-Maximilians-Universität zu München, 1983) leans heavily on a story about a translation of al-Ghazālī’s Deliverer being found among Descartes’ personal belongings, but nobody seems to be able to produce evidence to corroborate the claim. Elschazlī’s introduction to the German translation of al-Ghazālī’s Deliverer has a more careful account about the Orientalist Jacob Golius—or else Fabri de Peiresc or Marin Mersenne—as a scholar from whom Descartes, even if he was not able to read al-Ghazālī’s actual text, might have learned about the latter’s Muhammad alideas: see ʿAbd-Elsamad ʿAbd-Elhamīd Elschazlī, “Einleitung,” in Abū Hāmid



Ghazālī, Der Erreter aus dem Irrtum, tr. ʿAbd-Elsamad ʿAbd-Elhamīd Elschazlī (Hamburg: Felix

Meiner Verlag, 1988), xxxiv–xxxv. For my part, I remain skeptical.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

113

In the second part of the essay I probe the significance of this double movement in al-Ghazālī for the purposes of understanding better each of the three thinkers under review—al-Ghazālī himself, but also Teresa and Descartes. I am not so much interested in precursorism or genealogies as I am in the conceptual possibilities afforded by philosophy done in a comparative vein. If Descartes shapes a meditative tradition predicated on moral transformation into a more intellectual exercise, then alGhazālī’s argumentation appears to move in the opposite direction. Any inferential certainty based on intellectual first principles for alGhazālī turns out to be of secondary significance when compared to the kind of conviction that allows one to live in accordance with one’s professed beliefs. If this is a correct analysis of the Deliverer’s underlying logic, then it helps to distinguish the relative positionings of al-Ghazālī, Teresa, and Descartes vis-à-vis the kinds of transformations each strives to effect in their readership and the kinds of epistemic orders each wishes to destabilize, on the one hand, and reinstate, on the other. All this ties in with our understanding of what the history of philosophy can do for us, and whether broadening our horizons can aid us in the task(s) we designate for it. Mercer’s stance is clear: “In the end, it is indubitable that inclusive history of philosophy pays off, big time.”⁵ Many, I think, would agree on an intuitive level. Yet I also suspect that when it comes to justifying the effort we must expend when expanding the philosophical canon, it is only the promise of more interesting arguments, brought to bear on present and future philosophical reflection, that can carry the day. Fortunately, this is just what a three-way comparison between al-Ghazālī, Teresa of Ávila, and Descartes promises to deliver.

1. Descartes, Teresa, al-Ghazālī Mercer’s account of the skeptical sequence in Descartes’ Meditations comes in six steps. The refashioning is deliberate and meant to highlight Descartes’ affinities with sixteenth-century meditative literature in ⁵ Mercer, “Descartes’ Debt,’ 2554.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

114   general and Teresa’s Interior Castle in particular. Here are the steps, as characterized by Mercer: 1. Recognizing the problem. (“The meditator recognizes that many of her past and current beliefs need to be reevaluated.”) 2. Beginning anew. (“The meditator sees the need to set aside current beliefs as a first step in the discovery of fundamental truths.”) 3. Severing ties. (“The only proper way to begin anew and attain secure knowledge of fundamental truths is for the meditator to refrain from assenting to present and former beliefs about the external world.”)⁶ 4. Former beliefs return. (“Immediately after the meditator commits to withholding assent from her former beliefs, those beliefs return so that she is tempted to recommit to them.”) 5. The evil deceiver. (“A deceiving demon further confounds the meditator and stops her progress.”) 6. The move to self-knowledge. (“The meditator sees the need to rethink her way forward, which requires self-exploration.”) Not everyone will find this reconstruction equally compelling. The manifestly spiritual genre in which Teresa writes, and the many flourishes she furnishes in order for her writing to function within that genre, conspire to make her line of argument less than transparent to a readership primed on analytic philosophy. And of course, Descartes’ reasoning has been schematized any number of ways. Yet it is a great strength of Mercer’s presentation that it abstracts sufficiently from the particular content of Descartes’ and Teresa’s treatises to allow the underlying similarities to shine through. Even if Descartes and Teresa seek to test the ultimate viability of very different kinds of fundamental truth—the one

⁶ The above descriptions, culled from Mercer, do not make very clear the difference between the second and third steps, but I believe that Mercer’s meaning can be reconstructed from her use of the primary sources. In the second step, what is under discussion are the criteria for what would constitute a secure foundation; in the third, it is recognized that prior commonplaces not only do not meet these criteria, but that holding on to them will actively undermine the seeker’s purpose.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

115

cognitive and intellectual, the other predominantly moral and spiritual— the moves the two make appear appreciably similar. For present purposes, I will adopt the six steps without quarrel or argument. In what follows I gather Mercer’s evidence concerning Descartes in a single paragraph, then do the same for Teresa, followed by a lengthier exposition of al-Ghazālī’s text. I hope the reader will forgive the resulting imbalance. The added value of the present investigation, I take it, comes from my injection of al-Ghazālī into the conversation, and it is here that I must document the primary sources in close detail.⁷

a. Descartes (a1) Descartes’ plunging into the question is cast in first-personal terms and succinctly stated: “Some years ago I was struck by the large number of falsehoods that I had accepted as true in my childhood, and by the highly doubtful nature of the whole edifice that I had subsequently based on them” (tr. John Cottingham). (a2) Once in a lifetime, Descartes determines, one should establish the certainty of everything one holds, starting right at the foundations, if the sciences are to be constructed reliably. (a3) This entails withholding assent initially from even the most obvious propositions, whether of a physical and sensory or mathematical nature (“I am here, sitting by the fire,” “2+3=5”). (a4) Descartes records the continued pull of longstanding commonplaces and attributes this to custom and habituation. (a5) Descartes’ demonic deceiver is well known, although interpretations of its precise role and operations continue to proliferate. Mercer identifies as Descartes’ specific contribution the observation that one need not assume actual demonic forces to be operative in the world for radical doubt to be adduced; it is enough to point out that the existence of such a malicious deceiver is not precluded.

and ⁷ Page references to the Deliverer come from Al-Munqidh min al-.dalāl, ed. J. Salībā
K. ʿAyyād, 7th printing (Beirut: Dār al-Andalus, 1967) (henceforth Munqidh). I avail myself of Richard McCarthy’s translation in Freedom and Fulfillment: An Annotated Translation of alGhazālī’s al-Munqidh min al-.dalāl and Other Relevant Works, tr. R. J. McCarthy (Boston: Twayne Publishers, 1980) (henceforth “McC’), with reference to McCarthy’s numbered paragraphs.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

116   (a6) Finally, it is Descartes’ fresh self-identification as a thinking being that presents him with a way forward.

b. Teresa On Mercer’s reading, Teresa shows a similar epistemological bent to Descartes, even though her authorial strategies are different.⁸ (b1) Instead of plunging into self-examination at the outset, Teresa begins from a critique of third parties whose moral decrepitude is met by the wrongness of their beliefs. Taking the reader into her confidence, Teresa fixes her sights on the benighted state of those “blighted souls” whose undeveloped faculties trap them in a miserable, brutish existence that is not only “blindness,” but “darkness itself.”⁹ (b2) There is, therefore, a need to take “the right road from the beginning,” else one will remain relegated to a world “filled with falsehood.” (b3) Again, while a denouncement of worldly ties is standard fare in spiritual literature, “Teresa is unusual in the epistemological spin she gives this task . . . the exercitant must refrain from assenting to past and current beliefs about the external world.” (b4) This, however, proves easier said than done: customary beliefs and concerns flood back, blocking the meditator’s progress. (b5) The problem is compounded here by the deceitful demons, those that heighten the soul’s awareness of, and appreciation for, the present world: For here the devils once more show the soul these vipers – that is, the things of the world – and they pretend that earthly pleasures are almost eternal: they remind the soul of the esteem in which it is held in the world, of its friends and relatives, of the way in which its health will be endangered by penances . . . and of impediments of a thousand other kinds.¹⁰

⁸ For this exposition of Teresa I rely on Mercer, “Descartes’ Debt,” 2549–53. ⁹ Interior Castle 1.1, in The Complete Works of Avila, tr. E. A. Peers, 3 vols. (London: Burns & Oates, 2002), 2:204; Interior Castle 1.2 [Complete Works 2:206]. ¹⁰ Interior Castle 2.1 [Complete Works 2:214].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

117

(b6) The correct assessment is that “the world is full of falsehood,” yet this will only become firmly established in the heart once one has committed fully to a process of self-exploration. This act of the will, however, depends on divine grace for its success.

c. The Youthful al-Ghazālī—The Cognitive Problem Those familiar with al-Ghazālī will know the outlines of his purported youthful crisis. Al-Ghazālī, intent on securing certain knowledge, plunges into doubt concerning the reliability not only of his senses, but of his intellect as well. Yet when the story is interpreted in light of Mercer’s six steps, interesting added details emerge. (c1) Recognizing the Problem. The first step in each of our works is an awakening to the question of whether what is generally regarded as a secure understanding of things really is such. Al-Ghazālī’s opening gambit in the Deliverer takes a page from the Muslim heresiographers. Al-Ghazālī marvels at the great profusion of sects and beliefs, then evokes formulaically the trouble in identifying where in their midst salvation might lie: You should first of all know – God give you good guidance and gently lead you to the truth! – that the diversity of men in religions and creeds, plus the disagreement of the Community of Islam about doctrines, given the multiplicity of sects and the divergency of methods, is a deep sea in which most men founder and from which few only are saved. Each group alleges that it is the one saved, and “each faction is happy about its own beliefs.” This is the state of affairs which the truthful and most trustworthy Chief of God’s envoys – God bless him! – ominously promised us when he said: “My Community will split into seventy-odd sects, of which one will be saved.” And what he promised has indeed come to pass! (Munqidh, 61.5–12/McC, §4)

Al-Ghazālī points out that the children of Jews grow up to be Jews and the children of Christians, Christians. Is the acquisition of Islamic faith similarly predicated only on the authority of one’s parents and one’s community? In a manner redolent of Galen, al-Ghazālī underlines his

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

118   personal willingness to delve into the teachings of the various schools in an unprejudiced manner while paying fealty to none.¹¹ (c2) Beginning Anew. Al-Ghazālī’s initial observation about a multitude of sects is next parleyed into a methodological or meta-level argument about how one can know that one knows: So I began by saying to myself: “What I seek is knowledge of the true meaning of things. Of necessity, therefore, I must inquire into just what the true meaning of knowledge is.” Then it became clear to me that sure and certain knowledge is that in which the thing known is made so manifest that no doubt clings to it, nor is it accompanied by the possibility of error and deception, nor can the mind even suppose such a possibility.¹² [ . . . ] I realized, then, that whatever I did not know in this way and was not certain of with this kind of certainty was unreliable and unsure knowledge, and that every knowledge unaccompanied by safety from error is not sure and certain knowledge. (Munqidh, 63.12–64.11/McC, §§7–8)

(c3) Severing Ties. With the bar established that any putative certain knowledge must be clear, al-Ghazālī proceeds to review the sources of actual knowledge. Propositions accepted on the basis of authority or prior inferential reasoning are quickly set aside (taqlīdiyyāt, nazariyyāt),
leaving two kinds of basic beliefs: those based on sense-perception (hissiyyāt) and those deemed necessary (d. arūriyyāt). Judgments based
on sense-perception are soon shown to be unreliable. Al-Ghazālī does

¹¹ Al-Ghazālī labels his innately inquisitive nature a God-given gift, thereby blocking the putative objection that he is merely some objectionable troublemaker: Munqidh, 62.2–63.12/ McC, §§5–6. The benchmark study positing a connection between al-Ghazālī, Descartes’ carefully crafted self-image in the Discourse, and the tradition of intellectual autobiography stretching back to Galen is by Stephen Menn, “The Discourse on 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. On innate nature (fitra)
as an explanatory principle in al-Ghazālī see further Frank Griffel, “Al-Ghazālī’s Use of ‘Original Human Disposition’ (Fitra) and Its Background in the Teachings of al-Fārābī and Avicenna,”
The Muslim World 102 (2012), 1–32. ¹² Al-Ghazālī’s framing of certain knowledge in terms both psychological and modal (incapacity to question or doubt) echoes earlier Islamic theological discussions, for which see Binyamin Abrahamov, “Necessary Knowledge in Islamic Theology,” British Journal of Middle Eastern Studies 20 (1993), 20–32.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

119

not in fact claim that sensation itself would err when it comes to its proper objects (something Aristotle, too, would have denied, De an. 3.3, 427b11–12): but beliefs relating to the so-called common sensibles such as motion and size frequently turn out to be mistaken, so that we do not for instance realize that a shadow is in constant if imperceptible motion.¹³ Intellect is therefore needed to judge and—where necessary—correct hasty judgments arising on the basis of sense-perception.¹⁴ (c4) Deceiver. In a departure from Mercer’s established sequence, there is no immediate return in al-Ghazālī’s youthful exploration to beliefs once discarded. Instead, further devastation is visited upon al-Ghazālī’s established beliefs by sense-perception (hiss) personified—an adversary,
to be sure, but not one that would be painted in malevolent or demonic hues. Sense-perception, after it is pointed out how often judgments based on its findings are found to be in need of correction, effectively exclaims et tu quoque and charges the intellect with objections very similar to what Descartes has his evil demon voice. If the intellect is capable of correcting the errors made on the basis of sense-perception, then who is to say that a higher judge will not one day announce itself and declare the judgments of reason to be defective?¹⁵ And besides, is it not possible that all of life is but a dream? (c5) Former Beliefs Return. At first blush, the notion of former beliefs returning is scarcely to be found in al-Ghazālī’s account of his first crisis. Al-Ghazālī is simply plunged ever deeper into doubt, swept into a vortex of his own making. However, when al-Ghazālī says that what resulted from the accumulation of doubts was a period of two months during which al-Ghazālī was a ‘sophist’ in thought if not in outward appearance ¹³ The example of the shadow is first brought up in al-Ghazālī’s Revival of the Religious Sciences, written up to a decade prior to the Deliverer: see Ihyāʾ ʿulūm al-dīn, 10 vols. (Jeddah:
Dār al-minhāj, 2011), XXVI, 6:56. ¹⁴ Munqidh, 65.8–66.7/McC, §10; on why al-Ghazālī refrains from questioning sensation as such see Taneli Kukkonen, “Al-Ghazālī’s Skepticism Revisited,” in H. Lagerlund (ed.), Rethinking the History of Skepticism: The Missing Medieval Background (Leiden: Brill, 2010), 48. ¹⁵ The developmental view al-Ghazālī holds regarding the psychic faculties is relevant here. For al-Ghazālī, the power of discernment accompanies the dominance of the soul’s spirited part, while the power of reason properly awakens at the onset of puberty: see Ihyāʾ XXXI,
7:33–4, 7:57–9 and Taneli Kukkonen, “Al-Ghazālī on Error,” in F. Griffel (ed.), Islam and Rationality: The Impact of al-Ghazālī, Volume Two (Leiden: Brill, 2016), 4–8. The proper order of things is for the later-developing and more refined power always to assume dominance over the previous ones.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

120   (Munqidh, 67.14–17/McC, §15), it bears noting that in the Islamic letters of the period ‘sophistry’ or safsatah signifies Protagorean relativism.
Succumbing to sophistry carries the imputation of licentiousness, the kind of worldliness that from “we don’t know what is to become of us in the hereafter” infers “let us eat, drink, and be merry.” So perhaps a heedless cleaving to old ways was not far from al-Ghazālī’s mind. (We shall have occasion to return to this theme in the second section of this study.) (c6) Move to Self-Knowledge. Al-Ghazālī declares that it was an act of God that eventually brought him out of the darkness of sophistry and into the light again, not any intellectual effort on his own part: At length God Most High cured me of that sickness. My soul regained its health and equilibrium and once again I accepted the self-evident data of reason and relied on them with safety and certainty. But that was not achieved by constructing a proof or putting together an argument. On the contrary, it was the effect of a light which God Most High cast into my breast. And that light is the key to most knowledge. (Munqidh, 67.17–68.1/McC, §15)

The point is that once the necessity of the so-called necessary truths has been set in question, no secure naturalistic grounds remain from which to argue for their necessity. The earnest seeker after truth must recognize that the needed fortitude of mind can only issue directly from the divine domain: it is not as if, say, a humanly constructed proof from God’s necessary goodness could force things back into a familiar shape. What is noteworthy in light of Mercer’s analysis is how even these early findings are framed in a psychological light, so that they become understood in the context of an exploration of self: The aim of this account is to emphasize that one should be most diligent in seeking the truth until he finally comes to seeking the unseekable. For primary truths are unseekable, because they are present in the mind; and when what is present is sought, it is lost and hides itself. But one who seeks the unseekable cannot subsequently be accused of negligence in seeking what is seekable. (Munqidh, 68.11–14/McC, §17)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

121

Those who never enter into introspection will continue to enjoy the luxury of possessing the primary and necessary truths in a direct and naive manner, but only on pain of never coming to understand their nature or provenance. Those driven to delve deeper, meanwhile, cannot help but have the glass of their naive conformity shattered.¹⁶ But the process will at least lead these people—who, besides al-Ghazālī himself, will presumably include a cross-section of the Deliverer’s readership—to a deeper understanding about those inner resources on which all must draw in their search for truth. The Deliverer ostensibly charts its author’s explorations in this vein. The reinstitution of necessary truths, or even of direct perceptual cognitions, does little to settle any of the nettlesome disagreements between the different schools of thought. As it turns out, their application only leads al-Ghazālī deeper into introspection.

d. Al-Ghazālī’s Second Crisis—The Sufi Challenge Al-Ghazālī in the Deliverer describes a second crisis of knowledge. This latter crisis, in contrast to the first, is linked to events that verifiably occurred: in late 488/1095, al-Ghazālī unexpectedly vacated his professorial chair in Baghdad and assumed the mantle of a religious reformer. According to the Deliverer this was due to al-Ghazālī realizing that he was headed for perdition, seeing as how none of his cleverness and enterprise as a dialectically-minded defender of Sunni orthodoxy had brought him any closer to God. Both al-Ghazālī’s contemporaries and recent commentators have questioned whether al-Ghazālī’s repentance was real or feigned,¹⁷ but that need not concern us here. It is enough to note that in the Deliverer, the two crises are deliberately positioned so as to signpost different ¹⁶ Munqidh, 69.10–15/McC, §19; similarly al-Ghazālī, Iljām al-ʿawāmm ʿan ilm al-kalām, in Majmūʿāt rasāʾil al-Ghazālī (Beirut: al-Maktaba al-asriyya, 2012), 375–6; discussed in depth by
Richard M. Frank, “Al-Ghazālī on taqlīd: Scholars, Theologians, and Philosophers,” Zeitschrift für Geschichte der Arabisch-Islamischen Wissenschaften 7 (1991–2), 207–52. ¹⁷ See Kenneth Garden, “Coming Down from the Mountaintop: Al-Ghazālī’s Autobiographical Writings in Context,” The Muslim World 101 (2011), 581–96; on tawba or repentance as a trope in Muslim narratives of spiritual authority, see Atif Khalil, Repentance and the Return to God: Tawba in Early Sufism (Albany: State University of New York Press, 2018).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

122   phases in the intellectual seeker’s journey. The first crisis (the one we already countenanced) provides the intellectual tools with whose aid a reasoned evaluation of the merits and demerits of the various schools may be mounted. The second invites reflection on the possibility of a different set of criteria emerging in whose light the first approach may appear misguided, or at least partial and incomplete. (d1) Recognizing the Problem. Al-Ghazālī’s second tour of epistemic duty is not announced in the manner of the first and is therefore easier to miss. Yet if we shadow Mercer’s moves around the perimeter of Teresa’s Interior Castle, we can detect the epistemic character of al-Ghazālī’s second crisis in the way he introduces his readers to the Sufi path, the fourth and final stop on al-Ghazālī’s tour of the classes of seekers (tālibūn) after truth.
Everything here looks different from the three previous schools (in brief: theologians, philosophers, Shiʿī missionaries). Rather than dealing solely with the verification of theoretical propositions, the Sufis are distinguished by knowledge and works (ʿilm wa-ʿamal), something that will prove challenging to al-Ghazālī on a personal level. What is more, the Sufis lay claim to terrain that even as a matter of principle cannot be mapped through the means to which al-Ghazālī had hitherto grown accustomed and in which he had exhibited mastery:¹⁸ It became clear to me that their most distinctive characteristic is something that can be attained, not by study (taʿallum), but rather by fruitional experience (dhawq) and the state of ecstasy and “the exchange of qualities.” (Munqidh, 101.4–5/McC, §82)

(d2) Beginning Anew. Reading further into al-Ghazālī’s account of the events of 488/1095, we witness the protagonist tacitly accepting the epistemological claims made by the Sufis and recognizing the need to follow in their footsteps. Already by definition, after all, the apparatus he had hitherto used to weigh the doctrines of the various schools was unfit

¹⁸ For ‘cannot’ al-Ghazālī has lā yumkin, an expression explicitly borrowed from Aristotelian modal terminology.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

123

for the task of determining the truth or falsity of the knowledge the Sufis had claimed for themselves: I knew with certainty that the Sufis were masters of states, not purveyors of words, and that I had learned all I could by way of theory. There remained, then, only what was attainable, not by hearing and study, but by fruitional experience and actually engaging in the way. (Munqidh, 102.9–11/McC, §83)

In labeling the Sufis ‘masters of states’ (arbāb al-ahwāl) al-Ghazālī refers
to the notion that the Sufis’ ascetic exercises, introspection, and devotion to contemplation reliably grant them access to cognitive states and visions that lend themselves to linguistic description only barely and only in their lowliest stages. The verification of these experiences occurs along different lines from the inferentially based disciplines, although firsthand experience (tajriba) and second-hand reports (tawātur) each can play a role in how the standing of those who claim to have had such experience is investigated (Munqidh, 106.19–108.16/McC, §§95–9). The idea that the Sufis would have access to self-validating otherworldly experiences not accessible to the masses was a standard trope long before al-Ghazālī’s day.¹⁹ So why make it into a fresh realization on al-Ghazālī’s part? Because, for the purposes of presenting his own autobiography as a model for others to follow, the recognition of specialist Sufi insight into the divine has to be yoked to a commitment to follow the practical path of self-interrogation, self-regulation, and self-abnegation, as we shall see in step 6. (d3) Severing Ties. From here the steps taken by al-Ghazālī map closely on to Teresa’s. First comes the need to renounce all worldly ties: It had already become clear to me that my only hope of attaining beatitude in the afterlife lay in piety and restraining my soul from passion. The beginning of all that, I knew, was to sever my heart’s

¹⁹ See, e.g., the materials collected in John Renard (ed.), Knowledge of God in Classical Sufism (New York: Paulist Press, 2004).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

124   attachment to the world by withdrawing from this abode of delusion and turning to the mansion of immortality and devoting myself with total ardor to God Most High. That, I knew, could be achieved only by shunning fame and fortune and fleeing from my preoccupations and attachments. (Munqidh, 103.1–5/McC, §84)

Note the parallel between al-Ghazālī’s “abode of delusion” (dār alghurūr) and Teresa’s “world of falsehood,” penned nearly five centuries later. There is even an intriguing parallel to Teresa’s “interior castle” in al-Ghazālī’s “mansion of immortality” (dār al-khulūd), which some time later is described as having an antechamber (dihlīz: Munqidh, 107.3/ McC, §95) in the purification of the soul. For al-Ghazālī as for Teresa, the way up is simultaneously the way in.²⁰ The exercise involves discarding the kinds of beliefs that would motivate a lifetime seeking fame: I attentively considered my circumstances, and I saw that I was immersed in attachments which had encompassed me from all sides. I also considered my activities – the best of them being public and private instruction – and saw that in them I was applying myself to sciences unimportant and useless in this pilgrimage to the hereafter. Then I reflected on my intention in public teaching, and I saw that it was not directed purely to God, but rather was instigated and motivated by the quest for fame and widespread prestige. So I became certain that I was on the brink of a crumbling bank and already on the verge of falling into the Fire, unless I set about mending my ways. (Munqidh, 103.6–12/McC, §85)

(d4) Former Beliefs Return. Similar to what we find in Teresa, during alGhazālī’s crisis of 488/1095, the conventional truths that reassert them-

²⁰ The parallel mansion/castle metaphors in Islamic mysticism and in Teresa have led scholars since Miguel Asín Palacios to postulate a possible historical connection: the exact metaphor of seven concentric castles goes back at least to the third/ninth century, for which see Luce López-Baralt, Islam in Spanish Literature: From the Middle Ages to the Present (Leiden: Brill, 1992), 107–21. (I thank Mohammed Rustom for the pointer.) To my knowledge, no scholar mentions the Deliverer in connection with Muslim–Christian mystical exchanges, though some point to al-Ghazālī’s Niche of Lights (Mishkāt al-anwār).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

125

selves are of the type that lead one to prize material security and worldly esteem, family and friends, and these are the ones that prove most difficult to uproot. I therefore reflected unceasingly on this for some time, while I still had the freedom of choice. One day I would firmly resolve to leave Baghdad and disengage myself from those circumstances, and another day I would revoke my resolution. In the morning I would have a sincere desire to seek the things of the afterlife; but by evening the hosts of passion would assail it and render it lukewarm. Mundane desires began tugging me with their chains to remain as I was, while the herald of faith was crying out: “Away! Up and away!” At such thought the call would reassert itself and I would make an irrevocable decision to run off and escape. (Munqidh, 103.3–21/McC, §§86–7)

(d5) The Evil Deceiver. The epistemic dimension to the impasse reached by al-Ghazālī is highlighted by his introduction of an ill-intentioned deceiver, this time clearly Satanic in character. Al-Ghazālī’s Satanic whisperer effectively conveys the temptations faced by the academic grown cozy with even a moderate level of prestige and material comfort: Then Satan would return to the attack and say: “This is a passing state: beware, then, of yielding to it! For it will quickly vanish. Once you have given in to it and given up your present renown and splendid position from vexation and renounced your secure situation untroubled by the contention of your adversaries, your soul might again look longingly at all that – but it would not be easy to return to it!” (Munqidh, 103.22–104.2/McC, §87)

Note that these are the very same temptations Teresa evokes, right down to the temporal distortions imposed by the Satanic deceiver. The deceiver appeals to things that are treasured in the course of everyday life— esteem, family and friends, material security—and inflates the imagined cost of their loss, as if one were otherwise able to hold on to them forever. This of course is a fallacy, since all earthly goods are vanishing by definition. The correct calculation is to prioritize what is truly eternal,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

126   i.e., God’s final judgment and the consequent bliss or torment of the soul.²¹ (d6) Move to Self-Knowledge. Al-Ghazālī’s hemming and hawing is resolved in dramatic fashion, as he first finds himself unable to speak or even to eat. God thereafter relieves him of his agency altogether, making what was previously difficult subsequently easy. This effects a perspective shift whereby al-Ghazālī switches his mode of instruction from the kind of knowledge that enhances one’s worldly status to the kind that leads to renouncing the world (Munqidh, 123.3–7/McC, §139). Al-Ghazālī devotes close to eleven years of his life to seclusion and solitude (al-ʿuzla wa l-khulwa: Munqidh, 115.3/McC, §121, Munqidh, 122.14–15/McC, §138). While scholars have debated both the significance and the duration of this self-proclaimed isolation, what is clear is that al-Ghazālī links his newfound mode of teaching to a process of selfexploration and self-improvement. The larger part of al-Ghazālī’s literary output post-1095 spins out of his peculiar appropriation of the Delphic maxim under its Islamic guise: “whosoever knows himself, knows his Lord.’²² The following passage from the Deliverer illustrates the introspective turn that al-Ghazālī’s epistemic project takes post-1095: I now earnestly desire to reform myself and others, but I do not know whether I shall attain my desire or be cut off by death short of my goal. Yet I believe with a faith as certain as direct vision that there is no might for me and no power save in God, the Sublime, the Mighty; and that it was not I who moved, but He moved me; and that I did not act, but He acted through me. I ask Him, then, to reform me first, then to use me as an instrument of reform; to guide me, then to use me as an instrument of guidance; to show me the true as true, and to grant me

²¹ The correct weighing of temporary goods against eternal wages is a standing theme in alGhazālī’s principal work, the Revival of the Religious Sciences, in particular its section detailing a Condemnation of the World (Ihyāʾ XXVI). In another section, which notably focuses on
delusion and self-deception, al-Ghazālī anticipates Pascal’s wager by nearly six centuries. See Ihyāʾ XXX, 6:611–15.
²² See Taneli Kukkonen, “The Self as Enemy, the Self as Divine: A Turning Point in Islamic Anthropology,” in P. Remes and J. Sihvola (eds.), Ancient Philosophy of the Self (Dordrecht: Springer, 2008), 205–24; on the Delphic Maxim in the Islamic tradition Alexander Altmann, Studies in Religious Philosophy and Mysticism (London: Kegan Paul, 1969), 1–40.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

127

the grace to follow it; and to show me the false as false, and to grant me the grace to eschew it! (Munqidh, 123.8–14/McC, §139)

e. Further Lessons Our application of Mercer’s analysis suggests that in two separate passes, al-Ghazālī’s autobiography traces the steps both of Descartes’ Meditations and of Teresa’s meditations, respectively. This is remarkable enough in its own right, insofar as it further challenges any outmoded way of framing Descartes’ work as a unique and autochthonous eruption. But are there further lessons to be gleaned here beyond what alGhazālī had already said, that “it is not farfetched that ideas should coincide, just as a horse’s hoof may fall on the print left by another” (Munqidh, 87.20–88.1/McC, §55)? After all, al-Ghazālī’s anticipation of Descartes’ skepticism—Hume’s as well, for that matter—has been a matter of public record ever since George Henry Dewes’ 1845–1846 publication of his Biographical History of Philosophy (itself likely building on Auguste Schmölders’ 1842 French translation of al-Ghazālī). Yet despite this, and despite the further fact that Leo Groarke prominently reports the same parallels in a well-placed 1984 Journal of the History of Philosophy article,²³ one scans in vain for so much as a footnote on alGhazālī in today’s Descartes scholarship.²⁴ Without more grist for the philosophical mill, it seems unlikely that either al-Ghazālī or Teresa will ²³ See Auguste Schmölders, Essai sur les écoles philosophiques chez les arabes et notamment sur la doctrine d’Alghazzali (Paris: Firmis Didot Frères, 1842); George Henry Dewes, The Biographical History of Philosophy, From Its Origin in Greece Down to the Present Day (New York: Appleton, 1857), 2:363; Leo Groarke, “Descartes’ First Meditation: Something Old, Something New, Something Borrowed,” Journal of the History of Philosophy 22 (1984), 288–92. I owe my knowledge of Dewes to Mohammad Azadpur, “Unveiling the Hidden: On the Meditations of Descartes and Ghazzali,” in Anna-Teresa Tymieniecka (ed.), The Passions of the Soul in the Metamorphosis of Becoming (Dordrecht: Springer, 2003), 219–40. ²⁴ One searches in vain for al-Ghazālī’s name in David Cunning (ed.), The Cambridge Companion to Descartes’ Meditations (Cambridge: Cambridge University Press, 2014) or in Karen Detlefsen (ed.), Descartes’ Meditations: A Critical Guide (Cambridge: Cambridge University Press, 2013); nor do Rorty’s, Tweyman’s, or Gaukroger’s older essay collections mention him; nor do, say, Gary Hatfield in The Routledge Guidebook to Descartes’ Meditations (London and New York: Routledge, 2014), Catherine Wilson in Descartes’s Meditations: An Introduction (Cambridge: Cambridge University Press, 2003), or Janet Broughton in Descartes’s Method of Doubt (Princeton, NJ: Princeton University Press, 2002). To be fair, many of these works do not purport to perform any archeological or contextualizing work at all.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

128   ever end up as anything more than footnotes to Descartes, seeing as how—as I have argued in another context—the rewards of precursorism are soon reaped and often prove thin.²⁵ What can we do with the knowledge that al-Ghazālī’s, Teresa’s, and Descartes’ skeptical strategies line up closely yet imperfectly? What should we do? My proposed course of action is that we pay attention to differences as much as similarities, beginning with the observation that Descartes’ and Teresa’s evil deceivers take aim at different kinds of presumed certainties, the one cognitive and intellectual, the other moral and experiential. If alGhazālī’s intellectual journey in the Deliverer from Error passes through the two resultant types of skepticism, then a question arises as to what the pairing is meant to accomplish and why it is sequenced in the manner that it is. And if the answer turns out to be at all interesting, then this may have implications for our understanding of Teresa’s and Descartes’ respective argumentative strategies as well.

2. Cognitive and Moral Certitude In a nutshell, in the Deliverer, al-Ghazālī erects the entire project of seeking certainty through proofs only to discount it at a later stage. Through recounting his own life’s experience, al-Ghazālī seeks to illustrate first the kind of truth-seeking that proceeds according to established standards of discursive reasoning and inferential procedures; crucially, this includes a lengthy accounting of al-Ghazālī’s own irreproachable career as a public intellectual. The turning point arrives when al-Ghazālī realizes that no amount of dialectical or deductive labor will secure for him the kind of conviction that is required to commit to a path of religion that is inwardly lived rather than merely outwardly professed, and that might therefore come accompanied with a deeper understanding of the profundities of faith.²⁶ ²⁵ Kukkonen, “Al-Ghazālī’s Skepticism Revisited,” 30. ²⁶ Al-Ghazālī repeatedly resorts to the trope of faith having an inner kernel as well as an exterior husk, with the inner layers representing deepening levels of understanding and conviction: see, e.g., Al-Maqsad ed. Fadlou Shehadi,
al-asnā fī sharh
maʿānī asmāʾ Allāh al-husnā,
2nd ed. (Beirut: Imprimerie Catholique, 1982), 42–4.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

129

In working out how these successive moments map on to Teresa of Ávila and Descartes, a useful place to start is the Book of Knowledge, the opening section to al-Ghazālī’s massive Revival of the Religious Sciences. In this post-1095 work, which announces al-Ghazālī’s project of sketching a revitalized “science of the path of the hereafter,” al-Ghazālī distinguishes two ways of understanding the very term “certitude” (yaqīn). One signifies the absence of doubt (ʿadam al-shakk), while the other “has nothing at all to do with considerations of admissibility or doubt” (iʿtibār al-tajwīz wa-lshakk).²⁷ Instead, the second type of certitude is measured by one’s active commitment to one’s convictions, i.e., an ability to act consistently with what one professes to believe. Unlike intellectual certitude, which only admits of clarity and vagueness, the latter kind of certainty is characterized also by strength and weakness, as manifested in the varying levels of fortitude people display when faced with tests to their faith. The second kind of certitude is the province solely of “those who draw near,” i.e., those who seek proximity to God. The two kinds of certitude described in the Revival map on to the two crises al-Ghazālī ascribes to himself with remarkable precision. Indeed, I would posit that, autobiographical elements aside, what we find in the Deliverer is an illustration of an analysis al-Ghazālī had developed a decade prior. (This is a fact that has not, to my knowledge, been much heeded in the modern scholarship.) We can use this conceptual framework to extract four lessons, one to do with al-Ghazālī, the other three with how his work compares with that of Teresa and Descartes. 1. The first definition of certitude according to the Book of Knowledge is the one operative in the deliberations of the various theorists and theologians (al-nuz
zār
wa-l-mutakallimūn). It intimates a considered weighing of propositions, culminating in “veridical understanding, attained by way of demonstration” (al-maʿrifa al-haqīqiyya al-hāsila bi

tarīq al-burhān).²⁸ While there are many pathways leading up to such
knowledge, ranging from immediate sense-perception (hiss) and experi
ence (tajriba) to speculation (al-nazar) and unanimous reports (tawātur),


²⁷ Ihyāʾ I, 1:273.4.


²⁸ Ihyāʾ I, 1:273.10.


OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

130   the common outcome is an absence of doubt (ʿadam al-shakk). This indeed explains the pre-eminence of demonstrative knowledge (burhān) when it comes to certitude, since “in it, doubt is inconceivable” (lā yutasawwar al-shakk fī-hi).²⁹
It is this kind of certitude that is in al-Ghazālī’s sights when in the Deliverer he describes his purported first, youthful crisis of knowledge. What he finds disconcerting is the permissibility of doubt when it comes to the fundamental building blocks from which all knowledge is constructed, the basic functionality of those very cognitive avenues by which propositions come to be held. What the divine light restores, correspondingly, is al-Ghazālī’s confidence in the power of human reasoning. It is the necessary propositions, such as the law of the excluded middle, whose veracity is henceforth considered beyond reproach: and what these first principles make possible is a reasoned examination of the different schools of thought. Carol Bargeron has it right when she remarks that in Deliverer’s first epistemological crisis and its eventual resolution, for al-Ghazālī, “God’s light had not illumined the truth and certainty of what was knowable as such through reason; rather, divine enlightenment had given him the confidence and mental well-being to believe that his reason was sound.” Which is simultaneously to say: Al-Ghazālī’s confirmation of man’s inherent ability to reach true judgements did not mean that he believed that all sense- or reasonbased knowledge-claims were authentic. On the contrary, he worked to define the methods and ways by or through which human beings are able to distinguish instances of true knowledge from “empty imaginings.”³⁰

Bargeron’s and my takes on al-Ghazālī differ in that Bargeron treats alGhazālī’s story of his youthful doubt as an account of some real-life events, while to me the Deliverer’s programmatic features make it likelier

²⁹ Ihyāʾ I, 1:271.11; see Michael E. Marmura, “Ghazali and Demonstrative Science,” Journal
of the History of Philosophy 3 (1965), 183–204. ³⁰ Bargeron, “Sufism’s Role,’ 76.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

131

that the story is a literary invention.³¹ Nevertheless, Bargeron’s point stands. In the argumentative context of the Deliverer, al-Ghazālī’s first crisis is meant to secure for the protagonist the means for reasoning upon which an examination of the various schools of thought can be erected. Let us furthermore stipulate that in spinning the tale of his first crisis, al-Ghazālī reverses Mercer’s steps 4 and 5: first comes the insistent interrogator (not really ‘evil deceiver’ in this case) and only then do old beliefs return, albeit in the form of ‘sophistry,’ which is to say Protagorean relativism. What al-Ghazālī wants to say, it seems to me, is that the lack of a sound intellectual foundation for arguing the merits and demerits of the various sources of knowledge will result not so much in confusion as in complacency. The illness that he attributes to his youthful self is not one that would physically immobilize him, the way his second crisis does. Instead, by casting aspersions on the power of the intellect to adjudicate between right and wrong belief, the part of him that is still in thrall to the sensory world (i.e., sense-perception reified) succeeds in keeping al-Ghazālī effectively a sensualist.³² Seen in this light, the Deliverer’s story of a youthful bout with skepticism pairs up with al-Ghazālī’s later engagement in the same work with the ‘esotericists’—the Fātimid Ismāʿīlī missionaries, whose
bag of intellectual tricks included the inducement of an all-consuming skepticism, one that would deliver the listener into the waiting arms of the authoritative (or authoritarian, if you will) teacher, i.e., the Imām or his representative. For al-Ghazālī, materialist worldliness and authoritarianism both represent an abrogation of duty when it comes to the

³¹ While Teresa’s and Descartes’ skeptical meditations unquestionably play up the personal aspect in a manner not common to the majority of the Western philosophical tradition, neither author embeds them in the genre of autobiography the way al-Ghazālī does. As a consequence, many of al-Ghazālī’s readers have taken his writings at face value, as referring to a personal wrestling with skepticism, in a way that is absent from the scholarship on Descartes. I believe this to be a mistake: the confessional aspect of al-Ghazālī’s writing at all times serves a literary and polemical purpose. On the uses of Muslim autobiography see the materials in Dwight F. Reynolds (ed.), Interpreting the Self: Autobiography in the Arabic Literary Tradition (Berkeley: University of California Press, 2001). ³² See also section 1.c. (c5) above.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

132   reasoned examination of one’s beliefs: both result from giving up too early and too easily.³³ 2. In one important respect, (the youthful) al-Ghazālī and Descartes are more similar to one another than they are different. Both al-Ghazālī’s first deceiver and Descartes’ hypothetical demon call into question propositions that will in the end be given back to the inquiring mind, even if this is done in a framework that has been profoundly transformed: I see blue, ten is more than three, I am sitting in front of a fire. The purpose of these deceivers is to spur the questioner to delve deeper into what grounds one’s confidence in such beliefs—not, in the end, to have those beliefs overturned. So the old label of methodological skepticism seems warranted in both cases. Teresa’s deceiver and al-Ghazālī’s second deceiver, by contrast, are evil deceivers—they are, quite literally, designated demons and Satanic whisperers—for a reason: their purpose is to keep the meditator clinging on to materially false beliefs about the worth of status, creature comforts, and the like, and the doubts that they sow concern changes that do in fact need to be implemented. This is also why their appearance follows on the heels of such earthbound thoughts making a return independently. The deceivers essentially act as amplifiers for false evaluations that are already seen as exercising a hold on the corrupt soul. (Notice also that because of this, we may interpret both al-Ghazālī’s and Teresa’s vacillations in strictly psychological terms, should we choose to do so. Although each portrays the demonic deceiver as a real force operative in the world, the construct leaves open the possibility of regarding the whisperings as issuing from an interaction of the corrupted soul with its surroundings.³⁴) The mature al-Ghazālī and Teresa agree that the kind of self-knowledge that leads to true tranquility of mind, and thereby a receptivity to the

³³ On this point see Kukkonen, Al-Ghazālī’s Skepticism Revisited, 33–8; more generally Farouk Mitha, Al-Ghazālī and the Ismailis (London: I. B. Tauris, 2001). ³⁴ A deflationary interpretation of al-Ghazālī’s demonology has been suggested by Frank Griffel, “Al-Ghazālī at His Most Rationalist,” in G. Tamer (ed.), Islam and Rationality: The Impact of al-Ghazālī, Volume One (Leiden: Brill, 2015), 89–120 and by Kukkonen, “Al-Ghazālī on Error,” 12–20; a stoutly realist reading of Teresa’s demonology is offered by Stina Busman Jost, “The Devil in the Details: How Teresa of Avila’s Description of the Devil Assured and Liberated Women,” Medieval Mystical Theology 26 (2017), 6–19.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

133

higher and more transcendent kinds of truth, is personal and ethical in character. They both appeal to what is essentially the late antique Platonist model of the purificatory virtues as a necessary propaedeutic to philosophical enlightenment. And they both insist that the impetus for selfrenewal must ultimately issue from a divine source, seeing as how a corrupted soul cannot muster independently the requisite strength to initiate a comprehensive soul-searching.³⁵ These elements are signally missing from Descartes. And surely this is telling: for while Descartes may rightly be hailed as an early proponent of epistemic virtue,³⁶ he certainly gives no indication in the Meditations that moral virtue or spiritual self-examination should precede or attend optimal insight into the world’s workings. Such a perspective, meanwhile, is central to the meditative tradition represented by the mature alGhazālī and Teresa. In fact, this may turn out to be one of the stronger arguments to be made in favor of regarding Descartes as the founder of the modern philosophical subdiscipline of epistemology. Much of the epistemological work done in Western philosophy up to the time of Descartes is more properly placed under something like cognitive psychology, on the one hand (call this the De anima tradition), or philosophy of science, on the other (the Posterior Analytics tradition). And while the question What can we know? in the Western philosophical tradition goes back at least to Parmenides, its treatment in terms of epistemic dispositions— rather than a call to identify correctly the proper objects of knowledge— is a much rarer occurrence pre- than post-Descartes. So to view Descartes as coming out of the meditative tradition, as Mercer suggests we do, is in part to witness a secularization (so to speak) of the notion, traceable back to late antique Platonism, that a purification of the soul and a better understanding of its operations must necessarily precede any attempt to come to grips with the world as it is.

