In this book, Richard Sorabji defends his preferred conception of a self, surveys a large number of rival conceptions, reflects on the nature of self-knowledge, and considers how all of this ought to inform our attitudes toward life and death. Sorabji argues that the concept of an embodied self, a subject that “owns” its body, actions, and experiences, is the fundamental notion of a self on which all others depend. He is primarily concerned to advocate this view in opposition to Derek Parfit’s Lockean conception of a self in terms of psychological continuity. Sorabji’s mode of exposition alternates between philosophical argument and exegesis. The philosophical arguments are clearly stated, if not fully developed. The exegetical work serves mainly to exhibit the staggering variety of concepts Sorabji surveys from the Western, Hindu, and Buddhist traditions. As in his some of his prior works, Time, Creation and the Continuum (1983), Animal Minds and Human Morals: The Origins of the Western Debate (1993), and Emotion and Peace of Mind: From Stoic Agitation to Christian Temptation (2000), Sorabji aims neither for detailed dilation of philosophical argument nor expansive scholarly commentary. The virtue of his work, in general and in the present case, is the adroit combination of sharp philosophical insight with encyclopedic knowledge of the history of philosophy.
No doubt this is a work for philosophers.
Contemporary philosophers interested in the nature of the self will find Sorabji’s discursive overview of ancient, Modern and Eastern philosophical concepts of the self an excellent resource for further investigations, and his account of the self suggestive. Similarly, scholars in the relevant fields of the history of philosophy and comparative philosophy will benefit from Sorabji’s explanations of and responses to the debates about the self in contemporary philosophy and will enjoy his insights in their own scholarly fields. Specialists of either sort who hope for detailed philosophical argument or elaborate scholarly excurses may be disappointed by the scarcity of such.
After the introductory chapter, the main argument of the book is divided into seven parts. The development of the argument is labyrinthine and so I will summarize the main points of the chapters in each part, and I will note chief strengths and weaknesses.
Part I, “Existence of Self and philosophical development of the idea”, comprises chapters 1 and 2. In chapter 1, Sorabji briefly recapitulates (without answering) a handful of the main arguments against the existence of a self from the Modern and contemporary periods of Western philosophy. The remainder of the chapter is devoted to stating and arguing that a self is, fundamentally, an embodied subject that “owns” its (1) body, (2) bodily characteristics, (3) psychological states, and (4) actions. I shall call this the Embodied Owner Conception of the Self (hereafter referred to as ‘EOCS’). (21) Sorabji’s main argument for EOCS may be summarized as follows:
1. Individual human beings need to employ EOCS (consciously or otherwise) in order to perceive, in order to develop intentions and agency, and in order to acquire language.
2. Hence, individual human beings need to employ EOCS in order to survive.
3. Hence, EOCS accurately represents the self as it exists in reality unless compelling arguments to the contrary are proffered.
Over the course of the rest of the book, Sorabji argues that no such compelling arguments have been made. I will address my criticisms of this main argument below.
In part I, chapter 2, Sorabji provides a brief sketch of some the major and distinct conceptions of the self he finds in the texts of ancient Western philosophers — thereby refuting the claim that the ancients had no concept of the self — and offers a chronology of the development of the different conceptions. He differentiates among four main kinds of conceptions: the concepts of a “true” self, of an enduring self, of an individual persona, and of an inner self of which we can be directly aware. He dismisses, without argument, the idea that we could substitute without loss his EOCS for all of the concepts that fall within these categories.
Part II, “Personal identity over time”, comprises chapters 3-5 and focuses on what makes an individual the same individual over a period of time. Chapter 3 begins with a discussion of Aristotle’s account and then shifts to how the Stoics and the early Christians understood persistence in the next life. While the historical and exegetical work is interesting (albeit limited), none of Sorabji’s substantive claims hinges on the material presented in this chapter, and he seems content simply to elaborate some of the various possibilities. Chapter 4 evaluates whether or not a person’s identity can depend upon what happens to someone else, addressing ancient and modern puzzles according to which either one individual splits into many (cases of personal “fission”) or many individuals merge into one (personal “fusion”). Sorabji briefly addresses a few of the contemporary philosophical approaches to these topics, with the main focus being Parfit’s account. Chapter 5 is devoted to relating Locke’s account of personal identity to historically antecedents. Sorabji doesn’t devote much space to developing Locke’s view but sketches earlier views that explained the self in terms of psychological continuity, and especially in terms of memory. In the main, the chapter aims to undermine the common view that Locke’s account in terms of psychological continuity was revolutionary.
Sorabji considers two alternatives to EOCS in Part III, “Platonism: impersonal selves, bundles, and differentiation”, comprising chapters 6 and 7. In Chapter 6 he considers a difficulty facing Platonists who hold that the true self is the rational soul. The bulk of the chapter is devoted to how Plotinus deals with these difficulties, with the final section focusing on Avicenna, Averroës and Thomas Aquinas. In chapter 7, Sorabji outlines various ancient proposals for how to differentiate among concrete individuals. His main target is the bundle theory of individuals, an alternative that explains what an individual person is without recourse to a subject that owns the various properties of the person. On the bundle theory account, an individual person is identical to a particular “bundle” of properties. Sorabji does not explain why the bundle theory fails where his succeeds; this is a major shortcoming of the book.
