Who or What Is the Self?

Adam Robbert
Dec 4, 2019 · 19 min read

An excerpt from a chapter that never was.

There are many questions that orbit the central inquiry of who or what is the self. For instance, is the self a necessary condition for experiences of thinking, feeling, and perceiving? Is the self fully caught up in these experiences or does it stand apart from them? Does it alternate between more than one of these poles? Is the self an unchanging substance or is it a variable activity or process? Are the processes of experience identical to the subject of those experiences? Is the self just a concept, a necessary but illusory fiction? Or is the self equivalent to the body, simply a first-person way of talking about physiology? And if it is not just the body, what else is it and what does this something else mean for scientific and naturalistic worldviews? The point of this chapter is not to treat each of these questions separately but to practice approaching them from different points of view, as shades of possibility revealed or created through different practices of inquiry, though I will also offer answers to each of them.

To get at these questions and the practices that might answer them, I return again to the basic approach to philosophical practice laid out by Hadot, namely, that the key exercise, the primary askēsis, is to engage in a mode of self-examination that is presupposed by any and every attempt to navigate into, through, and beyond the question, who or what is the self? My approach is to treat the question itself as a practice and to treat the answer to the question — whatever that answer turns out to be — as a result of that practice. In this sense, Hadot’s practices of the self bare a strong resemblance to what Evan Thompson, borrowing from Buddhist traditions in Indian philosophy, calls “I-making” (ahamkāra), that is to say, at minimum, the self is enacted by a variety of sense-making and self-specifying processes that must be deployed and maintained within different environments across time, and through different practices of self making.

Thompson’s enactivist view of sense-making provides a general framework of which Hadot’s treatment of philosophy as embodied perceptual transformation is one higher-level human instantiation. If as Thompson says “living is sense-making in precarious conditions,”[1] then philosophy is one of the ways humans act upon the sense-making apparatus itself. In other words, the practices Hadot foregrounds in his work can be read as sets of actions individuals execute upon themselves to modify their sense-making capacities. Thompson’s work is a key reference point here not only because his work on Indian Buddhist philosophy bares a strong resemblance to the practices Hadot emphasizes, but also because his version of enactivism offers a framework for understanding a self at its most basic level. As David Fiordalis, editor of the comparative anthology Buddhist Spiritual Practices: Thinking with Pierre Hadot on Buddhism, Philosophy, and the Path notes, “Hadot provides an alternative framework for understanding philosophy as a practice and a discipline, and in doing so, he gives us a different model for comparing exemplars of philosophical discourse from the Western tradition with those from Buddhist and other non-Western intellectual traditions.”[2]

Thompson’s framework, drawing from Western philosophy (and especially phenomenology), cognitive science, and Indian Buddhism, suggests that whatever else a self is it is at minimum rooted in an organism’s rudimentary self/nonself distinction, though this basic capacity does not necessarily by itself imply abilities for conscious self-reference. In more complex organisms such as human beings, however, sense-making capacities include capacities for self-reference and for the generation of multiple kinds of self, for multiples kinds of I-making achieved through acts of self-specification and self-perpetuation. This is a process-oriented view of the self. Here’s how Thompson’s describes his view, “The self isn’t a thing or an entity at all; it is brought forth or enacted in the process of living.”[3] As he continues to frame his position, he writes, this “feeling that ‘I am’ . . . is a process that enacts an ‘I’ and in which the ‘I’ is no different from the I-ing process itself.”[4]

While Thompson’s account of I-making in an Indian Buddhist context is similar to Hadot’s accounts of askēsis in the Ancient Greek one, a comparative assessment of these traditions is outside the scope of this chapter, but it is worth noting that the central askēsis of a self relating to itself, which Hadot diagnosis as central to Platonism, Cynicism, Epicureanism, Stoicism, and so on, is central also to Indian Buddhist contemplative practice, makes the askēsis of the self a seemingly widespread, cross-cultural practice, at least in its most basic and general forms. This larger analysis aside, what is essential for my purposes in this chapter is to investigate further the possible philosophical positions one might take with regards to I-making, the organism, and the world. This is the task I turn to in the next section, where I survey the works of philosophers and cognitive scientists for empirical, phenomenological, and metaphysical responses to the question of the self. In particular, I draw on Thompson’s work in the context of his debates with the philosopher of mind Thomas Metzinger and the phenomenologist Dan Zahavi. What’s at stake in these debates is how one should read the relations among the self, the organism, and the world, and what amount of reality, if any, one should afford the self in relation to these other domains.