³⁵ In al-Ghazālī’s case, the seeds to this turn are sown already in the resolution to his youthful crisis. After all, it is a light issuing from God that gives al-Ghazālī the conviction to proceed even with his intellectual deliberations. ³⁶ See Linda Trinkaus Zagzebski, Virtues of the Mind (Cambridge: Cambridge University Press, 1996), 34–5; Ernest Sosa, A Virtue Epistemology: Apt Belief and Reflective Knowledge, Volume I (Oxford: Oxford University Press, 2007), 126–33.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

134   3. But this then points to the single greatest difference between al-Ghazālī and Descartes. Both rhetorically speaking and in philosophical terms, al-Ghazālī’s first crisis is merely prologue to the proper story he wants to recount, as the chapter titles of the Deliverer already make clear (the crisis is recounted under the heading of “preliminaries,” muqaddimāt). One might even say that it is preliminary only to the first half of al-Ghazālī’s journey. For while the resolution to his initial wrestling with skepticism helps to ground and motivate a rational investigation into the purported truths propounded by the various schools of thought, that investigation itself is found to be sorely lacking and dissatisfying in key respects by the end. In portraying himself as coming face to face with the realization that an altogether different type of conviction—the certitude of the saints, as per the Revival of the Religious Sciences—is required for the authentically religious life, al-Ghazālī effectively downgrades the very pursuits that the resolution to his first crisis had made possible. This again accords with the wider aims of al-Ghazālī’s Revival project, which had sought to constrain the claims to intellectual and religious authority made by theologians, legal scholars, preachers, and philosophers alike. In insisting that the purificatory virtues can lead to a contemplation of supernal truths directly, with the pursuit of dialectics and demonstration more of a hindrance than a help for most people, al-Ghazālī puts a twist on the venerable Platonic tradition of cognitive ascent just as surely as Descartes will later do, albeit in a rather different fashion.³⁷ Descartes’ deliberations proceed along a comparatively linear path. Descartes’ difficulties in letting go of old beliefs are what grind his

³⁷ See Taneli Kukkonen, “Al-Ghazālī on the Origins of Ethics,” Numen: International Journal for the History of Religions 63 (2016), 271–98. The essential point to grasp is that in the Deliverer, as well as in isolated remarks strewn across his authorship, al-Ghazālī carves out a space for naturally inquisitive souls such as himself to explore intellectually, and it is those select few who shall not cease from exploration, according to the full protocols of the different modes of valid inference, until they have circled back to a faith that recognizes both the uses of reason and its necessary limitations. (See section 1.c. (c6) above.) The true cognitive elite, meanwhile, may just be the saints whose perfected practice allows them direct insight into the supernal world; and the majority of people will be best served by a simple adherence to the articles of faith and an incompletely informed religious practice that protects it: see Frank, “Al-Ghazālī on taqlīd.’

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

135

explorations to a halt; be that as it may, even as the deceiver’s queries serve to heighten the sense of crisis in the narrative, they are also what allow for its forceful and—what is more important—final resolution. While there is much work left to do over the rest of the Meditations, once one has progressed past the skeptical crisis, the pieces begin to fall into place, for good and for better. With the principles of reason restored, we are invited to regard Descartes’ reasoning as both compelling and complete. And there will certainly be no need for an overthrow of the new epistemic order on the basis of moral considerations. If I am not mistaken, Descartes’ project is no less grounded in the intellectual trendlines of his times than was al-Ghazālī’s, but it is more streamlined. It simply moves from one inferential order of knowledge (the scholastic) to another (Descartes’ own). 4. For all that the post-1095 al-Ghazālī gets to appear the antiacademic rebel in comparison with Descartes (who himself is frequently portrayed as inciting revolution against the old scholastic order), when stood next to Teresa, both have a conservative profile. The moves made in the Interior Castle replicate something Elena Carrera has demonstrated in the context of Teresa’s autobiography, namely, that by foregrounding the need for inner reformation, and by putting forward innumerable perceptive observations concerning the movements of the soul, its vicissitudes as well as its triumphs, Teresa creates space for a different kind of religious authorship, and indeed authority, one predicated on personal spiritual insight.³⁸ The psychological journey traced by the Interior Castle is all the more remarkable for the way it trades in a lack of certitude and a candid admission of human fallibility, moral as well as intellectual, even as the author guides the reader with a sure hand through the many mansions of her castle. To be sure, al-Ghazālī’s mature authorship also shows a preoccupation with ‘technologies of the self,’ to employ Foucauldian parlance. In the vein of much of the ascetic/contemplative tradition, the very acts of self-

³⁸ Elena Carrera, Teresa of Avila’s Autobiography: Authority, Power, and the Self in MidSixteenth Century Spain (London: Legenda, 2005).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

136   study, self-interrogation, and self-discipline create the conditions for a new kind of self-constitution.³⁹ Yet the reproduction of such technologies on a public stage always at the same time constitutes an exercise in intellectual legitimation. And it is here that an interesting difference opens up between al-Ghazālī and Teresa. Al-Ghazālī’s navigation of the existing schools shows his continued willingness (and perhaps wiliness) when it comes to co-opting existing structures of knowledge production. How else to explain al-Ghazālī’s ostensible deference to the “masters of [inner] states”—Sufi teachers whose discoveries, however, he says he studied from books, as part of an established epistemic order, a school among schools? Al-Ghazālī’s dialoguing with the various claimants to spiritual power and authority in his time is nothing if not complex. But even if a tension between searching criticism and a re-affirmation of tradition cuts through his authorship as a theme, the end result looks more like a re-inscription of Sunni Islam and its intellectual disciplines, done on al-Ghazālī’s terms.⁴⁰ In marked contrast, Teresa neither trades on status nor does she make sweeping claims about having studied comprehensively all there is to know about the various schools and their respective claims to knowledge, the way we find in al-Ghazālī or Descartes. Instead, Teresa creates space for her topical explorations by affecting humility throughout and by drawing deferential contrasts where necessary between her teachings and those of “learned, clever men.” This approach, which of course is quite deliberate, accords with the authorial strategies of the mystical writings of women (also contemporaneous writings concerning them) in European medieval times, as Grace Jantzen and others have documented.⁴¹ ³⁹ See Ebrahim Moosa, Ghazālī and the Poetics of Imagination (Chapel Hill: University of North Carolina Press, 2005), 237–60; Wael B. Hallaq, The Impossible State: Islam, Politics, and Modernity’s Moral Predicament (New York: Columbia University Press, 2013), 129–35; Robert Landau Ames, “The Technology of Happiness: Philosophy, the Body, and Ghazālī’s Kīmiyā-yi saʿādat,” Comparative Islamic Studies 9 (2013), 121–39. ⁴⁰ See, e.g., Timothy J. Gianotti, “Beyond Both Law and Theology: An Introduction to al-Ghazālī’s “Science of the Way of the Afterlife’ in Reviving Religious Knowledge (Ihyāʾ
ʿulūm al-dīn),” The Muslim World 101 (2011), 597–613; for the complexities of al-Ghazālī’s engagement with Sufism in particular, Kenneth Garden, The First Islamic Reviver (Oxford: Oxford University Press, 2014). ⁴¹ See Grace M. Jantzen, Power, Gender, and Christian Mysticism (Cambridge: Cambridge University Press, 1995), 157–92.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

137

Beyond observations grouped around the sociology of knowledge, which are valuable in their own right, these alternative epistemologies— less inferential and less textually based; more experientially grounded and involved with analyses of embodied cognition—are of independent philosophical interest, as several recent studies show, even if they do require more work from us as historians of philosophy (much as they required more work from their proponents at the time).⁴² In the case of Teresa, what we notice even on a first pass is that her moment of radical doubt appears in a different light when its leveling effects are acknowledged. All those benighted souls whose eyes are occluded by the dust of material riches and earthly concerns can only hope to be divested of such cares through divine agency: no one is exempt from their temptations, and the battle against them is hardly won in a single pass or with a single act of will and understanding. If there are institutional safeguards against the demons of avarice, jealousy, and pride, or against any of the attendant incorrect evaluations that could stall the exercitant’s inward progress, then they are to be found in an unwavering focus on the love of God and hewing close to the monastic order, with an attendant emphasis on (female) solidarity. This certainly looks starkly different from the lessons drawn by Descartes from the skepticism of his lone philosophical genius. The relation of Teresa’s deliberations to al-Ghazālī’s, meanwhile, merits closer study.

3. Towards an Unsettled Historiography The philosophical benefits conferred by a more inclusive historiography are not to be found merely in the act of welcoming more people into the ⁴² See, e.g., Barbara Simerka, “Feminist Epistemology and Pedagogy in Teresa of Ávila,” in A. Weber (ed.), Teaching Teresa of Ávila and the Spanish Mystics (New York: Modern Language Association of America, 2009), 107–13 and Sarah Coakley, “Dark Contemplation and Epistemic Transformation: The Analytic Theologian Re-Meets Teresa of Ávila,” in O. D. Crisp and M. C. Rea (eds.), Analytic Theology: New Essays in the Philosophy of Theology (Oxford: Oxford University Press, 2009), 280–312; for recent comments on the relevance of these materials for historiographies of philosophy, Christina Van Dyke, “Many Know Much, but Do Not Know Themselves: Self-Knowledge, Humility, and Perfection in the Medieval Affective Contemplative Tradition,” in G. Klima and A. Hall (eds.), Consciousness and Self-Knowledge in Medieval Philosophy (Newcastle upon Tyne: Cambridge Scholars Publishing, 2018), 89–106 and Peter Adamson, “Philosophae in medias, pt. II” (accessed October 21, 2019).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

138   fold. If that were the case, why, outside of considerations of representation, would we ever consider swapping in an al-Ghazālī where Descartes once stood, when the scholarship regarding the latter will always remain unmatched in scope and depth? And why would we be looking more closely into an al-Ghazālī or a Teresa, if all that we can hope to find is a (more or less conforming) precursor to Descartes? Rather, the value of a more broad-based historiography lies in identifying cracks in the convenient fiction according to which we would already know which problems we should teach and in what manner, or what kinds of things we should be looking for in our research into thinkers past and present.⁴³ When subtle yet profound differences open up between trains of thought that on the surface appear quite similar, that is when the prospect of continued progress opens up. To illustrate: it is thanks to Mercer’s investigations that we can now identify Teresa of Ávila as a relevant precursor to Descartes, in terms of the presentation of skepticism and its overcoming in the Meditations. But what really excites the imagination—mine, at any rate—is her contention that Teresa’s deliberations involve an altogether different set of propositions, one that has a moral and existential dimension rather than a merely cognitive one. What this leads us to recognize is that Descartes’ choice of redirecting the meditative genre in the direction of considering the truths of science and metaphysics is just that—a choice. His recasting in non-moral terms the inward turn required of the seeker after knowledge, a trope that goes back to late antiquity if not all the way to Plato’s First Alcibiades, can be seen as more remarkable rather than less once one appreciates how closely Descartes otherwise traces the steps of this more traditional approach. Furthermore, Mercer’s work casts a new light on the previously identified parallel between Descartes’ and al-Ghazālī’s skeptical arguments and their resolution. The impact goes in both directions: without reaching back to look at al-Ghazālī’s contribution to the history of skeptical arguments, as I have done, we might not be aware of the option of treating the two kinds of propositions—factual and moral—in ⁴³ See Lisa Shapiro, “Revisiting the Early Modern Philosophical Canon,” Journal of the American Philosophical Association 2 (2016), 365–83.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

139

conjunction. At the same time, without Mercer’s contribution, it would probably escape the attention of Arabists as well as European readers that al-Ghazālī’s two skeptical episodes are treated dissimilarly by the man himself. A study that simply compared al-Ghazālī’s first crisis with Descartes’ Meditations would probably be easier to render watertight—through a process of stipulation and a circumspect manipulation of the textual evidence, of course. So would a separate comparison between alGhazālī’s second crisis and the Interior Castle, which would then take on a different character, perhaps that of an inquiry within the discipline of religious studies, focused on the ways in which spiritual legitimacy is conferred upon an author through rhetorically underlining the need for divine guidance in matters of religion. By contrast, any attempt to triangulate the three thinkers—al-Ghazālī, Teresa, Descartes—will be unstable by design, veering almost inevitably between the meditative/ transformational and the inferential/epistemological axes. But that very unsettled and unsettling quality is the precise reason why such a multi-way comparison can prove useful. If the past is a storehouse whose riches we can harness to “refresh current conversations and reframe solutions,” as Mercer puts it, then not only can a wider catchment secure us more materials from which to work. It is also the case that what we thought were overly familiar problems, perhaps even stale ones, have dimensions to them we did not previously perceive. And I strongly suspect that researching philosophy in a global manner—temporally as well as geographically, and looking especially into arguments produced by marginalized groups—will everywhere open up similarly disruptive and unsettling lines of questioning. Perceived binaries are always easier either to assimilate to one another or to organize into neat polarities; but with three or more thinkers, lines of thought manifest similarities as well as dissimilarities that can no longer be so neatly catalogued. And if this can be done with the most shop-worn of all source texts—Descartes’ Meditations—through the eyes of a female mystic and a Muslim theologian, then how much more can be accomplished once we venture farther afield? New York University Abu Dhabi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

140  

Bibliography Abrahamov, Binyamin. “Necessary Knowledge in Islamic Theology,” British Journal of Middle Eastern Studies 20 (1993), 20–32. Adamson, Peter. “Philosophae in medias, pt. II” . Accessed October 21, 2019. Altmann, Alexander. Studies in Religious Philosophy and Mysticism (London: Kegan Paul, 1969). Ames, Robert Landau. “The Technology of Happiness: Philosophy, the Body, and Ghazālī’s Kīmiyā-yi saʿādat,” Comparative Islamic Studies 9 (2013), 121–39. Ariew, Roger, John Cottingham, and Tom Sorell (eds.). Descartes’ Meditations: Background Source Materials (Cambridge: Cambridge University Press, 1998). Azadpur, Mohammad. “Unveiling the Hidden: On the Meditations of Descartes and Ghazzali,” in A.-T. Tymieniecka (ed.), The Passions of the Soul in the Metamorphosis of Becoming (Dordrecht: Springer, 2003), 219–40. Bargeron, Carol L. “Sufism’s Role in al-Ghazālī’s First Crisis of Knowledge,” Medieval Encounters 9 (2003), 32–78. Broughton, Janet. Descartes’s Method of Doubt (Princeton, NJ: Princeton University Press, 2002). Carrera, Elena. Teresa of Avila’s Autobiography: Authority, Power, and the Self in Mid-Sixteenth Century Spain (London: Legenda, 2005). Coakley, Sarah. “Dark Contemplation and Epistemic Transformation: The Analytic Theologian Re-Meets Teresa of Ávila,” in O. D. Crisp and M. C. Rea (eds.), Analytic Theology: New Essays in the Philosophy of Theology (Oxford: Oxford University Press, 2009), 280–312. Cunning, David (ed.). The Cambridge Companion to Descartes’ Meditations (Cambridge: Cambridge University Press, 2014). Detlefsen, Karen (ed.). Descartes’ Meditations: A Critical Guide (Cambridge: Cambridge University Press, 2013). Dewes, George Henry. The Biographical History of Philosophy, From Its Origin in Greece Down to the Present Day (New York: Appleton and Co., 1857).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

141

Elschazlī, ʿAbd-Elsamad ʿAbd-Elhamīd. “Einleitung,” in Abū Hāmid


Muhammad al-Ghazālī, Der Erreter aus dem Irrtum, tr. ʿAbd-Elsamad

ʿAbd-Elhamīd Elschazlī (Hamburg: Felix Meiner Verlag, 1988).


Frank, Richard M. “Al-Ghazālī on taqlīd: Scholars, Theologians, and Philosophers,” Zeitschrift für Geschichte der Arabisch-Islamischen Wissenschaften 7 (1991–2), 207–52. Garden, Kenneth. “Coming Down from the Mountaintop: Al-Ghazālī’s Autobiographical Writings in Context,” The Muslim World 101 (2011), 581–96. Garden, Kenneth. The First Islamic Reviver (Oxford: Oxford University Press, 2014). al-Ghazālī, Abū Hāmid. Al-Munqidh min al-d. alāl, ed. J. Salībā and K.

ʿAyyād, 7th printing (Beirut: Dār al-Andalus, 1967). al-Ghazālī, Abū Hāmid. Freedom and Fulfillment: An Annotated Translation
of al-Ghazālī’s al-Munqidh min al-d.alāl and Other Relevant Works, tr. R. J. McCarthy (Boston: Twayne Publishers, 1980). al-Ghazālī, Abū Hāmid. Al-Maqsad

al-asnā fī sharh
maʿānī asmāʾ Allāh al-h
usnā, ed. F. Shehadi, 2nd ed. (Beirut: Imprimerie Catholique, 1982). al-Ghazālī, Abū Hāmid. Ihyāʾ ʿulūm al-dīn, 10 vols. (Jeddah: Dār al-minhāj,

2011). al-Ghazālī, Abū Hāmid. Iljām ʿan ʿawāmm ʿan ʿilm al-kalām, in Majmūʿāt
rasāʾil al-Ghazālī (Beirut: al-Maktaba al-asriyya, 2012), 341–76.


Gianotti, Timothy J. “Beyond Both Law and Theology: An Introduction to al-Ghazālī’s “Science of the Way of the Afterlife” in Reviving Religious Knowledge (Ihyāʾ
ʿulūm al-dīn),” The Muslim World 101 (2011), 597–613.

Griffel, Frank. “Al-Ghazālī’s Use of ‘Original Human Disposition’ (Fitra)
and Its Background in the Teachings of al-Fārābī and Avicenna,” The Muslim World 102 (2012), 1–32. Griffel, Frank. “Al-Ghazālī at His Most Rationalist,” in G. Tamer (ed.), Islam and Rationality: The Impact of al-Ghazālī, Volume One (Leiden: Brill, 2015), 89–120. Groarke, Leo. “Descartes’ First Meditation: Something Old, Something New, Something Borrowed,” Journal of the History of Philosophy 22 (1984), 281–301. Hallaq, Wael B. The Impossible State: Islam, Politics, and Modernity’s Moral Predicament (New York: Columbia University Press, 2013).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

142   Hatfield, Gary. The Routledge Guidebook to Descartes’ Meditations (London: Routledge, 2014). Jantzen, Grace M. Power, Gender, and Christian Mysticism (Cambridge: Cambridge University Press, 1995). Jost, Stina Busman. “The Devil in the Details: How Teresa of Avila’s Description of the Devil Assured and Liberated Women,” Medieval Mystical Theology 26 (2017), 6–19. Khalil, Atif. Repentance and the Return to God: Tawba in Early Sufism (Albany: State University of New York Press, 2018). Kukkonen, Taneli. “The Self as Enemy, the Self as Divine: A Turning Point in Islamic Anthropology,” in P. Remes and J. Sihvola (eds.), Ancient Philosophy of the Self (Dordrecht: Springer, 2008), 205–24. Kukkonen, Taneli. “Al-Ghazālī’s Skepticism Revisited,” in H. Lagerlund (ed.), Rethinking the History of Skepticism: The Missing Medieval Background (Leiden: Brill, 2010), 29–59. Kukkonen, Taneli. “Al-Ghazālī on Error,” in F. Griffel (ed.), Islam and Rationality: The Impact of al-Ghazālī, Volume Two (Leiden: Brill, 2016), 3–31. Kukkonen, Taneli. “Al-Ghazālī on the Origins of Ethics,” Numen: International Journal for the History of Religions 63 (2016), 271–98. López-Baralt, Luce. Islam in Spanish Literature: From the Middle Ages to the Present (Leiden: Brill, 1992). Marmura, Michael E. “Ghazali and Demonstrative Science,” Journal of the History of Philosophy 3 (1965), 183–204. Menn, Stephen. “The Discourse on 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. Mercer, Christia. “The Methodology of the Meditations: Tradition and Innovation,” in D. Cunning (ed.), The Cambridge Companion to Descartes’ Meditations (Cambridge: Cambridge University Press, 2014), 23–47. Mercer, Christia. “Descartes’ Debt to Teresa of Ávila, or Why We Should Work on Women in the History of Philosophy,” Philosophical Studies 174 (2017), 2539–55. Mitha, Farouk. Al-Ghazālī and the Ismailis (London: I. B. Tauris, 2001). Moosa, Ebrahim. Ghazālī and the Poetics of Imagination (Chapel Hill: University of North Carolina Press, 2005).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

-¯ ¯ ,   ´ , 

143

Ormsby, Eric L. “The Taste of Truth: The Structure of Experience in al-Ghazālī’s Al-munqidh min al-d. alāl,” in W. B. Hallaq and D. P. Little (eds.), Islamic Studies Presented to Charles J. Adams (Leiden: Brill, 1991), 133–52. Renard, John (ed.). Knowledge of God in Classical Sufism (New York: Paulist Press, 2004). Reynolds, Dwight F. (ed.). Interpreting the Self: Autobiography in the Arabic Literary Tradition (Berkeley: University of California Press, 2001). Schmölders, Auguste. Essai sur les écoles philosophiques chez les arabes et notamment sur la doctrine d’Alghazzali (Paris: Firmis Didot Frères, 1842). Shapiro, Lisa. “Revisiting the Early Modern Philosophical Canon,” Journal of the American Philosophical Association 2 (2016), 365–83. Simerka, Barbara. “Feminist Epistemology and Pedagogy in Teresa of Ávila,” in A. Weber (ed.), Teaching Teresa of Ávila and the Spanish Mystics (New York: Modern Language Association of America, 2009), 107–13. Sosa, Ernest. A Virtue Epistemology: Apt Belief and Reflective Knowledge, Volume I (Oxford: Oxford University Press, 2007). Teresa of Avila. The Complete Works of Avila, tr. E. A. Peers, 3 vols. (London: Burns & Oates, 2002). Van Dyke, Christina. “Many Know Much, but Do Not Know Themselves: Self-Knowledge, Humility, and Perfection in the Medieval Affective Contemplative Tradition,” in G. Klima and A. Hall (eds.), Consciousness and Self-Knowledge in Medieval Philosophy (Newcastle upon Tyne: Cambridge Scholars Publishing, 2018), 89–106. Van Dyke, Christina. “What Has History to Do with Philosophy? Insights from the Medieval Contemplative Tradition,” in M. Van Ackeren (ed.), Philosophy and the Historical Perspective (Oxford: Oxford University Press, 2018), 155–70. Wilson, Catherine. Descartes’s Meditations: An Introduction (Cambridge: Cambridge University Press, 2003). Zagzebski, Linda Trinkaus. Virtues of the Mind (Cambridge: Cambridge University Press, 1996). Zakzuk, Mahmud H. Al-Ghazālīs Grundlegung der Philosophie: mit einer Erörterung seines philosophischen Grundansatzes im Vergleich mit Descartes, 2nd ed. (Munich: Ludwig-Maximilians-Universität zu München, 1983).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

Thomas Aquinas on How the Soul Moves the Body David Cory

Interpreters of Thomas Aquinas’s hylomorphic theory tend to focus on the problems that arise in his attempt to reconcile a hylomorphic account of substantial form with the individual immortality of human intellectual souls—a flagship claim often labeled “Thomistic hylomorphism.” But although the difficulties there are indeed great, it would be a mistake to proceed as if Aquinas’s hylomorphic theory is smooth sailing up until his account of the human soul. One important challenge for interpreters is his view that the soul moves the body, and that this ability to move the body distinguishes souls (including even the souls of plants) from other substantial forms. A good example is found in Summa contra gentiles 2.65 where he says that “an animal moves itself: the mover in it is the soul, while the moved is the body. The soul therefore is a mover which is not moved.”¹ To hold that the soul, which is the form of the body, moves ¹ SCG 2.65. For discussion see n. 49 below. For discussions of the various media through which the soul moves the body see SCG 2.71 [Leon. 13.454: 1b–9b], QDDA 9 [Leon. 24/I.82: 278–95], and the De motu cordis. For the ability of the soul to move the body as that which distinguishes it from other substantial forms, see DV 22.3, n. 56 below. The commentary on the Liber de causis is cited according to the edition by H.-D. Saffrey (Louvain: Nauwelaerts, 1954). The commentary on the Sentences, books I–III according to the edition by R. P. Mandonnet and R. P. Moos (Paris: Lethielleux, 1929–47), and book IV according to the Parma edition (Parma: Typis Petri Fiaccadori, 1852–73)—abbreviated respectively as “Mand.,” “Moos,” and “Parma.” The remaining texts are cited according to the Leonine edition of Aquinas’s Opera omnia (abbreviated “Leon.”). Titles of commonly-cited works are abbreviated as follows: InPhys = In octo libros Physicorum Aristotelis; DV = Quaestiones disputatae de veritate; InDA = Sentencia libri De anima; InDeSensu = Sentencia libri De sensu et sensato; QDDA = Quaestiones disputatae de anima; QDSC = Quaestio disputata de spiritualibus creaturis; DP = Quaestions disputatae de potentia; Sent. = Scriptum super libros Sententiarum; SCG = Summa contra gentiles; ST = Summa theologiae. Works are cited by their primary internal divisions and all translations into English are mine. David Cory, Thomas Aquinas on How the Soul Moves the Body In: Oxford Studies in Medieval Philosophy Volume 8. Edited by: Robert Pasnau, Oxford University Press (2020). © David Cory. DOI: 10.1093/oso/9780198865728.003.0005

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

145

the body at all might seem like a departure from the basic hylomorphic role of form. At least in the most basic cases (inanimate motion), it is the composite which properly moves and acts through its form, and not form which moves. Aquinas’s insistence that the soul moves the body seems to take a step away from hylomorphism, and towards a dualist account of the soul. This point has been nicely problematized by Robert Pasnau. Pasnau’s idea is that there is an historical shift from conceiving of the soul in an abstract and metaphysical way as that which is responsible for the unity and identity of a being over time, to conceiving of it in a more concrete and even perhaps mechanistic way, as an internal efficient cause of motion in the body. (Presumably, the further one moves away from the “metaphysical” account, the harder it becomes to apply to non-human souls.) Pasnau puts Aquinas in the center of this historical trajectory (both chronologically and thematically) as someone who attributes both roles to the soul, and offers the passing suggestion that he has not managed to integrate the two roles together cleanly.² In this paper, I propose a new reading of (1) what Aquinas means in calling the soul the mover of the body, and (2) how the soul plays a causal role in vital operations in his view more generally. The key to my interpretation will be a close analysis of vital motions as self-motions, which is the basis for Aquinas’s attributing a mover role to the soul in the first place. The study will be divided into two main sections. The first will consider how the soul’s role in the motion of living beings differs from his account of the role of inanimate substantial forms in the motion of non-living beings. The second section will consider the kind of body required for self-motion and then make the case that Aquinas calls the soul a mover because of its relation to this special kind of selfmoving body. My explicit focus in what follows will be on non-human souls both because I take them to offer the best case study with which to approach Aquinas’s general soul theory without getting lost in the particular complications brought by the human intellectual soul, and because animal and plant souls are closer to the contrast case of inanimate

² See Robert Pasnau, “Form, Substance, and Mechanism,” Philosophical Review 113 (2004), 42.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

146   substantial forms, which are not movers of their bodies. Nevertheless, once the general account of how the soul moves the body is made clear, it will be easy to see the implications for the human soul, at least in a general way.

1. Self-Motion and the Grades of Life a. Action among Inanimate Bodies Aquinas holds that souls are principles of self-motion, while inanimate substantial forms are only principles of the motion of other things. In the second half of the paper, I will argue that this self-motive role of souls is the basis for Aquinas’s willingness to call the soul a mover. The goal is just to establish what it is in virtue of which he calls vital operation selfmotion. After a brief clarification regarding the standards of motion and self-motion which Aquinas is applying, I will consider the motions of inanimate bodies as a kind of base case against which we can compare the self-motive operations of life. For Aquinas, motion requires both an agent and a patient, and the motion is always in the patient and from the agent.³ Thus, for something to have a motion is just for it to be moved rather than for it to move. As David Twetten has pointed out, it is easy (especially for the English speaker) to be misled by sentences describing motion which only involve one being. For example, the sentence “the rock is moving” picks out only one item (the rock). Presumably, the sentence seems well-formed because the motion of the rock is evaluated relative to an assumed fixed point of reference, but for Aquinas there is no such neutral sense of motion. Such a sentence would have to be reduced to either “The rock is being moved” (by something else) or “The rock is moving (transitively) something else.”⁴ Cases of self-motion are not exceptions to the rule that motion is always between two things (an agent and a patient), because a ³ For a rigorous account of action, passion, and motion see Gloria Frost, “Aquinas’ Ontology of Transeunt Causal Activity,” Vivarium 56 (2018), 1–36. ⁴ Here I agree with Twetten that inanimate bodies are moved (passively) in cases of natural local motion, in opposition to Weisheipl’s view that Aquinas endorsed a kind of being in motion

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

147

mover–moved relationship can pertain to two aspects of a single substance.⁵ The principal difference between inanimate and animate motion is that the former is fully specified by something extrinsic while the latter is in some way self-specifying. I will clarify what Aquinas has in mind by what I am calling specification momentarily, but the general point is that inanimate bodies are not responsible for their own motion. Accordingly, although they act on each other, they never act on themselves. In De veritate 22.3 Aquinas cashes out this other-oriented action in hylomorphic terms, as a claim about the substantial forms of inanimate bodies: Bodies, however, are moved in a certain way and although one thing is able to move another, none of them is able to move itself . . . Their forms cannot be movers, although they can be a principle of motion – as in the motion of earth, heaviness is the principle by which it is moved, but nevertheless not the mover (motor).⁶

There are two closely related claims about inanimate bodily agency here: (1) Bodies cannot move themselves; and (2) The forms of bodies cannot be movers. I will argue later on that Aquinas takes these claims to be mutually entailing, but the point for now is just that the substantial form of an inanimate body contributes to the motions (of other things), but can never serve as the principle of a body’s self-motion. As examples, consider the three kinds of motions in which fire might be involved. Fire can (i) heat (transitively, e.g., wood) or generate more fire; (ii) move upwards; and (iii) be cooled or quenched. None of these cases are cases of self-motion, and in none of them is the form a mover. We should consider why each is not self-motion. Fire specifies the motion in (i) by determining the end (some hot body in the case of alteration and fire in the case of generation). Here, the heating is not a that was neither moving (transitively) nor being moved. See David Twetten, “Back to Nature in Aquinas,” Medieval Philosophy and Theology 5 (1996), 205–43. ⁵ Here, I am focusing on self-motions which are motions in the strict sense. As we will see, Aquinas also considers immanent operation to be self-motion in an extended sense. ⁶ DV 22.3, quoted at n. 56 below.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

148   motion in fire itself, but rather in the body which becomes hot. Even though the form of fire is responsible for the motion, the motion is not a self-motion because the patient is something other than the fire itself. Cases (ii) and (iii) are motions in fire itself, but the fire does not specify those motions. In case (ii), lightness is an internal principle of being (passively) moved upwards rather than a principle of motion. The upward motion of fire is not a self-motion because the fire is not the mover. Aquinas thinks that properly speaking what moves the fire upwards is its generator (e.g., another previous fire from which it derives its own form); this is an intriguing suggestion which, however, is too complex to discuss here and is not ultimately useful for our purposes.⁷ Likewise in case (iii), the mover is whatever external agent is altering or corrupting the fire. In none of these cases is the form the mover, although it is a principle of motion. It is an active principle in the first case and a passive principle specifying how some motion is received in (ii) and (iii). As I will argue in section 2, the reason that forms are not movers in the examples just given is that forms can only be movers in cases of selfmotion. Crucially, when an inanimate agent acts, it acts through its form.⁸ The immediate active principles of nature are identified by Aristotle in the De generatione: heat, cold, moisture, and dryness. Importantly for our account, Aquinas holds that inanimate bodies are limited in their action to the proper actions of those four active and passive elemental qualities, which he calls the dispositions of material bodies.⁹ For example fire is by nature hot, and because heat is one of the active qualities of nature, fire is

⁷ See discussions in Aquinas see ST 1.18.1 ad 2 [Leon. 5.225–6]; SCG 3.23; De potentia 5.5; De caelo 3.2.7; InPhys 2.1.1; InPhys 8.4.8. ⁸ SCG 2.47 [Leon. 13.377: 5a–7a]: “The principle of any operation whatsoever is the form through which the thing is in act, for everything acting acts insofar as it is in act.” See also SCG 3.69. ⁹ For example QDSC 2 [Leon. 24/II.28: 277–80]: “The form of an element does not have any operation except that which occurs through active and passive qualities which are dispositions of corporeal matter.” I am passing over the somewhat more complicated case of the mixed bodies under the influence of the heavenly bodies. The short story is that their action too is limited to what can be brought about through the action of the elemental qualities (which is in fact also true of plants), but (unlike plants) they do not determine their own motion. Instead, they are disposed to be influenced in their action by the heavenly bodies such that they can bring about effects which they could not bring about on their own.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

149

permanently disposed to heat whatever less-hot substance is near it. This disposition can produce two kinds of effects: alteration (in which a cold body becomes a hot body), and generation (in which non-fire becomes fire). But beyond this disposition to heat, fire has no effect. Even though the accidental forms are the immediate active principles, the substantial form plays the principal role in the agency of inanimate bodies: And therefore it is to be said that because anything whatever acts according as it is in act, as is clear from De generatione book I, it follows that just as the being of the qualities of the elements derives from the principles of their essences, so also the power of acting follows upon these qualities from the power of their substantial forms; for everything that acts in virtue of another makes the likeness of that in virtue of which it acts, just as the saw makes a house by the power of the house in the soul, and the heat of nature makes living flesh from the power of the soul. And in this way, matter is transmutated to substantial forms through the action of elemental qualities.¹⁰

For any case of production where the agent is a body, both the substantial form and its accidents work together in bringing about the effect. Aquinas emphasizes just how widely this principle is supposed to apply by explaining the case under discussion (the production of the elements) to two seemingly dissimilar cases: the production of flesh by bodily heat, and the production of a house by the power of the house in the soul of the builder. In working towards the ultimate goal of asking how the soul’s role in agency differs from that of inanimate substantial forms such that Aquinas calls souls movers, one possibility is already off the table. It cannot be the case that inanimate substantial forms play a mere “metaphysical” role as mere markers of identity, while souls play both this role and an additional role as a principle of agency. Even in the inanimate case, the substantial form plays a role in the agency of the substance. In all cases of bodily action, the immediate active qualities derive their power from the substantial form.