Part IV, “Identity and persona in ethics”, comprises chapters 8-10 and traces the development of the concept of an individual persona in ancient philosophy and its role in practical reasoning. Sorabji considers the views of Cicero (chapter 8), Plutarch (chapter 9), Epictetus (chapters 8 and 10) and Aristotle (chapter 10), and he notes the relationship between these views and contemporary accounts of the “practical self” in the works of Charles Taylor and others. Sorabji apparently thinks all of these notions of the self presuppose his more fundamental EOCS, but it is not clear why this is so.
In Part V, “Self-awareness”, comprising chapter 11-14, Sorabji argues that the unity of self-awareness is a function of a single owner and not of a single faculty. In chapter 11, various ancient arguments for the claim that self-awareness is impossible are refuted. Some ancient arguments for the infallibility of self-awareness are reconstructed in chapter 12, with particular emphasis on those proposed by Plotinus, Augustine, and Avicenna. Chapter 13 is devoted to ancient arguments for the claim that we know ourselves through knowing others. All of these arguments are of general philosophical interest, and the texts Sorabji discusses are crucial. Nevertheless, it is unclear what Sorabji would have us think about infallible self-awareness or knowing ourselves through others, or how these arguments relate to the work as a whole. After a survey of ancient accounts of the faculty by means of which our awareness is unified, Sorabji concludes chapter 14 by asserting that all such approaches are misguided and that, instead, unity is provided by the subject described in EOCS. While this may be conceivable alternative, no argument is provided for the claim.
Part VI, “Ownerless streams of consciousness rejected”, comprises chapters 15 and 16 and constitutes the philosophical core of the book. In chapter 15, Sorabji argues against Derek Parfit’s account of persons in terms of psychological continuity. In Chapter 16 he surveys some important concepts of persons from the Buddhist and Hindu traditions.
In Part VII, “Mortality and the loss of self,” comprising chapters 17-19, Sorabji examines whether or not the fear of personal annihilation is irrational. He considers in chapter 17 various ancient and contemporary accounts of how we might survive death through reincarnation, some form of disembodied existence, or resurrection. Returning to themes and arguments he discussed in his earlier book Matter, Space, and Motion, in chapter 18 he explains how we can make sense of the claim that time goes in a circle and how such a cosmological context would inform our understanding of mortality. In chapter 19, he addresses head-on the irrationality of the fear of personal annihilation. After rejecting various arguments for thinking that such fear is irrational, he endorses a version of what he calls the ‘asymmetrical horror’ argument, according to which we ought not feel horror about our future non-existence since we don’t feel horror about our past non-existence.
Although the whole book is concerned with Sorabji’s EOCS, the most careful defense of is found in chapters 1, 15 and 16. There are a number of weaknesses. First, Sorabji never explains the crucial concepts involved in his account of the self. This is the chief philosophical demerit of the account: we never learn what it means for an embodied subject to “own” its various characteristics. What, exactly, is the ownership relation? Sorabji is also silent about the nature of the “embodied subject”. Presumably, on his account, an individual person is identical to a particular embodied subject. Does this entail that a person is identical to a particular human organism (as Eric Olson has argued) or is it some peculiar kind of “bare particular” (of the sort the David Armstrong discusses)? This is particularly pressing since Sorabji claims both that a person is identical to an embodied subject and that a person owns its body. Thus, an embodied subject owns its body. On the one hand, this suggests that the embodied subject cannot be the same as the body it owns, which may be a virtuous implication. On the other hand, if the embodied subject is not the same as the body, but embodiment is essential to the subject, what is this subject? Second, Sorabji doesn’t make an adequate case against Parfit’s alternative account of persons. Recall that Sorabji’s main argument for EOCS (see above) depends upon the claim that alternative concepts cannot do the work that EOCS does for us. In particular, then, he needs to show that Derek Parfit’s account can’t do this work, since Sorabji considers Parfit’s account the most formidable alternative. I am not persuaded, yet, that Parfit’s account can’t be used to explain perception, intention, and language acquisition and, hence, human survival.
Turning to the exegetical and comparative historical material, my chief complaint is that it is uneven. For example, Sorabji will provide relatively detailed exegetical argument in support of one attribution to an historical figure, and then make bald assertions about another. It is not always obvious why the claims are weighted differently.
The range of Sorabji’s knowledge is breathtaking. He weaves together substantive philosophical argument, textual exegesis and comparative work in the history of philosophy and illuminates the relevance of all. Anyone interested in the nature of the self, or in the historical debate surrounding the self, will find this book stimulating.