The Metaphysics of the Self Construct

In his work Being No One, Metzinger observes that what one might take to be a self is better seen as a shifting representational construct, a self-model rather than a pre-established, independently existing self-fact. A self for Metzinger is not some given universal substance shared by all humans everywhere; it is rather the contingent and virtual offspring of ecologically situated subpersonal cognitive architectures. For Metzinger, this phenomenal self is more properly understood as a cluster of automatic information processing systems engaged in acts of fictitious self modeling, where the representational construct merely obscures the subpersonal functioning of the organism.[5] Instead of a universal subject, Metzinger refers to the wider phenomenal state space, the arena in which all conceivable self-models are constructed.[6] This arena yields an incredibly large number of possible self-conceptions, a situation that leads Metzinger to deride the reality of the self-construct itself, giving it only an illusory, epiphenomenal metaphysical status. As Metzinger famously stated, “No such things as selves exist in the world: Nobody ever was or had a self.”[7]

In Metzinger’s account, then, there is no real thing called a self, there are only processes of self-modeling, processes that have the central property of being completely transparent as models. In other words, the self-model’s transparency is invisible in one’s experience of the world. One sees with and through the self-model, but one never sees the model for what it is — a fictitious representation. As Metzinger puts it, “[The conscious self-model of human beings] is a wonderfully efficient two-way window that allows an organism to conceive of itself as a whole, and thereby to causally interact with its inner and outer environment in an entirely new, integrated, and intelligent manner.”[8] The self for Metzinger is better seen as a transparent model that de facto cannot be seen or experienced as the representation that it is. The self-model on this view is an undetectable medium by means of which experience is shaped in consciousness. Metzinger writes, “The transparency of the self-model is a special form of inner darkness. It consists in the fact that the representational character of the contents of self-consciousness is not accessible to subjective experience.”[9]

It is reasonable to ask what role an unreal and inaccessible self-model could play in experience or in the larger casual system of nature. Metzinger’s point here is not that the self-model has no purpose, but that it is not what we take it to be in our felt sense first-person experience. For example, Metzinger notes that one crucial purpose of the self-model is to internally represent the inner states of the organism, to predict its own future needs and actions, and to anticipate the behaviors of other phenomena in its environment.[10] The self-model is on this account heavily invested in self-investigation and prediction. “Self-consciousness,” Metzinger writes, “is the capacity to become the object of one’s own attention.”[11] All other representations are on Metzinger’s view embedded in this primary self-representation, and more precisely, in the representation of an inner and outer distinction. As Metzinger notes, “For beings like ourselves, the representation of a self–world boundary is the most fundamental characteristic of our representational state space: all other forms of representational content always belong to at least one of those two categories, which, in standard situations, form an exhaustive distinction within reality itself. Everything belongs to either the self or the world.”[12]

That the self-model is a process and not a thing, a transparent and inaccessible representation, leads Metzinger to his no-self stance. He writes, “Under a general principle of ontological parsimony it is not necessary (or rational) to assume the existence of selves, because as theoretical entities they fulfill no indispensable explanatory function. What exists are information-processing systems engaged in the transparent process of phenomenal self-modeling. All that can be explained by the phenomenological notion of a ‘self’ can also be explained using the representationalist notion of a transparent self-model.”[13]

To be sure, there is good empirical support for Thompson’s and Metzinger’s view of the self as process brought forth by an organism within its life cycle. The issue here is over how to think about the ontological status of the self-model, rather than about how to read the process-oriented and situated status of the self per se. For example, the cognitive scientist Lisa Barrett observes that the human organism is able to generate multiple self-conceptions and self–world understandings.[14] On Barrett’s account, the brain can generate different kinds of minds from the same set of physical organs, neuronal networks, and hemispheric and cortical physiologies through an ongoing process of wiring and rewiring, giving rise to highly diverse sets of mental events.[15] “The human brain evolved,” Barrett writes, “in the context of human cultures, to create more than one kind of mind.”[16]

The diverse possibility space within which multiple kinds of minds emerge is on Barrett’s account a result of the fact that the brain is a complex system replete with plasticity (referring to the brain’s ability to change over time), degeneracy (through which different sets of neurons are able to produce the same outcomes), and multipurpose circuitry (where the same networks of neurons are capable of executing multiple different tasks in distinct settings). In Barrett’s words, “The brain is high-complexity because, within one physical structure, it can reconfigure its billions of neurons to construct a huge repertoire of experiences, perceptions, and behaviors.”[17] For Barrett, then, a mind is not equivalent to a brain. Rather, a brain is a necessary but insufficient condition for having a mind.