¹⁰ InDeSensu 9 [Leon. 45/2.55: 177–89].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

150   In sum, then, although Aquinas assigns an agential role to the forms of inanimate bodies, this role is limited. But the very fact that he grants any role at all is significant, because for Aquinas and his contemporaries, the agency of inanimate bodies was not to be taken for granted. In fact, in holding that bodies are movers but not self-movers, Aquinas is offering an alternative way to explain the passivity of bodies—a phenomenon propelling the arguments of certain occasionalist interlocutors.¹¹ As Aquinas understands them, these interlocutors denied that bodies could act at all. They held instead that bodies are purely passive receivers of action, and that spiritual/mental beings alone can act. Against his opponents, Aquinas holds that even the lowest body (i.e., inanimate bodies) can act, because even these bodies have forms.¹² Fire heats insofar as it is hot, and it is hot because it is fire, and not through the compulsion of some external agent. (Of course, whatever was corrupted in the production of fire did become first hot and then fire through compulsion. After all, corruption is violent to the nature being corrupted, even though natural to material beings insofar as they are material.) But the forms of inanimate bodies are very limited in their agency. These bodies are fully determined ad extra, and so their forms play no active role in specifying the motions which they undergo, and play only a determinate and invariable role in the motions which they cause other things to undergo.

b. The Self-Motion of Plants With the framework of inanimate agency in place, we can turn to the contrasting cases of living agency. Aquinas considers all vital operations to be cases of self-motion, but they are not all self-motive in the same way. Rather, the higher grades of soul contribute more to their actions than the lower grades contribute to theirs, with the result that they are ¹¹ For Aquinas’s treatments see Sent 2.1.1.4, SCG 3.69, and ST 1.115.1. In his Summa theologiae treatment, Aquinas gives a short history of the view, associating it primarily with the historical figures Avicebron, Avicenna, and Plato, but it seems likely that he had contemporary opponents in mind as well. ¹² See SCG 3.69.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

151

capable of higher grades of activity. Ultimately in my conclusion, I will argue that all soul activities have some features in common which will allow us to say what is proper to the agential role of the soul, and to distinguish souls from both inanimate substantial forms and spiritual substances. But here I want to emphasize the differences between the various grades of soul, and show why each level of animate activity is from the soul in its own way. The lowest level of vital operation belongs to plants, which perform the operations of nutrition, growth, and generation. All of these actions depend on the plant as well as on something external to the plant. In order to understand why Aquinas claims that these motions are selfmotions, then, we should ask how the interactions between living beings and their environment differ from those between inanimate things and their environment. The challenge is particularly sharp in the case of nutrition. On a certain way of parsing the activity, it looks as though nothing categorically different is happening in cases of plant nutrition than is happening in, say, the interaction between fire and a dry log. At the level of the active principles, the story seems the same in both cases. Fire heats through its heat, and is simultaneously affected by what it heats. Likewise, a living body heats its food through its heat, and is simultaneously affected by the food. To understand why Aquinas takes the two cases to be different, we have to detour through his account of nutrition, which draws heavily from Aristotle’s treatment in Book II chapter 4 of De anima. The account there is largely schematic: it is less concerned with the empirical details, and more with correctly identifying which agents are doing what work. The view is that there are three movers, i.e., the soul as principal agent, and two instruments of nutrition acting in the service of the soul. One of these instruments (the heat) is “conjoined” to the agent, i.e., an accident of the agent, while the other (the food) is an external instrument. Both instruments bring with them their own causal powers.¹³ ¹³ It is worth noticing that, following Aristotle, Aquinas is willing to call the soul (as opposed to the composite) the agent here. I do not take this to undermine the dictum actiones sunt suppositorum. The point in these contexts is not to identify the proper subject of the action, but rather which form is the principal active principle. For example InDA II.8 [Leon. 45/I.101: 138–43]: “It is necessary that all food must be cooked, which happens in a certain way through

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

152   Aquinas sets out the relationship between the three movers in his question on life in God at ST 1.18.3: For first of all, the end moves the agent; but the principal agent is that which acts through its form. And it [the principal agent] sometimes acts through some instrument, which does not act by the power of its form, but by the power of the principal agent, the instrument of which contributes only the execution of a motion.¹⁴

Both the instrument and the end for which the action is being done can legitimately be called movers, but the principal mover (the agent) is that which acts through its form.¹⁵ The claim is clear enough, but in this text Aquinas does not give much guidance about how to tell that it is the plant which acts through its form, and not either of the instruments. What special marker is there in the act of nutrition such that we can tell that it is the plant, and not the food or the heat, which principally acts?

fire. Whence fire is in a certain way operative in nutrition and consequently in growth. But not as a principal agent, for this is the soul, but rather as a secondary and instrumental agent.” See also InDA II.8 [Leon. 45/I.101: 148–62 . . . 165–84]: “The principal in any action is that by which the terminus and nature (ratio) is imposed on it. This is clear in artificial beings because the terminus and nature of a box or a house is not imposed by the instruments, but from the art itself. For the instruments are related indifferently as cooperating with this form or quantity or that one. For a saw is as such fit to the cutting of wood either for a door, or a bench, or a house, and in whatever quantity. But that the wood be thusly cut, that it be fit to such a form and to such a quantity, belongs to the power of the artist. But it is clear that in everything according to nature there is a certain limit, and determinate proportion (ratio) of magnitude and growth . . . . (Not all men are the same size, but there is rather some large quantity above which the species of humans does not extend and another small quantity below which man is not found.) Therefore that which is the cause of the determination of magnitude and growth is the principal cause of growth. This however cannot be the fire. For it is clear that the growth of fire does not extend to a determinate quantity, but extends infinitely (in infinitum), if there is infinite matter for combustion to be found. It is therefore clear that fire is not the principal agent in growth and nutrition, but rather the soul is. And this is reasonably granted, for the determination of quantity in natural things is from the form which is the specific principle rather than from the matter. Now the soul is related to the elements which are in the body of a living thing as form is to matter. Therefore the terminus and nature of magnitude and growth is from the soul rather than from the fire.” ¹⁴ ST 1.18.3 [Leon. 4.228]. ¹⁵ The end is a mover in the order of final causality rather than agent causality, and so its contribution can be set aside in our analysis. The story will be different when we come to the case of animals, in which the agent cognizes the end.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

153

Aquinas is here relying on a more general principle of causality, that every agent makes its patient like itself (omne agens agit sibi simile).¹⁶ We can tell which form is acting by considering the likeness of the product to the original. In cases of ordered causality, the end will always be similar (or at least more similar) to the primary agent rather than to the instrumental one.¹⁷ In nutrition, the living thing introduces its own likeness into the food which it assimilates. The heat also introduces its own likeness, but we can tell that its contribution is instrumental because rather than producing the usual effect of heat (dissolving), it produces flesh.¹⁸ Thus the heat is adequate as an instrument but it is not up to the task of setting the end,¹⁹ because the nutrition-action does not terminate in the production of heat, but rather in the maintenance of the whole body. In our example, the terminus of the honeysuckle’s nutritive motion is its continued existence. Likewise, in growth, the terminus is the production of more honeysuckle. Thus Aquinas says in his discussion of nutrition in the De anima that when a plant (or an animal) takes in

¹⁶ The slogan appears at least 25 times in Aquinas’s opus, in a variety of contexts. Of course, Aquinas also endorses the related view expressed in proposition one of the Liber de causis that the first cause produces its effect more powerfully than that of the secondary cause. ¹⁷ SCG 3.149 [Leon. 14.439: 16a–21a]: “The instrumental agent does not dispose to the perfection induced by the principal agent except insofar as it acts by the power of the principal agent: for the heat of fire does not prepare the matter to be the form of flesh more than any other form except insofar as it acts by the power of the soul.” Earlier, in SCG 3.69 [Leon. 14.201: 22a– 26a] Aquinas gives the somewhat softer formulation that the instrumental cause does produce an effect like itself, but that the effect is more like the principal cause than the instrumental cause: “What acts through the power of another produces an effect similar not only to itself, but especially to that in the power of which it acts, as from the action of an instrument there is the likeness of the form of the maker in an artifact.” ¹⁸ SCG 2.89 [Leon. 13.542: 26b–39b]: “If they are ordered to one another, it is necessary that they have one effect. For the cause acting first acts in the second agent’s effect even more strongly than the second agent itself. Whence we see that an effect which is brought about through an instrument by a principal agent is more properly attributed to the principal cause than to the instrumental. For it happens that sometimes the action of the principal agent achieves something in the operation that is not achieved in the action of the instrument: as when the vegetative power produces the species of flesh, to which the heat of fire, which is its instrument, is not able to lead, although it is at work in the disposing of it by dissolving and consuming.” ¹⁹ For example, Quodlibet VIII, q. 2 a. 1 co. [Leon. 25/1.56: 44–53]: “For the agent by itself suffices in a certain way for the inducing of its form in the patient, just as fire is sufficient of itself for heating. But in a certain way, the agent does not suffice by itself for the inducing of the form in the patient, except through the supervenience of some other agent; as the heat of fire does not suffice for the completion of the action of nutrition except through the power of nutritive soul; whence the power of the nutritive soul is the agent principally, and the heat of fire is the instrumentally.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

154   nutrients, there is a coincidence between the agent and the patient: “That which is fed and acts in the food is also the terminus into which the food is changed.”²⁰ The causal power of the instrument is still required, but as an instrument (which does not set the end) rather than as the principal agent (which does). Interestingly, despite this analysis, Aquinas recognizes that the plant case seems similar enough to the inanimate action case that the natural philosopher might not notice the role of the soul at all, at least if he is focused on the immediate active principles: [Aristotle] shows that the works of the vegetative power are from the soul, which was necessary because since the active and passive qualities contribute to their operations, it may seem to someone that they are from nature and not from the soul. And it is especially so because in plants life is obscure and hidden.²¹

The point here is programmatic to the project of natural philosophy. At the level of the mechanism (the active and passive qualities), there is no difference between this lowest level of animate agency and inanimate agency. That is, the active principles of the nutritive soul are no different from those of inanimate agents. But as we saw before, the immediate active principles by themselves cannot act and are never sufficient for producing an effect.²² Accordingly, the natural philosopher must go beyond the enumeration of the active principles if the explanation is to be complete. Because the action terminates in the product of plant (rather than anything else), the conclusion to be reached is that the plant sets the end through its own remote causal principle (the soul) using the heat and the food as instrumental causal principles. Having clarified what Aquinas takes the soul to contribute in the nutritive operations of a living body, we are ready to consider what he

²⁰ InDA II.9 [Leon. 45/I.104: 92–4]: “That which is fed and acts in the food is also the terminus into which the food is changed.” ²¹ InDA II.7 [Leon. 45/I.96: 54–9]: “[Aristotle] shows that the works of the vegetative power are from the soul, which was necessary because since the active and passive qualities contribute to their operations, it may seem to someone that they are from nature and not from the soul. And it is especially so because in plants life is obscure and hidden.” ²² See n. 10 above.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

155

thinks is distinctive about vital motion. In contrast to inanimate things, which do not contribute at all to their own motions, the living body moves itself in nutrition: It nourishes itself. But it moves itself only with respect to what Aquinas calls the “execution” of the motion: There are therefore found certain things which move themselves, not with respect to the form or an end (which is in them by nature) but only with respect to the execution of the motion. But the form through which they act and the end on account of which they act are determined for them by nature. And such things are plants, which move themselves according to a form put into them by nature with respect to growth and diminution.²³

Aquinas is deliberate in his use of the term “execution” here. As executors of their motions, plants have forms within themselves (e.g., heat and ultimately the substantial form or soul) which are the active principles of the specific motion which the plant undergoes (e.g., growth). Unlike in the case of the inanimate moved body, the motive forms are within the plant rather than in an external agent. For example, in the inanimate case of fire heating water, the water does not have the immediate active principle (heat) before it is heated. Nor does it have the ultimate principle (fire-form), which gives rise to heat, although it is in remote potency (through substantial change) to becoming fire in virtue of its matter. But the plant has the forms which are the active principles of its own growth, nutrition, and reproduction. The point here is just that the forms which are the active principles are internal; the sense in which any of these forms (especially the substantial form) is the mover will become clear in the following section. Yet despite being executors of their own motion, plants have both the form through which they act and the end on account of which they act “determined for them by nature,” by which Aquinas means that they play no role in selecting the end towards which they work. Consider the motion of a honeysuckle creeping up a trellis. Aquinas’s point is that the honeysuckle implements (executes) an action of twining and climbing,

²³ ST 1.18.3 [Leon. 4.228].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

156   but the specific, honeysuckle-type pattern of that action is already set for it (determined) by its nature. Like fire at the level of the elements, the honeysuckle has an internal active principle. Unlike fire, the honeysuckle uses this principle instrumentally in its own motion (growth according to a specific pattern). Thus, the honeysuckle itself is the primary mover in its growth according to this particular pattern (up the suitable object, i.e., the trellis), even though many external conditions (the nutrients, the trellis etc.) are required. The honeysuckle is the executor of its motion insofar as the motion is brought about by an internal active principle, but this falls short of the self-determining self-motion, which as we will see, is the kind proper to cognitive beings.

c. The Self-Motion of Animals and Humans Aquinas considers cognition and local motion in animals to be more perfect self-motions than the kind of self-motion proper to plants. Like plants, animals contribute through their substantial form to the execution of the motion which they undergo, but they also play a role in determining themselves to that motion. Nature does not entirely specify their action, and leaves it up to each animal to direct itself to its end relative to its environment: But higher up there are things that move themselves not only with respect to the execution of the motion, but even with respect to the form which is the principle of the motion, which it acquires for itself. And such things are animals, the principle of the motion of which is a form not put in them by nature, but received through sense. Whence insofar as they have a more perfect sense, to such a degree do they move themselves more perfectly. Those that only have the sense of touch, move themselves only by a motion of dilation and constriction, such as oysters, [and] barely surpass the motion of plants. But those which have the sensitive power perfectly not only up to the cognizing of conjoining and touching, but also even to the cognition of distant things, move themselves at a distance by progressive motion. But although such animals receive through sense the form which is the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

157

principle of motion, they are nevertheless not able to predetermine themselves to the end of their operation of their motion; but this is given to them by nature, by whose instinct they are moved to some action through the form apprehended in sense.²⁴

Animals move themselves in two ways, through cognition and through progressive motion (i.e., local motion from one place to another). There are two assumptions here which are necessary to understand why Aquinas believes that in both cases the animal determines its own motion through its form. The first (and easier) of these two assumptions is that cognition is more fundamental that progressive motion because the latter is based on the cognition of the end towards which the animal moves itself. The second is that cognition is a self-motion, and this will require some unpacking. As we just saw, the reason Aquinas understands animal local motion to be self-motive is that it depends on sensation. The point here is that sensitive beings not only have within themselves a formal principle by which they execute their own local motion, but also an acquired formal principle by which they determine their motion toward this or that object in their environment, i.e., the sensible species. For example, in order for a giraffe to feed on acacia leaves, it must be aware of the tree, of its own body, and of the leaves as desirable. Were the giraffe to move towards the acacia tree without cognition, it could not be internally determining its own local motion; rather, the direction of its own local motion would be set from an external source, e.g., in the manner of elemental motion (e.g., perhaps it is being driven in a truck by a zookeeper).²⁵ The point is that in the local motions of the giraffe there is some sensible and imaginative species of acacia present in it, whereby its acts of desiring can be directed acacia-wards. This self-directing is not found in plants because they have by nature an externally determined

²⁴ ST 1.18.3 [Leon. 4.228]. ²⁵ But possibly Aquinas would allow that the giraffe could be moved in the manner of plant motion (were some external agent to take advantage of its internal principle of self-execution, e.g., by stimulating its muscles such that it is compelled to move). Of course, Aquinas does not deal with this sort of hypothetical. At any rate, such a motion would fall short of progressive local motion.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

158   intrinsic principle of the execution of their motions. The difference between the two kinds of self-motion is that in the lower kind of selfmotion, the organism merely executes its own motions (as opposed to undergoing motions caused by some external agent, as happens in the motion of inanimate bodies), while in the higher kind of self-motion, the organism both executes and intentionally directs its own motions. The more difficult and fundamental point is that animal cognition is itself a self-motion. The difficulty here is that Aquinas is sometimes held to have a purely passive account of sensation. More precisely, because sensation (as well as other cognitive animal actions and human intellectual cognition) is something caused by an extramental object, seemingly it cannot proceed from within. I cannot here give a full account of Aquinas’s sensation theory, but I can at least briefly suggest the way in which the animal ought to be considered an agent of its own sensation. As I have pointed out elsewhere, Aquinas distinguishes the reception of the sensible species (a purely passive reception of form) from the operation of sensation itself.²⁶ The latter is an action, and specifically an immanent operation. The operation of sensation immediately follows upon the reception of the sensible species, but the two are distinct, with the received species serving as an instrument of sensation.²⁷ The likely reason why this active aspect of sensation has gone largely unnoticed despite sustained attention to sensation in Aquinas is that scholarly focus has been on whether and how Aquinas holds sensation to be immaterial, and not on how sensation ought to be understood in comparison to other motions. This is not the place for a detailed account of sensation; the point for now is only that in the case of sensation, the extramental object is sufficient to cause the sensitive animal to receive the form, but not sufficient for the performance of the operation of sensation itself. And it

²⁶ David Cory, “Agency and Materiality in Aquinas’s Soul Theory” (PhD dissertation, The Catholic University of America, 2018), 112–20. Jean-Luc Solère has also independently noticed the active character of perception in Aquinas in his address to the Society for Medieval and Renaissance Philosophy at the 2019 Eastern American Philosophical Association, “Perceptual Activity According to Thomas Aquinas.” ²⁷ Aquinas says that the spiritual immutation is required for the operation at ST 1.78.3 [Leon. 5.254]: “But for the operation of sense, a spiritual immutation is required, by which the intention of the sensible form comes to be in the organ of sense.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

159

is the operation which Aquinas takes to be self-motive insofar as it is immanent, i.e., the action remains within the agent. Although all cognitions are immanent operations, they are not all selfmotive to the same degree. All created beings have something given to them by nature, e.g., the first principles or reason and the inclination to the ultimate end.²⁸ In addition, bodily cognizers depend on extramental objects for their species. Within the genus of sensitive beings, different animals determine their own motions to varying degrees. The more perfect the cognition, the more intrinsic the motion. The lowest animals respond only to touch (Aquinas gives oysters as his example), and so determine themselves to the lowest degree. Those with all the senses determine themselves to the highest degree.²⁹ Intellectual animals have the power to relate means to ends, and thus enjoy an even more perfect kind of self-motion: Whence above these animals there are those which move themselves even with respect to the end which they are able to predetermine. For this is not possible except through reason and intellect, to which it belongs to know (cognoscere) the relation (proportionem) of the end and the means (quod est ad finem), and to order one to the other. Whence the more perfect mode of living belongs are those which have understanding, for these more perfectly move themselves.³⁰

As before, at each successive level of perfection, less and less is supplied by nature, and more is accomplished by the agent’s acquisition of various cognitive perfections.³¹ Of course, even at the level of intellectual ²⁸ Aquinas makes this point about the human intellect at ST 1.18.3 [Leon. 4.228]: “But although our intellect moves itself (se agat) to something, there is nevertheless something given to it by nature, such as the first principles, about which it is not possible to be related any other way, and the ultimate end, which it is not possible not to will. Whence although in a way it moves itself to something, nevertheless it is necessary that in a way it be moved to something by another.” ²⁹ See also ST I.78.1 [Leon. 5.251] for a parallel distinction between perfect animals, which move themselves according to place and imperfect animals, which do not. There again Aquinas turns to sea-animals for his example. ³⁰ ST 1.18.3 [Leon. 4.228]. ³¹ See also SCG 2.47 [Leon. 13.377: 5a–15b], in which Aquinas adds more detail to the higher divisions of self-movers, namely those with appetite. His goal in that text is to establish “dominium” or mastery as a technical category signifying a certain kind of completeness to the self-motion belonging to creatures with will, and correspondingly, the text de-emphasizes the self-motive character of plants. Despite this difference in emphasis, the criterion used to

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

160   creatures, there is still something supplied by nature, namely the first principles which the intellect cannot doubt, and the universal inclination to the good, to which the will is directed: But although our intellect moves itself (se agat) to something, there is nevertheless something given to it by nature, such as the first principles, about which it is not possible to be related any other way, and the ultimate end, which it is not possible not to will. Whence although in a way it moves itself to something, nevertheless it is necessary that in a way it be moved to something by another.³²

At each higher level of self-motion, then, the active principle is intrinsic in a higher way, with the result that each level of self-motion is more internally directed than the previous. Moreover, each higher degree of agency more perfectly exemplifies the independence that characterizes self-motion, in that less and less is given to it by nature, while more and more comes from the agent’s own cognitive perfection. At no level of created being, however, is the source of motion entirely intrinsic, because some first principle of motion (e.g., the universal inclination to the good, in the case of the will) is always given by nature. How far does this trajectory of increasingly perfect self-motion extend? Although Aquinas does not approve of the formulation, there is a certain sense in which perfect self-motion belongs to God. In a reply to an objection at Summa theologiae 1.18.3 ad 1, Aquinas explains that Plato’s doctrine that God is self-moving should be understood to indicate that God performs an act of understanding because speaking imprecisely, intellectual operations can be called motions. Speaking precisely, God is an unmoved mover rather than a self-mover, but Aquinas is willing to employ a broader notion of “self-motion” which captures both immanent cognitive operations as well as motions in the strict sense. It is this broader notion of self-motion that Aquinas takes to be the essential

distinguish the grades is the same: self-movers are ranked based on the way in which they contain the principles of their own motion within themselves. ³² ST 1.18.3 [Leon. 4.228].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

161

feature of life, and which will allow him to say that God is most properly called living. What we have seen, then, is that living beings all move themselves according to a broad sense of self-motion, which includes both motions in the strict sense as well as immanent operations. While the active principles of motion are wholly external to the motions of inanimate bodies (because they only move each other), the active principles of living beings are all intrinsic in some way. The distinctions within the genus of living beings are based on the various ways in which agent principles are intrinsic, i.e., by having an intrinsic principle of execution or an intrinsic principle of execution along with an intrinsic principle of determination.

2. The Diversity of Parts and the Grades of Physical Life We have seen that something is self-moved if it has the active principle of its own motion through its form. We have also seen that this is a deliberately general approach to self-motion intended to allow for a range of self-directed activities, including some which are not motions at all (but rather immanent operations). We have not yet seen why the fact that the soul supplies different kinds of active principles from inanimate substantial forms means that it ought to be called a mover. What I want to argue is that the mover-role of the soul stems from the need for bodily complexity in self-motion. Self-movers require a certain degree of diversity of bodily parts, and the soul’s moving of the body is tied to its ability to supply the body with parts which are distinct from one another. This requirement for diverse bodily parts is of course connected with Aristotle’s dictum in Physics 8 that a self-mover is always divided into two parts (one of which moves the other), but as we will see, Aquinas interprets this passage in a peculiar way. To understand the soul’s role as mover, then, we must first understand the kind of body which is capable of such self-motion, and then ask why the distinctive character of such a form-principle leads Aquinas to call it the mover of the body.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

162  

a. The Diversity Proper to Living Beings In order to show that the soul’s power to move the body is manifested in the diversity of bodily parts, we must first see the connection Aquinas draws between bodily complexity and life. What distinguishes the bodies of living beings from inanimate bodies is that the former have distinct parts, as Aquinas explains in QDDA 9: But it must be considered that the grade of forms in the power of operating corresponds to their grade in the perfection of being, because an operation belongs to something existing in act (in actu). And therefore, to the extent that some form is more perfect in giving being, to that extent does it have greater power in operating. Whence more perfect forms have more operations and a greater diversity than less perfect forms. And therefore, it is that the diversity of accidents suffices for the diversity of operations [found] in less perfect things. But in more perfect things a greater diversity of parts is required; and it must be as much greater as the forms are more perfect. For we see that to fire belong diverse operations according to diverse accidents, such as to be carried upwards according to lightness, to heat according to heat, and so on for the others. But still, fire has all of its operations according to every part of itself. In living beings, by contrast, which have more noble forms, different operations are given to different parts: e.g., the operations of the root, the trunk, and the branches differ from each other. And insofar as a living being is more perfect, to that degree and on account of its greater perfection are to be found a greater diversity in the parts. Whence since the rational soul is the most perfect of material forms, in man is found the greatest distinction of parts, on account of [man’s] diverse operations.³³

Inanimate things have substantial forms that are able to give rise to diverse accidents. For example, fire is both hot and light, and as the text goes on to explain, this diversity of proper accidents allows fire to be

³³ QDDA 9 [Leon. 24/I.81–2: 250–78].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

163

moved in diverse ways. It is able to heat in virtue of heat, and rise in virtue of lightness. But although the elements and the inanimate bodies are diverse in a certain limited way, they are not diverse at the level of parts. Both lightness and heat inhere in fire, but this does not undermine the homogeneity of their parts.³⁴ Speaking precisely, Aquinas says that in these beings, the form is the same both in whole and in part.³⁵ Animate beings, in contrast, enjoy a complexity of parts corresponding to their grade of nobility. The need for organs, of course, picks up on Aristotle’s definition of soul as the first act of an organic body.³⁶ In this respect, artifacts are less like inanimate beings and more like living beings in that they have diverse parts juxtaposed in order to perform some function, although of course artifacts do not have this diversity of parts by nature. The point is not that it turns out empirically true that things with more noble operations have more diverse parts, but rather that diversity of organic parts is the bodily manifestation of a certain grade of soul. Souls with more powers and more diverse powers have bodies with more parts, as Aquinas explains in De spiritualibus creaturis: In those corruptible things, imperfect forms which are weaker in power have fewer operations, for which the dissimilarity of parts is not required, as is clear in all inanimate bodies. But a soul, since it has a higher and more powerful form, is able to be the principle of diverse operations, the execution of which requires dissimilar parts of the

³⁴ See ST 1.11.2 ad 2 [Leon. 4.110]: “there are two kinds of wholes, one of which is homogeneous, which is composed of similar parts, the other of which is heterogeneous, which is composed of dissimilar parts. In a homogeneous whole, the whole is constituted from parts which have the form of the whole. For example, every part of water is water, and so is the constitution of a continuum from its parts. In a heterogeneous whole, each part lacks the form of the whole: for no part of a house is a house, and neither is some part of a man a man. And such a whole is a multitude.” ³⁵ See QDSC 4, especially what is cited in n. 37 below. In that question, Aquinas asserts in a qualified way that non-living bodies are homogeneous. He contrasts the forms of living bodies, which require dissimilar parts, with those of non-living bodies, which “have matter which is similar or mostly similar in whole and in part (habent materiam similem aut fere similem et in toto et in parte).” ³⁶ Aquinas gives a helpful example of the sort of thing he has in mind in InDA II.1 [Leon. 45/ I.72: 334–57] where he calls attention to a plant which uses its leaf to protect its pericarp and the pericarp to cover its fruit. The larger context is also interesting in that it explains that to have “organic parts” is just to have parts which are specified by diverse operations.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

164   body. And therefore, every soul requires a diversity of organs in the parts of the body to which the act belongs, and the greater the diversity, the more perfect the soul will be. In this way, therefore, the forms of lower things perfect their matter uniformly, but souls [perfect their matter] in multiform ways, such that the integrity of a body whose first act is primarily and per se a soul is constituted from dissimilar parts.³⁷

In contrast to inanimate substantial forms, souls perfect their matter in a non-uniform way. And they do so in such a way that the parts of the body are able to act on one another. Heterogeneous parts enable the soul to harness the instrumental power of contrary qualities by arranging them harmoniously, and this in turn allows for specific parts of the body to have specific functions. The need for diverse parts in order to have self-motion becomes obvious upon considering the inability of simpler bodies to move themselves. Homogeneous bodies cannot be self-movers because motion arises through contact of opposed parts, as Aquinas explains in his commentary on Physics 8: That no continuum moves itself is proved thusly. The mover is related to the moved as agent to patient. But if the agent is to be contrary to the patient, it is necessary that what is naturally able to act be divided from that which is naturally able to be affected. Accordingly, those things which are not in contact with each other, but are entirely one and continuous in quantity and form, are therefore not able to be affected by one another. Thus, it follows that no continuum moves itself, but it is necessary that the mover be divided from that which is moved, as is clear when inanimate things are moved by animate ones, e.g., a stone by a hand. Whence also in animals which move themselves there is a connection of parts rather than a perfect continuum. For in this way one part is able to be moved by the other, which is not found in the heavy and the light.³⁸

³⁷ QDSC 4 [Leon. 24/II.52: 196–210].

³⁸ InPhys 8.4.7 [Leon. 2.389].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

165

Motion arises through contact of opposed parts, but in a homogeneous body there are no parts with contrary qualities in contact with one another. In the case of the vegetal soul’s operations, however, diverse organs permit the parts of the body to act on one another through contraries. Again, in the case of bodily cognitive operations, the bodily dispositions required by one sense power differ from those required by the other. Generally, a diversity of organic parts is the necessary condition for self-motion, and a more perfect self-mover requires a more complex arrangement of parts than does the body informed by a less powerful form.³⁹ Bodily complexity, in other words, tracks the hierarchy of bodily agents.

b. An Objection from Divine Simplicity To a reader familiar with Aquinas’s arguments for divine simplicity, it might seem strange that the nobility of form tracks the complexity of bodily parts, because this move seems to set up simplicity as a sign of ontological inferiority, and complexity as a sign of ontological perfection. Indeed, Aquinas explicitly blames a body’s simplicity and the ignobility of its substantial form for its inability to move itself, and places more complex, heterogeneously-organized bodies higher on the scale of being.⁴⁰ It might seem at least unfitting that bodily complexity is a sign of life and that higher grades of life enjoy a greater bodily complexity than lower ones, given that life is preeminently said of God (who is simple), and also said of separate substances (which have no bodily parts). Aquinas is not unaware of the tension. In Summa theologiae 1.3.7, he cites the nobility of creaturely complexity as the basis for an objection against divine simplicity. The argument is straightforward: Whatever is better must be attributed to God. We see from creatures that ³⁹ Paul Hoffman notices, but does not explain this connection between the diversity of corporeal parts and its relative immateriality. “The less similar the parts of a body, the less the form has extension in matter.” Paul Hoffman, “St. Thomas Aquinas on the Halfway State of Sensible Being,” Philosophical Review 99 (1990), 82. ⁴⁰ See DV 22.3, n. 56.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

166   composition of parts is better than simplicity. Therefore composition (rather than simplicity) must be attributed to God.⁴¹ Aquinas’s strategy in responding seems unsatisfactory at first. He says that perfection in God is different than perfection in created things. Divine perfection consists in simplicity, but it is not possible for creatures to have all perfection in a simple way, and so they must have parts.⁴² As it stands, this response seems like a reformulation of the problematic assertion that gives rise to the problem in the first place, or a bald assertion that God is different (with the implication being that therefore there is no problem). What is assumed in this text but explicit elsewhere, however, is that there are two kinds of simplicity in play, one pertaining to something’s essence and the other pertaining to the scope of its power. (How this distinction exactly relates to bodily parts, I will explain shortly.) The distinction is brought out clearly in Aquinas’s response to the 14th argument of QDDA 8. The objection is that because homogeneous bodies approach simplicity more than bodies with diverse parts, it follows that more noble souls should be united to more homogeneous bodies. Aquinas responds: “Although a soul is simple in essence, it is nevertheless multiple in power, and to the extent it is greater [in power], so much more perfect will it be. And therefore, it requires an organic body which is of dissimilar parts.”⁴³ Something which is simple in essence, i.e., uncomposed, can be multiple in power, i.e., enjoy a wide scope of action and extend to more things in its action. The kind of simplicity that indicates nobility pertains to the being of a thing, while the kind of simplicity which indicates a limitation (at least in material beings) pertains to something’s powers and operations. What makes the case of animate bodies confusing is that extension in quantity and the multiplicity of bodily parts mixes features of both kinds of simplicity. Bodies, insofar as they are extended and composed, are

⁴¹ ST 1.3.7 arg. 2 [Leon. 4.46]: “Further, everything that is better must be attributed to God. But, for us, composed things are better than simple things, as the body is a mixture of the elements, and the elements are its parts. Therefore, it should not be said that God is in every way simple.” ⁴² See n. 48 below. ⁴³ QDDA 8 ad 14 [Leon. 24/I.73: 460–3]. See also Sent 2.1.2.5 ad 3 [Mand.-Moos 2.56] and QDDA 10 ad 15 [Leon. 24/I.94: 383–91].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

167

spread out in their being, which means both that they are more susceptible to the action of other bodies, and that their power is divided from itself. This makes them weak in comparison to separate substances, which immediately act in whatever way they will, whenever they will. On the other hand, through being spread out, animate bodies attain a wider scope of action than those bodily things without any organic parts. Thus, among bodies (inanimate) homogeneous bodies are the weakest because their operation extends to the narrowest range of objects. More perfect beings, in contrast, extend in their actions to a greater range of objects, and in the case of bodily beings, this greater range of operation requires a greater diversity of parts. But although these beings are not simple in quantity, they are still “simple in essence,” because by extending their power to more things, they bring a wider range of objects under the power of a single essence. For example, an animal’s operations extend to all sensible beings (insofar as it can sense), while a plant’s operations extend only to its own being (insofar as it can nourish itself and grow). Insofar as the animal’s simple essence extends to more objects, more power is unified under a single essence. The animal’s greater nobility in comparison to a plant is not a function of it having fewer parts than the plant (it has more parts!), but rather of having more power unified by a single essence.⁴⁴ What holds true of the complexity of parts holds true of quantity in general: assuming equal intensity of form, something which is lesser in quantity is weaker than something which is more spread out, because it is in contact with less than what is extended in quantity. But the reverse is true with respect to power, as Aquinas explains in De veritate 8.15 ad 6 sc:

⁴⁴ See SCG 2.72 [Leon. 13.457: 11b–24b]: “But it is not unfitting that the soul, although it be a certain simple form, is nevertheless the act of diverse parts. For matter is always fitted to form in a way that is fitting to matter. To the extent that some form is more noble and simpler, to that degree is it greater in power. Whence the soul, which is the most noble among lower forms, although simple in substance, is multiple in power and in its many operations. Whence it requires diverse organs for the completion of its operations, of which the diverse powers of the soul are said to be the act, as vision is the act of the eye, hearing of the ear, and so on of the rest. For this reason, perfect animals have the greatest diversity among their organs, but plants have the least.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

168   The more simple something is, the more it extends by power to many things but by dimensive quantity to fewer. And thus, the extension of dimensive quantity to many indicates composition, while the extension of power indicates simplicity.⁴⁵

Bodies affect other bodies by contact of quantitative extremities, and thus any increase of quantity allows an active principle to extend further. But this spreading out also means that their power is diluted among the various parts into which they are spread. Of course, extension in quantity also means that a body is more subject in more parts to the actions of other bodies, as Aquinas makes clear using the example of fire, in SCG 3.69. Presuming equal intensity, a greater quantity of fire is able to burn more than a smaller quantity is.⁴⁶ But because of the symmetry between matter and form, any spreading out of corporeal power is also a spreading out into matter. Fire which is spread into a larger area is both active in a larger area and susceptible in a larger area. Thus, there is a kind of vulnerability or weakness accompanying an extension in quantity.⁴⁷ Correspondingly, among bodies composed of matter and form, larger bodies are both more powerful and more susceptible than smaller ones, but in different respects. Being extended in quantity as such is a weakness, but for any given extended body, quantity adds power. The weakness is not the quantity itself but the susceptibility (via matter) which quantity entails. Given that all material beings are composed of matter and form, and that they are all equally susceptible to contraries, they all equally suffer from the limitations accompanying composition. But because they also act through their extended bodies, greater complexity affords some of them greater power than others.

⁴⁵ DV 8.15, ad s.c. 6 [Leon. 22/2. 267: 385–90]. ⁴⁶ SCG 3.69 [Leon. 14.201: 53a–8b]. ⁴⁷ Nevertheless, it is not extension into quantity per se which gives rise to the weakness, but rather the matter which always accompanies quantity. See SCG 3.69 [Leon. 14.201: 44a–49a]: “It is not true that quantity impedes the action of forms except accidentally, insofar as all continuous quantity is in matter. But because a form which exists in matter has less actuality, it consequently has less power in acting.”

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

169

We are now in a position to understand why the complexity of bodily parts indicates the perfection of a corporeal body (at least with respect to agency) but why simplicity is a perfection in God: With us, composed things are better than simple things because the perfection of the goodness of creatures is not found all together in one thing, but in many things. But the perfection of divine goodness is found all together in one thing.⁴⁸

In the material world, being spread out into parts is the mechanism by which bodies have power over other bodies. But because this being spread out into parts also makes bodies susceptible to the actions of other bodies, the most powerful agents are not at all bodily, and can affect lower agents without themselves being affected.

c. How the Soul Moves the Body Now that we have seen that in bodies (as opposed to other kinds of life) increased organizational complexity is correlated with an increase in power, we are ready to propose an answer to the central question: What does Aquinas intend by calling the soul a mover of the body in addition to being its form? The first challenge is to bring this motor-role of the soul into line with the technical framework for self-motion that Aristotle gives in book 8 of the Physics, and that Aquinas adopts. The stipulation in Physics 8 is that a self-mover is always divided into two parts, an unmoved mover and a moved thing which does not move (itself). Thus, in SCG 2.65 Aquinas says: As was proved above, and is proved in Physics 8, everything moving itself is composed of two [things], of which one is a mover and not moved, and the other is moved. But an animal moves itself: the mover in it is the soul, while the moved is the body. The soul therefore is a mover which is not moved.⁴⁹

⁴⁸ ST 1.3.7 ad 2 [Leon. 4.47].

⁴⁹ SCG 2.65 [Leon. 13.435: 28a–2b].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

170   Aquinas’s interpretation is ambitious, given that Aristotle has in mind the one and eternal unmoved mover rather than the soul (though due to Aristotle’s dialectical approach at the beginning, the scope of some of his claims is ambiguous). In his commentary on that passage, he takes advantage of the ambiguity in some of Aristotle’s formulations in order to interpret at least some of the references to the ‘unmoved mover’ to refer indifferently to bodies and non-bodily movers (i.e., movers with magnitude and movers without magnitude).⁵⁰ The claim that the soul moves the body cannot be dismissed as a case of Aquinas using ‘mover’ liberally, because even in the most technical discussion of bodily selfmotion from the Physics, he interprets the text with a view to accommodating the soul’s motor-role. This creates a problem, at least if the two roles are really distinct. If the soul is a mover of the body in addition to being its form, Aquinas seemingly ought to have a story to tell about how the soul acts on the body. Indeed, if the soul acts on the informed body in the way that one body part acts on another, the “motor” role of the soul must overcome the kinds of interface problems facing cruder forms of dualism. Crucially, however, Aquinas does not think that the soul’s moving of the body is really different from its informing the body. Rather, he holds that they differ in account (ratio) only: To the seventh it should be said that the governing of the body pertains to the soul insofar as it is the mover, not insofar as it is the form. And although those things whereby the soul governs the body are necessary for its being in the body, such as the proper dispositions of this matter, it nevertheless does not follow from this that there be the same account of governing and of formal union. For just as the soul as mover and the soul as form are the same in substance but different in account, so also

⁵⁰ According to Aquinas, InPhys 8.5.11 [Leon. 2.405–6] there is a stage in Aristotle’s text at which he has not yet made clear that the “unmoved mover” is the one, eternal mover of all. Because Aristotle might be referring dialectically to other cases in which something is an unmoved mover in an analogous sense, Aquinas interprets the passage broadly. The text is very sparing with examples, but one might consider the case of walking in which some part of my body must remain stationary in order for the rest to move forward. In this case, the stationary part is the moving part, which is divided from the moved part.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

171

what is necessary for the formal union and for the governance is the same, but not the same in account.⁵¹

The key claim for my purposes is that although the two roles of the soul—as form and as mover—are distinct (at least in ratio), what is required by the soul to perform these two roles is the same. Here we can draw on some helpful analogies. Consider the case of an axe with two “roles,” namely crushing and cutting. This case is unlike that of the soul because the two roles require different qualities of the axe, namely weight (for crushing) and sharpness (for cutting). But now consider the contrasting case of a musically obsessed town which has as a rule that it will elect its mayor solely on the basis of which candidate is the best pianist. In this case, to be elected mayor differs in ratio from being the best pianist, but what is required for one is identical with what is required for the other. In the case of the soul, the point is that there is no separate causal mechanism by which the soul governs (administrat) the body. Everything that it needs to govern the body it has by informing it. Now this response is both clever and problematic. On the one hand, cleverly, it allows Aquinas to embed the motor-role of the soul into the same metaphysical framework common to all corporeal substances. Every natural body, animate ones included, has some substantial form by which it is the kind of thing that it is. By arguing that the soul’s role as form and its role as mover differ in ratio but not in substance, he can fit the soul into the purview of his general theory of the role of form in bringing about operations. Just as fire has heat and therefore heats in virtue of its substantial form, so an animal has heat and therefore nourishes itself in virtue of its soul. But on the other hand, problematically, his response makes it difficult to see why only some forms, i.e., souls, are movers. If being a form and being a mover differ only in ratio, how can Aquinas deny that the forms of inanimate things are movers? The problem of explaining the soul’s

⁵¹ QDSC 3, ad 7 [Leon. 24/II.45: 493–504]. Here Aquinas is responding to an objection found in the Liber de spiritu et anima, a text which he knew to be falsely attributed to Augustine, and for which he reserves his harshest criticism. The objection is fascinating because the crude model of interaction between the soul and the body which it presupposes gives Aquinas a foil against which to develop his more nuanced theory.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

172   motive role is pushed back one step: Although Aquinas does not need to explain why the soul has a special sort of causal influence, he instead has to explain why being a form of a certain sort qualifies a soul as a mover. In considering a solution, it is important to remember what we have already established, namely that the soul does not do anything “extra” beyond informing the body in order to make it move. That is, the soul is called a mover in virtue of what it already does by informing the body, and this is clear from the fact that the form-role and the motor-role are distinct in ratio only. Accordingly, what makes souls be “movers” cannot lie in some additional role they take on that is really distinct from their status as forms, or some special extra influence of the soul on the body. I propose that Aquinas calls souls movers of the body because they inform organic bodies, i.e., bodies composed of non-homogeneous parts, and composed in such a way that one part moves another in the performance of vital operations. My contention, then, is that Aquinas calls the soul a mover precisely because it informs a body such that the parts differ from one another in such a way that one part moves the next. To clarify, it is sufficient for a substantial form to be a mover that it inform this sort of body, i.e., a body with organic parts, in virtue of which the informed body moves itself. In other words, the soul deserves to be called a mover not because it does something in addition to informing, but because of the kind of body to which it gives rise as a form. On my reading, then, the difference between souls and inanimate substantial forms lies principally in the fact that souls are able to inform complex bodies, the parts of which can act on one another. The soul can be called the principal mover of the body in that it is the ultimate formal principle responsible for the body’s self-motion; in informing, the soul orchestrates the harmony of contraries required for the corporeal parts to move one another. Those parts move one another in a derivative way. It is not surprising, then, that Aquinas insists that souls are distinct from inanimate forms and artificial forms precisely in that only souls inform both a whole body as well as the parts of that body, each of which has its own specific kind of being as a part different from the other parts. Aquinas credits the soul with the bestowal of specific being (esse specificum) on each of the parts of an organic body. I take this to mean not only that it makes those parts belong to the body (i.e., makes a heart to be

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

173

a part of a horse), but also that it gives each part its own character (i.e., makes one part to be a heart, and another an eye). In QDSC 2, he says that the “body and its parts” have esse specificum through the soul: For if the soul were not united to the body as form, it would follow that the body and its parts would not have esse specificum through the soul. This is clearly false, for when the soul departs, the eye and the heart are not called such, except equivocally, as a painted or a stone eye.⁵²

For an eye or a hand to have specifically human being is for it to have the being of an eye or the being of a hand (esse oculi, esse manus). The same point also appears when Aquinas contrasts souls with artificial forms, e.g., the form of a house.⁵³ For example, in Summa contra gentiles 2.72 he says: Thus, however the soul is the form of the whole body in such a way that it is the form of all the single parts. For if it were the form of the whole and not the parts, it would not be the substantial form of such a body, as the form of a house, which is the form of the whole and not of all the single parts, is an accidental form. But because it is the substantial form of the whole and of the parts, it is clear from this that from it are drawn the species of both the whole and the parts.⁵⁴

Similarly, in De spiritualibus creaturis, a. 4 ad 4 he says: The form of a house, because it is accidental, does not give specific existence (esse specificum) to the several parts of the house, as the soul gives to each part of the body.⁵⁵

While the forms of artifacts take advantage of the tendencies and dispositions that each part has in virtue of its own substantial form, the soul is responsible not only for the very being of the whole, but also for the being of each single part. Thus, the soul does not merely integrate

⁵² QDSC 2 [Leon. 27–8/II.53: 264–9]. ⁵³ In addition to the texts cited here, see also ST 1.76.8. ⁵⁴ SCG 2.72 [Leon. 13.456–7: 3b–3a]. ⁵⁵ QDSC 4, ad 4 [Leon. 24/II.53: 275–8].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

174   pre-existing or even ontologically prior parts into a whole, but rather immediately gives being to both the parts and to the whole. It gives the specific character to the parts that it unites. So far I have focused on indirect evidence for my view that the soul moves the body by informing it so as to move itself. My point has been that for Aquinas, only souls are capable of giving being to a complex whole as well as its parts, in contrast to artificial forms (which do not give being to their parts) and to inanimate substantial forms (which can only inform homogeneous wholes). But there is also some direct evidence for grounding the soul’s mover-role in its ability to inform organically constituted wholes. In a very dense passage from De veritate, we catch Aquinas describing the soul’s role as mover in terms of the kind of body which it informs: Things that are absolutely spiritual have such a nature as to impart motion (ut moveant) but not as to be moved (ut moveantur). Bodies, however, are moved in a certain way and although one thing is able to move another, none of them is able to move itself; for that which moves itself, as proved in Physics 8, is divided into two parts, of which one is the mover and the other something moved. But this cannot occur in purely corporeal things; for their forms cannot be movers, although they can be a principle of motion – as in the motion of earth, heaviness is the principle by which it is moved, but nevertheless not the mover (motor). And the reason for this is the simplicity of inanimate bodies, which do not have the diversity in their parts such that one part is able to move and the other to be moved, as well as on account of their ignobility and the materiality of their forms. For these [forms] are far removed from separate forms, to which it belongs to move, [and] do not retain the ability to move, but only to be a principle of motion. Animate things, however, are composed from a spiritual and a corporeal nature; whence there can be in them one part that is moving and the other that is moved, in both local motion as well as in other motions. And therefore, insofar as to be moved is made to be the proper action of animate things (such that they move themselves to the determinate

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

175

species of motions), there are found in animate things special ordained powers . . .⁵⁶