While Metzinger and Thompson both offer process-oriented accounts of the self, they disagree on the metaphysical weight such an account should be afforded. Their disagreement also opens out into an important and subtle question, namely, what is the relation between the self-model and the organism who generates it? Following the Indian Buddhist account of dependent origination — the so-called “middle way” view that sees the self neither as independent substance nor as lacking reality altogether — Thompson argues against what he describes as the “neuro-nihilism” of philosophers like Metzinger. Here is how Thompson depicts this position, “We find this nihilist extreme today among those neuroscientists and neurophilosophers who, realizing that the brain offers no home for a substantially real self, come to the conclusion that there is no self whatsoever and that our sense of self is a complete illusion.”[18] On Thompson’s reading, the neuro-nihilist imports a particular view of the self — the self as independent substance — and takes aim, assassinating the validity of this position by bringing to bear the full power of modern cognitive neuroscience. Thompson continues, “Neuro-nihilism assumes that were the self to exist, it would have to be an independently real thing or indivisible entity. The problem is that there is no such thing or entity in the brain. So, if it seems to us that we have or are an independently real self, then our sense of self must be created by our brain.”[19]

Thompson for his part counters the neuro-nihilist image with Tibetan Buddhist concepts of dependent origination, concepts which he sees as consistent with his own accounts of enactivism, self-specification, and sense-making, and with aspects of the Western philosophical tradition, especially as articulated by the phenomenologist Maurice Merleau-Ponty, whose approach to phenomenology centers the body’s sensorimotor relation to its environment.[20] Buddhism, enactivism, and phenomenology, then, all support a similar understanding of dependent origination while drawing on different vocabularies, methods, and traditions.

In Thompson’s summary, there are three primary aspects to dependent origination. The first aspect is that all beings depend for their existence on sets of interdependent causes and conditions that lead to the emergence and perishing of that being. The second aspect is that all beings are mereologically dependent in the sense that each one depends on constituent part­–part relations and part–whole relations that allow the being to continue existing in time and place. These relations are complex and multidirectional in the sense that organisms rely on organs, organs rely on cells, cells rely on organelles, and so on, but also that organelles rely on cells, cells rely on organs, and organs rely on organisms, making mereological dependence a two-way street between bottom-up and top-down dependency. The third aspect is that all phenomena are conceptually dependent in the sense that the identification of what counts as a single whole depends upon the concepts one brings to bear in the identification process.[21] In summary, dependent origination includes causal, mereological, and conceptual dependence.

The aspect of concept dependence also implies scale dependence: Wholes are conceptually partitioned in consciousness across scales of being (from micro to macro), meaning wholes emerge or disappear depending on the scale applied. As Thompson notes, conceptual dependence is a more subtle form of dependence than the other two. Thompson writes, “What we mark off as a system depends on our cognitive frame of reference and the concepts we have available . . . [but] such conceptual dependence doesn’t mean that nothing exists apart from our words and concepts, or that we make up the world with our minds.”[22] The reality of conceptual dependence is what links practices of sense-making with practices of I-making, and it is the conceptual dependence of notions of the self that give rise to the great variety of self-concepts one finds in the literature, and, consequently, of the different self–world relations that one finds across history and culture, a point I’ll return to in more detail below.

The phenomenologist Dan Zahavi also offers his own criticism of Metzinger’s position, joining Thompson in interrogating Metzinger’s metaphysical stance on the self. Zahavi writes of Metzinger, “In his view, a phenomenological account of selfhood has no metaphysical impact,”[23] and then immediately counters with his objection,

If we were to wholeheartedly endorse [Metzinger’s] metaphysical principle, we would declare illusory most of the world we live in and know and care about. Why not rather insist that the self is real if it has experiential reality and that the validity of our account of the self is to be measured by its ability to be faithful to experience, by its ability to capture and articulate (invariant) experiential structures.[24]

Again, while Zahavi’s and Thompson’s metaphysical stances to the reality of the self differ from Metzinger’s, the differences among the three philosophers are rooted less in strong disagreements about cognitive science research than they are in the appropriate ontology of experience; the issue is distinctly metaphysical more than it is empirical.[25]

Zahavi’s own account of what a self is, like Thompson’s, is not a retreat into an independent substance view of the self but rather towards what he calls a “minimal” or “core self.” Zahavi describes the minimal self in this way, “In short, the self is conceived neither as ineffable transcendental condition, nor as a mere social construct that evolves through time; it is taken to be an integral part of our conscious life with an immediate experiential reality.”[26] By affording immediate experiential reality a metaphysical status in his philosophical image of the self, Zahavi offers what one could describe as a naturalism that includes in its metaphysics the aesthetics, capacities, and consequences of different representational constructions, including the impact and affects of different self-constructions, in whatever forms they finally take.

For Zahavi, then, his image of the self is a situated, process-oriented construct, but it is not an entirely relative fabrication either. It is precisely the self’s status as universally and structurally situated and constructed that unites the various self-conceptions one might generate. Here’s how Zahavi describes it, “Whereas we live through a number of different experiences, the dimension of first-personal givenness remains the same. In short, although the self, as an experiential dimension, does not exist in separation from the experiences, and is identified by the very first-personal givenness of the experiences, it may still be described as the invariant dimension of first-personal givenness throughout the multitude of changing experiences.”[27]

Self-conceptions on Zahavi’s view thus emerge within a broader phenomenal state space in much the way that Metzinger describes, but the self-concept’s status — the self’s conceptual dependence, as Thompson would say — as interdependent with the first-person architecture does not mean it is not a real thing; it simply means that its ontology should be understood as part of a broader experiential domain that includes experience, self-awareness, and selfhood.

Zahavi’s account of the self-concept is in this way not different in kind from his phenomenological account of experience and perception as such. Here is how in general terms Zahavi describes phenomenology, “Phenomenology pays attention to the givenness of the object, but it does not simply focus on the object exactly as it is given; it also focuses on the subjective side of consciousness, thereby illuminating our subjective accomplishments and the intentionality that is at play in order for the object to appear as it does . . . when we investigate appearing objects, we also disclose ourselves as datives of manifestation, as those to whom objects appear.”[28] The subject “to whom objects appear” is precisely the kind of self-model that Metzinger suggests is essential for representing and predicting the inner and outer environments that every human must navigate, regardless of the specific goals or needs.

Understanding the terms of art in Zahavi’s quote — and especially the notions of intentionality, givenness, and subjective accomplishment — is thus essential to understanding Zahavi’s account of the self, and his view of the nature of experience in general. Phenomenologists do not use the word “intentionality” in its common sense definition of “doing something on purpose” but rather to center the aboutness or directedness of ordinary conscious experience. In other words, ordinary episodes of lived experience are always directed towards some object, thought, event, belief, or feeling. How these phenomena appear in the experience of the individual is described in terms of their “givenness” to consciousness. Episodes of conscious experience are related to the ways acts of intentionality render phenomena in perception, and thus the skills of intentionality one possesses are best read as subjective accomplishments; they are efforts of perception that give phenomena to experience in different ways.

For a phenomenologist like Zahavi, then, experiences of understanding a phenomena to be a certain way — to mean what it does to the individual having the experience — vary in accordance with the skills of perception the individual brings to the encounter. As Zahavi observes, “The same object can be given in a variety of modes.”[29] The phenomenologist responds to this dynamic by investigating the gap between the intentional representations of a phenomenon as it is given to consciousness and the intrinsic properties of the phenomenon itself, reasoning that often times what one takes to be differences attributed to the properties of objects are better understood as differences in intentional attitudes issuing from the subject. This is why, as Zahavi suggests, “The same object, with the exact same wordly properties, can present itself in a variety of manners.”[30]

While Zahavi’s account of the self-construct is not immediately different in kind from other intentional or subjective accomplishments, it does differ from other intentional relations in one crucial way: Zahavi suggests that the minimal self is “pregiven.”[31] To say that the self is pregiven simply means that whatever presentation an appearing object takes in our intentional achievements, that object has as its precondition of appearing a subjectivity before which the object appears. The minimal self for Zahavi is thus a precondition for experiencing any kind person, object, event, or world whatsoever. Insofar as we humans are organisms that encounter phenomena in a world, argues Zahavi, we have if we investigate the plurality of such possible experiences a minimal self that accompanies each of these experiences, and it is only through the variety of these intentional stances that we encounter and discover the world in which we live and the aims, values, and purposes that guide our actions as we move through it. Contra Metzinger, Zahavi suggests that a human has as a constituent part of its being a pregiven minimal self, and its status as an ever-present background of other intentional states makes including the minimal self a necessary feature in the ontology of experience.