How does the soul’s role as form of a body with organs entail that it is a mover? Aquinas sets out to explain why inanimate substances are incapable of moving themselves. The question here is ambiguous, and an attentive reader who has not yet seen Aquinas’s answer might guess that he is interested in answering one of the following two questions. The question might be (1) why are souls able to have an effect on the actions of bodies, while inanimate substantial forms are not? Or (2) what is it about the composition of inanimate beings, considered as composite substances, that makes them incapable of self-motion? An answer to the first question would seemingly address questions about the interface between the soul and the body, while an answer to the second question would seemingly address the bodily mechanisms which make selfmotion possible. Looking more closely, however, we can see that Aquinas is not interested in the question of the interface between soul and body, but rather in the question of why souls deserve to be called movers, while other substantial forms do not. Aquinas does not acknowledge the ambiguity in the question, but simply gives answers to both questions together, as though one were the exposition of the other. He says that the bodies of these substances are not divided into parts (i.e., they lack the relevant sort of complexity), and that the souls of these bodies cannot be movers. The first answer is at the level of the composite substance, and the second at the level of the form. But even more strikingly, he explicitly links these two answers and says that the division into parts cannot happen in purely corporeal bodies. This sudden shift from the level of the substance to the level of the substantial form is hard to explain if the question he is attempting to answer is one about the interface between the soul and the body, but makes sense in light of the question I have posed: Why are only some forms (and not others) able to be movers? The answer to that question is now becoming clear: some forms are the sorts of forms that

⁵⁶ DV 22.3 [Leon. 22/3. 618–19: 58–88].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

176   inform an organic body, i.e., a body with parts sufficiently diverse for self-motion. This parallel exposition at two levels (that of substance and that of substantial form) continues as the text goes on. Aquinas then considers why inanimate things cannot move themselves and why their forms cannot be movers. Again, he gives the same answer on two levels: inanimate beings (a) lack organic parts, and therefore the diversity of parts necessary for self-motion and (b) have a relatively “ignoble” and material substantial form, and therefore are far “distant” from separate forms. This ongoing parallel exposition is best understood as Aquinas’s attempt to reconcile two assumed commitments regarding motion. The first formulation is a nod to Aristotle’s constraint on bodily self-motion, i.e., that the self-mover must be divided into two parts, one of which is moved and the other moves. The second, which has a more Neoplatonic cast, satisfies the constraint that the forms that are higher in the hierarchy of being are more powerful agents than those lower down. Taken together, they point to the same fact about the difference between nonliving and living bodies: living bodies are self-movers because they are composed of complex organic parts which are capable of moving one another, and they are so composed because their substantial forms are sufficiently powerful to cause organic diversity. In contrast to other substantial forms, only souls are movers of their bodies. I have argued that the reason why the soul deserves to be called a “mover” of the body is not because the soul has some extra efficient causal role over and above its “form” role, but rather because the specific kind of body which a soul is able to inform is an organically constituted body which is able to move itself. More precisely, the soul gives both the whole and the parts of such an organic body their specific character. It orchestrates motion within the body by being the form of the agent-parts and the patient-parts, as well as the whole of which they are parts. The soul’s moving the body simply reduces to its being a form of a certain sort.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

177

3. Conclusion: Situating the Soul Relative to other Principles of Action The question posed at the outset of the paper is whether Aquinas’s account of how the soul moves the body makes the soul be something like an internal efficient cause. In the first section of the paper, I called attention to the aspects of vital operation which distinguish actions which flow from soul from those which flow from inanimate substantial forms. In particular, I showed that for Aquinas ensouled bodily beings can contribute to the execution and sometimes even the determination of their own motions. By contrast, the substantial forms of inanimate bodies can only be principles of motions of other bodies insofar as they make their own bodies have the relevant active principles for changing something else. This ability of souls to contribute to the motions of their own bodies is what leads Aquinas to call the soul a mover of its own body, but also what leads interpreters to wonder whether soul as he understands it is something like an internal efficient cause. Yet Aquinas’s own account of how the soul moves the body is very far from the model of an efficient cause moving its patient, and indeed much of his theory is set up in response to “the Platonists” who see body as a pure passive receiver of the action of soul. Correspondingly, in the second part of the present study, we saw an alternative way in which the soul could be called a mover, i.e., by informing a body such that its organic parts move one another. All this opens up the possibility for a more nuanced perspective on the role of souls within Aquinas’s taxonomy of principles of action. By way of concluding, let me briefly sketch the big picture of what could be called his “stepped view” of forms, and show how the higher grades of substantial form (souls) approximate in their mover-role the fully independent agency of spiritual creatures. Within this stepped view, what all substantial forms have in common is that they give their bodies esse specificum, i.e., they make it to be what it is. Correspondingly, as we saw, substantial forms are the ultimate principles responsible for the motions which we see in the physical world, because the accidental qualities through which bodies act are always acting through the power of some

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

178   substantial form. In this respect all substantial forms bear the same fundamental relationship to their bodies. Yet despite this commonality, the esse specificum which different grades of forms give to their bodies differs in kind from one grade to the next. Bodies with higher grades of life perform higher kinds of action, which in turn require principles of action which differ in kind from those found in the inanimate world. We saw this implicitly already in the idea that only souls can be principles of self-execution and self-determination, but Aquinas makes the point explicitly in his Quaestiones disputatae de anima: The action of the soul transcends the action of nature operating in inanimate things, but this happens in two respects, namely with respect to the manner of acting and with respect to that which is done. It is necessary however that with respect to the manner of acting, every action of the soul transcend the operation or action of inanimate nature; for since the action of the soul is an act of life, and what moves itself to operate is called life, it is necessary that every operation of the soul be acting according to something intrinsic. But with respect to that which is done, not every action transcends the action of inanimate nature . . . . There are, however, higher actions of the soul which transcend the actions of natural bodies, even with respect to that which is done, namely inasmuch as it is given to the soul to be all things according to an immaterial existence. And thus, the soul is in a certain way everything according to immaterial existence.⁵⁷

The two ways in which the actions of soul transcend those of inanimate substantial forms correspond to the execution/determination distinction we saw earlier. The difference in the “manner of acting” (modum agendi) refers to the fact that the soul is the principle of the execution of its own motion, while the difference in the “that which is done” (id quod agitur) refers to the soul’s being a principle of an immanent operation, which is a kind of action distinct from (and more noble than) a motion. In these cases, the soul does not function as an efficient cause; the claim is rather

⁵⁷ QDDA 13 [Leon. 24/I.116: 188–98 . . . 213–18].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

179

that as a principle of action, it transcends the action of nature (i.e., of the elemental qualities alone). The conclusion to draw is that while inanimate substantial forms, the souls of plants, and the souls of animals are all forms, there are three very distinct ways to be the form of a body, each of which has its own grade of agency. In other words, Aquinas subscribes to a stepped view of substantial form, in which each higher grade of form more closely approximates the independent agency of spiritual substances and God. The lowest grade of substantial form is not a mover at all, but only a principle of motion. For example, the substantial form of fire is the principle of its accidental form heat, by which fire acts, but it does not govern the specific effect which is brought about by heat, and so falls short of being a mover. By contrast, the souls of living beings all give rise to specific effects, even though they sometimes use the same active and passive qualities of nature which are brought about by inanimate substantial forms. Aquinas’s example is the heat of digestion, which acts as an instrument of the soul. A plant or an animal uses the active qualities of nature to break down food and grow according to the proportions of its nature. The difference is that the nutritive soul sets the terminus of its own motion, whereas the fire can only determine some effect in a body extrinsic to it, and can only produce its proper effect (the production of more fire) if the patient happens to be an appropriately disposed receiver (i.e., something combustible). In other words, the difference in the “manner of acting” has to do with the source of active principle governing the motion, e.g., the soul of the acacia tree in the operation of nutrition. The souls of animals give rise to a different sort of action altogether, e.g., the activity of sight in a cat. Strictly speaking, the soul does not move the body to see because sight is not a motion, but in making the eye to see, the soul of the cat is the principle of both the harmonious constitution of the organ required for sight as well as the power of sight itself. But even though sensitive souls are principles of a more perfect kind of self-actualization which transcends motion, their action remains within the body. But the soul is not only the highest of the substantial forms; it is also a principle of the lowest grade of life. If we compare the soul to the lives which are above it (spiritual substances), we can see that the soul holds a

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

180   specific place in the hierarchy of agents. Just as the comparison between soul and other forms helps clarify what is essential to Aquinas’s understanding of form, the comparison between souls and spiritual substances reveals what he takes to be essential about life. And interestingly, this comparison between souls and spiritual substances reveals an oftenoverlooked Neoplatonic dimension to Aquinas’s account of soul and of life. The essential feature of life is self-motion, but souls only have selfmotion (the essential feature of life) in the lowest possible way. Thus, while our understanding of life begins with the most evident kinds of self-motion (the biological, and especially animal local motion), these lower levels of life are imperfect participations in the kind of independent agency found among the higher grades of being. In discussing the divine name of life in Summa theologiae I.18.3, Aquinas says that the name life is taken from things which move themselves to their own motion or operation: From this it is clear that those things are properly living which move themselves according to a certain species of motion, whether this be motion in the proper sense, insofar as motion is called the act of the imperfect, that is, of that which exists in potency; or whether motion is taken generally, insofar as it is called the act of something perfect, insofar as to understand and to sense are called “to be moved,” as is said in De anima 3. Whence whatever things move themselves (se agunt) to some motion or operation are said to be alive, but those things to whose nature it does not belong to move themselves to some motion or operation cannot be said to be living things except by a certain similitude.⁵⁸

Of course, the term “self-motion” must be understood broadly so as to include immanent operations as well as self-motions which are motions in the strict sense. As we have seen, souls are the principles of both kinds of self-motion, but always (with the exception of the intellectual soul) by being the form of the body. Correspondingly, they are called movers of their bodies because they are the primary formal principles of things ⁵⁸ ST 1.18.1 [Leon. 4.225]. See also ST 1.78.1 [Leon. 5.250].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

181

which move themselves. Thus, in contrast to spiritual substances, souls are principles of self-motion in the lowest possible mode, namely by unifying the diverse parts of a self-motive body. The soul is the principle of a unity which is a harmony, i.e., a unity which presupposes diversity. This is especially evident in the case of souls of plants, which can only produce a self-motion based on the juxtaposition of parts. Its operation remains within the agent only because the parts which act on one another are opposed parts of a single substance, and so this vital operation only barely qualifies as self-motion. At a higher level, cognitive operations do not produce any product, but they still actualize a potency of the body, i.e., of the organ of cognition. Accordingly, sensitive souls are self-motive in a higher way than vegetal souls. But it is proper to all souls to be a principle of self-motion of a harmoniously organized body. Aquinas gives a principled reason to rank cognitive instances of life higher than non-cognitive instances, based on his general principle that an action is more noble to the extent to which it is universal. He introduces the principle in Summa theologiae I.77.3 ad 4: “A superior power is ordained to the object in a more universal way than an inferior power, because to the degree to which the power is superior, it extends to more things.”⁵⁹ And the details get worked out subsequently, in 1.78.1: The kinds of powers are distinguished according to their objects. To the extent that a power is higher, to that extent it concerns a more universal object. The object of the operation of a soul is able to be considered according to a threefold order. For in any [soul] there is a power of the soul the object of which is only the body united to the soul. And this genus of powers of soul is called “vegetative,” for the vegetative power does not act except in the body to which it is united. But there is some genus of the powers of soul which concerns universal objects, namely the entirety of sensible body, and not only the body that is united to [the creature’s] soul. And there is a genus of powers of the soul which regards still more universal objects, namely not only of sensible body, but of all being universally. From this it is clear that

⁵⁹ ST 1.77.3 ad 4 [Leon. 5.241].

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

182   these two genera of powers of the soul have an operation that extends not only to what is conjoined to it but to external things as well.⁶⁰

By “universal” here, Aquinas does not mean “abstract,” but rather “extending to a larger swathe of being.” Each higher grade of soul has a more universal object, and this scale of universality complements the order of internal active principles considered above. Because each higher action has a more universal object than the next, the agent is able to act towards a more universal good. In vegetal operations, which are noncognitive, the object is only the individual (e.g., the soul’s own body in growth). In sensation, sensible beings cognize some external things and so can determine their action in relation to them. The power of understanding, which regards all being, makes it possible to choose ends (or at least ends short of the ultimate end, at which the will aims by nature). This scale of universality might initially seem to undermine Aquinas’s use of self-motion as the way to rank the grades of life, because it might seem as though more universal actions are less self-motive than less universal actions, because their effects extend farther beyond the agent. For example, the honeysuckle’s action of growth is clearly self-motion because the terminus is some change in its own body. A city planner’s planning high-rise zoning around transit hubs seems less self-motive to the extent that it seems to extend so far beyond the individual agent: it produces some external product (the city plan and ultimately the buildings themselves). But this superficial tension can be resolved, if we recall that actions are specified by their ends. At each of the levels of self-motion which we have seen the end is somehow present in the agent. Plants achieve the lowest level of universality, because the objects of their actions are their own bodies. Sensate beings have the entirety of sensible being as their object, with some having more sense modalities (and thus a greater scope of action) than others.⁶¹ And beings with intellectual powers (humans) as ⁶⁰ ST 1.78.1 [Leon. 5.251]. ⁶¹ This is why in DV 22.3, Aquinas says that all things tend or desire (appetere), but that tending belongs especially to animals. There is just more intelligibility in the motions of living things, and more in animals than in plants, because each higher kind of action is end-directed in a higher way.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

183

well as intellectual beings have all being as their object. What the scale of universality tracks is not how far beyond itself an agent is able to reach in its transitive actions, but rather the scope of what the agent is able to take into itself in performing an action, whether it is an immanent operation or a productive action. Thus, even though all motion is essentially teleological, not all actions are directed at the good of the agent, and some actions have more specified ends than others. Souls are the less self-motive of living beings, then, because they are directed to what is least universal—either just their own bodies (in the case of plants), or to the objects of the sensitive appetites in the case of animals. Only at the level of intellect can cognition have a truly universal object (i.e., all of being), and for this reason intellectual beings have life “more perfectly” even than other animals: Whence above these [the plants] animals there are those which move themselves even with respect to the end which they are able to predetermine. For this is not possible except through reason and intellect, to which it belongs to know (cognoscere) the relation (proportionem) of the end and the means (quod est ad finem), and to order one to the other. Whence the more perfect mode of living belongs to those which have understanding, for these more perfectly move themselves. (ST 1.18.3 [Leon. 4.228])

All cognitive operations are more universal (in the sense of extending to a greater range of being) than other vital operations, but the ability to possess being in a universal mode means that intellectual beings possess an especially perfect degree of life,⁶² and a correspondingly perfect degree of mastery over their own actions, which Aquinas calls dominium.⁶³ Of course, if life belongs more perfectly to that which has its operation more perfectly from itself, the fullness of life belongs to God. It is this most complete grade of independent activity which Aquinas seems to find most interesting, as indicated by the fact that his most detailed ⁶² See Super Evangelium S. Ioannis lectura c. 14 lectio 2. ⁶³ See an explanation of dominium in terms of having an intrinsic source of motion, see for example SCG 2.47. Jamie Spiering explicates the grade of self-motion found in free creatures in her “ ‘Liber est causa sui’: Thomas Aquinas and the Maxim ‘The Free Is the Cause of Itself,’ ” The Review of Metaphysics 65 (2011), 351–76.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

184   treatments of biological life occur in the contexts of discussing life in God. For example, we already saw the discussion of the divine name of life in Summa theologiae I.18.3. In the commentary on the Divine Names, he mentions both plant and animal life as the base case from which we learn about life in God.⁶⁴ And in commenting on the Gospel of John, in glossing Jesus’ revelation that he is the way, the truth, and the life, Aquinas says that everything which has an activity from itself is called living.⁶⁵ In all these cases, the move is from the kind of self-motion obvious to us (that of living animate bodies, and especially the local motions of animals)⁶⁶ to God’s own completely independent act of being. The self-motion of animate bodily beings attains only the lowest levels of independence, but in this small way imitates the truly independent act of the divine life.⁶⁷ University of Notre Dame

Bibliography Cory, David. “Agency and Materiality in Aquinas’s Soul Theory” (PhD dissertation, The Catholic University of America, 2018). Frost, Gloria. “Aquinas’ Ontology of Transeunt Causal Activity,” Vivarium 56 (2018), 1–36. Hoffman, Paul. “St. Thomas Aquinas on the Halfway State of Sensible Being,” Philosophical Review 99 (1990), 73–92. Pasnau, Robert. “Form, Substance, and Mechanism,” The Philosophical Review 113 (2004), 31–88. Spiering, Jamie. “ ‘Liber est causa sui’: Thomas Aquinas and the Maxim ‘The Free Is the Cause of Itself,’ ” The Review of Metaphysics 65 (2011), 351–76. Thomas Aquinas. Commentum in quartum librum in Sententiarum magistri Petri Lombardi, in Opera omnia, vol. 7/2 (Parma: Typis Petri Fiaccadori, 1858). ⁶⁴ De divinis nominibus 6.1 [Marietti 254–5]. ⁶⁵ See also Super Ioannis 14.2 [Marietti 351]. ⁶⁶ For animal local motion as the most evident form of self-motion see ST 1.18.1. ⁶⁷ I note with gratitude that the present study has been greatly improved by the criticism and feedback received at various stages from Thérèse Cory, Jeffery Brower, Gloria Frost, Michael Gorman, Susan Brower-Toland, and an anonymous reviewer.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

        

185

Thomas Aquinas. In octo libros Physicorum Aristotelis, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 2 (Rome: S. C. De Propaganda Fide, 1884). Thomas Aquinas. In octo libros De Caelo Aristotelis, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 3 (Rome: S. C. De Propaganda Fide, 1886). Thomas Aquinas. Summa theologiae, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vols. 4–12 (Rome: S. C. De Propaganda Fide, 1888–1906). Thomas Aquinas. Summa contra gentiles, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vols. 13–16 (Rome: Typis Riccardi Garroni, 1908–48). Thomas Aquinas. Scriptum super libros Sententiarum magistri Petri Lombardi, vols. 1–2, ed. R. P. Mandonnet; vols. 3–4 ed. R. P. Maria Fabianus Moos (Paris: Lethielleux, 1929–47). Thomas Aquinas. Super librum beati Dionysii, De divinis nominibus expositios, Thomae Aquinatis, opera omnia (Turin: Marietti, 1951). Thomas Aquinas. Super Evangelium S. Ioannis lectura, Thomae Aquinatis, opera omnia (Turin: Marietti, 1952). Thomas Aquinas. Super Librum de causis expositio, ed. H. D. Saffrey (Louvain: Nauwelaerts, 1954). Thomas Aquinas. Quaestiones disputatae de potentia, in P. Bazzi et al. (eds.), Quaestiones disputatae, vol. 2 (Turin: Marietti, 1965). Thomas Aquinas. Quaestiones disputatae de veritate, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 22/1–3 (Rome: Sancta Sabina, 1970–6). Thomas Aquinas. De motu cordis, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 43 (Rome: S. C. De Propaganda Fide, 1976). Thomas Aquinas. Sentencia libri de anima, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 45/1 (Rome: S. C. De Propaganda Fide, 1984). Thomas Aquinas. Sentencia libri de sensu, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 45/2 (Rome: S. C. De Propaganda Fide, 1985). Thomas Aquinas. Quaestiones de quolibet, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 25/1–2 (Rome: Commissio Leonina, 1996).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

186   Thomas Aquinas. Quaestiones disputatae de anima, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 24/1 (Rome: Commissio Leonina, 1996). Thomas Aquinas. Quaestiones disputatae de spiritualibus creaturis, in Sancti Thomae Aquinatis, Doctoris Angelici, opera omnia, iussu impensaque Leonis XIII P.M. edita, vol. 24/2 (Rome: Commissio Leonina, 2000). Twetten, David. “Why Motion Requires a Cause: The Foundation for a Prime Mover in Aristotle and Aquinas,” in R. J. Long (ed.), Philosophy and the God of Abraham: Essays in Memory of James A. Weisheipl, OP (Toronto: Pontifical Institute of Mediaeval Studies, 1991), 235–54. Twetten, David. “Back to Nature in Aquinas,” Medieval Philosophy and Theology 5 (1996), 205–43.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

Thomas of Sutton’s Intellectualist Doctrine of the Will’s Self-Motion Michael Szlachta

1. Introduction Along with William Hothum and Richard Knapwell, the Dominican friar Thomas of Sutton (ca. 1250–1315) was one of the earliest defenders of Aquinas at Oxford.¹ Not much is known about Sutton’s life, but his writings have been widely praised by scholars of medieval philosophy. Described as “one of the most astute defenders of Thomas Aquinas’s characteristic theological and philosophical doctrines,”² and representing “a high point in the progress of Thomism,”³ it is not surprising that Sutton’s views on topics like the essence–existence distinction and the analogy of being have been (and continue to be) the focus of scholarly attention.⁴

¹ For Sutton’s life and work, see Frederick J. Roensch, Early Thomistic School (Dubuque, IA: Priory Press, 1964), 44–51. See further Gyula Klima, “Thomas of Sutton,” in H. Lagerlund (ed.), Encyclopedia of Medieval Philosophy: Philosophy between 500 and 1500 (Dordrecht: Springer, 2011), 1294. For an overview of Sutton’s thought and his opposition to Henry of Ghent, see Francis E. Kelley, “Two Early English Thomists: Thomas Sutton and Robert Orford vs. Henry of Ghent,” The Thomist 45 (1981), 345–87. ² Gyula Klima, “Thomas of Sutton on the Nature of the Intellective Soul and the Thomistic Theory of Being,” 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), 436. ³ Roensch, Early Thomistic School, 51. ⁴ For the essence–existence distinction, see Kelley, “Two Early English Thomists,” 360–71; Mark D. Gossiaux, “Thomas of Sutton and the Real Distinction between Essence and Existence,” Modern Schoolman 83 (2006), 263–84. For the analogy of being, see Domenic D’Ettore, “Thomas Sutton’s Doctrine of Analogy: Revisiting a Continuator of Thomas Aquinas,” Nova et Vetera 14 (2016), 1153–74; Mark G. Henninger, “Thomas Sutton on Univocation, Equivocation, and Analogy,” The Thomist 70 (2006), 537–75; Gyula Klima, Michael Szlachta, Thomas of Sutton’s Intellectualist Doctrine of the Will’s Self-Motion In: Oxford Studies in Medieval Philosophy Volume 8. Edited by: Robert Pasnau, Oxford University Press (2020). © Michael Szlachta. DOI: 10.1093/oso/9780198865728.003.0006

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

188   What is surprising is that his view on the will, the human faculty of volition, has not received the same attention.⁵ As with many other topics, the view that Sutton adopts here is broadly that of Aquinas. Aquinas’s view on the will, however, has long been the source of controversy. Following the publication of William de la Mare’s Correctorium fratris Thomae,⁶ the last quarter of the thirteenth century witnessed much debate concerning whether Aquinas held a view that would be condemned by the Bishop of Paris, namely, that the will is necessitated by reason.⁷ More recent literature has been dominated by debates concerning whether Aquinas’s view on the will is “intellectualistic” or “voluntaristic,” and whether he is a “compatibilist” or “incompatibilist” about the will’s freedom.⁸ The will is perhaps the one area of Aquinas’s philosophy about which there is the most disagreement. Is it possible that the means to address some of the confusion surrounding Aquinas’s view on the will might be found in Sutton, a thinker

“Thomas Sutton and Henry of Ghent on the Analogy of Being,” Proceedings of the Society for Medieval Logic and Metaphysics 2 (2002), 34–44. ⁵ D. E. Sharp makes some remarks about the will in “Thomas of Sutton, O.P. His Place in Scholasticism and an Account of His Psychology,” Revue néoscolastique de philosophie 36 (1934), 352–4. François-Xavier Putallaz gives a brief outline of Sutton’s psychology of human action and its historical context in “Thomas de Sutton, ou la liberté controversée,” Revue thomiste 97 (1997), 31–46. ⁶ In particular, see William de la Mare, Correctorium fratris Thomae, In primam partem Summae, art. 24 (ed. Glorieux, 106); In primam secundae, art. 7 (ed. Glorieux, p. 232). ⁷ Bishop Tempier condemned the thesis that, in general, the will is necessitated in Article 3 of his 1270 condemnation. See the Chartularium Universitatis Parisiensis (ed. Denifle and Châtelain, 487). In Articles 159 (164) and 163 (163) of his 1277 condemnation, he condemned the thesis that, in particular, the will is necessitated by reason. See La condamnation parisienne de 1277 (ed. Piché, 126–8). Articles 129 (169), 130 (166), 131 (160), 135 (161), 158 (165), and 194 (151) are also relevant here. For the so-called “Correctoria controversy,” see Mark D. Jordan, “The Controversy of the Correctoria and the Limits of Metaphysics,” Speculum 57 (1982), 292–314. ⁸ For an overview of the debate between intellectualists and voluntarists, and relevant scholarship, see Tobias Hoffmann, “Intellectualism and Voluntarism,” in R. Pasnau and C. Van Dyke (eds.), The Cambridge History of Medieval Philosophy (Cambridge: Cambridge University Press, 2009), 414–27. For Aquinas’s (in)compatibilism about the freedom of the will, see Eleonore Stump, “Aquinas’s Account of Freedom: Intellect and Will,” The Monist 80 (1997), 576–97; Scott MacDonald, “Aquinas’s Libertarian Account of Free Choice,” Revue Internationale de Philosophie 52 (1998), 309–28; Brian J. Shanley, “Beyond Libertarianism and Compatibilism: Thomas Aquinas on Created Freedom,” in R. L. Velkley (ed.), Freedom and the Human Person (Washington, DC: The Catholic University of America Press, 2007), 70–89.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

189

known for his “championship” of Aquinas’s cause?⁹ My aim in this paper is to answer this question by looking at one particular feature of Aquinas’s view on the will, namely, that there is a distinction in the will between a potency to the “specification” of its act and a potency to the “exercise” of its act. I will begin by explaining Aquinas’s distinction and describing some of the problems it seems to present. Then, focusing on the sense in which the will is in potency to the “exercise” of its act, I will consider how Sutton adopts Aquinas’s distinction and develops it in a way that addresses these problems. And, finally, I will motivate some doubts concerning whether Sutton succeeds in preserving what makes Aquinas’s distinction so attractive in the first place.

2. Thomas Aquinas on Specification, Exercise, and the Will’s Self-Motion A well-known feature of Thomas Aquinas’s mature view of the will is that it has two potencies. First, it is in potency to the “specification” of its act, that is, to willing this or that thing. Second, it is in potency to the “exercise” of its act, that is, to willing or not willing.¹⁰ A human being wills when the potencies of her will to the specification and to the exercise of its act of willing are both actualized. Put differently, since movement in Aristotelian metaphysics is the “reduction” of potency to act, Aquinas thinks that, for a human being to will, it is necessary that her will be moved both to the specification and to the exercise of its act of willing. How is the will so moved? Aquinas explains in the Prima secundae (1271–2) of his Summa theologiae that, in general, the object of a power is responsible for the specification of that power’s act.¹¹ In particular, the ⁹ Sharp, “Thomas of Sutton,” 322. ¹⁰ Thomas Aquinas, Prima secundae Summae theologiae (= ST, I-II), qu. 9, art. 1, co. (Leonine VI, 74); Quaestiones disputatae de malo (= QDM), qu. 6, co. (Leonine XXIII, 148–50). Aquinas seems to present a version of this distinction in Quaestiones disputatae de veritate (= QDV), qu. 22, art. 12, co. (Leonine XXII/III.1, 642). ¹¹ Thomas Aquinas, ST, I-II, qu. 9, art. 1, co. (Leonine VI, 74). For the dating of Aquinas’s Summa, see Jean-Pierre Torrell, Saint Thomas Aquinas, tr. R. Royal, vol. 1 (Washington, DC: The Catholic University of America Press, 1996), 145–8.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

190   good apprehended by reason is responsible for the specification of the act of the will. When it apprehends an object as good and presents it to the will, reason “reduces” the will’s potency to willing this or that thing, such that it wills this thing rather than that one. What about the will’s potency to the exercise of its act? What Aquinas says here is puzzling. For while the will is moved to the specification of its act by the good apprehended by reason, it is moved to the exercise of its act by itself. As Aquinas says in the Disputed Questions on Evil (1269–72): And if we should consider the movement of the soul’s powers regarding exercise of the act, then the source of the movement comes from the will. For the power to which the chief end belongs always moves to action the power to which the means to the end belongs. For example, the art of war causes a bridle maker to make bridles. And thus does the will move both itself and all the other powers.¹²

What makes this claim puzzling is that self-motion—the reduction of some potency to act by the very thing that has that potency—is not possible in Aristotelian metaphysics, for it violates the principle that nothing can be in act and potency at the same time and in the same respect. Following John Wippel, I call this the “act–potency axiom.”¹³ Aquinas upholds this axiom throughout his career, and it follows from the axiom that, in particular, the will’s self-motion is impossible.¹⁴ How, then, is the will supposed to move itself? Aquinas’s answer to this question is problematic. He explains in the Prima secundae that the will moves itself, but this [ . . . ] lies in the fact that by willing an end it reduces itself to willing the means to it (ea quae sunt ad finem). This cannot be done except through a process of deliberation; when someone wills to be cured he begins to think about how health may be ¹² Thomas Aquinas, QDM, qu. 6, co. (Leonine XXIII, 149). Translation, with some modification: On Evil, tr. R. Regan, ed. B. Davies (Oxford: Oxford University Press, 2003), 258. For the dating of Aquinas’s QDM, see Torrell, Saint Thomas Aquinas, vol. 1, 201–7. ¹³ For a discussion of the axiom in the writings of Godfrey of Fontaines, see John F. Wippel, “Godfrey of Fontaines and the Act-Potency Axiom,” Journal of the History of Philosophy 11 (1973), 299–317. ¹⁴ See, for example, QDV, qu. 19, art. 1, co. (Leonine XXII/II, 565); Pars prima Summae theologiae (= ST, I), qu. 2, art. 3, co. (Leonine IV, 31); ST, I-II, qu. 1, art. 2, co. (Leonine VI, 9).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

191

secured, and arrives at the decision to consult the doctor, and then wills to do this.¹⁵

Suppose, for example, that I will to be healthy. Since I am not presently healthy, I deliberate about the means to my health. Suppose further that the conclusion of my deliberation is a judgment that consulting the doctor is the best means to my health. So, I will to consult the doctor. Here, my will does not move itself directly, but rather indirectly, “through a process of deliberation” (consilio mediante). That is, by willing some end and moving reason to deliberate about the means to that end, my will moves itself to willing some means to that end. Since my will, which wills the end in act, moves itself from willing the means in potency to willing the means in act, it does not, as Aquinas says, move itself “according to the same [respect]” (secundum idem) but rather “according to a different [respect]” (secundum aliud).¹⁶ The act–potency axiom is preserved, for my will is not in act and potency at the same time and in the same respect. Why is this answer problematic? Well, notably, it fails to explain how the potency of the will to the exercise of its act is actualized. According to Aquinas, the will is supposed to move itself to the exercise of its act by moving reason to deliberate about the means to some end. Reason, having concluded its deliberation, makes a judgment that this means should be willed. The will then wills that means. Since the will wills the means that reason judges should be willed, and reason judges that this means should be willed because it was moved by the will to deliberate about the means to the end, the will is said to move itself to the exercise of its act of willing that means indirectly (the will moves itself by moving reason). But how exactly is the will’s potency to the exercise of its act of willing the means being actualized? When the will moves reason to deliberate about the means to some end and reason judges that this means should be willed, reason merely specifies the will’s act, such that it wills this means rather than some other ¹⁵ Thomas Aquinas, ST, I-II, qu. 9, art. 4, co. (Leonine VI, 78). Translation, with some modification: Summa Theologiae: Volume 17, Psychology of Human Acts: 1a2ae. 6–17, ed. T. Gilby (Cambridge: Cambridge University Press, 1970), 73. See further QDM, qu. 6, co. (Leonine XXIII, 149). ¹⁶ Thomas Aquinas, QDM, qu. 6, ad 20 (Leonine XXIII, 152).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

192   one. That the will moves reason to deliberate about the means and reason judges that this means should be willed makes it clear how the will’s potency to the specification of its act is actualized, but not its potency to the exercise of its act. If reason’s judgment merely specifies the will’s act, Aquinas has not explained how the will’s potency to the exercise of its act of willing this means to some end is actualized by its moving reason to deliberate about the means to that end and judge that this means should be willed. “When someone wills to be cured,” Aquinas says, “he begins to think about how health may be secured, and arrives at the decision to consult the doctor, and then wills to do this”; but what happens between his arriving at the decision to consult the doctor and his willing to consult the doctor is a mystery.¹⁷ By claiming that the will is directly moved to the specification of its act by the good apprehended by reason and indirectly moved to the exercise of its act by itself, Aquinas does avoid violating the act–potency axiom. But we should still want to know how exactly its potency to the exercise of its act is being actualized. Perhaps it is not surprising that many thinkers in the last quarter of the thirteenth century were critical of Aquinas’s distinction in the will between a potency to the specification of its act and a potency to the exercise of its act. Henry of Ghent (ca. 1217–93) famously challenged Aquinas’s distinction in Question 5 of his ninth Quodlibet (Lent of 1286) by arguing that it is not possible for the will to be moved by the good apprehended by reason with respect to the specification of its act and not also be moved by it with respect to the exercise of its act.¹⁸ Henry of Ghent’s challenge takes the form of a dilemma. Consider the (supposed) movement of the will with respect to the specification of its act by the apprehended good. This specification is either (i) a “mere presentation” (pura ostensio) of that good to the will, or (ii) an inclination made in the will.¹⁹ The first option is not on the table, ¹⁷ Thomas Aquinas, ST, I-II, qu. 9, art. 4, co. (Leonine VI, 78, my emphasis). ¹⁸ Henry of Ghent, Quodlibet (= Quodl.) IX, qu. 5 (ed. Macken, Opera omnia, vol. 13, 122–5). See further Quodl. X, qu. 9 (ed. Macken, Opera omnia, vol. 14, 238–48); Quodl. XII, qu. 26 (ed. Decorte, Opera omnia, vol. 16, 150). For the dates of Henry’s writings, see Gordon A. Wilson, “Henry of Ghent’s Written Legacy,” in G. A. Wilson (ed.), A Companion to Henry of Ghent (Leiden: Brill, 2011), 3–23. ¹⁹ Henry of Ghent, Quodl. IX, qu. 5 (ed. Macken, Opera omnia, vol. 13, 122).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

193

since nothing is moved by something unless some impression is made in it by that thing. If, therefore, the specification of the will is a “mere presentation,” then, contra Aquinas, the will is not moved to the specification of its act by the apprehended good (or by reason itself).²⁰ What about the second option? Say that an inclination is made in the will by the apprehended good. That inclination is either (i) “some impression” (aliqua impressio) inclining the will like a weight (sicut pondus) to willing that good, or (ii) “some volition” (aliqua volitio).²¹ Again, the first option is not on the table. For suppose that the good apprehended by reason makes an impression in the will inclining it to willing that good. If the impression merely inclines the will, then the will is free to not will that good. So, if the will is moved to willing that good, then it seems that it has to be moved by itself, and indeed, moved by itself both to the specification and to the exercise of its act. Where does that leave us? When the good apprehended by reason moves the will to the specification of its act, it makes an inclination in the will, and this inclination is “some volition,” that is, a “certain willing” (quoddam velle).²² But, as Henry of Ghent argues, when it wills something, the will “carries that [willing] out” (illud exsequitur) unless it is impeded. So, if the will is not willing something that it can be impeded from carrying out, then the “certain willing” made in the will by the apprehended good is simply the exercise of the will’s act.²³ The upshot of Henry of Ghent’s challenge is that, if the good apprehended by reason moves the will to the specification of its act, it also moves the will to the exercise of its act; Aquinas’s specification–exercise distinction does not seem to hold up. Considering that Godfrey of Fontaines (ca. 1250–1306/9), who generally followed Aquinas and opposed Henry of Ghent,²⁴ was also critical ²⁰ Henry of Ghent, Quodl. IX, qu. 5 (ed. Macken, Opera omnia, vol. 13, 122–3). ²¹ Henry of Ghent, Quodl. IX, qu. 5 (ed. Macken, Opera omnia, vol. 13, 123). ²² Henry of Ghent, Quodl. IX, qu. 5 (ed. Macken, Opera omnia, vol. 13, 123). ²³ To be precise, Henry of Ghent’s qualification is “if there is not some performance to be done externally” (si exterius non sit aliqua exsecutio facienda). Such a performance is precisely one which the will could be impeded from carrying out. ²⁴ See Odon Lottin, Psychologie et morale aux XIIe et XIIIe siècles, vol. 1, 2nd ed. (Gembloux: J. Duculot, 1957), 331–9. See further Maurice De Wulf, Un théologien-philosophe du XIIIe siècle: Étude sur la vie, les œuvres et l’influence de Godefroid de Fontaines (Brussels: M. Hayez, 1904), 116–18.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

194   of Aquinas’s distinction,²⁵ it is tempting to think that the distinction might not be one worth upholding. But there does seem to be something attractive about Aquinas’s claim that the will is moved to the specification of its act by the good apprehended by reason and to the exercise of its act (of willing some means to some end) by itself. On the one hand, it is plausible that, when a human being wills means to her end, there is some relationship between her act of appetition and some act of cognition. Consider why I will to consult the doctor as a means to my health. I think that the doctor is a health expert, and since I will my own health, I think that it would be good for me to consult her. If I did not think that she is a health expert, or if, for whatever reason, I did not think that it would be good for me to consult her, then I would not will to do so. Aquinas’s distinction makes room for a relationship between willing and thinking; reason’s apprehension and presentation of some object as good to the will is a cause of the will’s act, in that it actualizes a potency of the will, namely, its potency to the specification of its act. If reason’s apprehension and presentation of some object as good to the will were not a cause of the will’s act, that is, if the will were the only cause of its own act, then its act would seem to be arbitrary.²⁶ (Why does it will? The will simply wills because it wills.) On the other hand, although Aquinas makes room for a relationship between willing and thinking, he does not conceive of this relationship in such a way that willing is merely a consequence of thinking. When an object is apprehended and presented to the will as good, actualizing the will’s potency to the specification of its act, it does not actualize the will’s potency to the exercise of its act; the will (somehow) moves itself to the exercise of its act. And since the will moves itself to the exercise of its act, Aquinas seems to avoid a species of psychological determinism, where the acts of my will are not, as it were, “up to” me, but rather “up to” how things appear to me (intellectually). By drawing a distinction between potencies in the will to the specification and to the exercise of its act, the former of which is actualized by the good apprehended by reason, the

²⁵ Godfrey of Fontaines, Quodlibet (= Quodl.) XV, qu. 2 (PB 14, 6–11). ²⁶ Perhaps the will’s act is sometimes arbitrary, but I think it would be a mistake to suppose that the willing of some means to some end is ever arbitrary.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

195

latter by the will itself, Aquinas arrives at a view that makes room both for the non-arbitrariness of some acts of the will and for free will (two desiderata that, I think, are not so clearly satisfied by Henry of Ghent and Godfrey of Fontaines’s views on the will). The challenge, then, is to address the problematic aspects of Aquinas’s distinction. Assuming that the will is moved to the specification of its act by the good apprehended by reason, how does it move itself to the exercise of its act? Aquinas’s claim that it moves itself indirectly does not explain how its potency to the exercise of its act is actualized; it does not explain how the will, by willing some end, “reduces itself to willing the means to it.”²⁷ Moreover, if the good apprehended by reason does move the will to the specification of its act, then should it not, as Henry of Ghent argues, also move the will to the exercise of its act? Aquinas owes us a clear story about how the will moves itself in such a way that its potency to the exercise of its act is actualized. Although Aquinas does not tell us such a story, Sutton does. Sutton adopts Aquinas’s view on the will and develops it, taking more care to explain the way in which the will is in potency to the exercise of its act and the way in which the will moves itself with respect to this potency than Aquinas does in the Prima secundae and the Disputed Questions on Evil. In the following two sections, I would like to consider the details of what Sutton says about the will, focusing on his Quaestiones ordinariae.²⁸

3. Thomas of Sutton on Specification and Exercise Sutton thinks that every passive power of the soul is in potency both to the specification and to the exercise of its act.²⁹ In things that do not have cognition, he says that a passive power is moved both to the specification and to the exercise of its act by the same object.³⁰ For example, ²⁷ Thomas Aquinas, ST, I-II, qu. 9, art. 4, co. (Leonine VI, 78). ²⁸ In general, dating both Sutton’s Quodlibeta and his Quaestiones ordinariae is difficult. Schneider groups them all around 1300 (see the introduction of his edition of Sutton’s Quaestiones ordinariae). ²⁹ Thomas of Sutton, Quodlibet (= Quodl.) III, qu. 13 (ed. Schmaus and GonzálezHaba, 424). ³⁰ Thomas of Sutton, Quaestiones ordinariae (= Quaest. ord.) 6 (ed. Schneider, 169).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

196   something that is able to be heated is moved by something that heats in such a way that its change, namely, becoming hot, is both specified and “impressed on it” (imprimitur sibi) by the thing that heats. Put differently, the thing that heats explains why the thing that is able to be heated is heated to some temperature rather than another and is actually heated. Sutton argues that, since such a passive power is moved by the same object “with respect to everything” (quoad omnia), it is moved necessarily.³¹ Similarly, since it is a passive power of the soul, “having no actuality from itself” (nihil habens de se actualitatis), the will is in potency to the specification and to the exercise of its act.³² If the will were moved both to the specification and to the exercise of its act by the same object, as happens with passive powers in things that do not have cognition, then it would be moved necessarily by that object and therefore not will freely. But if the will were not to will freely, then Sutton says that all the principles of moral philosophy would be subverted. There would be no merit or demerit, punishment or reward, praise or blame, and so on, which is absurd.³³ Supposing that it does will freely, how is the will moved to the specification and to the exercise of its act? Sutton follows Aquinas by arguing that the will is moved to the specification of its act by the good apprehended by reason and to the exercise of its act by itself.³⁴ The will, then, is similar to other passive powers inasmuch as it is moved to the specification of its act by its object, but differs from them inasmuch as it is moved to the exercise of its act by itself.³⁵ And since the will is moved to the exercise of its act by itself, it is possible for reason to apprehend some object as good, but the will not will that object. Therefore, the will is not necessitated by the good apprehended by reason but wills freely.

³¹ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 169). ³² Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 168). See further Quaest. ord. 7 (ed. Schneider, 211–2); Quaest. ord. 15 (ed. Schneider, 441). ³³ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 164–5). ³⁴ Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 212). See further Quaest. ord. 24 (ed. Schneider, 671); Quaest. ord. 6 (ed. Schneider, 168–70); Quodl. III, qu. 13 (ed. Schmaus and González-Haba, 426–8). ³⁵ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 169).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

197

Sutton is aware that he faces a difficulty here, for he rigidly adheres to the act–potency axiom,³⁶ according to which it is not possible for something to be in act and potency at the same time and in the same respect. The act–potency axiom has the consequence that, strictly speaking, nothing can move itself.³⁷ This is because movement, in general, is the actualization of potency, or the “reduction” of potency to act. So, for something to move itself, it has to actualize its own potency. But for something to actualize its own potency, it has to be in act and potency at the same time and in the same respect, which violates the act–potency axiom. Sutton, like Aquinas, has the task of explaining how the will is able to move itself to the exercise of its act. On the one hand, he thinks that it is not possible for the will to move itself; the act–potency axiom has no exceptions. On the other hand, since the will is a passive power of the soul, and a passive power that is moved both to the specification and to the exercise of its act by the same object is moved necessarily, he thinks that the will does move itself to the exercise of its act. So, how does the will move itself to the exercise of its act if everything moved is moved by another?

4. Thomas of Sutton on the Will’s Self-Motion We know Aquinas’s answer to our question. He draws a distinction between direct and indirect self-motion, that is, between something’s reducing its own potency directly and something’s reducing its own potency indirectly. Direct self-motion violates the act–potency axiom, but indirect self-motion does not, for when something moves itself indirectly, it actualizes its own potency by actualizing the potency of something else. According to Aquinas, the will moves itself to the ³⁶ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 166). See further Quaest. ord. 2 (ed. Schneider, 57–8); Quaest. ord. 8 (ed. Schneider, 258); and Quaest. ord. 15 (ed. Schneider, 436–7). For Aristotle’s arguments, see Physics VIII, 257a33–b13. ³⁷ A composite thing can move itself in the sense that one of its parts moves another, but Sutton denies that it can move itself “as a whole” (secundum se totam). Since it is simple, the will cannot be divided into a part that moves and a part that is moved. See Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 165–6). Cf. Sharp, “Thomas of Sutton,” 352–3.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

198   exercise of its act by moving itself indirectly, that is, by moving reason. But, as I argued, since reason moves the will by actualizing its potency to the specification of its act, it is not clear how the will is supposed to actualize its potency to the exercise of its act by moving reason. Aquinas, then, answers our question of how the will moves itself to the exercise of its act by drawing a distinction between ways that a thing is able to actualize its own potency. We will find that Sutton’s answer is both similar to and different from that of Aquinas. For in Quaest. ord. 7, he argues that the will “moves itself in a twofold mode” (dupliciter movet se): It is evident from what has been said that the will moves itself in a twofold mode. First, [it moves itself] “accidentally” by moving reason to deliberate through counsel about what should be willed. [Second,] with deliberation having been concluded, the will, by willing the end, moves itself to willing that which was arrived at through deliberation, not “effectively,” but “consecutively.”³⁸

I would like to unpack the twofold mode through which the will moves itself, starting with the first mode. The will, according to Sutton, moves itself per accidens, or “accidentally,” by moving reason to deliberate about what should be willed. While the appetite of a brute is moved immediately to pursue or avoid an object that is grasped as good or bad, respectively,³⁹ the will, a rational appetite for the good, is not necessarily moved to will an object grasped by reason, unless that object is grasped as good “according to every consideration” (secundum omnem considerationem).⁴⁰ Outside of the beatific vision, however, an object is only grasped as good “under some aspect” (sub aliqua ratione boni). I might apprehend going to the beach as good for relaxation, for example. And if I will relaxation, then my will moves reason to consider whether or not going to the beach should be willed, that is, whether going to the beach is appropriate (conveniens) for my end. If I judge that going to the beach should be willed, then I will it. By moving reason, then, the will is able to indirectly move itself. The will’s accidental self-motion is not a

³⁸ Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 218). ³⁹ Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 214). ⁴⁰ Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 214).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

199

violation of the act–potency axiom, for that axiom, again, merely has the consequence that nothing is able to actualize its own potency directly. On the one hand, Sutton’s answer to our question, “How does the will move itself to the exercise of its act?” is similar to that of Aquinas; Aquinas thinks that the will moves itself indirectly by moving reason, and Sutton that the will moves itself accidentally “by moving reason to deliberate through counsel about what should be willed.”⁴¹ Both think that, if the will, willing some end, moves reason to deliberate about the means to that end and then wills some means to it, then the will is moving itself to the exercise of its act of willing that means; the will, as Aquinas says, moves itself “through a process of deliberation.”⁴² That the will moves itself indirectly (or accidentally) to the exercise of its act is a good answer to our question insofar as this kind of self-motion does not violate the act–potency axiom, but it is not good enough, since it does not explain how the will’s potency to the exercise of its act is actualized by its moving reason. On the other hand, Sutton’s answer to the question differs from that of Aquinas; he does not think that the will merely moves itself accidentally (or indirectly): “with deliberation having been concluded, the will, by willing the end, moves itself to will the means to that end, not effectively, but consecutively.”⁴³ What does he mean? Whereas Aquinas draws a distinction between different ways that a potency is reduced, namely, directly or indirectly, and accordingly distinguishes between two different kinds of self-motion, namely, direct and indirect, Sutton draws a distinction between different kinds of potency, namely, “essential” and “accidental,” and accordingly distinguishes between two different kinds of self-motion, namely, “effective” and “consecutive.” In general, self-motion is the reduction of some potency to act by the thing having that potency. Effective self-motion is when the potency that is reduced to act is “essential,” and consecutive self-motion is when the potency that is reduced to act is “accidental.” ⁴¹ Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 218). ⁴² Thomas Aquinas, ST, I-II, qu. 9, art. 4, co. (Leonine VI, 78). ⁴³ Sutton does sometimes use ‘illative’ instead of ‘consecutive,’ but the two are synonymous. Also, note that, in general, the consecutive mode of self-motion is supposed to be effective “according to a similitude.” See Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 213).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

200   Sutton will argue that effective self-motion violates the act–potency axiom, but consecutive self-motion does not, and since the will, willing some end, moves itself to the exercise of its act of willing some means to that end consecutively, its self-motion does not violate the act–potency axiom. To better understand the distinction between essential and accidental potency, the one between effective and consecutive self-motion, and how the will moves itself to the exercise of its act consecutively, I would like to examine some of Sutton’s examples of consecutive self-motion. Consider the following passage: In the generation of fire, there is no middle potency that must be actualized between the induced substantial form and the heat that is consequent to that form, but the generator giving the substantial form [also] gives heat and the other accidents that are consequent to that form. It also happens in heavy and light objects that the generator giving weight gives the consequent motion, unless it is impeded. And if it is impeded, as soon as that impediment is removed, the same generator gives that motion effectively, as is said in Physics VIII. Thus, a heavy object does not move itself effectively, by drawing itself from potency to act. And therefore, that a heavy object moves itself consecutively is neither inappropriate, nor against the Philosopher, since [if it moves itself consecutively] it does not follow that it draws itself from potency to act.⁴⁴

Sutton is using a passage from Physics VIII that medieval philosophers generally understood as drawing a distinction between two kinds of potency: essential and accidental.⁴⁵ Consider, for example, an object that is cold, but able to become hot. The object is in one kind of potency, namely, essential potency to being hot. If something hot is made present to the object, then the object’s essential potency to being hot is actualized; the object undergoes an “essential reduction” from potentially being hot to actually being hot.⁴⁶ The now hot object is able to make other things ⁴⁴ Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 216). ⁴⁵ Physics, VIII, 255a27–255b23. See further De Anima, II, 5, 417a22–417b1. ⁴⁶ Peter Hartman, “Causation and Cognition: Durand of St. Pourçain and Godfrey of Fontaines on the Cause of a Cognitive Act,” in A. Speer et al. (eds.), Durand of St. Pourçain

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

201

hot. But suppose that there is nothing around for it to heat. The object, then, is in a second kind of potency, namely, accidental potency to heating, that is, “it is now something able to make other things hot but (provided there is nothing around for it to heat) it is not something actually making other things hot.”⁴⁷ If something that is able to be heated is made present to the object, then the object’s accidental potency to heating is actualized; the object will undergo an “accidental reduction” from potentially heating to actually heating.⁴⁸ The two kinds of potency are not reduced the same way. In our example, when the object undergoes an essential reduction, becoming hot, it receives a form; the object is changed from cold to hot. When it undergoes an accidental reduction, heating a thing that has been made present to it, the object does not receive a form. For the object is hot and thereby has everything that is necessary for it to bring about an act of heating. But it will not heat if there is nothing around for it to heat. When a thing that is able to be heated is made present to the object, however, its accidental potency to heating is actualized. With this distinction between essential and accidental potencies in hand, I would like to come back to what Sutton says in Quaest. ord. 7 about the generation of fire and the generation of heavy and light objects. He continues his explanation of these examples by noting that, when fire is generated, there is no “middle potency” (media potentia) between the induced substantial form and the heat that follows upon that form.⁴⁹ What Sutton means is that it is impossible to generate fire that is not hot, for being hot is a necessary accident of fire’s substantial form. That is, when the fire is generated, there is no potency that must be reduced for it to be hot. Moreover, the fire will naturally heat in virtue of its substantial form, but if there is nothing present to the fire that is able to be heated, then the fire will be in accidental potency to heating. Similarly, a heavy object will naturally move downwards in virtue of its substantial form,

and His Sentences Commentary: Historical, Philosophical, and Theological Issues (Leuven: Peeters, 2014), 245–7. ⁴⁷ Hartman, “Causation and Cognition,” 246. ⁴⁸ Cf. Thomas of Sutton, Quaest. ord. 2 (ed. Schneider, 55–6). ⁴⁹ Sutton argues similarly about risibility in Quaest. ord. 24 (ed. Schneider, 675).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

202   but if there is an obstacle that impedes it from moving downwards, then it will be in accidental potency to moving downwards. The fire and the heavy object do not have to receive forms to heat and move downwards, respectively. That is, neither has to undergo an essential reduction. When a thing that is able to be heated is made present to the fire, the fire will naturally heat that thing in virtue of being hot, and when the obstacle impeding the heavy object from moving downwards is removed, the heavy object will naturally move downwards in virtue of being heavy. The fire and the heavy object move themselves consecutively inasmuch as heating and moving downwards naturally follow being hot and being heavy, respectively; neither moves itself effectively, that is, neither reduces itself from essential potency to actuality, but rather from accidental potency to actuality. Whereas effective self-motion is “essential self-reduction,” consecutive self-motion is “accidental self-reduction.”⁵⁰ Insofar as effective selfmotion is essential self-reduction, it violates the act–potency axiom.⁵¹ Consider that, for an object that is potentially hot to become actually hot, it has to receive a form, reducing its potency to act. If the object reduces this potency to act by itself, then it receives the form from itself. But if it receives the form from itself, then it already has the form to which it is (supposed to be) in potency to receiving. Put differently, the object is both potentially hot and actually hot at the same time and in the same respect, which violates the act–potency axiom. Insofar as consecutive self-motion is accidental self-reduction, it does not violate the act– potency axiom. Consider that the fire is now actually hot and potentially heating. When an object that is able to be heated is made present to the fire, the fire’s actually heating that object naturally follows its actually being hot; since the fire’s actually heating that object naturally follows its actually being hot, nowhere in its self-motion is the fire potentially heating that object and actually heating that object. In general, since an

⁵⁰ Sutton does not explicitly distinguish between essential and accidental potency in Quaest. ord. 7 but seems to use ‘potency’ to signify essential potency. While he affirms that it is possible for something to move itself consecutively, he denies that it is able to reduce itself from potency to act. What he means is that a thing is able to reduce itself from accidental potency to act but not from essential potency to act. ⁵¹ Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 219).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

203

object moves itself consecutively when one of its acts naturally follows another of its acts, consecutive self-motion preserves the act–potency axiom. The will’s consecutive self-motion is supposed to be like the consecutive self-motion of the fire and the heavy object: the will, Sutton says, moves itself “in such a way that one of its acts [willing the means] is consequent to another of its acts [willing the end].”⁵² As there is no “middle,” or essential, potency between being hot and heating or between being heavy and moving downwards, so there is no essential potency between willing the end and willing the means; the will, when it actually wills the end, does not have to receive a form to actually will the means.⁵³ And as the fire, by being hot, moves itself consecutively to heating, and the heavy object, by being heavy, moves itself downwards consecutively, so the will, by willing the end, moves itself consecutively to willing the means, since willing the means naturally follows willing the end. Since, when it moves itself consecutively from actually willing the end to actually willing the means, the will does not receive a form from itself, nowhere in the will’s self-motion is the will potentially willing the means and actually willing the means; one act of the will, namely, willing the means, naturally follows another of its acts, namely, willing the end.⁵⁴ The act–potency axiom is therefore not violated. I have to make two clarifications. First, I called consecutive selfmotion, in general, “accidental self-reduction,” that is, the reduction of an accidental potency by the thing having that potency. Of course, it is necessary that something be in accidental potency for it to reduce that potency to act by itself. Although the fire’s heating naturally follows its being hot, the fire will not heat if there is nothing around for it to heat; it will be in accidental potency to heating since there is an impediment, namely, the absence of something around for it to heat, that blocks its heating from naturally following its being hot. Similarly, the heavy object’s moving downwards naturally follows its being heavy, but it will not move downwards if there is something blocking it from doing so; it

⁵² Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 215). ⁵³ Thomas of Sutton, Quaest. ord. 24 (ed. Schneider, 675). ⁵⁴ Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 216).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

204   will be in accidental potency to moving downwards since there is an impediment that blocks its moving downwards from naturally following its being heavy. Again, in consecutive self-motion, one act of something naturally follows another act of that thing. If there is no accidental potency, then no accidental potency can be actualized, and if no accidental potency can be actualized, then there can be no consecutive self-motion. But there can be no accidental potency if there is nothing that blocks one act of something from naturally following another act of that thing; the presence of an impediment is what makes something be in accidental potency, and when that impediment is removed, the potency is actualized. When the will, willing the end, moves itself consecutively to willing the means to that end, it is actualizing its accidental potency to willing the means. But for the will to be in accidental potency to willing the means there has to be some impediment that blocks its willing the means from naturally following its willing the end; and that there would be such an impediment makes sense, for when we will some end, it does not happen that we will some means to that end immediately. What, then, is that impediment? In replying to the eleventh objection of Quaest. ord. 24, Sutton argues as follows: The apprehended good does move the will per se, but the will is blocked from willing it because of the bad that is joined to that good. Thus, the will, since it wills the end, moves reason to judge that that good is to be pursued, regardless of the bad joined to it. When this judgment has been made, the obstacle is removed, and the act of willing that good, which is for the end, follows. So, the will, through its act, does not move itself per se, but moves [itself] accidentally, as if removing an obstacle.⁵⁵

Sutton affirms that the good apprehended by reason moves the will, and does so per se. In particular, the apprehended good moves the will to the specification of its act (the “per se” meaning that it moves the will directly, for the apprehended good does not actualize the will’s potency to the specification of its act by actualizing the potency of something else). But there is something that “blocks” the will from willing that good, namely, the bad joined to it. ⁵⁵ Thomas of Sutton, Quaest. ord. 24 (ed. Schneider, 675–6).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

205

Suppose that the will wills some end, then. What makes the will be in accidental potency to willing the means to that end, that is, what blocks the will’s act of willing the means from naturally following its act of willing the end is that the means (a particular good) has some bad joined to it. Consider Sutton’s example in Quaest. ord. 6. The will has a volition for health and reason apprehends bitter medicine. The bitter medicine is not immediately willed as a means, for it is not apprehended as good “according to every circumstance.”⁵⁶ Indeed, the bitter medicine is apprehended as good because it contributes to health, but it is also apprehended as bad because it tastes unpleasant. The will, then, moves reason to deliberate. While reason deliberates, it is not necessary for the will to will the particular good about which reason deliberates. But when deliberation has been concluded and reason makes a “complete judgment” (plenum iudicium), or a judgment that the good should be pursued or should be avoided here and now, then the will necessarily wills or nills that good.⁵⁷ Suppose that reason deliberates and concludes that the bitter medicine is to be taken, that is, reason makes a complete judgment that, although it tastes unpleasant, it is better to take the bitter medicine than not; this particular good, as Sutton says in Quaest. ord. 24, should be pursued non obstante malo adiuncto.⁵⁸ When reason makes a complete judgment, the impediment blocking the actualization of the will’s accidental potency is removed, and the will’s act of willing the means naturally follows its act of willing the end; the will thereby moves itself consecutively. And since the will is what moves reason to deliberate and make the complete judgment that removes this obstacle, Sutton affirms that, in one mode, the will does move itself by moving reason; the will moves itself accidentally, that is, “as if removing an obstacle” (per accidens tamquam removens prohibens).⁵⁹

⁵⁶ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 170). ⁵⁷ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 171). ⁵⁸ Sutton does not, to my knowledge, use the language of “plenum iudicium” in Quaest. ord. 24, but I adopt it as shorthand for a judgment that some particular good is to be pursued non obstante malo adiuncto. ⁵⁹ Cf. Thomas of Sutton, Quaest. ord. 7 (ed. Schneider, 218).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

206   When the will wills some end, for it to be in accidental potency to willing some means to that end, does reason have to think that the means is bad in some way or other? Well, there seem to be other possibilities here that Sutton does not consider. For example, when I will some end, I might think that there are many means to my end that are equally worthy of choice. Here, the obstacle that blocks my willing this means from naturally following my willing the end is not that I think that, in some way or other, the means is bad, but that I do not think it is more worthy of choice than any other means; an impediment such as this is not removed by reason’s making a judgment that the means should be pursed non obstante malo adiuncto but rather by its making a judgment that, for example, the means should be pursued “come what may.” As I have argued, since consecutive self-motion is accidental self-reduction, for something to move itself consecutively, it must be in accidental potency, and what makes something be in accidental potency is that there is an impediment blocking one of its acts from naturally following another of its acts. Sutton suggests that, in the will’s consecutive selfmotion, the impediment that makes the will be in accidental potency to willing the means is the bad joined to that means. But there is nothing in what he says that precludes there being other kinds of impediment. The second clarification that I have to make is that, although Sutton thinks that the will moves itself to the exercise of its act non-accidentally, for it moves itself consecutively, he does not think that the will moves itself in the “voluntaristic” way that, for example, Henry of Ghent does. According to Henry of Ghent, for whom the act–potency axiom does not hold for incorporeal things as it does for corporeal things,⁶⁰ the will moves itself through the actuality of its essence; since it is free, when it wills, the will is moved “from itself, directly.”⁶¹ But, for Sutton, the will is moved through the actuality of its act, namely, willing the end.⁶² Consider this passage from Quaest. ord. 24: The will moves itself with respect to the exercise of its act, and so it draws itself from potency to act in a certain way. But the per se motive ⁶⁰ Henry of Ghent, Quodl. IX, qu. 5 (ed. Macken, Opera omnia, vol. 13, 138). ⁶¹ Henry of Ghent, Quodl. I, qu. 17 (ed. Macken, Opera omnia, vol. 5, 127). ⁶² Thomas of Sutton, Quaest. ord. 24 (ed. Schneider, 671).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

207

principle (principium motivum) here is not the will itself, but the act of the will, which is willing the end (velle finem). Through this, that the will wills the end, it is drawn from potency to the act of willing that which is for the end, with respect to the exercise of its act. And it does not follow that the will draws itself from potency to act per se, as if through an active principle, but only per accidens, namely, through its act. Nor does it follow that, secundum se, the will is in potency and in act with respect to the same thing. As risibility follows humanity, so the exercise of [the will’s] act with respect to that which is for the end naturally follows volition of the end, and thus, properly speaking, there is no drawing from potency to act here, except from an accidental potency.⁶³

By willing the end, the will moves itself to the exercise of its act of willing the means to that end. But it does not follow that the will moves itself per se, or directly, since the per se “motive principle” of this motion is not the will itself, but rather its act, namely, willing the end. The will does draw itself from (accidental) potency to the exercise of its act of willing the means to the end, but it does so by willing the end, and moving reason to deliberate about the means to that end, that is, it does so accidentally, or indirectly. What is directly responsible for the actualization of the will’s potency to the exercise of its act of willing the means is the will’s act of willing the end, from which willing the means naturally follows. Since the will’s act of willing the means naturally follows its act of willing the end, Sutton calls willing the end the per se “motive principle” of the will’s being moved to the exercise of its act of willing the means, and since willing the end and willing the means are both acts of the will, when the latter naturally follows the former, the will is said to move itself consecutively. The difference between Henry of Ghent and Sutton, between the will moving itself through the actuality of its essence (per actualitatem, quae sit de eius essentia) and the will moving itself through the actuality of its act (per actualitatem suae operationis) is easy to see if we consider what happens when the will, willing some end, moves reason to deliberate ⁶³ Thomas of Sutton, Quaest. ord. 24 (ed. Schneider, 671).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

208   about the means to that end, and reason makes a complete judgment that this means should be willed. For Henry of Ghent, when reason makes such a judgment, it is possible for the will not to will the means that reason judges should be willed; again, the will wills “from itself, directly,” and so reason might apprehend and present some means to the will as good (or best) but it is “up to” the will whether or not it wills that means.⁶⁴ For Sutton, however, when reason makes a complete judgment that this means should be willed, the impediment that blocks the will’s act of willing that means from naturally following its act of willing the end is removed; the will cannot but will the means if reason makes such a judgment.⁶⁵ Put differently, whereas for Henry of Ghent the will’s act of willing the means is freely caused by the will itself, for Sutton it is naturally caused by the will’s act of willing the end. Sutton nicely summarizes his complex view of the self-motion of the will in the following passage from Quaest. ord. 24: Nevertheless, inasmuch as the end is the principle of that which is for the end, through this, that it wills the end, the will is moved consequently (illative) to willing that which is for the end, not accidentally, but per se by [this] volition of the end. Or rather, since the will willing the end removes an obstacle, as if moving accidentally, volition of that which is for the end follows per se from volition of the end by a certain natural consequence. And I say [that the will moves itself] to this extent, because nothing moves itself consecutively de novo, unless the consequence of one thing to another is blocked by something else, with which having been removed [the first] follows naturally. Therefore, I said now that the will, by willing the end, removes an obstacle, and the exercise of its act of willing that which is for the end follows. And so, the will, through its act, draws itself from accidental potency to act, but not from essential potency, and it does this not by moving itself immediately, but by moving reason to judging, upon which judgment the exercise of the act of the will, to which [the will] was in accidental potency, follows.⁶⁶ ⁶⁴ Henry of Ghent, Quodl. I, qu. 16 (ed. Macken, Opera omnia, vol. 5, 103). ⁶⁵ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 171). ⁶⁶ Thomas of Sutton, Quaest. ord. 24 (ed. Schneider, 675–6).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

209

Suppose that the will has a volition for some end and that reason apprehends some means to that end. Although willing the means is a natural consequence of willing the end, the will is in accidental potency to the former because the means are not apprehended as good according to every circumstance. Rather, there is some bad joined to the means, which blocks the will from willing it. The will, then, by moving reason to judge that the means is to be pursued non obstante malo adiuncto, removes that impediment. With the impediment having been removed, the act of willing the means naturally follows the act of willing the end. On the one hand, since it wills the means by moving reason, the will is said to move itself accidentally, or indirectly. On the other hand, since the will undergoes an accidental self-reduction when reason judges that the means is to be pursued non obstante malo adiuncto, in which one of its acts, willing the means, naturally follows another of its acts, willing the end, the will is said to move itself consecutively, or “consequently,” to the act of willing the means.⁶⁷ And since, when it moves itself accidentally, the will moves itself by moving reason and, when the will moves itself consecutively, its act of willing the end (not the will itself) is the direct, or per se, motive principle of its accidental reduction, the act–potency axiom is not violated by the self-motion of the will. In this way, the will moves itself in two, related modes.⁶⁸

5. A Development of Aquinas’s View? Sutton has a clearer view than Aquinas about what happens between willing some end and willing some means to that end. The will, when it ⁶⁷ Again, Sutton uses ‘illative’ and ‘consecutive’ synonymously. Here, I think he prefers the former because he presents willing the means as the conclusion, or illation, and willing the end as the premise; self-motion is the consequence between the two. ⁶⁸ The reader will likely have noticed that, in Quaest. ord. 24, Sutton says that the will moves itself “as if” (tamquam) accidentally, whereas elsewhere he says that the will moves itself per accidens. Why the tamquam? Since the motive principle of the will’s self-motion is not the will itself, but rather its act of willing the end, it is not unusual that Sutton would qualify any attribution of self-motion to the will using tamquam. But it also possible that his use of tamquam is not doing much work at all. He might be using tamquam as part of his specification of the manner in which the will draws itself from potency to act (it does so “as if accidentally”). For a similar use of tamquam, see Quaest. ord. 15 (ed. Schneider, 452).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

210   wills some end, is in potency both with respect to the specification of its act and the exercise of its act of willing some means to that end. The will is moved to the specification of its act by reason and moved both accidentally and consecutively to the exercise of its act by itself. Sutton’s view of the will’s self-motion allows him to preserve the act– potency axiom, to explain how the will’s potency to the exercise of its act is actualized, and to conclude that the will is not necessitated by its object, but rather wills freely. I said that Sutton’s accidental self-motion is simply Aquinas’s indirect self-motion. But his consecutive self-motion is novel. Moreover, I think it is the “missing piece,” as it were, in the puzzle that is Aquinas’s view on the will’s self-motion. For Aquinas, again, the will moves itself to the exercise of its act “through a process of deliberation.”⁶⁹ Put differently, when it wills some end, the will moves itself to willing some means to that end by moving reason to deliberate about how to bring that end about. But, as I argued, the deliberation of reason concludes in a judgment that this means should be willed, moving the will to the specification of its act. What about the exercise of the will’s act? How is the will’s potency to the exercise of its act supposed to be actualized by its moving reason? Sutton has an answer: when reason’s deliberation concludes in such a judgment, the will’s act of willing that means naturally follows its act of willing the end. That is, the will moves itself consecutively. By attributing consecutive self-motion to the will, Sutton is able to tell a story about how the will’s potency to the exercise of its act is actualized without positing a mysterious power for essential self-reduction in the will, that is, without violating the act–potency axiom. According to Sutton, the will moves itself in what is fundamentally the same (and familiar) way that fire moves itself to heat, for example. I think that Sutton’s view of the will’s twofold self-motion also gives Thomists the means to address some of the arguments of those who were critical of Aquinas’s distinction between the specification and the exercise of the will’s act. In particular, consider Henry of Ghent’s challenge in Question 5 of his ninth Quodlibet. Again, that challenge supposes that, by moving the will to the specification of its act, the apprehended good ⁶⁹ Thomas Aquinas, ST, I-II, qu. 9, art. 4, co. (Leonine VI, 78).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

211

causes a “certain willing” (quoddam velle) in the will, which the will “carries out” (exsequitur) if it is not impeded.⁷⁰ By actualizing the will’s potency to the specification of its act, then, the good apprehended by reason also seems to actualize the will’s potency to the exercise of its act. Put differently, contra Aquinas, the will seems to move itself neither to the specification nor to the exercise of its act. Aquinas’s claim that the will is moved to the specification of its act by the apprehended good and to the exercise of its act by itself means that there are two potencies in the will that are actualized whenever the will acts. But it does not mean that these potencies have to be actualized the same way. For potencies come in two flavors: essential and accidental. While an essential potency is actualized through the reception of a form, an accidental potency is actualized through the removal of an impediment that blocks one act of something from naturally following another act of that thing. The distinction between these two species of potency (and how they are actualized) is drawn in Sutton’s Quaestiones ordinariae and suggests a possible reply to Henry of Ghent’s challenge. For Sutton, the will’s potency to the specification of its act is not accidental, but essential; the good apprehended by reason is a per se cause with respect to the specification of the will’s act, “impressing” the act of willing this thing on the will, which is to move it effectively.⁷¹ In the specification of the will’s act, the apprehended good, then, does cause an inclination in the will, a “certain willing.” If, moreover, the will is not blocked from willing that good, that is, if reason judges that it should be pursued non obstante malo adiuncto, then the will does “carry out” the act of willing it. Does this mean that, by moving the will to the specification of its act, reason also moves the will to the exercise of its act, as Henry of Ghent argues? If we take “movement” to be the reduction of essential potency to act, then the answer to the question is clear. Supposing that it is willing some end, the will is not in essential potency to the act of willing some means ⁷⁰ Henry of Ghent, Quodl. IX, qu. 5 (ed. Macken, Opera omnia, vol. 13, 123). ⁷¹ Thomas of Sutton, Quodl. III, qu. 13 (eds. Schmaus and González-Haba, 428). See further Quaest. ord. 7 (ed. Schneider, 218). Note that aplenum iudicium is necessary for the will to be moved to the specification of its act, as Sutton explains in Quaest. ord. 6 (ed. Schneider, 178). See further Quaest. ord. 15 (ed. Schneider, 451).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

212   to that end. There is therefore no essential potency with respect to the exercise of the will’s act for reason to reduce by means of its specification. The answer to the question is also clear if we take “movement” to be the reduction of accidental potency to act. Supposing that it wills some end, the will is in accidental potency to the act of willing some means to that end, which has the consequence that, if there is no impediment, the latter naturally follows the former. When reason makes a judgment that some means is to be pursued non obstante malo adiuncto, and thereby specifies the will’s act, it removes the impediment that blocks the will’s act of willing that means from naturally following its act of willing the end. But, again, the per se motive principle of the will’s act of willing the means is its act of willing the end, as the per se motive principle of fire’s heating something is its being hot. By reducing the will’s essential potency to the specification of its act, reason makes it possible for the will’s act of willing the means to follow its act of willing the end as a natural consequence. To the extent that the act of willing the means is a natural consequence of the will’s own act of willing the end rather than some act of reason, however, it is the will that moves itself to the exercise of its act, undergoing an accidental self-reduction. By distinguishing between senses of “potency,” then, Sutton shows that, contra Henry of Ghent, reason moves the will to the specification but not the exercise of its act. There is more. Sutton’s view of the will’s twofold self-motion does not only “fill the gap” in Aquinas’s own view of the will, explaining how the will’s potency to the exercise of its act is actualized, and give Thomists the means to address Henry of Ghent’s criticism of Aquinas. At first glance, it also seems to preserve what makes Aquinas’s distinction between the will’s being moved to the specification of its act (by the good apprehended by reason) and its being moved to the exercise of its act (by itself) attractive. I claimed that, by drawing such a distinction, Aquinas makes room for both the non-arbitrariness of some acts of the will (in particular, the willing of some means to some end) and free will. Consider the first desideratum. Sutton’s view on the will’s self-motion has the consequence that the willing of some means to some end is never arbitrary. Indeed, the will only ever wills some means if reason judges that it should be pursued

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

213

non obstante malo adiuncto (that is, if reason makes a complete judgment that the means should be willed). Since the act of the will and the act of reason are so related, whenever the will wills some means to some end, there is a non-trivial reason that explains why the will wills that means; the will does not simply will because it wills, but rather wills because reason judges that it should so will. Sutton thereby makes room for the non-arbitrariness of some acts of the will. What about the second desideratum? Since the will wills some means only if reason makes a complete judgment that the means should be pursued, it is possible that reason apprehend some means as good, but the will not will that means; in particular, if reason apprehends some means as good, but does not judge that it should be pursued non obstante malo adiuncto, the will does not will that means. Since every particular good has some bad joined to it, and every means is a particular good, the will does not necessarily will any means. If, however, reason judges that some means should be pursued non obstante malo adiuncto, then the will necessarily wills that means.⁷² But is such necessity opposed to the freedom of the will, a rational appetite for the good? Consider that reason judges that some means is to be pursued non obstante malo adiuncto only because it is moved by the will to deliberate about how to bring about some end that the will wills. As Sutton says, if the will were to oppose the complete judgment of reason and not will the means reason judges should be pursued non obstante malo adiuncto, then the will would just be opposing itself.⁷³ Moreover, when it does will the means that reason judges should be pursued non obstante malo adiuncto, the will is moved to the exercise of its act of willing that means by nothing other than itself. And, according to Sutton, that the will, by willing some end, moves itself to the exercise of its act of willing the means specified by reason is “sufficient for the freedom of willing” (sufficiens ad libertatem volendi).⁷⁴ ⁷² Put differently, the will’s willing some means is not absolutely, but conditionally necessary. ⁷³ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 171). To the extent that it is a rational appetite for the good, if the will were not to will the means that reason judges should be pursued non obstante malo adiuncto, then it would seem to be deficient in its activity. And as Aquinas argues in his Disputed Questions on Evil, it does not pertain to the ratio of a power that it is deficient in its activity. Thomas Aquinas, QDM, qu. 16, art. 5, co. (Leonine XXIII, 305). ⁷⁴ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 172).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

214   But is it? That the will moves itself to the exercise of its act matters for Sutton, for if the will did not move itself, then it would be just like a passive power in a thing lacking cognition, that is, not free.⁷⁵ But how the will moves itself to the exercise of its act also matters. By attributing consecutive self-motion to the will, Sutton is able to explain how the will, by willing the end, actualizes its potency to the exercise of willing some means to that end (without violating the act–potency axiom). Is he also able to explain how it wills freely? For the sake of comparison, consider the fire’s consecutive selfmotion. When something is made present to the fire, it moves itself consecutively to heating that thing; the fire changes from not heating the thing to heating the thing and it is because the fire is hot that it heats that thing. But does the fire have power over this change? No. That fire heats is a natural consequence of its being hot; when something is made present to the fire, heating “just happens.” What about the will? When reason judges that some means to some end should be pursued non obstante malo adiuncto, the will moves itself consecutively to willing that means; the will changes from not willing the means to willing the means and it is because the will wills the end that it wills the means. But does the will have power over this change? If we consider that the fire’s consecutive self-motion is one of Sutton’s models for understanding the will’s consecutive self-motion, the answer seems to be “No.” The will’s willing the means is a natural consequence of its willing the end; when reason judges that the means should be pursued non obstante malo adiuncto, willing “just happens.” Now, there is a difference between the consecutive self-motion of the will and that of the fire, namely, that the fire does not do anything with respect to the removal of the impediment that blocks its heating from naturally following its being hot; the fire has to “wait,” as it were, for something to be made present to it. But the will does do something with respect to the removal of the impediment that blocks its willing the means from naturally following its willing the end. For when it wills some end, the will moves reason to deliberate about how to bring that end about and judge that some means should be pursued non ⁷⁵ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 168–9).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

215

obstante malo adiuncto. Although the fire and the will each moves itself consecutively, the former, but not the latter, also moves itself accidentally, that is, the will moves itself indirectly by moving reason. Since the will does something with respect to the removal of the impediment that blocks its willing the means from naturally following its willing the end, it might seem that, when reason makes a complete judgment that some means should be pursued and the will then wills that means, the will has some power over its act of willing the means; it is the will, after all, that moves reason to deliberate about how to bring the end about. But consider the will’s moving reason more closely. Why does the will move reason? Well, it moves reason because it wills the end. Willing the end, however, is not an act to which the will moves itself.⁷⁶ For, as Sutton says, the will moves itself accidentally and consecutively. But the will cannot move itself accidentally to willing the end, since the will moves itself accidentally by moving reason to the act of deliberation, which is not about the end, but the means.⁷⁷ Moreover, the will cannot move itself consecutively to willing the end, since the will moves itself consecutively when one of its acts naturally follows another of its acts. But there is no act which willing the end naturally follows (the way that willing the means naturally follows willing the end). The consecutive self-motion of the will, then, does differ from that of the fire (and that of the heavy object). But it is not obvious that it differs in such a way that the will has power over the act to which it moves itself consecutively. When it wills some end, the will moves reason to deliberate about how to bring that end about and make a complete judgment that some means should be pursued. But the will does not move itself to willing that end. Considering that the will does not move itself to willing the end, and willing the means follows willing the end the way that fire’s heating follows its being hot, what reason do we have to think that the will has any power over willing the means? Sutton’s claim that it is

⁷⁶ Sutton suggests in Quaest. ord. 7 (ed. Schneider, 218–22) that, in general, the will is moved effectively to its act of willing the end by the apprehension of reason. ⁷⁷ Thomas Aquinas, ST I-II, qu. 14, art. 2, co. (Leonine VI, 106).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

216   sufficient for the freedom of willing that the will move itself to the exercise of its act seems dubious.⁷⁸

6. Conclusion I asked whether it is possible that the means to address some of the confusion surrounding Aquinas’s view on the will might be found in Sutton. Now that we have examined Sutton’s doctrine of the will’s twofold self-motion, I think it is clear that the answer is “yes and no.” Yes, for supposing that the good apprehended by reason actualizes the will’s potency to the specification of its act, by attributing accidental and consecutive self-motion to the will, Sutton is able to make sense of how the will, when it wills some end, actualizes its potency to the exercise of its act of willing some means to that end by itself; the will’s act of willing the means is a natural consequence of its act of willing the end. Since the potency of the will actualized by the apprehended good is essential and the potency of the will actualized by the will itself is accidental, Sutton has a reply to Henry of Ghent’s criticism that, by moving the will to the specification of its act, the apprehended good moves the will to the exercise of its act. The movement that pertains to the specification of the will’s act and the movement that pertains to the exercise of its act are different species of movement. But also no, for the modes of self-motion that Sutton attributes to the will do not matter for the freedom of willing; that the will moves itself accidentally and consecutively to the exercise of its act does not seem to have the consequence that the will has power over that act. Indeed, when the will moves itself accidentally, it moves itself by moving reason, but the will moves reason because it wills some end, and the will does not move itself to the act of willing that end. Moreover, when the will moves itself consecutively, one of its acts (willing the means) naturally follows another of its acts (willing the end). But if the former act naturally follows the latter act, the way that fire’s heating is a natural consequence of its being hot, and the will does not move itself to that latter act, then ⁷⁸ Thomas of Sutton, Quaest. ord. 6 (ed. Schneider, 172).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

217

why think that it has power over the former act? Supposing that it is necessary for the freedom of willing that the will move itself to the exercise of its act, it is not enough to show that the will moves itself, for self-motion is not, by itself, special; one should also like to show how the will’s moving itself makes its activity free. Some modification to the Thomistic doctrine seems to be necessary. But what is it? Here are two proposals. One could, as Henry of Ghent does, argue that there is a distinction to be drawn between corporeal and incorporeal motion, and that the act–potency axiom does not hold for the former as it does for the latter.⁷⁹ On this ground, one could argue that, although it is not possible for corporeal things to move themselves directly, it is possible for incorporeal things to so move themselves. Since the will is incorporeal, then, it is possible for it to move itself directly. And if the will does move itself directly, then it has power over its act of willing. The task for those who attribute such self-motion to the will would be to explain how the activity of the will and that of reason are related. What role does reason play in the direct self-motion of the will, and is that role sufficient to explain how some acts of the will are not arbitrary but motivated by reason? Alternatively, one could, as Godfrey of Fontaines does, reject the supposition that it is necessary for the freedom of willing that the will move itself to the exercise of its act, and affirm that, by moving the will to the specification of its act, the good apprehended by reason also moves the will to the exercise of its act.⁸⁰ If the will is moved both to the specification and to the exercise of its act by the apprehended good, then no act of the will is arbitrary; whenever it wills, it wills because reason judges that it should so will. The task for those who deny the selfmotion of the will and its being necessary for the freedom of willing would be to explain how the will has any power over its act of willing. If my will does not move itself, then in what sense are my acts of willing “up to” me? The question is difficult,⁸¹ but for those who wish to follow ⁷⁹ Henry of Ghent, Quodl. IX, qu. 5 (ed. Macken, Opera omnia, vol. 13, 138). ⁸⁰ Godfrey of Fontaines thinks that there is merely a distinction of reason between the will’s being moved to the specification of its act and its being moved to the exercise of its act. Godfrey of Fontaines, Quodl. XV, qu. 2 (ed. Lottin, PB 14, 8–9). ⁸¹ For Godfrey of Fontaines’s answer, see his Quodl. XV, qu. 4 (ed. Lottin, PB 14, 29–31).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

218   Aquinas in thinking that the will is a rational appetite moved by reason, that not all necessitation is opposed to freedom, and that the act–potency axiom holds in general, it is a question worth trying to answer.⁸² St. Francis Xavier University

Bibliography Denifle, Henri and Émile Châtelain. Chartularium Universitatis Parisiensis, vol. 1 (Paris: Delalain, 1889). D’Ettore, Domenic. “Thomas Sutton’s Doctrine of Analogy: Revisiting a Continuator of Thomas Aquinas,” Nova et Vetera 14 (2016), 1153–74. De Wulf, Maurice. Un théologien-philosophe du XIIIe siècle: Étude sur la vie, les œuvres et l’influence de Godefroid de Fontaines (Brussels: M. Hayez, 1904). Godfrey of Fontaines. Quodlibet XV, ed. O. Lottin, Les Philosophes Belges [= PB], vol. 14 (Louvain: Institut Supérieur de Philosophie de l’Université, 1937). Gossiaux, Mark D. “Thomas of Sutton and the Real Distinction between Essence and Existence,” Modern Schoolman 83 (2006), 263–84. Hartman, Peter. “Causation and Cognition: Durand of St. Pourçain and Godfrey of Fontaines on the Cause of a Cognitive Act,” in A. Speer et al. (eds.), Durand of St. Pourçain and His Sentences Commentary: Historical, Philosophical, and Theological Issues (Leuven: Peeters, 2014), 229–56. Henninger, Mark G. “Thomas Sutton on Univocation, Equivocation, and Analogy,” The Thomist 70 (2006), 537–75. Henry of Ghent. Quodlibet I, in R. Macken (ed.), Henrici de Gandavo Opera omnia, vol. 5 (Leuven: Leuven University Press, 1979). Henry of Ghent. Quodlibet X, in R. Macken (ed.), Henrici de Gandavo Opera omnia, vol. 14 (Leuven: Leuven University Press, 1981). Henry of Ghent. Quodlibet IX, in R. Macken (ed.), Henrici de Gandavo Opera omnia, vol. 13 (Leuven: Leuven University Press, 1983). Henry of Ghent. Quodlibet XII (qq. 1–30), in J. Decorte (ed.), Henrici de Gandavo Opera omnia, vol. 16 (Leuven: Leuven University Press, 1987). ⁸² I would like to thank Martin Pickavé and the anonymous referees who commented on earlier drafts of this paper.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

   ’  

219

Hoffmann, Tobias. “Intellectualism and Voluntarism,” in R. Pasnau and C. Van Dyke (eds.), The Cambridge History of Medieval Philosophy (Cambridge: Cambridge University Press, 2009), 414–27. Jordan, Mark D. “The Controversy of the Correctoria and the Limits of Metaphysics,” Speculum 57 (1982), 292–314. Kelley, Francis E. “Two Early English Thomists: Thomas Sutton and Robert Orford vs. Henry of Ghent,” The Thomist 45 (1981), 345–87. Klima, Gyula. “Thomas of Sutton on the Nature of the Intellective Soul and the Thomistic Theory of Being,” 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), 436–55. Klima, Gyula. “Thomas Sutton and Henry of Ghent on the Analogy of Being,” Proceedings of the Society for Medieval Logic and Metaphysics 2 (2002), 34–44. Klima, Gyula. “Thomas of Sutton,” in H. Lagerlund (ed.), Encyclopedia of Medieval Philosophy: Philosophy between 500 and 1500 (Dordrecht: Springer, 2011), 1294. Lottin, Odon. Psychologie et morale aux XIIe et XIIIe siècles, vol. 1, 2nd ed. (Gembloux: J. Duculot, 1957). MacDonald, Scott. “Aquinas’s Libertarian Account of Free Choice,” Revue Internationale de Philosophie 52 (1998), 309–28. Piché, David. La condamnation parisienne de 1277 (Paris: J. Vrin, 1999). Putallaz, François-Xavier. “Thomas de Sutton, ou la liberté controversée,” Revue thomiste 97 (1997), 31–46. Roensch, Frederick J. Early Thomistic School (Dubuque, IA: Priory Press, 1964). Shanley, Brian J. “Beyond Libertarianism and Compatibilism: Thomas Aquinas on Created Freedom,” in R. L. Velkley (ed.), Freedom and the Human Person (Washington, DC: The Catholic University of America Press, 2007), 70–89. Sharp, D. E. “Thomas of Sutton, O.P. His Place in Scholasticism and an Account of His Psychology,” Revue néoscolastique de philosophie 36 (1934), 322–54. Stump, Eleonore. “Aquinas’s Account of Freedom: Intellect and Will,” The Monist 80 (1997), 576–97.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

220   Thomas Aquinas. Opera omnia iussu Leonis XIII P.M. edita [= Leonine] (Rome: Ex Typographia Polyglotta S. C. de Propaganda Fide, 1882–). Thomas Aquinas. Summa Theologiae: Volume 17, Psychology of Human Acts: 1a2ae. 6–17, ed. T. Gilby (Cambridge: Cambridge University Press, 1970). Thomas Aquinas. On Evil, tr. R. Regan, ed. B. Davies (Oxford: Oxford University Press, 2003). Thomas of Sutton. Quodlibeta, ed. M. Schmaus and M. González-Haba (München: Bayerischen Akademie der Wissenschaften, 1969). Thomas of Sutton. Quaestiones ordinariae, ed. J. Schneider (München: Bayerischen Akademie der Wissenschaften, 1977). Torrell, Jean-Pierre. Saint Thomas Aquinas, tr. R. Royal, vol. 1 (Washington, DC: The Catholic University of America Press, 1996). William de la Mare. “Correctorium fratris Thomae,” in P. Glorieux (ed.), Les premières polémiques thomistes: I. Le Correctorium Corruptorii “Quare” (Kain: Le Saulchoir, 1927). Wilson, Gordon A. “Henry of Ghent’s Written Legacy,” in G. A. Wilson (ed.), A Companion to Henry of Ghent (Leiden: Brill, 2011), 3–23. Wippel, John F. “Godfrey of Fontaines and the Act-Potency Axiom,” Journal of the History of Philosophy 11 (1973), 299–317.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

Walter Burley on Co-Signification in Opaque Contexts Nathaniel Bulthuis

Say that I know that Hesperus is Hesperus. And say that Hesperus is Phosphorus. Do I thereby know that Hesperus is Phosphorus? Puzzles of this sort are quite familiar to students of analytic philosophy. That sort of puzzle even has a pithy name attached to it: Frege’s Puzzle, so-named because Gottlob Frege famously raises it in On Sense and Reference.¹ But philosophers in the late nineteenth and early twentieth centuries were far from the first philosophers to recognize the issues raised by co-referring terms in opaque contexts, nor were they the first to engage in serious philosophical analysis of those issues. This paper is focused on one such earlier analysis: that of Walter Burley (born c. 1275; died c. 1345). Like Frege and others beginning in the late nineteenth century, Burley in the fourteenth century recognizes that opaque contexts raise serious questions about the nature of meaning. In particular, Burley recognizes the potential force that opaque contexts have in undermining his own preferred semantics, one according to which names directly signify things in the world. In response, Burley develops a sophisticated account of our noetic states, one which, when paired with a semantics of propositional attitude reports, is readymade to accommodate his commitment to direct signification even in the face of opacity considerations.

¹ See Gottlob Frege, “Sense and Reference,” The Philosophical Review 57 (1948), 209–30. Frege in fact raises two related puzzles about meaning and reference in that work: one about opaque contexts, and another about identity statements. Nathaniel Bulthuis, Walter Burley on Co-Signification in Opaque Contexts In: Oxford Studies in Medieval Philosophy Volume 8. Edited by: Robert Pasnau, Oxford University Press (2020). © Nathaniel Bulthuis. DOI: 10.1093/oso/9780198865728.003.0007

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

222   This paper will proceed in four parts. The first part presents Burley’s articulation and analysis of a case involving co-signifying terms in an opaque context. The second part examines Burley’s theory of direct signification, and the role of concepts in that account. The third part explores his theory of the real proposition, and the relationship of real propositions to propositions in mental language. The final part applies the results of parts two and three to the case at hand, articulating a semantics of propositional attitude reports that can accommodate typical intuitions about these cases while retaining a commitment to direct signification.

1. The Challenge of Opacity Consider the following case: (1) Nate believes that Marcus is a great orator (2) Marcus is Tullius (3) Nate believes that Tullius is a great orator Does (3) follow from (1) and (2)? It might not seem so. For if I were to fail to recognize that Marcus is Tullius, then, when queried, I might assent to the claim that I believe that Marcus is a great orator but might not assent to the claim that I believe that Tullius is a great orator. And self-reports of one’s belief-states, I take it, constitute good (even if defeasible) evidence of what one believes and of what one does not. Burley himself recognizes the challenge that cases of this sort pose. In his treatise, De obligationibus, for example, Burley considers the following: Another difficulty is where many names signify the same thing, as happens in the case of synonyms. For example, suppose that ‘Marcus’ and ‘Tullius’ are names of the same thing. Suppose that you doubt this, and suppose that you know whose name is ‘Marcus,’ and suppose you know that he runs, and suppose that you do not know whose name is ‘Tullius.’ Now, you know that Marcus runs, but Marcus is Tullius,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

223

therefore you know that Tullius runs. Additionally, therefore, you know who is called ‘Tullius.’ But the positum is that you do not know [that Tullius runs].²

The context of this passage is a disputation de obligatione in which the positum—that is, the proposition initially posited by the opponent in the dispute, which the respondent in this case accepts—is that the respondent does not know that Tullius runs. While ‘Marcus’ and ‘Tullius’ signify the same person (namely, Cicero), the respondent does not take herself to know this, in part because she does not take herself to know who Tullius is. In contrast, the respondent does know (and takes herself to know) who Marcus is, and that Marcus runs, and so will concede these facts in the disputation. Since ‘Marcus’ and ‘Tullius’ do in fact signify the same person, however, it seems she does know that Tullius runs; Tullius just is Marcus, after all, and the respondent knows that Marcus runs. But this is contrary to the positum. How ought we to respond to cases of this sort? Perhaps the most parsimonious response is to accept that the respondent does in fact know that Tullius runs, despite what she takes herself to know and to not know. Call this the resolute response. Proponents of this response typically favor a semantics of names according to which a name directly picks out the thing in the world that it names. ‘Marcus,’ for example, ² Walter Burley, De obligationibus, 39–40: “Alia difficulta est ubi plura nomina imponuntur eidem rei, ut accidit in synonymis. Verbi gratia: sint Marcus et Tullius nomina eiusdem, et hoc dubites, et [scias] cuius nomen est Marcus et scias eum currere, et [nescias] cuius nomen est Tullius. Deinde, tu scis Marcum currere, et Marcus est Tullius, igitur tu scis Tullium currere. Et ultra igitur, tu scis quis vocatur Tullius. Sed positum est quod tu ignores.” I have transposed the nescias/scias in Green’s transcription, since it seems to me likely to be a corruption of the text, either by Green or by the scribe. For one of the points the opponent in the disputation makes at the end of this passage is that, given that Marcus is Tullius, it seems that the respondent does in fact know who is called ‘Tullius’ (“Et ultra igitur [ . . . ]”). But that point only makes sense in a context in which the respondent had initially assumed that she did not know who is called ‘Tullius.’ Likewise, in his own analysis of the case in the following paragraph, Burley writes: “and this [i.e., ‘You know that Tullius runs’] can be true although you do not know what is signified through this term, ‘Tullius’ ” (De obligationibus, 40). But that analysis, likewise, assumes that the respondent does not know who is named ‘Tullius.’ So the initial claim must be that the respondent does not know who is called ‘Tullius.’ (Related arguments can be made that the passage must read such that you do know who is called ‘Marcus,’ contrary to Green’s transcription.) Nothing I claim in this paper is undermined by accepting Green’s transcription, however; the passage concerns co-signification in opaque contexts in either case.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

224   directly picks out Cicero; ‘Tullius’ likewise. Their contributions to meaning are thus the very thing that they each pick out. On this semantics, (4) Marcus is a great orator has the very same meaning as (5) Tullius is a great orator In the current philosophical landscape, for example, neo-Russellians hold that proper names directly refer, such that (4) and (5) express the same thing: namely, a singular proposition partly constituted by Cicero. Moreover, they argue, the objects of our propositional attitude verbs (in the case of singular thought, at least) just are propositions of this sort. Consequently, given that Marcus is Tullius, what I believe when I believe that Marcus is a great orator just is what I believe when I believe that Tullius is a great orator. And so, according to many of these neoRussellians, one is right to infer (3) from (1). As we will see in the next two sections, Burley finds many of the assumptions that drive the resolute response quite attractive. Yet, unlike those who favor a resolute response, he denies that one is always licensed to infer (3) from (1) and (2). For he argues that thinking involves a relationship to a dictum, and that such a relationship can block an inference from (1) to (3): [‘You know that Tullius runs’] signifies that you know this dictum, ‘Tullius runs,’ and this is false, and so the consequence put forth does not hold.³

While ‘dictum’ had been used by others in the medieval period more expansively, to denote things other than meaningful expressions in natural language, Burley uses that term here and in other works in its more traditional sense.⁴ What does it mean, in this context, to know a dictum? Perhaps the simplest reading of that claim is that the immediate

³ Walter Burley, De obligationibus, 40. ⁴ Abelard, for example, is well-known for arguing that statements in natural language signify dicta, which are for him not things, that is, not items falling within any of the ten Aristotelian categories.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

225

object of one’s belief is a dictum—here, an uttered sentence in natural language. This interpretation faces serious difficulties, however. First, such an interpretation would seem to require that natural language precedes thought, since utterances in natural language would, on this interpretation, constitute the objects of our beliefs and other attitudes. But Burley himself argues that public language is a tool developed for expressing our thoughts to one another.⁵ If natural language is a tool developed for expressing our thoughts, then thoughts (at least some thoughts, at any rate) must have objects other than dicta. Likewise, this interpretation seems to slice our attitudes too finely. Assume that I know no English, but that I understand Mandarin and consent to the Mandarin equivalent of ‘Marcus is a great orator.’ Surely it is appropriate to report, ‘Nate believes that Marcus is a great orator.’ And yet this would be false, since I would not in fact bear any sort of cognitive relationship to the dictum, ‘Marcus is a great orator’ (indeed, to any dictum of English).⁶ A better way to understand this claim, it seems to me, is focus not on the dictum itself so much as on a way of representing, which the dictum expresses. Plausibly, ‘Marcus is a great orator’ and ‘Tullius is a great orator’ differ not just orthographically but also semantically. They both represent the very same thing, but the way in which they represent it differs. ‘Marcus is a great orator’ represents Cicero in a Marcus-like way, we might say, whereas ‘Tullius is a great orator’ represents him in a Tullius-like way. And since, unlike the dicta that express them, ways of representing need not be features or elements of natural language, they won’t fall prey to the worries I mentioned above. In fact, Burley is himself committed to these ways of representing, calling them “formal aspects,” rationes formales, of signification. When addressing another case of co-signification in opaque contexts (this one concerned with co-signifying definitions and definita in demonstrative contexts), Burley argues that co-signifying terms can differ in their formal aspects: ⁵ See, e.g., Exp.Perih, k4ra. ⁶ On a contemporary treatment of the puzzles these sorts of cases give rise to, see Saul A. Kripke, “A Puzzle about Belief,” in A. Margalit (ed.), Meaning and Use: Papers Presented at the Second Jerusalem Philosophical Encounter April 1976 (Dordrecht: Springer, 1979), 239–83.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

226   [T]he same thing is signified through a definition and a definitum – for example, the same thing is signified through ‘rational animal’ and ‘human’ – but it is taken up under a different formal aspect. For consider that thing [i.e., humanity]. That thing under one formal aspect is signified through its definition, and under a different formal aspect it is signified through its definitum.⁷

Both ‘human’ and ‘rational animal’ signify the same thing, namely, humanity. But, Burley argues, a definition such as ‘rational animal’ can be more fundamental to one’s noetic structure than ‘human’ because these terms differ in the formal aspects of their signification. We’ll provide a bit more analysis of what these formal aspects are later in this paper. But, for now, let’s focus on the way in which the notion of a formal aspect might help resolve concerns about opacity. Just like the dicta that express them, formal aspects can be individuated more finely than in terms of what they represent. For two different formal aspects can represent the very same thing, though in different ways. For example, the formal aspect expressed by ‘Marcus’ differs from the formal aspect expressed by ‘Tullius,’ even though what they both represent is the same. And since, as we will see in the next section, these formal aspects play a role for Burley not just in signification but in thought as well, we can distinguish thoughts themselves more finely than merely according to the things that those thoughts are about.⁸ For example, the belief expressed in (1) is about the very same thing as the belief expressed in (3). But the belief expressed in (1) can be distinguished from the belief expressed in (3), since the belief expressed in (1) involves a relationship to a formal aspect expressed by ‘Marcus is a great orator,’ whereas the belief expressed in (3) involves a relationship to a formal aspect expressed by ‘Tullius is a great orator.’ And those formal aspects themselves differ from one another. Consequently, my having the belief that Marcus is a great orator need not tell us anything about whether I have the belief that Tullius is a great orator, despite that fact that Marcus is Tullius. ⁷ Quaes.Post, para. 2.38–9 (59–60). ⁸ Burley argues that signs can be distinguished according to their formal and material aspects. But concepts are themselves signs, namely, signs in mental language. See Quaes.Post, para. 2.49 (62–3). And concepts play a central role in Burley’s account of cognition. See §2 infra. So these formal aspects play a role in cognition as well.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

227

What role, exactly, do these formal aspects play in thought? Perhaps the most natural suggestion is that these formal aspects are the immediate objects of our propositional attitudes. On such an account, our thoughts are about things in the world (about Cicero, for example) only in the sense that our thoughts are directed at formal aspects, which serve to pick out those things. However, as we will see in the following sections, Burley himself rejects this account. He argues instead that our propositional attitudes are directed at the very things in the world that our thoughts are about. Burley adopts rather what we can call a conciliatory response. An advocate of a conciliatory response accepts that it is infelicitous (in most contexts, at least) to infer, for example, (3) from (1) and (2). At the same time, she is attracted to many of the assumptions that drive the resolute response. For example, she is attracted to the position that names directly signify things in the world, and that the immediate objects of our propositional attitudes can be such things, or entities partly constituted of them. The challenge for any conciliatory response with regard to cases of opacity is to explain how such a position is tenable. Burley’s semantics, as well as related positions he adopts in the philosophy of mind, provide him the resources to respond to this challenge.

2. Direct Signification To understand Burley’s adoption and defense of the conciliatory response, we need to begin by considering his account of the signification of names. Like most others in the medieval period, a central concept in Burley’s philosophy of language is signification. And, again like most others in the medieval period, Burley’s account of signification is heavily shaped by the Aristotelian tradition.⁹ A thing signifies, as Boethius’s translation of Aristotle records, just in case it constituit intellectum, just in case it establishes an understanding.¹⁰ To understand

⁹ See, for example, Walter Burley, De Puritate Artis Logicae, tr. 1, pt. 3 (p. 54): “Now every name that does not by itself establish an understanding is a syncategorema.” ¹⁰ Aristotle, De interpretatione, tr. Boethius, ch. 3 (p. 7).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

228   Burley’s account of signification, therefore, we need to understand his account of understanding.¹¹ And, to do that, we need to investigate his concept of the concept. Concepts are central to Burley’s theory of language and mind throughout his career, though he comes to call them conceptus, ‘concepts,’ only later in his career.¹² These concepts are intelligible species (hereafter ‘species’), that is, real, representational qualities in the mind.¹³ These species have a dual role in Burley’s philosophy: one semantic, the other cognitive; and the former is dependent upon the latter: Just as in an act of thinking, so in an act of signifying. But in an act of thinking, we must consider three things, namely, the thing thought, and the thinking intellect, and the species by whose mediation the thing is thought [ . . . ]. So too in an act of signifying three things appear: the signifying utterance, the thing signified, and the species of the thing by whose mediation the thing is signified.¹⁴

According to Burley, when we engage in an act of thinking, we require not just a thinker and a thing that is thought, but also some mental representation of that thing; for we can only think about some thing

¹¹ This psycho-causal articulation of signification has led some scholars—most notably Jennifer Ashworth—to argue that we should not confuse signification with contemporary notions of meaning. On Ashworth’s discussion of the differences between signification and meaning, see E. Jennifer Ashworth, “Medieval Theories of Signification to John Locke,” in M. Cameron and R. J. Stainton (eds), Linguistic Content: New Essays on the History of Philosophy of Language (Oxford: Oxford University Press, 2015), 160–3. I believe Prof. Ashworth is right to be cautious on this front. However, certain medieval accounts of signification track quite closely accounts of meaning and reference developed and defended today, enough that I believe important parallels can be drawn; Burley’s account of signification strikes me as one such account. ¹² Burley begins to call intelligible species “concepts” in his questions commentary on the Posterior Analytics, and he maintains that practice throughout the rest of his career. Prior to this Posterior Analytics commentary, he had used ‘concept’ to refer to ficta, that is, products of mental activity that are mere beings of reason. Since he denies the existence of ficta, he denied the existence of concepts in his early works. See Quaes.Perih, para. 3.8 (259). ¹³ For Burley’s account of species, see his De Potentiis Animae. In his literal commentaries on the Perihermeneias, Burley argues that a species is a likeness of a thing (similitudo rei) insofar as it mediates cognition and an affection of a thing (passio rei) insofar as it mediates signification. See Comm.Perih, para. 1.16 (55–6); Exp.Perih, k3va. Burley himself doesn’t consistently use his terminology in this way, however. ¹⁴ Quaes.Perih, para. 1.7 (212). The mediating role of concepts is one that Burley stresses repeatedly throughout his career. See, for example, Quaes.Perih, para. 1.7 (211–12); Comm. Perih, para. 1.16–1.110 (56–7); Exp.Perih, k3rb–va.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

229

because the mind employs some mental item—a species—that represents that thing. Moreover, since names are imposed onto things only insofar as those things are cognized, the mediating role that species play in cognition becomes reflected in signification as well. This account of our concepts and their role in cognition and meaning provides Burley with a reductive analysis of what a formal aspect is. On this analysis, the formal aspect of a meaningful expression in natural language is nothing other than the concept that mediates its signification. For two expressions in natural language, signifying the same thing, differ in formal aspect if and only if their respective significations are mediated by different concepts. Focusing again on demonstrative contexts, Burley makes this point with respect to a definition of a subject and its definitum—or, more precisely, a name of that definitum. According to Burley, the middle term in a demonstration needs to explain why its subject has the attribute that it does, by “expressly conveying” that subject’s nature: In a demonstration [ . . . ], the middle is the cause of cognition in respect to the conclusion, since through a middle in such a demonstration the nature of the species is conveyed expressly.¹⁵

A mere name of the subject does not expressly convey that nature, however, because it does not convey the principles of that nature: But a name of the thing defined, which is the subject of the conclusion, does not signify the thing except implicitly, nor through such a name are conveyed the principles of the thing defined.¹⁶

Unlike a mere name, however, a definition does convey the principles of the subject’s nature: [I]f ‘animal’ is made to signify the genus of human, and ‘rational’ to signify its difference, then this expression, ‘rational animal’ [ . . . ] expressly signifies the principles of humanity.¹⁷

¹⁵ Quaes.Post, para. 11.52 (161). ¹⁷ Quaes.Post, para. 11.53 (161–2).

¹⁶ Quaes.Post, para. 11.52 (161).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

230   Consequently, a definition can do important demonstrative work that a mere name of the subject cannot—even though that definition and name signify the very same thing. Why can a definition do that work? Because a definition is a complex expression, the parts of which are each themselves meaningful expressions that signify some principle of the nature that that definition signifies as a whole. As we have seen, a meaningful expression signifies as it does only because it is mediated by a concept. The parts of a definition therefore signify what they do only because they are mediated by concepts that themselves signify those things. Of course, the signification of the entire definition itself, as a meaningful expression, must likewise be mediated by a concept. But, unlike in the case of a mere name, where the name signifies what it does via a brute act of imposition, the signification of a definition is rather a function of the signification of its constituent parts. Consequently, the signification of a definition is itself mediated by a complex concept, composed of those very concepts that mediate the signification of the elements of the definition itself.¹⁸ ‘Human’ and ‘rational animal’ differ in the formal aspects of their signification, for example, because the signification of ‘human’ is mediated by the simple concept human and that of ‘rational animal’ by the complex concept rational animal. Likewise, the concepts human and rational animal, which are themselves meaningful expressions in mental language, differ in the formal aspects of their signification, but that difference in formal aspect is to be articulated by nothing other than the differences between those concepts themselves.¹⁹

¹⁸ On Burley’s account of the mind’s production of complex concepts, see Walter Burley, In Physicam Aristotelis Expositio et Quaestiones (Hildesheim: Georg Olms, 1972), 8rb–va. See also Quaes.Post., para. 11.52 (161): “[S]uch a middle term [i.e., a term that conveys the principles of the subject’s nature] is composed out of many utterances or out of many concepts, which concepts or utterances signify expressly the principles of the thing defined.” ¹⁹ In demonstrative contexts, Burley can distinguish between the formal aspects of a definition and its definitum by appeal to differences in semantic structure, since definitions are mediated by complex concepts whereas mere names are mediated by simple concepts. He can’t appeal to that sort of difference in the Ciceronian case, however, since ‘Marcus’ and ‘Tullius’ are both semantically simple. But I assume that, in this case, Burley could differentiate between the relevant concepts by appeal to differences in their causal histories.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

231

While concepts, or species, mediate thought and signification, however, Burley is at pains to stress that a species is not what itself is thought, nor consequently what is signified: A thing outside the soul is understood before an affection which is in the soul [i.e., a species], because a thing outside the soul is understood directly and an affection of the soul is understood indirectly, through reflection, and direct cognition precedes indirect or reflective cognition. Therefore, in the first way in which a thing outside the soul or an affection of the soul is understood [i.e., as an object of thought], the intellect can impose a name on a thing outside of its cognitive faculties, although that name is not imposed onto an affection of the soul, since an affection of the soul is not then known, and a noun is not imposed except on what is known.²⁰

Burley here argues that, in cases of “direct” or first-order cognition, the object of a cognitive act cannot be a species, because species, as qualities of the mind, are only ever known via reflection on one’s mental states. Rather, in cases of first-order cognition, the object of a cognitive act is a thing outside the mind that a species represents. Consequently, since we only impose names onto what we cognize, and since what we cognize in first-order cognitive acts are things outside the mind, names themselves (most, anyway) are made to signify things outside of the mind. On this view, then, thought is a mediated relation—mediated by species—but it is also, as Burley tells us here and elsewhere, a direct (directe) relationship that a mind bears to the thing that that species represents, such that when we think about some thing, our thought is directed at that very thing, and not at the species in the mind which represents it.²¹ And since, for Burley, signification is parasitic on ²⁰ Exp.Perih, k3va. Burley uses ‘affection of the soul’ (passio animae) and ‘species’ equivalently in this work. In his 1301 commentary on the Perihermeneias, Burley uses ‘passio’ to denote things outside the soul insofar as they are suited to cause a thought in us. Consequently, in that work, passiones are quite different from species. See Quaes.Perih, para. 1.7 (211–12). ²¹ I suggest we think of mediation here in terms of a kind of grounding relation. On this interpretation, a mind’s conceiving of some thing is grounded in another, more fundamental relationship (some real, non-intentional relationship, in this case) that the mind bears to one of its species. Likewise, an utterance’s signifying a thing is grounded in some other relationship (a causal relationship, for example) that that utterance bears to a species in the mind that represents that thing.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

232   cognition, the signification of a thing by a name evidences a similar direct relationship. Consequently, while our words and thoughts only hook onto reality because of the species we possess, in most cases, species are not what our words signify, nor are they the immediate objects of our thoughts. Rather, our words directly signify and our thoughts are directed at the things that those species represent. In fact, since Burley is an Aristotelian and a realist about universal forms, such that universals exist concretely within individual substances such as you and I, these direct semantic and cognitive relationships that our words and thoughts bear to the world apply not just to singular terms but to common ones as well: to ‘human,’ ‘cat,’ ‘red,’ and so on—and more fundamentally to their conceptual analogs—each of which directly signifies some concrete, universal form out in the world.²²

3. Real Propositions and Mental Language Terms for Burley constitute the most basic building blocks of meaning. We can combine terms with one another to form more complex subpropositional expressions. Moreover, with the aid of the copula, we can even combine terms to form propositions. We can do this with terms in natural language: we can combine ‘cat’ and ‘black,’ for example, to produce ‘black cat’ or, with a copula, ‘The cat is black.’ At a more fundamental level, we can do this with our concepts: the concepts cat and black can be combined to produce the complex concept black cat or, via a mental act of “combining and dividing,” to produce the proposition in mental language the cat is black.²³ This is, in many ways, familiar territory for scholars of fourteenthcentury philosophy of mind. What is peculiar about Burley’s view,

²² To address cases of reference failure—for example, kind-terms where the kind isn’t presently instantiated—Burley argues that, if an expression fails to signify something that exists subjectively in extra-mental reality, that expression still signifies something that exists merely objectively in the mind. These intra-mental things are distinct from concepts, however, which exist subjectively in the mind. On Burley’s account of objective being, see Exp.Praed, g7ra–b; Walter Burley, Tractatus de Universalibus, 60–6. ²³ On Burley’s discussion of mental language and its relationship to natural language, see Quaes.Post, para. 2.49 (62).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

233

however, is that, in addition to propositions in natural and mental language, Burley argues that we have the capacity to form a further kind of proposition: a real proposition (propositio in re). Real propositions are, like propositions in mental language, propositions formed by the mind. Just as in the case of propositions in mental language, real propositions have as their copula a mental act of “combining and dividing” that joins the terms of the proposition together. But, crucially, they differ from propositions in mental language with respect to the nature of their terms. Whereas propositions in mental language have concepts as their terms, real propositions have as their terms things out in the world that those concepts signify. Burley’s defense of real propositions can be found throughout his career; the central elements of the theory are already present in one of his earliest commentaries, a questions commentary from 1301 on the Perihermeneias, for example.²⁴ His most mature treatment of them, however, occurs in his 1337 commentary on the Categories. In that commentary, Burley writes that: Whatever the intellect can put together or set apart from one another can be parts of a sentence, and consequently can be a subject or a predicate. But the intellect is able to put together things, by asserting that they are the same, and can set things apart from one another by asserting that they are not the same [ . . . ]. Therefore some proposition is composed of things outside the soul.²⁵

Using a hylomorphic analogy, Burley argues that the “form” of a real proposition is a mental act.²⁶ For a mental act is what gives the real proposition its propositional, or truth-conditional, structure, in virtue of the fact that real propositions are constructed, or composed, by an act of

²⁴ See Quaes.Perih, para. 3.55–3.554 (248–50). Note that in this and other early works, Burley calls a real proposition a ‘mental proposition’ (enunciatio in mente/oratio in mente/propositio in mente). Part of the reason for this terminology is that, in those early works, Burley hasn’t yet developed a full-fledged theory of mental language. When, beginning in his questions commentary on the Posterior Analytics, he adopts a theory of mental language, Burley begins to use the more familiar expression ‘real proposition’ (propositio in re) to refer to what he had up to that point called a ‘mental proposition,’ and sticks with that terminology throughout the rest of his career. ²⁵ Exp.Praed, c3vb. ²⁶ See Exp.Praed, c4rb.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

234   the mind in which the mind represents one thing’s being another. Consequently, Burley states, real propositions are (at least partly) in the mind, on account of the role that mental acts play in forming real propositions.²⁷ But, Burley argues, the “matter” of a real proposition— that is, the subject and predicate terms themselves—is rarely in the mind.²⁸ For the terms of a real proposition are things, according to Burley, and things are rarely in the mind. A thing is what is signified by a significant utterance, and, as we’ve already seen, Burley holds that signification is, in standard cases, a direct relationship between utterances and things in the world. ‘April,’ for example, signifies April, a giraffe in upstate New York that recently was, as the kids say, “internet famous.” Likewise ‘giraffe’ signifies giraffehood, if I may: a universal substantial form, out in the world, that all giraffes possess.²⁹ So Burley’s claim that things can be combined by the mind to form a proposition amounts to the claim that, for example, the mind can combine April, the giraffe herself, and giraffehood itself, a certain universal form out in the world, into the real proposition that April is a giraffe, a proposition in which April plays the role of subject and giraffehood that of predicate.³⁰ To drive this point home, Burley remarks throughout his career that the things which the mind combines into a real proposition are extra animam, outside of the soul.³¹ Indeed, in one of his last works, he will go so far as to stress that these things are totaliter extra animam, completely outside of the soul.³² ²⁷ See Exp.Praed, c4rb. See also Quaes.Perih, para. 3.622–3 (251–2): “[I] concede that a proposition is an accident, according to which an accident is distinguished from a substance, and has subjective being in the soul [ . . . ]. A proposition has subjective being in the soul and yet its parts do not.” The “parts” here are the things outside the mind that the mind combines with one another or divides from one another. That a real proposition is mental, then, is due not to its terms but to the mental act, a real accident of the mind, which combines those things together into a proposition. ²⁸ See Exp.Praed, c4rb. ²⁹ On Burley’s account of universals and its relationship to his semantics, see Nathaniel Bulthuis, “Properties in Walter Burley’s Later Metaphysics,” in C. Rode (ed.), A Companion to Responses to Ockham (Leiden: Brill, 2016), 140–72. ³⁰ Burley argues repeatedly that things can play the role of subjects and predicates. See, e.g., Quaes.Perih, para. 3.621 (251); Exp.Praed, c4rb. This should be expected, however, given his claim that the mind can form a proposition out of things, and that a proposition is composed of a subject and a predicate held together by a copula, that is, by the mind’s combining one thing with another. ³¹ See Quaes.Perih, para. 3.553 (249); Quaes.Post, para. 3.53 (63); Exp.Praed, c3vb–c4va. ³² See Exp.Praed, c4rb.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

235

The claim that the mind can produce a proposition composed of things existing outside of it might sound bizarre.³³ But it can be made more plausible when seen in the light of Burley’s account of thinking that we explored in the last section. On that account, thinking is a direct intentional relationship between thinker and thing, such that my thought is directed at that thing. To think of April is to have April as an immediate object of my thought; likewise in the case of giraffehood. I can of course think of April and giraffehood serially: first a thought of April, then of giraffehood. But—and this is what motivates the theory of the real proposition, I think, at least in part—I can also think of those things in combination with one another: of April’s being a giraffe, for example. The mode of thought here differs, of course; this form of thinking is complex, or propositional, in contrast to the cases of simple cognition we’ve been discussing so far. But the nature of the relationship itself—a direct intentional relationship between the mind and thing(s) in the world—is the same with respect to both kinds of cognition. A real proposition, on this interpretation, is just a mind’s being directed at certain things in the world, thinking of those things as related to one another in a certain way: as April’s being a giraffe, for example. Burley draws an analogy with sense-perception in an attempt to make this point about the nature of the real proposition clearer. Writing that anything put together by the mind, such as a real proposition, can be called an ens copulatum, Burley writes that: An ens copulatum can come about not only through the intellect but also through one’s sense faculty, or through the attention [intentio] of that sense faculty. For Augustine says that attention joins together a sense or the operation of a sense with a sensible object. For oftentimes

³³ Scholars of Burley have generally shied away from this sort of interpretation of Burley’s theory of the real proposition, favoring instead accounts according to which real propositions are, alternatively, states of affairs existing entirely outside of the mind, or intentional objects of our attitudes existing entirely within it. Reasons for this are myriad, but one important reason seems to me to be worries about prima facie plausibility of an account according to which real propositions are partly in the mind and partly outside of it. (Some scholars of Burley have expressed this worry to me in personal correspondence, for example.) A notable exception is Elizabeth Karger, who defends an interpretation of the account similar to the one I articulate here. See Elizabeth Karger, “Mental Sentences According to Burley and to the Early Ockham,” Vivarium 34 (1996), 192–230.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

236   something visible is offered to a sense, yet that which is present to that sense is not sensed if the sense neither turns nor attends to the object. [ . . . ] Attention couples the act of seeing or sight with an object. For example, this utterance, ‘seen stone,’ signifies a certain being composed out of a stone which is outside the soul and the act of seeing which is in the eye, nor is it incongruous that from such things separated by place and subject there comes about an ens copulatum.³⁴

To have occurrent thought or occurrent sight requires not just a mental power such as intellect or vision, nor even that power and some potential object(s) towards which the power could be directed. Rather, cases of occurrent thought or occurrent sight require that the power is actualized such that it in fact attends to, or is directed towards, its object(s). Vision, for example, only occurs when there is a thing, such as a stone, at which our visual power is directed. The mere presence of something in our visual field, if we don’t attend to it, isn’t sufficient for sight. Oftentimes, in cases of occurrent vision, we’re only interested in the object of the vision itself—in the stone itself, for example—quite apart from our seeing it. But, Burley argues, we can also talk about that object qua something seen. For example, as awkward as such an expression might be, we can talk about that stone qua seen, through the use of the expression “seen stone.” A similar distinction holds true at the level of occurrent cognition. Occurrent cognition only occurs when our cognitive power is actually directed at some thing(s) in the world. Like in the case of occurrent vision, we might at times simply be interested in talking about the things themselves that that cognitive activity is directed towards. For example, I can talk about April by uttering “April” and giraffehood by uttering “giraffe.” Far more often, however, we are interested in discussing those objects qua objects of a certain sophisticated cognitive power of ours, and in particular interested in discussing them qua thought of as related to each other in a certain way.³⁵ We are so interested, in fact, that, unlike in the vision case, we have a very standard

³⁴ Exp.Praed, c4va. ³⁵ We could also, I suppose, talk about those things qua objects of simple cognition, by way of expressions such as ‘cognized April’ or ‘thought April,’ for example. Burley himself does not consider such cases, however.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

237

and familiar way of doing this: by uttering a statement (propositio).³⁶ For statements in natural language signify real propositions, complexes consisting of things in the world (signified by the terms of our statements) and an act of thinking, directed at those things, by which we conceive of those things as related in a certain way (an act represented by the copula—what Burley calls a “principle of thought”—which has the effect of bringing about a similar mental act, directed at those things, in the mind of the audience).³⁷ It might be tempting to think of these real propositions themselves as that towards which our beliefs and other attitudes are directed. But Burley rejects this suggestion. For, Burley argues, the mind’s use of things in the world as subjects and predicates relative to one another in a real proposition always carries with it assertive force, since the mind produces a real proposition composed of things by asserting that those things are the same, or that they are different.³⁸ Consequently, our attitudes are not relations to real propositions independent of those attitudes, but rather are productive of their propositional contents. For Burley, we might say, our attitudes are a matter of content creation, not ³⁶ Burley is especially clear in his early works that propositions in natural language directly signify real propositions. See, e.g., Quaes.Perih, para. 3.55 (248); Comm.Perih, para. 1.2 (57–8), 1.24 (60). He is less clear about this point in his 1337 commentary on the Categories, writing as if propositions in natural language signify propositions in mental language, and that these propositions in mental language in turn signify real propositions, so that the signification of an uttered proposition would be indirect (see Exp.Praed, c4rb–c4va). But that position would be inconsistent with his general position that expressions in natural language directly signify things, not concepts—a position he references shortly after this passage and then provides a fuller articulation and defense of in his 1337 commentary on the Perihermeneias, a companion piece to the Categories commentary (see Exp.Perih, k3va–b). Moreover, those passages in which he seems to allow that propositions in natural language directly signify propositions in mental language have a clear purpose: to argue that, regardless of what a proposition in natural language directly signifies, the signification of a proposition in natural language must ultimately terminate in a proposition composed of things, and not (as he takes many of his contemporaries to suppose) in a proposition composed of concepts, since concepts themselves are signs. So I don’t take Burley himself to be committed to the claim that propositions in natural language only indirectly signify real propositions. ³⁷ Burley argues that ‘is’ doesn’t signify any thing but rather serves to mark out its associated subject and predicate terms to the hearer as subject and predicate terms relative to one another (see Exp.Perih, k8va). Comprehension of an uttered proposition thus amounts to thinking of things signified by the subject and predicate terms of that uttered proposition as subject and predicate relative to one another, which, for Burley, is nothing other than the mind’s thinking of the former thing’s being the latter. ³⁸ Burley argues throughout his career that the mind’s predicating one thing of another involves an assertion. See Quaes.Perih, para. 3.55 (248); 3.553 (249); Comm.Perih, para. 1.24 (60–1); Exp.Praed, c3vb.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

238   content relation. However, mental acts are for Burley a special kind of action, namely, an action that has no product different from itself.³⁹ And so, the real propositions that these actions produce are nothing beyond the actions themselves, either because the action is identical to the proposition, as Burley argues early in his career, or because the action is partly constitutive of the proposition, as its copula, a position which Burley defends in his 1337 commentary.⁴⁰ One way to understand the position that Burley adopts about the real proposition and its relationship to our attitudes is in terms of a distinction, introduced by Susan Brower-Toland, between the content object of an attitude and the referential objects of an attitude.⁴¹ According to that distinction, we distinguish between the referential objects of an attitude, which are what the attitude is about, and the content object of that attitude, which is an object towards which our attitude is directed that represents those things that our attitude is about as being situated in a certain truth-conditional way. Applying that distinction to Burley’s view, Burley holds that our attitudes are about things in the world; indeed, directly so. The belief that April is a giraffe, for example, is about April and giraffehood. Our attitudes thus clearly have referential objects. But our attitudes are not relations directed at something (such as a proposition) that represents those things as being related to one another in a certain way, so that our attitudes have truth-conditional structure only because they are related to something, distinct from them, that has that structure more primitively. Thus, for Burley, the structure of attitudes is not like this:

³⁹ See, e.g., Quaes.Perih, para. 1.61 (211): “No philosopher finds that, through an action remaining in an agent, something is produced in the agent really different from that action.” Burley here is clearly drawing from Aristotle’s account of immanent action. See, e.g., Metaphysics IX, 6 (104818–34); IX, 8 (1050a30–b5); De Anima I, 3 (407a33–4); III, 7 (431a1–8). ⁴⁰ In his earlier commentaries, Burley argues that, while the parts of a real proposition exist outside of the mind, the proposition itself exists within the mind. See, e.g., Quaes.Perih, para. 3.623 (252): “A [real] proposition has subjective being in the soul but its parts do not.” In his 1337 commentary on the Categories, in contrast, Burley argues that a real proposition exists partly inside the mind and partly outside of the mind. See Exp.Praed, c4rb. ⁴¹ Brower-Toland introduces this distinction in her discussion of the evolution of Ockham’s theory of judgment. See “Ockham on Judgment, Concepts, and the Problem of Intentionality,” Canadian Journal of Philosophy 37 (2007), 97–9.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

239

The real proposition that Marcus is a great orator Attitude

Marcus himself

as being

Mind

Quality of being a great orator itself

Rather, representing that those things that are the referential objects of our thought as related to one another in a certain way is simply something that the mind itself does, by asserting one thing’s being the other and thereby predicating one referential object of another. On this view, we don’t believe a proposition so much as we believe propositionally. That is, our attitudes are not relations to something that has its truth-conditional structure independently of them, but rather are actions that involve conceiving of things in the world in a truth-conditional way.⁴² While our attitudes are contentful, then, they have no content objects. Consequently, the structure of our attitudes looks rather something like this: The real proposition that Marcus is a great orator

Mind’s predicating one referential object of another, constituting its attitude

ty

nali

tio nten

i

int

ent

ion

alit

y

Marcus himself

Quality of being a great orator itself

⁴² Burley’s account of our attitudes bears some striking parallels to so-called act-based accounts of the proposition that have been developed and defended in the last decade and a half. See, for example, Peter Hanks, “Structured Propositions as Types,” Mind 120 (2011), 11–52; Friederike Moltmann, “Propositional Attitudes without Propositions,” Synthese 135 (2003), 77–118; and Scott Soames, What Is Meaning? (Princeton, NJ: Princeton University Press, 2010). On at least one interpretation of these accounts, our propositional attitudes are not relations to propositions, but rather are themselves acts whereby the mind forms a proposition.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

240   On this picture, our attitudes have referential objects and are contentful, but they have no content objects. Hopefully that suffices as a short summary of Burley’s theory of the real proposition. But it still leaves a piece of the puzzle unsolved. For what, exactly, is the relationship of these real propositions to propositions in mental language, that is, those propositions that the mind produces out of its own concepts? Indeed, what purpose does mental language even serve in Burley’s theory of mind, given that the theory of the real proposition explains cognitive content? We can begin to answer these questions by turning, again, to the fruits of our labor in the last section. Burley, as far as I am aware, nowhere spells out the cognitive relationship of these two propositions explicitly. But his account of simple cognition suggests something important about the role that concepts play in the formation of a real proposition. In cases of simple cognition, a thinker bears a direct cognitive relationship to some thing in the world. But that relationship is mediated, or explained, by the thinker’s use of a concept that represents that thing. So, for example, my thinking of April occurs only because I utilize the April concept. A similar story needs to be told at the level of our propositional attitudes. To think of April and giraffehood requires I utilize concepts that represent April and giraffehood. The nature of the use of those concepts, however, determines the nature of the cognitive activity itself. In the case of simple cognitive activity, all that’s required is the mere grasping of the relevant concept; in virtue of grasping the concept, I think about the thing that that concept represents. In cases of complex cognitive activity, in contrast, I need to use those concepts in a more complex way: as subject and predicate terms relative to one another, forming a proposition in mental language out of them. And, in fact, this is the position that Burley himself endorses. For, he argues, the mind’s activity of predicating one term of another occurs not just with respect to real propositions but with respect to propositions of every sort: In every proposition there is something formal and something material. The formal element in a proposition is the copula putting together (copulans) the predicate with the subject, and that copula is in the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

241

intellect, because it is a composition or division of the intellect. The material elements of a proposition, moreover, are the subject and the predicate [ . . . ]. A proposition composed out of concepts is totally in the intellect, and a proposition composed out of things is partly in the intellect and partly outside the intellect—with respect to its formal element it is in the intellect, but with respect to its material features it is outside the intellect.⁴³

In just the way that my mere grasping of a concept in the case of simple cognition explains my simple cognition of the thing it represents, in the case of my propositional attitudes, my more complex use of those concepts as subject and predicate relative to one another explains my more complex conception of the things that those concepts represent. It is in virtue of predicating the concept giraffe of the concept April, for example, that I conceive of April’s being a giraffe. As in the case of simple cognition, then, at the level of one’s attitudes, a thinker’s relationship to the things she thinks about is direct but mediated. She is in direct cognitive contact with, for example, April and giraffehood; they are the immediate objects of her thought, the things at which her thought is directed. Unlike many others in the fourteenth century, then, for whom propositions in mental language serve as the contents of our attitudes, for Burley, cognitive content is provided by a sophisticated account of the mind’s direct relationship to the things that it is thinking about. However, a thinker has that direct contact with April and giraffehood only in virtue of the fact that she uses the concepts April and giraffe. Moreover, the propositional character of her thought—her thinking of April’s being a giraffe—is explained by the way in which she utilizes those concepts. To have the belief that April is a giraffe, on this picture, consists in predicating the concept giraffe of the concept April, in which the proposition in mental language April is a giraffe consists. Consequently, though for Burley propositions in mental language do not constitute cognitive content, mental language still plays an essential role in his account of thinking, as that which constitutes the engine of cognition itself.

⁴³ Exp.Praed, c4rb.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

242  

4. The Semantics of Propositional Attitude Reports Let’s briefly take stock of the commitments Burley wishes to defend. Like those who adopt a resolute response, Burley argues that names directly signify things in the world, and that the contents of our propositional attitudes are constituted of such things. So, for example, ‘Marcus’ and ‘Tullius’ both directly signify the very same thing: Cicero; and, relatedly, the content of the belief that Marcus is a great orator just is the content of the belief that Tullius is a great orator, the contents of both beliefs involving the mind’s predicating the quality of being a great orator itself of Cicero himself. Commitments of these sorts have driven many to adopt a resolute response to cases of co-signification in opaque contexts. For, on the plausible assumption that our thoughts our individuated by their contents, and given that the beliefs ascribed to me in (1) and (3) have the very same content, the belief ascribed in (3) must just be the belief ascribed in (1). As we have seen, however, Burley rejects that response. In most contexts, at least, (3) isn’t entailed by (1) and (2). What Burley’s position requires is that our attitudes can be individuated more finely than in terms of their contents, since we need a way to discriminate between, for example, the beliefs ascribed in (1) and (3), but, given Burley’s other commitments, the beliefs ascribed in (1) and (3) have the very same content. The role that concepts, and mental language more generally, play in cognition gives Burley the resources he needs to defend this claim. For the way in which I represent to myself that Cicero is a great orator in the belief expressed in (1), for example, is explained by my predicating a certain concept—great orator, say—of the concept Marcus. And the way in which I represent that Cicero is a great orator to myself in the belief expressed in (3) is likewise explained by my predicating great orator of the Tullius concept. But, of course, the Marcus and Tullius concepts differ from one another, and so the belief expressed in (1) is different from the belief expressed in (3), though they have the very same content. With this analysis of our attitudes in hand, Burley can turn his attention to an analysis of the semantics of our propositional attitude

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

243

reports. Since, for Burley, our attitudes are not relations to propositions but rather actions that produce them, we can treat propositional attitude verbs as unary predicates, which can be further specified by their respective complement clauses. These complement clauses signify things, of course. ‘That Marcus is a great orator,’ Marcum esse magnum oratorem, for example, signifies Marcus and the quality of being a great orator.⁴⁴ But, beyond the things that they signify, Burley holds that these complement clauses also represent those concepts which mediate the signification of those things.⁴⁵ For, as a general matter, our utterances don’t just signify things but also represent the concepts, or species, by which those utterances signify what they do: A spoken name or a verb is a mark or a sign that the one who is speaking has a similitude [i.e., a species] in his intellect of that thing which is signified through the name or through the verb.⁴⁶

My uttering ‘Marcus,’ for example, does not just signify Cicero; it also marks out a concept of mine, Marcus, that mediates the signification of Cicero by ‘Marcus.’ This additional information that our utterances convey about the way we represent the content of our assertions to ourselves is typically semantically inert. When I say that Marcus is a great orator, for example, we are usually interested in what I have said, the content of my assertion: that Cicero is a great orator. However, in certain opaque contexts, we are not interested (at least not typically and primarily) in

⁴⁴ It is not clear to me what, if anything, the complement clause of a propositional attitude verb as a whole signifies. It cannot be a real proposition, for two reasons. First, cognitively, because real propositions are not objects of our attitudes. And, second, linguistically, because complement clauses (given the accusative-plus-infinitive construction Burley uses) do not contain copulae, that is, finite forms of the verb ‘to be,’ and so do not state anything. I am partial to the view that complement clauses, on their own, do not signify any thing at all for Burley, though their parts (e.g. ‘Marcus,’ ‘great orator’) certainly do each signify things in the world. ⁴⁵ This sort of approach, and its application to propositional attitude ascriptions, finds some parallels today. See Thomas McKay and Michael Nelson, “Propositional Attitude Reports,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Spring 2014 edition) , and in particular the second contextualist approach they highlight, a modified version of a view held by Mark Richard. See Mark Richard, Propositional Attitudes: An Essay on Thoughts and How We Ascribe Them (Cambridge: Cambridge University Press, 1990), 133–54. ⁴⁶ Comm.Perih, para. 1.16 (56). Burley uses ‘similitude’ and ‘species’ interchangeably in this passage.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

244   what the speaker asserts. We are rather interested in how the speaker represents the content of her assertion to herself. Since complement clauses such as ‘that Marcus is a great orator’ not only signify, for example, Cicero and the quality of being a great orator, but also mark out the concepts Marcus and great orator by which the speaker represents those things to herself, a speaker’s utterance provides to her audience information about how she represents to herself the content of her assertion. Besides assertion-ascription, belief-ascription is a paradigm case in this respect.⁴⁷ When someone reports that I believe that Marcus is a great orator, for example, they not only report the content of my belief, but also convey how I represent that content to myself: in part, via the Marcus concept. There are, perhaps, some reporting contexts where all we care about is that content itself; Burley himself acknowledges this, arguing that, on one reading of the Ciceronian case, one can in fact infer (3) from (1) and (2).⁴⁸ Consider, for example, situations in which we want to count two people as having the same belief merely because what they believe is the same, even though the way in which each represents what he believes to himself differs from the other. Assume, for example, that Abe would assent to (4) but not (5), and Bob would assent to (5) but not (4). It seems felicitous to say that Abe and Bob share a belief, or at least that they believe the same thing—this despite the fact they wouldn’t agree about the truth of either (4) or (5).

⁴⁷ I am inclined to think that, for Burley, verbs of saying operate in the same way as verbs of knowing and believing, though I know of no texts where he makes this point explicitly. ⁴⁸ Burley argues that ‘You know that Tullius runs’ is “multiplex,” or has many readings. We have so far focused on only one of those readings. But Burley argues that, on another reading of the case, it is true that you know that Tullius runs just in case Marcus is Tullius and you know that Marcus runs. See De obligationibus, 40: “Or it [‘You know that Tullius runs’] signifies that you know about he who is Tullius, that he runs, and this can be true, although you do not know what is signified through this term, ‘Tullius.’ ” And Burley can endorse this position precisely because our attitudes are directed at things, rather than the concepts that represent those things, such that the knowledge-state expressed by ‘knows that Tullius runs’ can be individuated coarsely, according to what is known—the things themselves out in the world, such as Cicero—and not just according to how the agent represents those things to herself. Such a position is not available to someone who holds that the contents of our attitudes are constituted by concepts rather than the things that those concepts represent, since, on that account, what is known when one knows that Tullius runs is different from what is known when one knows that Marcus runs, for example.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

245

Far more often, however, what we care about is how the agent represents the content of her belief to herself. In those cases, we need a more fine-grained analysis of our beliefs, and a way for our belief reports to express that more fine-grained analysis. Burley’s semantics, and the analysis of our attitudes that stands behind it, provides him with this. For when someone reports (1), for example, they ascribe to me the beliefstate of believing that Marcus is a great orator. That belief-state has the same content as the belief-state reported in (3). But it differs from the belief-state reported in (3) because it is individuated by the mind’s predicating great orator of Marcus, whereas the belief-state reported in (3) is individuated by the mind’s predicating great orator of Tullius. The complement clauses of (1) and (3), respectively, convey this difference in mental representation. Consequently, the predicate expressions ‘believes that Marcus is a great orator’ and ‘believes that Tullius is a great orator’ both signify belief-states with the very same content. But the belief-state signified by the former can be distinguished from the belief-state signified by the latter, because the former conveys that the belief-state it expresses is individuated by the mind’s use of the Marcus concept, among others, while the latter conveys that that belief-state it expresses is individuated by, again among other concepts, the mind’s use of a different concept, the Tullius concept. Therefore, (1) ascribes a different belief-state to me than (3), and so (3) need not be inferred from (1), even though Marcus just is Tullius.

5. Conclusion Opaque contexts have long been considered a problem for theories of direct reference—longer than perhaps most philosophers today even realize. Similar to contemporary advocates of direct reference, Walter Burley is committed to direct signification, but he recognizes the problem that opacity generates for such a position. In response, Burley develops a sophisticated account of our propositional attitudes and the ways in which we can talk about those attitudes. Central to that account is that our attitudes are individuated not merely by their contents but also by the way that we represent those contents to ourselves. And his

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

246   theory of mental language and its role in cognition explains how it is that we represent in that way. Paired with a general semantic theory about what our utterances express, and in particular a theory of how our propositional attitude reports operate, Burley provides a robust defense of his commitment to direct signification even while accepting the force of standard intuitions involved in cases of opacity. Saint Joseph’s University

Bibliography Aristotle. De Interpretatione Vel Perihermenias, tr. Boethius, ed. L. MinioPaluello (Bruges: Desclée de Brouwer, 1965). Aristotle. The Complete Works, ed. J. Barnes (Princeton, NJ: Princeton University Press, 1984). Ashworth, E. Jennifer. “Medieval Theories of Signification to John Locke,” in M. Cameron and R. J. Stainton (eds.), Linguistic Content: New Essays on the History of Philosophy of Language (Oxford: Oxford University Press, 2015), 156–75. Brower-Toland, Susan. “Ockham on Judgment, Concepts, and the Problem of Intentionality,” Canadian Journal of Philosophy 37 (2007), 67–109. Bulthuis, Nathaniel. “Properties in Walter Burley’s Later Metaphysics,” in C. Rode (ed.), A Companion to Responses to Ockham (Leiden: Brill, 2016), 140–72. Frege, Gottlob. “Sense and Reference,” The Philosophical Review 57 (1948), 209–30. Hanks, Peter W. “Structured Propositions as Types,” Mind 120 (2011), 11–52. Karger, Elizabeth. “Mental Sentences According to Burley and to the Early Ockham,” Vivarium 34 (1996), 192–230. Kripke, Saul A. “A Puzzle about Belief,” in A. Margalit (ed.), Meaning and Use: Papers Presented at the Second Jerusalem Philosophical Encounter April 1976 (Dordrecht: Springer, 1979), 239–83. McKay, Thomas and Michael Nelson. “Propositional Attitude Reports,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Spring 2014 edition) .

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

  -   

247

Moltmann, Friederike. “Propositional Attitudes without Propositions,” Synthese 135 (2003), 77–118. Richard, Mark. Propositional Attitudes: An Essay on Thoughts and How We Ascribe Them (Cambridge: Cambridge University Press, 1990). Soames, Scott. Beyond Rigidity: The Unfinished Semantic Agenda of Naming and Necessity (New York: Oxford University Press, 2002). Soames, Scott. What Is Meaning? (Princeton, NJ: Princeton University Press, 2010). Walter Burley. De Puritate Artis Logicae, 1st ed. (Louvain: The Franciscan Institute, 1951). Walter Burley. De Obligationibus, in R. Green (ed.), An Introduction to the Logical Treatise De Obligationibus: With Critical Texts of William of Sherwood and Walter Burley, vol. 2 (Louvain: Institut Supérieur de Philosophie, 1963). Walter Burley. Expositio super librum Isagoge in Super Artem Veterem Porphyrii et Aristotelis (Frankfurt: Minerva, 1967). [Exp.Porph] Walter Burley. Expositio super librum Perihermeneias in Super Artem Veterem Porphyrii et Aristotelis (Frankfurt: Minerva, 1967). [Exp.Perih] Walter Burley. Expositio super librum Praedicamentorum in Super Artem Veterem Porphyrii et Aristotelis (Frankfurt: Minerva, 1967). [Exp.Praed] Walter Burley. “The De Potentiis Animae of Walter Burley,” ed. M. J. Kitchel, Mediaeval Studies 33 (1971), 85–113. Walter Burley. In Physicam Aristotelis Expositio et Quaestiones (Hildesheim: Georg Olms, 1972). [In Phys] Walter Burley. “Walter Burley’s Middle Commentary on Aristotle’s Perihermeneias,” ed. S. F. Brown, Franciscan Studies 33 (1973), 45–134. [Comm.Perih] Walter Burley. “Walter Burley’s Quaestiones in Librum Perihermeneias,” ed. S. F. Brown, Franciscan Studies 34 (1974), 200–95. [Quaes.Perih] Walter Burley. Tractatus de Universalibus, ed. H. U. Wöhler (Leipzig: Sächsischen Akademie der Wissenschaften, 1999). Walter Burley. Quaestiones super Librum Posteriorum (Toronto: Pontifical Institute of Mediaeval Studies, 2000). [Quaes.Post]

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

Social Powers and Mental Relations William Ockham on the Semantics and Ontology of Lordship and Ownership Jenny Pelletier

The entities that occupy Ockham’s ontology are singular and few in kind. They are restricted to God, singular substances, their essential parts, and some of their singular qualities. The shape and heft of this ontology as well as the various strategies that Ockham employs for reducing the kind and number of entities to a rationally acceptable minimum have dominated the literature on Ockham’s metaphysics. He is, after all, most known as a nominalist who consistently denied the existence of universal entities and rejected any kind of created being other than a substance or a quality. Despite often taking his point of departure from an analysis of language—a common complaint is that philosophers multiply faux entities in step with proliferating words and concepts—Ockham’s approach to the study of reality chiefly focuses on the non-human world and is carried out from a non-human perspective. One wonders, then, how he might approach the metaphysics of the social world, a current line of research that has been gathering momentum in contemporary Anglo-metaphysics since the 1980s.¹ The question is not as counter to the spirit of Ockham’s metaphysics as it may seem at first glance since we can find some discussion of various social and political concepts in his earlier theological writings, which command far more ¹ For a recent introductory overview of what I have called the “metaphysics of the social world,” see Brian Epstein, “Social Ontology,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Summer 2018 edition) Epstein includes a rapid overview of historical antecedents to the contemporary debate, moving immediately from Aristotle to Hobbes. No medieval author is mentioned. Jenny Pelletier, Social Powers and Mental Relations: William Ockham on the Semantics and Ontology of Lordship and Ownership In: Oxford Studies in Medieval Philosophy Volume 8. Edited by: Robert Pasnau, Oxford University Press (2020). © Jenny Pelletier. DOI: 10.1093/oso/9780198865728.003.0008

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

249

attention in his later political writings. He conceives of lordship as a power (potestas) in his political writings, particularly in The Work of Ninety Days and Short Discourse, where he addresses the origin and nature of the social and political institutions of lordship, ownership, and property. In the earlier theological writings, the Ordinatio and Quodlibeta, he conceives of lordship as a mental relation (relatio rationis).² My principal question is whether ownership, which is a kind of lordship, is real. Ockham’s parsimonious ontology, with its resolutely non-human perspective, might lead one to think that there is no place within that ontology for ownership, indeed for any social entity, object, or phenomenon. I disagree; I think that there is a sense in which ownership is real. Within his general metaphysical framework, to ask if some x is real tends to translate to the question, “is this x a really distinct entity?” And this is to ask, in effect, whether x is a substance or a quality. The question about the ontological status of ownership, framed in these terms, is an important question to settle, and one that has not been asked before.³ I contend that ownership is a power that is really identical with some person, persons, or collectivity of persons, and in this sense it is not a real entity that adds to the ontology. However, what is notable about this power is that it is instituted through human agreement and it carries deontic force, permitting, obliging, or prohibiting owners to perform various acts with respect to the goods over which they exercise

² The Ordinatio (1317–18) and Quodlibeta (1324–5) date from the first half of Ockham’s philosophical career, where he focused exclusively on logic, theology, metaphysics, etc. with no interest whatsoever in political issues. The Work of Ninety Days (1332–4) and A Short Discourse on the Tyrannical Government Over Things Divine and Human, but Especially Over the Empire and Those Subject to the Empire, Usurped by Some Who are Called Highest Pontiffs (1341–2) date from the second half of his career, where he switched tracks entirely and only wrote political philosophy. The standard explanation for this abrupt turn is his involvement in the Franciscan poverty debate. Having been called to Avignon in 1324 under suspicion of heresy, he was asked by the head of the Franciscan Order to examine the papal position on poverty and ownership. Ockham was thus drawn into the main political crisis of his day and henceforth devoted his energies to political philosophy. ³ In his article “The Ontology and Scope of Human Rights: Forward with Ockham,” American Catholic Philosophical Quarterly 86 (2012), 527–38, A. S. McGrade argues that for Ockham a natural right is a capacity for reasonable action. But he does not explore the ontological dimension further and his focus falls on natural, i.e. human, rights. He holds that such rights meet a criterion of reasonableness, which extends their scope quite significantly, and he further thinks that this view is relevant for contemporary political rights discourse.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

250   ownership. Ownership is not, then, real in the sense of being an entity really distinct from substances and qualities, but it is real to the extent that the community authorizes some human beings to behave in certain ways with respect to certain goods, even though ownership is instituted purely by means of human intellective and volitional activity. I proceed as follows. In section 1, I give an account of the origin and nature of lordship and ownership that we find in The Work of Ninety Days and the Short Discourse. Having shown that Ockham conceives of ownership as a power in these political works, I turn to the ontological status of ownership in section 2. I argue for conceiving of ownership as being really identical with the substances that can be called ‘owners,’ first providing two reasons for why I think we should resist the conclusion that ownership is a really distinct quality. In 2.a I continue my argument by turning to mental relations and the semantics of mental relative terms in general and of ‘lordship’ and ‘ownership’ in particular that we find in the Ordinatio and the Quodlibeta. I then take ‘owner’ and ‘ownership’ as a pair of concrete and abstract relative terms in 2.b, arguing that ‘ownership’ does not pick out any entity that is really distinct from what ‘owner’ picks out. Finally, in section 3, I address the concern about the reality of ownership, and whether there is some sense in which we can take ownership to be real without positing it as an additional entity in Ockham’s ontology. I do this by focusing on the role of the community. I conclude that Ockham’s approach to lordship is as ontologically reductionist as one might expect, reducing ownership to the human beings that are owners, but he leaves room for a conception of social and political powers that are fundamentally instituted—one might say “grounded”—by human understanding and decision, and they affect the social world insofar as they authorize the performance of certain acts by members of the community.

1. The Origin and Nature of Lordship and Ownership The motivation for asking about the reality of ownership, and by extension lordship more generally, is a broader interest in the reality of social and political entities—institutions, groups, rights, powers, etc., which

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

251

owe their existence to human activity. Not merely human activity, but collective human activity. We lie at the origin of the social entities that constitute and structure the social world. In this regard, social entities seem to be markedly different from the more basic metaphysical and physical entities of reality like substances, qualities, atoms, and molecules. Naturally then, one wonders in what sense, or to what extent, social entities are real too. Why lordship? The concept of dominium, typically translated into English as ‘lordship’ and occasionally ‘dominion,’ is foundational in medieval political thought. The nature and origin of lordship was closely connected to the hotly debated issue over property and property rights of the Franciscan Order in the first quarter of the fourteenth century.⁴ Setting aside the case of divine lordship over his creation, Ockham distinguishes between two basic kinds of human lordship: an owner’s lordship over goods (property) and a ruler’s lordship over his or her subjects (political and legal jurisdiction). Ockham joins Scotus in conceiving of the fall as the decisive event in the history of humankind that irrevocably damaged human nature, with the consequence that the emergence and enforcement of property and jurisdiction was necessary for ensuring human social and political survival.⁵ While medieval political thinkers generally agreed that lordship was crucial for living and living well, Ockham tends to paint lordship as unfortunate but unavoidable in light of the reality of our fallen and sinful human condition. Lordship, both as property and jurisdiction, is foundational not only because it traces its origin back to the primeval beginnings of post-

⁴ The literature on this is vast. See, for instance, Virpi Mäkinen, Property Rights in the Late Medieval Discussion on Franciscan Poverty (Leuven: Peeters, 2001) and Roberto Lambertini, “Poverty and Power: Franciscans in Later Medieval Political Thought,” in J. Kraye and R. Saarinen (eds.), Moral Philosophy on the Threshold of Modernity (Dordrecht: Springer, 2005), 141–63. For lordship in Ockham’s political writings, see (amongst many others) Jürgen Miethke, Ockhams Weg zur Sozialphilosophie (Berlin: De Gruyter, 1969) 467–77, and more recently, Jonathan Robinson, William of Ockham’s Early Theory of Property Rights in Context (Leiden: Brill, 2013), ch. 4. ⁵ See Annabel Brett’s detailed and helpful introduction to Ockham’s political thought in her translation, On the Power of Emperors and Popes (Bristol: Thoemmes Press, 1998), esp. 14–15 for connections to Scotus. Her reading of Ockham’s account of temporal political power is particularly negative, arguing that dominium and domination are inextricably intertwined. Also see Joseph Canning, “Power and Powerlessness in the Poverty Debates,” in Ideas of Power in the Late Middle Ages, 1296–1417 (Cambridge: Cambridge University Press, 2011), 107–32.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

252   lapsarian human life but also because it is inseparable from the structure of our social and political world. The significance of lordship thus justifies it as an appropriate focus within a larger discussion on the ontology of the social world that we can extricate from Ockham’s political, and to a lesser extent theological, writings.⁶ In addition to its moral and political character, lordship has metaphysical salience. It is a kind of social entity insofar as it is a capacity and a right that, as I will show, depends for its existence on having been instituted and accepted by the community. Being the owner of a house or the prime minister of Canada are properties belonging to human beings but only insofar as they are persons who live and act within the social world. Ockham describes lordship as a power (potestas) in his political writings. Early in The Work of Ninety Days, he explains that the term ‘lordship’ can be taken in many ways.⁷ For the natural philosopher, lordship is “the power of changing another.” This sense evokes the notion of a power as the ability to perform a certain act or to bring about an effect that could not have been brought about without that power, as fire has the power to cause heat in a piece of wood, and the rational soul has the power to perform an intellective act. For the legal theorist, by contrast, lordship is the “power of laying claim to some temporal thing and defending, holding, controlling it.” This sense evokes the notion of a power as a right, and here the power in question has a deontic aspect, carrying with it a set of duties, responsibilities, freedoms, etc. that oblige, prohibit, or permit the holder of that right to perform certain acts.

⁶ In “Ockham on Human Freedom and the Nature and Origin of Lordship,” in P. Adamson and C. Rapp (eds.), State and Nature: Essays in Ancient Political Philosophy (Berlin: De Gruyter, forthcoming), I discuss lordship from a different perspective. The aim there is to show that the voluntary origin of lordship, instituted by acts of the will, places Ockham within a nonnaturalist tradition on the origin of social and political institutions. Using the same passages on lordship from the political and theological writings, I elaborate on Ockham’s position by introducing and applying his conception of final causality and purposive deliberation to his account of lordship. ⁷ See Opus Nonaginta Dierum [OND] ch. 2, OPol. I, 307: 30–5. Translation in A Letter to the Friars Minor and Other Writings, ed. and tr. A. S. McGrade and J. Kilcullen (Cambridge: Cambridge University Press, 1995), 27–8. All translations are my own unless otherwise stated.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

253

The two come together in the concept of lordship that Ockham develops over the course of The Work of Ninety Days and the Short Discourse. Lordship is not merely the ability to exercise control over goods and people but the right to control goods and people, and the latter seems to rely on it having been agreed upon that one may. Ockham holds that a right is a licit power,⁸ which begs the question, where does its licitness come from? God is surely one possibility, but Ockham believes that human beings first instituted lordship, suggesting that its licit-ness arises in part from the fact that it is forged by human pact (humana pacta) or agreement.⁹ Put roughly, the power of ownership is the ability to exercise a kind of control over goods that is consented to and thus authorized by the community. The role of the community in establishing ownership is present but unelaborated on in Ockham’s political writings. It finds more explicit support in the theological writings, which I discuss in 2.a. There, I defend the view that the institution, acceptance, and perpetuation of ownership require that the past and present members of the community decide to allow certain persons or collectivities of persons to be able to exercise a kind of control over goods. Before then, we must address Ockham’s account of the origin and nature of ownership, which is to say of lordship over goods. Political lordship falls outside the scope of this paper. He distinguishes between two forms of lordship over goods: common and exclusive lordship. Prior ⁸ E.g. OND, ch. 2, OPol. I, 304: 9–12 and ch. 61, OPol. II, 559: 46–50. On Ockham’s conception of right (ius) as a licit power, see Brian Tierney, “Villey, Ockham, the Origin of Individual Rights” and “Languages of Rights” in The Idea of Natural Rights: Studies on Natural Rights, Natural Law and Church Law, 1150–1625 (Atlanta, GA: Scholars Press, 1997), 13–42 and 118–30 respectively, and Annabel Brett, Liberty, Right and Nature: Individual Rights in Later Scholastic Thought (Cambridge: Cambridge University Press, 1997), ch. 2. Brett argues that Ockham’s understanding of rights as licit powers (potestas) relies on the Aristotelian sense of powers (potentia) as principles for acting. ⁹ I do not mean to suggest that human agreement alone is the only source of a power’s licitness. As McGrade shows in “Human Rights,” to be licit, and so qualify as a right, a moral and physical capacity for acting has to meet some criterion of reasonableness at least in the case of a natural right. As McGrade points out, Ockham was no relativist. Right reason, which human beings have and that they can deploy in response to contingent and variable situations, plays a crucial role in determining objective moral and political norms. Because my interest is in the ontology of social and political powers, the question of a right’s reasonableness falls outside the scope of this paper. Nevertheless, my emphasis on community consent is entirely consistent with the assumption that in consenting to the institution of ownership, the members of the community meet some standard of reasonableness. Consent is after all is a volitional act that follows upon a process of rational deliberation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

254   to the fall, God gave Adam and Eve common lordship over goods. This was the power to manage and use goods for one’s advantage,¹⁰ but without appropriating goods to any one individual as opposed to another. Adam and Eve enjoyed common lordship over goods because they were sufficiently loving and peaceable with one another.¹¹ They did not and could not divide goods up between themselves so that goods would fall under their respective, circumscribed control. With the fall, however, human nature was altered, introducing greed, violence, and the tendency to misuse goods against right reason.¹² In order to alleviate the chaotic and damaging consequences of the fall, God gave Adam and Eve exclusive lordship, the power “. . . of managing temporal things, appropriated to one person or to certain persons or to some particular collectivity.”¹³ For Ockham, exclusive lordship is ownership; common lordship is not ownership. The goods over which an owner exercises exclusive lordship is his or her property, and so all property is private property. Even if many persons own land, buildings, tables, etc., they do not have common lordship over those goods. Ockham now complicates his account because he wants to maximize human agency in the origin story of lordship while minimizing divine agency. So, he claims that common and exclusive lordship were both possible immediately after the fall for a short time. During this time, exclusive lordship was merely possible, not yet having been actually exercised, and there was no property yet.¹⁴ Cain and Abel first acted on this power, thereby introducing ownership and property into human history for the first time.¹⁵ The claim that Cain and Abel were the first human beings to act on the power of ownership allows Ockham to maintain that ownership is at once a God-given but human-instituted power. He explains that Abel was “compelled to make some such division of things by the malice of Cain, who wanted to oppress him

¹⁰ Breviloquium 3, ch. 7, OPol. IV: 178: 7–8. ¹¹ Breviloquium 3, ch. 9, OPol. IV: 182: 29–31. Cf. OND ch. 88, OPol. II: 656–7: 118–24. ¹² Breviloquium 3, ch. 7, OPol. IV: 179: 9–25. ¹³ Breviloquium 3, ch. 7, OPol. IV: 178: 14–17. Translation in A Short Discourse on Tyrannical Government, ed. and tr. A. S. McGrade and J. Kilcullen (Cambridge: Cambridge University Press, 1992), 88. ¹⁴ Breviloquium 3, ch. 7, OPol. IV: 179: 31–3. ¹⁵ Breviloquium 3, ch. 9, OPol. IV: 182: 8–10.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

255

violently and to appropriate everything to himself unjustly.”¹⁶ Therefore, Ockham thinks, ownership and property are established by human will in accordance with human law, the ‘human pact’ referred to above, and not divine law.¹⁷ In TheWork of Ninety Days, Ockham also describes lordship as the power to lay claim to a good in a court. On a narrower description, the lord or owner can further treat that good in any way he or she sees fit unless prohibited from doing so by natural law.¹⁸ Ownership, which is clearly the kind of lordship under discussion here, is not merely the dividing of goods between persons or collectivities of persons; it includes the right to go to court to defend one’s claim of ownership to those goods. Ownership also includes the right to transfer one’s ownership to another. This emerges in a later chapter where Ockham is concerned to show that the sole existence of one person does not necessarily mean that that person has exclusive lordship over some good. Prior to the creation of Eve, Adam was the only existing person, but Ockham thinks that we could not conclude from this that Adam owned the things that he managed and used. Ockham develops an analogy that he leans on in a number of the political writings: the case of a monk who finds himself the only resident of a monastery. That he is the lone surviving monk does not mean that he owns it. Ockham writes: . . . it must be known that lordship is not called ‘proprium’ [exclusive] because it belongs in fact to one by himself: for then, if all the monks of an abbey were to die or be killed except one, he by himself would have exclusive lordship of all the goods of the abbey and would then have to be regarded as its owner, which must be reckoned altogether false. Rather, it is called exclusive when it is appropriated to one person in such a way that it cannot belong to anyone else without his gift, sale, bequest, or some other human contract by which lordship of a thing is transferred to someone else, or at least by some act of his or by his death [my italics].¹⁹

¹⁶ Breviloquium 3, ch. 9, OPol. IV: 182: 12–16. Translation in A Short Discourse, 92. ¹⁷ OND ch. 88, OPol. II: 656: 97–9. ¹⁸ OND ch. 2, OPol. I, 308: 23–6. ¹⁹ OND ch. 27, OPol. II: 487: 59–63. Translation A Letter to the Friars Minor, 39–40.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

256   In this passage, Ockham makes clear that exclusive lordship is exclusive partly insofar as it includes the right to transfer one’s property as one choses, and not because it happens to be used and managed by one single person. In sum, exclusive lordship, or ownership, is a power that is given by God but instituted by human beings whereby we agree to allow certain members of the community—a person, persons, or collectivity of persons—to be able to exercise exclusive control over goods. In a more complex social and political order, where legal structures have been introduced to protect and regulate property rights, ownership eventually includes the right to defend one’s claim to one’s property in a court if necessary, the right to transfer exclusive control over that property to another person, or persons, or collectivity of persons (by sale, gift, etc.). To further elaborate on the deontic aspect of ownership, we can note also that an owner is obliged to treat their property in a way that is not contrary to natural law, and presumably all members of the community within which ownership has been instituted are obliged to respect the rights of owners in that community. They are prohibited, for example, from taking other people’s property except in highly extraordinary circumstances.

2. The Ontological Status of Ownership I now want to turn to the question of the ontological status of lordship and particularly ownership. To what extent is ownership real? Within the general framework of Ockham’s metaphysical commitments, the question about the ontological status of some putative entity tends to be articulated as follows: is this x a really distinct entity (a res distincta or absoluta)? If yes, then it must be either a substance or quality, since he holds that only God, singular substances, their essential parts, and some of their singular qualities count as really distinct entities.²⁰ If not, then ²⁰ Not all qualities are really distinct entities according to Ockham. Following Aristotle, medieval authors recognized four kinds of quality: (1) habits, e.g. virtue and knowledge; (2) natural powers of acting or being acted on, e.g. being a fast runner; (3) sensible qualities, e.g. being hot or white; (4) shape and form, e.g. being curved or rectilinear. Ockham is quite clear across his writings that sensible qualities are really distinct from the substances in which

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

257

Ockham tends to argue that the putative entity is in fact a term that signifies really distinct entities in various ways. I do not mean to suggest that anyone has thought that Ockham did hold that the power of ownership is a quality really distinct from the substances (i.e. lords or owners) that can exercise that power. To pose the question as I do is justified on the grounds that this is how Ockham himself generally proceeds when confronted with some putative entity. Moreover, I think that one might find it plausible that a right, a licit power, is something somehow separate from the bearer of that right. After all, we can transfer a right from one person to another, we can be “stripped” of a right, and we can be granted a right. This language suggests that rights are distinct from their bearers, and what better way to cash this out ontologically, with Ockham’s set of available entities, than thinking of ownership as a really distinct quality? This very ontology includes various such qualities that are powers, like the form of heat, which is naturally disposed to cause heat in another. Ockham admits that some natural powers (a potentia naturalis) are indeed really distinct qualities.²¹ So why not conceive of ownership as a quality in the same way? I argue that the power of ownership is a power that is really identical with the substances — the person, persons, or collectivity of persons — who can licitly appropriate goods for their own management and use to the exclusion of anyone else. My argument proceeds in two steps. First, I give two reasons to reject the alternative view that ownership is a quality that is really distinct from and inherent in the substances that are owners. The second reason leads me to the second step of my argument, where I turn to Ockham’s discussion of mental relations and mental relative terms, which includes ‘lordship,’ as well as of concrete and abstract counterpart terms, to argue that the power of ownership is really identical to owners and thus to substances. For, the ontological fact is reflected in the semantic synonymy or equivalence that holds between

they inhere while qualities of shape and form are not. Most but not all habits are really distinct qualities, and Ockham argues that some natural powers are habits, which presumably can in turn be really distinct qualities or not as the case may be. ²¹ See Expositio in librum Praedicamentorum Aristoteles, ch. 14, § 5, OPh. II, 275: 36–44, for Ockham’s discussion on natural powers.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

258   ‘owner’—‘ownership.’ ‘Ownership’ conveys precisely the same really distinct entities as ‘owner,’ and this includes substances, the goods over which they can exercise exclusive control, and the intellective and volitional acts by which we decide to allow certain human beings to be so-related to certain goods in the first place. ‘Ownership,’ in short, is a linguistic fabrication. One reason to think that Ockham would not agree that ownership is a quality is indirect but interesting and persuasive nonetheless. Ockham and his Franciscan confrère Walter Chatton (c. 1290–1343) conducted a well-documented disagreement about their respective principles of ontological commitment. Chatton defended his own so-called ‘anti-razor’ that stipulated the conditions for determining when one ought to posit entities. This roughly has to do with taking entities as the truth-makers for true and affirmative sentences, and the idea is that if two entities do not suffice then one must posit a third, and a fourth, and so on. The details of this disagreement are not relevant for our purposes, but in his Lectura d. 3, q. 1, Chatton reports the following objection to his own principle: the passage of time can be sufficient to account for a change in the truth-value of an affirmative sentence, in which case there is no need to posit a second (or third, or fourth) entity. The example that Chatton provides to illustrate the objector’s point is the sentence, “this castle is yours.” Suppose that a king gives you a castle but only for Sunday. The sentence “this castle is yours” is true on Sunday but false on any other day of the week, and no new additional entity need be posited.²² Chatton apparently grants the objection. What is significant for us is that the editors of Chatton’s Lectura attribute the objection to Ockham, providing a reference to Ockham’s Ordinatio d. 30, q. 2. It is indeed true that Ockham argues there that the passage of time and a change in place can both account for a change in the truth-value of an affirmative proposition beyond positing the

²² Walter Chatton, Lectura d. 3, q. 1, n. 75, ed. Wey and Etzkorn, 24. For commentary on this passage in the context of Chatton’s anti-razor, see Magali Roques, “Le principe d’écomonie d’après Guillaume d’Ockham,” Franciscan Studies 73 (2015), 169–97. For Chatton’s conception of dominium, see Jonathan Robinson’s “Walter Chatton on Dominium,” History of Political Thought 35 (2014), 656–82. I would like to thank Magali Roques for bringing this passage to my attention.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

259

production or destruction of an entity. Now, Ockham himself does not use the example “this castle is yours,” nor does he mention lordship or ownership. But Chatton’s use of the example is arresting, and in light of the close intellectual exchange that took place between the two Franciscans, I think we can assume that Chatton knew Ockham either did or at least would have approved of the example. If I am right, then the upshot of the objection with its example is that Ockham did not think that ownership is a really distinct quality that could be produced or destroyed in the substance of the owner. He would not have claimed that “this castle is yours” is true on Sunday but false on Monday because by Monday I have lost the quality of ownership. A second reason to think that Ockham would not agree that ownership is a quality is that he claims in the Ordinatio that lordship, along with slavery (servitus), spoken and written linguistic signs (signum), and money or value (pretium), is a mental relation. While he does not mention lordship explicitly in the Quodlibeta, he discusses money and signs in somewhat more detail there, and I see no reason why that same analysis would not extend to lordship. A relation, including a mental relation, is not a quality. So, if lordship is a mental relation, which he clearly states that it is, then it is not a quality. Here we finally turn to Ockham’s conception of lordship or ownership in the earlier theological writings, bringing me to the second and more important step in my argument.

a. Reducing Ownership I: Mental Relations In the Ordinatio, Ockham writes, “Such [mental relations] of this sort include: ‘value,’ ‘sign,’ ‘lordship,’ ‘slave,’ insofar as they apply to creatures.”²³ To understand why and how categorizing lordship as a mental relation is significant for determining its ontological status, we have to explain Ockham’s position on relations in general and on mental relations in particular as briefly as possible. Ockham famously holds that relations do not exist over and above the entities that are related to one ²³ Ordinatio [Ord.] d. 35, q. 4, OTh. IV, 472: 15–16.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

260   another, barring a few theological exceptions.²⁴ Before turning to his account of mental relations, we need to see that Ockham thinks of relations in two ways: first, as terms, and second, as entities that are related to one another. For the sake of clarity, I will refer to relative terms on the one hand, and to relations on the other but it should be noted that Ockham himself does not follow this practice. I will use inverted commas to indicate when I am mentioning a term as opposed to using it. A relative term like ‘similarity’ is a kind of connotative term. Connotative terms signify some things primarily (their significata) and others things secondarily (their connotata). The term ‘grey’ primarily signifies things that are grey, like thunderclouds and sidewalks, while connoting the patches of greyness that inhere in thunderclouds and sidewalks. Connotative terms have nominal definitions indicating their significata and connotata.²⁵ A relative term, as a connotative term, signifies some things primarily and other things secondarily, taken conjointly. The relative term ‘mother’ primarily signifies female animals while connoting their offspring. The nominal definition of ‘mother’ is something like “female animals having born offspring.” By contrast, a relation is the entities that are related to one another in some respect and can be signified by a relative term. All patches of greyness, for instance, are a similarity, and so they can be conjointly signified by the term ‘similarity,’ if such a term is coined. Ockham shows a preference for thinking of relations as relative terms, which he believes is the correct Aristotelian understanding of relations. But he himself always presents both as two viable alternatives. Indeed, Ockham applies the distinction to mental relations in the following statement: “. . . I say that a mental relation can be understood in two ways: in one way as an utterance [vox] or concept conveying ²⁴ For Ockham’s discussion on relations, see Marilyn McCord Adams, William Ockham, 2 vols. (Notre Dame, IN: University of Notre Dame Press, 1987), 261–5; Mark G. Henninger, Relations: Medieval Theories 1250–1325 (Oxford: Clarendon Press, 1989), 136–40; and Jeffrey Brower, “Medieval Theories of Relations,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2015 edition) . ²⁵ This distinction is introduced and discussed in Summa logicae [SL] I, ch. 10, OPh. I, 35–8. See Claude Panaccio, “Semantics and Mental Language,” in P. V. Spade (ed.), The Cambridge Companion to Ockham (Cambridge: Cambridge University Press, 1999), 53–73 for a clear presentation.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

261

something or some things; another way as the signified [things] themselves.”²⁶ Unlike real relations, mental relations fundamentally depend on an “operation of the intellect” in order to obtain.²⁷ I am really related to my son insofar as I am one of his producing causes, and this is true irrespective of whether anyone thinks that we are related to one another in this way. By contrast, coins cannot be called ‘money’ nor the owning of people ‘slavery’ unless it has been understood and willed that coins can be exchanged for goods and that some people can be the property of others, and then further that the one is dubbed ‘money’ and the other ‘slavery.’ These are examples of mental relations. More precisely, a mental relation is composed of the entities that are connected to one another and the mental act by which they are connected. A mental relative term connotes the mental act thanks to the fact that two or more entities can be said to be so-related in addition to whatever other connotata it also has. A real relative term, of course, has no such mental act that it would connote. ‘Lordship,’ ‘money,’ ‘slavery,’ and ‘sign’ are examples of mental relative terms taken in a strict sense. There is a broader sense in which other terms are mental relative terms as well, like ‘subject,’ ‘predicate,’ ‘genus,’ etc., since each of these requires that some mental act has been or could have been performed by which, say, one term can be the subject term of a sentence, or one term is predicable of many other species terms. In the strict sense, however, a term is “. . . said to be a ‘mental relation’ when a thing cannot be such as it is expressed to be by that term without a concurrent intellective or volitional act.”²⁸ A mental relation in this sense is a term that signifies things to be related to one another where their connection in the relevant respect is only brought about by a mental act, whether an intellective or volitional act. In the absence of one such intellective or volitional act, the entities in question would not be related, at least not in that respect, and could not therefore, be signified by that relative term, “And with neither of these [intellective or volitional acts having been performed] concurrently, the extremes can remain and

²⁶ Ord. d. 35, q. 4, OTh. IV, 470: 13–16. ²⁷ Quodlibeta [Quodl.] 6, q. 30, OTh. IX, 699: 14–16. ²⁸ Ord. d. 35, q. 4, OTh. IV, 472: 11–14.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

262   neither is such as to be denoted by a relative [term] of this kind.”²⁹ Simultaneous to, or perhaps immediately upon, the requisite mental act having been performed, the mental relation ensues, and the mental relative term, if coined at that point, conveys the relevant entities as they are thought or desired to be connected to one another by the mental act. I will use the term ‘to institute’ to refer to what the mental act, intellective or volitional, does in relating the mental relative term’s significata and other connotata to one another, authorizing, therefore, talk of the institution of mental relations and mental relative terms that transpires by means of the performance of a particular mental act. Ockham helpfully includes an illustration: For a spoken utterance is a sign and a coin is money or value only because, with a previous intellective act, we wish to use the spoken utterance or coin in this way. And given that there is such a volitional act in us – or there was and there was no contrary volitional act – immediately, without anything else added, the spoken utterance is a sign and the coin money.³⁰

This passage clarifies that the mental act is at least one intellective act followed by a volitional act, so that ‘concurrently’ should not be taken to mean ‘simultaneously.’ Is it plausible to think that one can perform an intellective act by which one grasps some content and perform a volitional act by which we give our assent to this content at precisely the same time? Surely the acts are sequential, even if they immediately follow one another. Notwithstanding Ockham’s succinct description of the process in the above-cited passage, the institution of the mental relative term ‘money,’ or say ‘ownership,’ quickly acquires a rather complex genealogy. The use of ‘ownership’ is complicated because it is an abstract term. I will argue in the next section that ‘ownership’ and its concrete counterpart, ‘owner,’ are semantically synonymous or equivalent, but for the moment suppose that they are. A reconstruction for ‘owner’ that makes use of the material from The Work of Ninety Days and the Short

²⁹ Ord. d. 35, q. 4, OTh. IV, 472: 14–15. ³⁰ Ord. d. 35, q. 4, OTh. IV, 472: 17–21. Translation from Henninger, Relations, 138 with modification. Cf. Quodl. 6, q. 29, OTh. IX, 698: 99–103 and Quodl. 6, q. 30, OTh. IX, 700: 34–6.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

263

Discourse might include the following sequence of distinct intellective and volitional acts (one can imagine that Cain and Abel perform something like this sequence respectively): (1) cognizing some good (intellective act) (2) desiring that good (volitional act) (3) understanding the function of exclusive control over a good, i.e. the private use and management of that good (intellective act) (4) cognizing an appropriate medium to which the function of exclusive control of a good could be ascribed, e.g. some person, persons, or collectivity of persons (intellective act) (5) understanding the ascription of exclusive control of a good to some person, persons, or collectivity of persons (intellective act) (6) willing the ascription of exclusive control of a good to some person, persons, or collectivity of persons (volitional act) At this point, we still haven’t yet arrived at the mental relative term ‘owner.’ Acts (1)–(6) get us to being an owner as a mental relation but without in fact coining the term ‘owner.’ This presumably takes place by some further distinct volitional act such as: (7) willing to name a certain person, persons, or collectivity of persons with respect to their being able to exercise the function of exclusive control over goods as ‘owner’ (volitional act) It is important to keep this distinction clear because our interest is in the institution of owners and ownership rather than in the coining of ‘owner’ and ‘ownership.’ The two are independent from one another. We can have a mental relation without the mental relative term for it. The example that Ockham uses is the mental relative term ‘sign’ and the proper name ‘Socrates.’ The name ‘Socrates’ is indeed a sign, having been chosen by parents to pick out a particular boy, and Socrates is the boy that is signified by ‘Socrates.’ But we do not need the mental relative term ‘sign’ in order for ‘Socrates’ to be the boy’s name. We can carry on quite well getting Socrates’ attention and talking about him to other people without ever having to coin the term ‘sign.’ We only need the term ‘sign’ if we want to express the fact that ‘Socrates’ is this boy’s name and is,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

264   therefore, a sign that picks him out.³¹ Similarly, people can own other people even if they do not have a term that names the owning of other human beings for labor of some sort, i.e. ‘slavery.’ So, the mental acts (1)–(6) get us to the mental relation of being an owner of some good, and here the mental relation is the following entities taken together: (a) a person, persons, or collectivity of persons, (b) the goods over which that person, persons, or collectivity of persons can exercise exclusive control, (c) and the acts (1)–(6) themselves by which it was understood and willed that person, persons, or collectivity of persons could be ascribed that function in the first place. The mental acts (1)–(7) get us to the mental relative term ‘owner,’ which primarily signifies (a) a person, persons or collectively of persons, secondarily signifies (b) the goods over which that person, persons, or collectivity of persons can exercise exclusive control, as well as (c) acts (1)–(6). I take it that ‘owner’ does not connote act (7), since this would mean that all spoken and written terms would connote the volitional acts by which they are coined, and they do not. The independence of the mental relation and the mental relative term means that the act by which being an owner was first instituted (6) and the act by which ‘owner’ was first coined (7) need not have been performed by the same agent or agents. In order to institute the mental relation of being an owner in the strict sense, act (6) must be actually performed. But Ockham specifies that this act could have taken place in the past. Recall that he writes, “And given that there is such a volitional act in us – or there was and there was no contrary volitional act – . . . the spoken utterance is a sign and the coin money” (my italics). In the case of a spoken and written sign, he is also clear. What is necessary for the English word ‘human being’ to signify human beings is “. . . the utterance, human beings, and the past or present volitional act by which we will this utterance to be used for all human beings.”³² As long as the same agent or another agent does not perform a counter-volitional act at a later time, then the mental relation continues to hold in the present. The trans-temporal aspect that Ockham evokes here suggests that at least more than one agent is involved if not in the original institution of ownership then certainly in the perpetuation ³¹ Ord. d. 35, q. 4, OTh. IV, 471: 15–17.

³² Ord. d. 35, q. 4, OTh. IV, 471: 4–8.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

265

of ownership over time. I think he is getting at something along these lines: ownership is first instituted by an original sequence of intellective and volitional acts performed by an agent or group of agents in the past, say Cain and Abel, and then accepted in subsequent sequences of intellective and volitional acts performed by an agent or group of agents in the present.³³ On my reading, the mental relative term ‘owner’ connotes the intellective and volitional acts (1)–(6) by which it was originally understood and decided to relate certain entities to one another for the first time by an original institutor(s). This sequence of mental acts constitutes the institution proper of the connection between persons (owners) and goods (their property), and it could have taken place at some point in the past, say at T₁. When present members of the community buy, sell, or loan goods, defend their claim to those goods in a court, or even go about respecting other people’s property rights by not stealing at T₂, they do not institute being an owner anew. Rather, the mental acts they concurrently perform at T₂ are acts by which they understand and accept the past mental acts (1)–(6) that were performed by the original institutor(s) at T₁. So, mental acts must be performed concurrently at both T₁ and T₂ but while we might call the first acts of institution, the second are acts of acceptance. One way to think of these acts of acceptance is as tacit cognitions and volitions, and they are performed in conformity with the original acts of institution. In this way, barring any deliberate

³³ In this regard, what I have called the ‘institution’ of a mental relation closely follows Panaccio’s analysis of subordination and imposition in the case of the conventional signification of spoken and written languages. Contending that Ockham endorses a strong form of linguistic externalism, Panaccio argues that for Ockham, a spoken word, for instance, inherits its signification purely by virtue of having been subordinated to a concept by its original impositor(s). In Ockham on Concepts (Aldershot: Ashgate, 2004), Panaccio writes, “What determines the semantical status of a given spoken word within a given linguistic community is the concept which the original impositor (or group of impositors) associated with it, not the concept that any particular speaker happens to have at the moment of utterance” (170) and further that, “the utterer of the new token must agree to use the sign as it was imposed . . . If a speaker is to use certain spoken sounds as linguistic signs, he or she must accept the authority of some past impositional acts” (172). Also see his “Ockham’s Externalism,” in G. Klima (ed.), Intentionality, Cognition and Mental Representation in Medieval Philosophy (New York: Fordham University Press, 2015), 166–85; and again in “Linguistic Externalism and Mental Language in Ockham and Buridan,” in G. Klima (ed.),Questions on the Soul by John Buridan and Others: A Companion to John Buridan’s Philosophy of Mind (Dordrecht: Springer, 2017), 225–37.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

266   modification of the community’s understanding and acceptance of ownership, the present community inherits ownership as it was forged by the past members of that same community. Alternatively, suppose the requisite intellective and volitional acts connecting lord and servant or lord and castle are concurrently performed at T1, instituting lordship at that time. At T₁, lordship is a mental relation strictly conceived. Now suppose that the subsequent mental acts are not performed at T₂. At T₂, lordship is a mental relation broadly conceived, where it need only be possible that another sequence of mental acts could be performed but without having been actually performed. On this reading, when the members of the community are not thinking of and willing to condone the connection of lord to servant or lord to castle, tacitly or not, then lordship is only a mental relation in the broad sense. Notice that if the mental acts are actually performed at T₃ then lordship becomes a mental relation strictly conceived once again. We can still draw the same distinction between original acts of institution performed at T₁ and later acts of acceptance at T₃, where the acts of acceptance are performed in conformity with the acts of institution, albeit with an intermediary lapse at T₂. On either reading, there is room here for establishing a connection between the individual intellective and volitional acts performed by the past and present members of a community. The institution, acceptance, and perpetuation of ownership depend on the members of the community having individually performed the intellective and volitional acts to do so. This means that we, the past and present community, consent to some members of the community being owners, thereby authorizing those individuals to be able to exercise exclusive control over the goods that are their property. In this regard, community consent is what renders licit the ability to exercise exclusive control over certain goods.

b. Reducing Ownership II: ‘Owner’—‘Ownership’ and the Semantics of Abstract and Concrete Terms I will now argue that ‘ownership’ is a linguistic fabrication, an abstraction of the term ‘owner.’ It is either synonymous with ‘owner,’ in which case

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

267

the two terms have precisely the same nominal definition, or it is equivalent in signifying to ‘owner,’ in which case it still conveys the same entities. Either way, the semantic relation between ‘ownership’ and ‘owner’ allows us to see that ‘ownership’ does not convey anything other than what ‘owner’ conveys. And, this in turn allows me to get to my conclusion that the power of ownership is in fact nothing other than a certain person, persons, or collectivity of persons who, insofar as it has been decided upon and willed by the members of the community, are able to exercise exclusive control over certain goods. Some of the intricacies of the semantics of abstract and concrete terms must be explained in order to understand what follows. I take it that the term ‘ownership’ is an abstract term and its concrete counterpart is ‘owner,’ whereas ‘owned’ is the concrete counterpart of the abstract term ‘property.’ So let us take the pair ‘owner’—‘ownership.’ Ockham identifies four modes by which abstract and concrete terms relate to one another.³⁴ ‘Owner’—‘ownership’ could fall under either the second or third modes. All abstract and concrete relative terms relate to one another on the second mode, and so too would ‘owner’—‘ownership’ since they are mental relative terms. On this mode, the terms of each pair are synonymous with one another, meaning broadly that each term signifies precisely the same thing (or set of things) in precisely the same way as the other term, such that nothing is signified by the one without being signified by the other in the same way.³⁵ On Ockham’s view, relations are not really distinct from the absolute entities that are related, so abstract and concrete relative terms like ‘father’—‘fatherhood,’ ‘potency’—‘potentiality,’ ‘capable’—‘capacity,’ etc., are synonymous. Take ‘father’—‘fatherhood.’ Rejecting the ontological view that relative entities exist in addition to related absolute entities, Ockham rejects the semantic view that ‘father’ signifies male animals that have generated offspring, while ‘fatherhood’ signifies the relative entity inhering in those animals either as one of their accidents or as one of their parts or as anything extrinsic to them. Rather, ‘fatherhood’ is a linguistic abstraction

³⁴ These modes are discussed in the early chapters of SL 1 and Quodl. 5, q. 9. ³⁵ See SL 1, ch. 6, OPh. I, 21: 65–72.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

268   of ‘father,’ with both terms primarily signifying male animals while connoting their offspring. As a synonymous abstract and concrete pair of mental relative terms, ‘owner’—‘ownership’ have the same nominal definition. A nominal definition makes explicit the defined connotative term’s significata and connotata, thereby indicating the various entities in the ontology that the connotative term conveys and how they are connected to one another. The advantage of correctly formulating a connotative term’s nominal definition is that it reveals just what underlies that term in the ontology.³⁶ We have already seen in the context of Ockham’s treatment of mental relative terms that both terms would connote the intellective and volitional acts by which their significata and whatever other connotata they have are decided to be related to one another. Now, the goal is to formulate a nominal definition of ‘owner’—‘ownership’ that indicates all its significata and connotata while specifying how they stand in relation to one another. This might be something like: someone who can appropriate things for their exclusive control in light of past or presently performed intellective and volitional acts by which it is agreed that a person, persons, or collectivities of persons can licitly appropriate things for their exclusive control.

On this nominal definition, ‘owner’—‘ownership’ primarily signifies human beings, either individually or as having formed unified collectivities, and connotes both the things that can be appropriated, i.e. goods, as well as the mental acts by which the members of a community decided that human beings can be permitted to be perform certain acts with respect to certain things. That ownership is a power is captured by the use of the modal ‘can’ in our proposed nominal definition. This dispenses with the need to posit any additional entity in the ontology that ‘ownership’ might directly pick out, just as ‘fatherhood’ does not pick out anything other than fathers and, indirectly, their offspring. Rather, our definition makes it explicit that to be an owner, to have the power of ownership, is nothing other than being able to act in certain ways, viz. exercising exclusive control over certain goods. Moreover, our proposed ³⁶ For this view of the role of nominal definitions, see Panaccio, Ockham on Concepts, ch. 5.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

269

definition includes the crucial reason for why this power is a right, that is to say a licit power: it is by virtue of the intellective and volitional acts performed in the past and present by the community that the ability to behave in certain ways with respect to goods, to be so-related to them, is authorized. To elaborate on the definition using material from The Work of Ninety Days, we can add that, to own goods entails being able to defend one’s claim to being the owner of those goods in a court, to transfer one’s goods to the exclusive control of another person, persons or collectivity of persons, and so on. In other words, the range of obligations, duties, and freedoms that one acquires as an owner and a member of a community within which ownership and property have been established. Notice, that Ockham’s inclusion of ‘capable’—‘capacity’ as an example of a synonymous pair of concrete and abstract relative terms can be adduced as yet another reason to reject the view that ownership is a quality and that ‘ownership’ would directly pick out that quality. We can infer from the fact that Ockham uses ‘capable’—‘capacity’ in this context that he does not think that ‘capacity’ refers to a really distinct ability (a power!) that inheres in whatever the term ‘capable’ signifies, whether as an accident or part or something entirely extrinsic. The two terms are synonymous because they refer to the same thing in the same way. Similarly, ‘ownership’ would not refer to really a distinct accident—a quality—that inheres in what ‘owner’ refers to, nor to one of their parts, nor to something entirely extrinsic to them. To be sure, the intellective and volitional acts by which it is agreed that a certain person, persons, or collectivities of persons can appropriate goods are extrinsic to them. But the ability to control goods, which they can licitly act on subsequent to this agreement, is not. Alternatively, ‘owner’—‘ownership’ could relate to one another on the third mode. In this case, they are not synonymous but nevertheless convey the same entities(s). On the third mode, some abstract and concrete pairs of terms that one might think are synonymous are in fact not, because they cannot be truly reciprocally predicated of one another, and cannot be truly predicated by the same term. The reason is that the abstract counterpart is equivalent to a complex expression including various syncategorematic elements or adverbial determinations, which it has acquired from the way it is used by language-users, and that its concrete counterpart does

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

270   not have.³⁷ The abstract term ‘humanity,’ for instance, is equivalent to ‘a human being as such’ or ‘a human being per se’; whereas its concrete counterpart, ‘a human being’ is not. These abstract and concrete counterpart terms are equivalent in signifying insofar as they signify the same thing(s), but fall short of synonymy. On this mode, it would remain the case that ‘ownership’ would convey human beings that can exercise exclusive control over goods, those goods themselves, and the mental acts by which it was agreed that certain human beings could do so in the first place. However, ‘ownership’ would be equivalent to a complex expression containing various syncategorematic elements or adverbial determinations, which would render it non-synonymous with ‘owner.’ Consequently, the sentences, “this owner is ownership,” and, “this ownership is an owner,” would be badly formed, as would, “this owner is Socrates,” and, “this ownership is Socrates.” Moreover, ‘ownership’ might not be able to function as a properly referential term, because it could be a non-genuine name. Non-genuine names, which have been fashioned for the purposes of poetical verse, literary ornament, or concision, cannot be the subject or predicate of a well-formed sentence without some interpretation. In this case, “ownership” would be a figurative term.³⁸

3. Is Ownership Real? If my argument holds, then one might naturally think that ownership is not real. ‘Ownership’ is a linguistic fabrication, an abstraction of its ³⁷ See Quodl. 5, q. 9, OTh. IX, 515: 56–60. ³⁸ On non-genuine names see Tractatus de Quantitate q. 1, OTh. X, 21–6 and then 29–35. When the complex expression that an abstract term is equivalent to is not a complete phrase, then its equivalent term is a non-genuine name, since neither complex expression nor abstract term can stand as the subject or predicate term of a well-formed sentence. When the complex expression is a complete phrase, and can stand as the subject or predicate term of a well-formed sentence, then so too can the abstract term. If ‘ownership’ were equivalent to “some person, persons, or collectivity of persons who can licitly exercise exclusive control over good(s) as agreed upon by the past or present members of the community,” then it does have definitive signification, and can refer to what it signifies as the subject or predicate term of a well-formed sentence. If ‘ownership’ were equivalent to an incomplete complex expression like “some person, persons, or collectivity of persons qua goods,” then it would not be able to refer to its significata in any literal sense, though it could still be used figuratively.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

271

concrete counterpart ‘owner,’ and a mental relative term. So why not just admit that ownership is better thought of as the term ‘ownership’? This, however, would threaten the notion that ownership is a power that belongs to the realm of reality rather than a term that belongs to the realm of language and thought. We cannot ignore the, “is this x a really distinct substance or quality?” question in the context of thinking about the reality of ownership, and I have not eschewed dealing with it. But it does not exhaust all reflection on the reality of social entities (objects might be a better term, since one should like to de-reify social phenomena as much as possible) including powers and rights, even considering Ockham’s austere ontology. Ownership is not real in the sense of being an entity that is really distinct from the human beings who can exercise the power of ownership. It cannot exist independently of being thought and willed by the members of a community, if not in the present then at least at some point in the past. We choose to allow certain members of the community to have the power of ownership and this entails that they can licitly perform certain acts with respect to certain good(s). But what is fundamental to Ockham’s account, as evidenced in both the political and theological writings, is that the community forges the connection between those who can be called ‘owners’ and the goods that can henceforth be called their ‘property,’ thereby authorizing the former to be able to perform certain acts with respect to the latter. The material from the Ordinatio and the Quodlibeta that I discussed in 2.a above, where lordship and ownership are conceived as mental relations, is significant in this regard. In the reconstruction that I gave there, I wrote of willing to ascribe exclusive control to an “appropriate medium,” i.e. to human beings, or a person, persons, or collectivity of persons. The performance of this volitional act is a necessary condition for certain members of the community to be called ‘owners,’ the goods they own ‘property,’ and the term ‘ownership’ being coined. Moreover, I argued that Ockham gives us some indication of a more precise account of the connection that might obtain between the intellective and volitional acts of the members of a past and present community, from Cain and Abel to us. On this account, later members of the community will in conformity with the intellective and volitional acts by which the earlier

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

272   members of the community willed, that is to say decided, to institute ownership. A more fine-grained analysis would unpack what Ockham might mean by one will willing in conformity with another will, which he discusses in the context of the created will willing in conformity with the divine will in Ordinatio 48, q. un. It is very tempting here to evoke a distinction between ownership as ontologically subjective but epistemically objective,³⁹ and both for the same reason: ownership is dependent on human agreement, grounded upon the intellective and volitional acts of the members of the community that instituted ownership for the first time and the subsequent intellective and volitional acts by which later members of the community continue to accept its original institution. This emphasizes the community, historical and current, as foundational for grounding ownership as something real beyond the immediate perspective of one individual person at one particular moment in history. There is no ownership, no property, and no owners in the absence of the intellective and volitional acts by which past and present members of the community decide that there should be, and in this sense, they are ‘ontologically subjective.’ But, once instituted, the members of the community can determine sentences like “this castle is mine” to be true or false, and in this sense, they are ‘epistemically objective.’ In TheWork of Ninety Days, Ockham is explicit that ownership includes the right to be able to go to court to lay claim to one’s property, to defend one’s claim to it. This means that one’s ownership is publicly ascertainable, according to whatever criteria the community has decided upon, and there are legal procedures in place to settle property disputes, handle property transference by sale, gift, or inheritance, punitive measures for those who break the law with respect to property, etc., and all of this is beyond, as it were, the private sphere of the person, persons, or collectivity of persons that is able to exercise exclusive control over certain goods. The reason lies in the role of the community. It is plausible that we have the bare ability to perform the various acts by which we can exercise control over goods irrespective of whether the

³⁹ The distinction is from John Searle, The Construction of Social Reality (London: Penguin, 1995), 7–9 and later in his Making the Social World: The Structure of Human Civilization (Oxford: Oxford University Press, 2010), 17–18.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

273

community has decided that we may. The consent of the community is presumably required in order that we are able to exercise exclusive control over goods at all, since the protection of and compliance with exclusive control typically necessitates the use of law and coercive force. The consent of the community is certainly required in order to authorize and render licit the exclusive control that we may be allowed to exercise over goods. Ownership as a right, a licit power, fundamentally rests on the approval and consent of the members of the community who condone and thus authorize the exercise of exclusive control over goods. To be sure, Ockham does not have the contemporary concept of collective intentionality or group action, and it is not immediately obvious where he might address the question of whether a group of people can jointly think, will, and act together with some shared common goal or purpose. Perhaps when he discusses the will willing in conformity with another will. He does intimate the possibility of the trans-temporal institution, acceptance, and perpetuation of ownership. And certainly in his account of the origin of exclusive lordship in The Work of Ninety Days and Short Discourse, he implies that Cain and Abel each perform their own individual intellective and volitional acts by which, nevertheless, they understand and decide to allow one another to exercise exclusive control over certain goods. The inchoate suggestion here, which I have discussed elsewhere, is that each man desires some goal, has beliefs about how best to achieve that goal, and then performs some intellective and volitional acts accordingly.⁴⁰ At each step, both men are confronted with the other, and the only way that either man will end up achieving his goal is by coming to some mutual agreement that, in this case, gives rise to owners, property, and ownership. Cain and Abel consent, although acrimoniously, to allow each other to claim exclusive control over goods. Finally, Ockham’s description of ownership and lordship in The Work of Ninety Days crucially though implicitly rely on human agreement, which is necessary for the authority of law, of legal courts, and for legitimate and enforceable claims to property ownership,

⁴⁰ See my “Ockham on Human Freedom” (forthcoming) for a detailed discussion of the deliberative and purposive process that Cain and Abel conduct in the institution of ownership and property after the fall.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

274   etc. Human agreement, or social convention, ultimately rests on the actual performance of the sort of past or present intellective and volitional acts described in the theological writings.

4. Conclusion Ockham’s metaphysical investigations are predominantly carried out without much consideration for questions and concerns pertaining to the social world. Some of the foregoing material in this paper is admittedly reconstructive but not woefully speculative. Ockham construes both lordship and ownership as licit powers with deontic force that are structurally pervasive within historical human communities. They were forged at a particular moment in history, in response to the circumstances in which the first human community found itself, and by means of agreement— voluntary consent—to permit certain individuals to be able to exercise exclusive control over certain goods. The claim that the power of ownership is not really distinct from the human being(s) upon whom it has been conferred should not be surprising for anyone acquainted with Ockham’s general metaphysical persuasions. But this does not mean that having the power of ownership and lordship more generally has no tangible consequences on the lives of the members of a community. Being an owner or being a lord, though utterly dependent on human agreement and convention, alters the way that members of the community can behave with respect to goods and to one another. On my reading, Ockham gives an account of lordship that is ontologically reductionist but receptive to the particularities and shared reality of the human social world.⁴¹ Research Foundation—Flanders (Belgium)/University of Gothenburg (Sweden)

⁴¹ Previous versions of this paper were presented at the Moody Workshop, UCLA (March, 2018) and the Cornell Summer Colloquium in Medieval Philosophy, NYC (June, 2018). I am very grateful to the participants of both for their helpful remarks. I would also like to thank Claude Panaccio and the anonymous reviewer. Their incisive comments and criticisms greatly improved the paper.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

     

275

Bibliography Adams, Marilyn McCord. William Ockham, 2 vols. (Notre Dame, IN: University of Notre Dame Press, 1987). Brett, Annabel. Liberty, Right and Nature: Individual Rights in Later Scholastic Thought (Cambridge: Cambridge University Press, 1997). Brett, Annabel. “Introduction,” in William Ockham, On the Power of Emperors and Popes (Bristol: Thoemmes Press, 1998). Brower, Jeffrey. “Medieval Theories of Relations,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2015 edition) . Canning, Joseph. “Power and Powerlessness in the Poverty Debates,” in Ideas of Power in the Late Middle Ages, 1296–1417 (Cambridge: Cambridge University Press, 2011), 107–32. Epstein, Brian. “Social Ontology,” in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Summer 2018 edition) . Henninger, Mark G. Relations: Medieval Theories 1250–1325 (Oxford: Clarendon Press, 1989). Lambertini, Roberto. “Poverty and Power: Franciscans in Later Medieval Political Thought,” in J. Kraye and R. Saarinen (eds.), Moral Philosophy on the Threshold of Modernity (Dordrecht: Springer, 2005), 141–63. McGrade, A. S. “The Ontology and Scope of Human Rights: Forward with Ockham,” American Catholic Philosophical Quarterly 86 (2012), 527–38. Mäkinen, Virpi. Property Rights in the Late Medieval Discussion on Franciscan Poverty (Leuven: Peeters, 2001). Miethke, Jürgen. Ockhams Weg zur Sozialphilosophie (Berlin: De Gruyter, 1969). Panaccio, Claude. “Semantics and Mental Language,” in P. V. Spade (ed.), The Cambridge Companion to Ockham (Cambridge: Cambridge University Press, 1999), 53–73. Panaccio, Claude. Ockham on Concepts (Aldershot: Ashgate, 2004). Panaccio, Claude. “Ockham’s Externalism,” in G. Klima (ed.), Intentionality, Cognition and Mental Representation in Medieval Philosophy (New York: Fordham University Press, 2015), 166–85. Panaccio, Claude. “Linguistic Externalism and Mental Language in Ockham and Buridan,” in G. Klima (ed.), Questions on the Soul by John Buridan

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, SPi

276   and Others: A Companion to John Buridan’s Philosophy of Mind (Dordrecht: Springer, 2017), 225–37. Pelletier, Jenny. “Ockham on Human Freedom and the Nature and Origin of Lordship,” in P. Adamson and C. Rapp (eds.), State and Nature: Essays in Ancient Political Philosophy (Berlin: De Gruyter, forthcoming). Robinson, Jonathan. William of Ockham’s Early Theory of Property Rights in Context (Leiden: Brill, 2013). Robinson, Jonathan. “Walter Chatton on Dominium,” History of Political Thought 35 (2014), 656–82. Roques, Magali. “Le principe d’écomonie d’après Guillaume d’Ockham,” Franciscan Studies 73 (2015), 169–97. Searle, John. The Construction of Social Reality (London: Penguin, 1995). Searle, John. Making the Social World: The Structure of Human Civilization (Oxford: Oxford University Press, 2010). Tierney, Brian. “Languages of Rights,” in The Idea of Natural Rights: Studies on Natural Rights, Natural Law and Church Law, 1150–1625 (Atlanta, GA: Scholars Press, 1997), 118–30. Tierney, Brian. “Villey, Ockham, the Origin of Individual Rights,” in The Idea of Natural Rights: Studies on Natural Rights, Natural Law and Church Law, 1150–1625 (Atlanta, GA: Scholars Press, 1997), 13–42. Walter Chatton. Lectura super Sententias, Liber 1, Distinctiones 3–7, ed. J. C. Wey and G. Etzkorn (Toronto: Pontificate Institute of Mediaeval Studies, 2008). William Ockham. Opus Nonaginta Dierum, in Opera Politica I and II, ed. J. G. Sikes and H. S. Offler (Manchester: University of Manchester Press, 1940 and 1963). William Ockham. Opera Philosophica et Theologica, 17 vols., ed. P. Boehner, G. Gál, et al. (St. Bonaventure, NY: The Franciscan Institute, 1974–86). William Ockham. A Short Discourse on Tyrannical Government, ed. and tr. A. S. McGrade and J. Kilcullen (Cambridge: Cambridge University Press, 1992). William Ockham. A Letter to the Friars Minor and Other Writings, ed. and tr. A. S. McGrade and J. Kilcullen (Cambridge: Cambridge University Press, 1995). William Ockham. Breviloquium de principatu tyrannico, in Opera Politica IV, ed. H. S. Offler (Oxford: The British Academy, 1997).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 11/8/2020, SPi

Briefly Noted Rudolf Schuessler, The Debate on Probable Opinions in the Scholastic Tradition (Brill, 2019), 527 pp. This magisterial volume opens up a whole field of neglected disputes from the late scholastic era on the proper method for adjudicating disagreement within practical reasoning. Schuessler’s principal focus is on the doctrine known as probabilism, according to which, in cases of disagreement, it is permissible to act on the basis of any line of reason that satisfies certain minimal conditions for plausibility, even if that opinion is overall less plausible than other options. This doctrine takes explicit form in 1577, with Bartolomé de Medina, and finds its greatest champion in the next century in Juan Caramuel y Lobkowitz. Ultimately it fell from favor after generations of criticism from moral theorists who regarded probabilism as the breeding ground for moral laxism. Schuessler’s coverage of these debates is exhaustive yet, because his material is so consistently fascinating, it is never exhausting. Topics discussed include authority, disagreement, probability, skepticism, doxastic voluntarism, and autonomy, set throughout within the context of moral reasoning. The whole debate, in fact, turns out to be interested in the very sorts of questions at the intersection of epistemology and ethics that are the subject of vibrant discussion in philosophy today. Yet these texts are, of course, virtually unknown except to a small group of experts, and little of that literature has previously been available in English. Time will tell whether this material, for all its interest, finds a home in our standard histories. Working against it is the way it fits awkwardly into the usual narratives: it is not medieval, but yet does not appear to be modern; it is concerned not with the familiar metaphysical questions that lie at the center of the canon, but with questions in applied ethics; the relevant texts are voluminous and have never been translated. But if anything could bring this material out of the shadows, it would be a book like this.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 11/8/2020, SPi

278   Sophia Vasalou, Virtues of Greatness in the Arabic Tradition (Oxford University Press, 2019), 169 pp. Within the great canon of the Aristotelian virtues, the ugly duckling has long been megalopsychia, greatness of soul, which Aristotle called “a kind of crown of the virtues” (NE 1124a1–2), but which has struck centuries of commentators as doubtfully a virtue at all. In this brief, elegantly written book, Sophia Vasalou shows how this ancient virtue has two counterparts in classical Arabic philosophy, greatness of soul (kibar al-nafs) and greatness of spirit (ʿizam al-himma). The first is the direct descendant of the Greek
notion, but it turns out to be the second, on Vasalou’s account, that exercises a deep influence on Arabic ethics. She follows the story through Yaḥya ibn ʿAdī and al-Rāghib al-Iṣfahānī, and into various mirrors for princes. Ultimately, the virtue that has often been dismissed as a product of provincial Greek customs emerges as a strong contender to rejoin the pantheon of the universal virtues, raising interesting questions about whether there indeed are universal virtues, and what it would look like to study them across periods and cultures. Nicolas Faucher, La volonté de croire au Moyen Âge: les théories de la foi dans la pensée scolastique du XIIIème siècle (Brepols, 2019), 412 pp. Medieval discussions of faith might not, on their face, look like the most fertile territory for fresh topics of philosophical reflection, but Nicolas Faucher proves in this volume just how much we still have to learn from this material. The book’s central focus is the problem now discussed under the heading of doxastic voluntarism: to what extent are we able to bring our beliefs under voluntary control? For medieval authors this question was particularly important in the context of faith. On the one hand, since faith is praiseworthy, it seems that it must be something we come to have voluntarily. On the other hand, since faith requires an assent that goes beyond the available evidence, we cannot engender faith in the way we ordinarily acquire beliefs, by considering the evidence. Faucher devotes lengthy chapters to Alexander of Hales, Bonaventure, Aquinas, Olivi, Henry of Ghent, Godfrey of Fontaines, and Scotus. Combining an encyclopedic knowledge of the texts with an understanding of the current debate in analytic philosophy, he shows how diverse and resourceful medieval discussions of this topic were. Interestingly, the topic is one that pulls apart authors who usually find themselves in sync.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 11/8/2020, SPi

 

279

Thus Godfrey takes a very different line than does Aquinas, and Scotus’s view is utterly distinct from Olivi’s. Accordingly, “voluntarism” in this context cuts against the grain of how that label is generally applied. For it turns out to be Aquinas who is the champion of voluntarism when it comes to acts of faith, whereas, in this limited context, Scotus emerges as the leading critic of voluntarism. Thomas Aquinas, Quodlibetal Questions, translated by Turner Nevitt and Brian Davies (Oxford University Press, 2020), lxii + 497 pp. In the world of Aquinas Anglicus, it’s long been an oddity that there has been no translation of the Quodlibeta. That gap has now been filled with this careful volume from Nevitt and Davies. In addition to a complete translation, and an extended introduction, it offers a useful glossary, detailed index, and brief biographies of authors cited. The volume chooses to print the questions in chronological order, and so e.g. begins with Quodlibet 7. This creates an immediate obstacle in locating a particular text within the volume, but repays that trouble by offering a constant reminder of which questions come early in Aquinas’s career and which come late. The reason it has taken this long for a translation to appear is that, of course, quodlibetal questions are an odd sort of genre, and Aquinas’s are no exception. It might be a fun parlor game to look for the strangest topic in the whole genre. One worthy entry from Aquinas’s corpus would be 11.9.1: “Does bewitchment impede marriage?” (Readers on the market for an annulment might consult Aquinas’s answer.) Connoisseurs will be aware that, amidst the curiosities, there are also scattered articles of considerable importance for Aquinas’s fundamental philosophical views. But perhaps the greatest payoff of sustained engagement with this text is that it shows Aquinas giving direct attention to questions of concrete practical import, such as 8.5.1: “Is praying for someone else as valuable as praying for yourself?” or 5.13.1: “Do [members of] religious [orders] have to patiently tolerate the insults directed at them?” In both cases, the answer depends on the details, which is of course the whole point of the exercise.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 11/8/2020, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 17/8/2020, 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, all papers will be refereed according to a triple-blind process, so that neither editor nor referee will know the identity of the author, nor will the author know the identity of the referee. 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 author’s name or any other information that would impede blind refereeing. Papers may be of any length, and in particular we welcome the submission of longer works that fall outside the parameters of most journals. Contributors should bear in mind, however, that the lengthier the work, the higher the 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, 17/8/2020, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 11/8/2020, 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. Abelard 224n.4 Al–Ghazālī 112–13, 115, 117–39 Anselm 36, 79–80, 88–108 Aristotle 7, 19, 22, 36–7, 118–19, 148–9, 151, 154, 161, 163, 170, 176, 190, 227–8 Armstrong, David 65–6 Ashworth, Jennifer 228n.11 Augustine 1–38, 41–3, 45–76

Jantzen, Grace 136 John Duns Scotus 36

Bargeron, Carol 112, 130–1 Brower–Toland, Susan 238 Brown, Peter 26–7, 30 Bussanich, John 11–12

Nietzsche, Friedrich 1n.1 Nussbaum, Martha 1–2, 6, 24

Carrera, Elena 135 Cicero 4n.11, 5, 7, 15–16, 19, 30, 81–2

Lactantius 104–6 Martin of Braga 104–5 McGrade, A. S. 249n.3, 253n.9 Mercer, Christia 111, 113–17, 119–20, 122, 127, 133, 138–9

O’Connell, Robert 11 Panaccio, Claude 265n.33 Pasnau, Robert 144–5 Plato 29–30, 160–1 Plotinus 6

Descartes, René 111–16, 127–9, 132–9 Dewes, George Henry 127–8

Richard Knapwell 187

Epictetus 27–8, 30

Seneca 19, 28, 80–8, 95–108 Schmölders, Auguste 127–8 Sotinel, Claire 26–7

Frege, Gottlob 221 Gilson, Etienne 3, 11–12, 15–16 Godfrey of Fontaines 193–5, 217–18 Groarke, Leo 127–8 Henry of Ghent 192–5, 206–8, 210–12, 216–17

Teresa of Ávila 111–17, 122–9, 132–3, 135–9 Thomas Aquinas 144–84, 187–200, 209–13, 216–18 Thomas of Sutton 187–9, 195–218 Tornau, Christian 15–16, 18 Twetten, David 146–7

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 11/8/2020, SPi

284    Walter Burley 221–46 Walter Chatton 258–9 William de la Mare 249–50 William Hothum 187

William Ockham 248–74 Wippel, John 190 Wolterstorff, Nicholas 26, 34–5