There are also important practical and therapeutic dimensions to Zahavi’s and Thompson’s positions that are not easily accounted for from Metzinger’s point of view. For example, of the Greek aphorism “know thyself,” which has followed us throughout these pages, Zahavi writes, “Only a being with a first-person perspective could make sense of the ancient dictum ‘know thyself’; only a being with a first-person perspective could consider her own aims, ideals, and aspirations as her own and tell a story about them.”[32] Even a move like Metzinger’s — a move to undermine the reality of first-person experience — happens always within the structures of awareness made possible only in the first person. This much Metzinger would agree with, as for him it is through the self-model that self-consciousness can take its own self as an object of investigation. But why, then, call this crucial developmental maneuver an illusion if it is only through acts of investigation deliberately taken up in the first-person mode that the self’s abilities, character, convictions, habits, and dispositions are acted upon?

Thompson has also noted the more specific link between volitional consciousness and contemplative training (of which the injunction “know thyself,” as we saw in the previous chapter, is surely an archetypical example).[33] Thompson’s contemplative neuroscience method approaches attention, awareness, and emotion regulation as trainable skills. As with the trainability of intentional skills in general, Thompson claims that contemplative practitioners are able to generate new data about conscious states unachievable without training; are able to enhance capacities for focused attention, one-pointed concentration, and metacognition; and are able to give more accurate first-person reports of their own mental states than their nontrained peers.[34] Moreover, Thompson argues, contemplative training provides a means by which a psychological action like contemplative practice can intervene on the biological or physical state of the organism, including by initiating changes to neural activity, immune function, and hormonal patterns.[35]

In other words, Metzinger’s claim ought to be countered not simply on the grounds that one has a romantic attachment to appearances, aesthetics, and untrue fictions, but more precisely because the minimal sense of self is an integral part of living and making one’s way through the world, and is especially crucial to acts and practices that involve I-making. There is, if we accept Thompson’s view, a causal relationship between the intentional acts of the contemplative practitioner, their modes of I-making, and the structure and functioning of the organism, and this causal relationship ought to lead one to re-think the ontological status of the self.

On these more general points, one can see how Thompson’s contemplative neuroscience method is but one branch of what he elsewhere calls neurophenomenology (a program to which Metzinger in Being No One also subscribes). Thompson describes this larger research program in this way:

The most important feature of this approach, for our purposes here, is that experience is not seen as an epiphenomenal side issue, but is considered central to any adequate understanding of the mind, and accordingly needs to be investigated in a careful phenomenological manner. Phenomenology and experimental cognitive science are thus seen as complementary and mutually informing modes of investigation. Neurophenomenology builds on this view with the specific aim of understanding the nature of consciousness and subjectivity and their relation to the brain and body. The working hypothesis of neurophenomenology is that phenomenological accounts of the structure of human experience and scientific accounts of cognitive processes can be mutually informative and enriching.[36]

For Zahavi and Thompson, then, there are practical, phenomenological, and metaphysical reasons for rejecting Metzinger’s view of the self as having no metaphysical impact.

To make the point in a more general way, I-making matters for aesthetic and epistemological reasons, because different types of self have different capacities; and I-making matters for political and ethical reasons, both in the sense that who or what a self is has changed over time, and because the types of self that have been coded with different kinds of legal and political rights have not been stable or equitably distributed across history either. These facts demand a view of the self-construct that accepts its legal, political, ethical, and metaphysical reality, without losing hold of its situated, process-oriented, and constructed character. As Thompson notes, I-making happens on biological, social, and psychological levels,[37] such that one is obliged to say that, while the circulations among these levels is no doubt complex, it is not the case that anyone level can operate with a freedom independent from the others, or that one level can totalize the rest. In other words, psychological freedom is limited and constrained by social and linguistics conditions, while social and linguistic conditions are themselves limited by the biology of human organisms, even as both psychological and social conditions can change the biological possibilities present within a human being’s life time, as Thompson’s neurophenomenological framework attests.

[1] Thompson, Waking, Dreaming, Being, 328.

[2] Fiordalis, “Introduction,” 9.

[3] Thompson, Waking, Dreaming, Being, 324.

[4] Thompson, Waking, Dreaming, Being, 325.

[5] Zahavi, Subjectivity and Selfhood, 128.

[6] Metzinger, “Are You Sleepwalking Now?” https://aeon.co/essays/are-you-sleepwalking-now-what-we-know-about-mind-wandering

[7] Metzinger, Being No One, 1.

[8] Metzinger, Being No One, 1.

[9] Metzinger, Being No One, 331.

[10] Metzinger, Being No One, 327.

[11] Metzinger, Being No One, 317.

[12] Metzinger, Being No One, 315.

[13] Metzinger, Being No One, 337.

[14] Barrett, How Emotions Are Made, 279.

[15] Barrett, How Emotions Are Made, 279–281.

[16] Barrett, How Emotions Are Made, 280.

[17] Barrett, How Emotions Are Made, 282.

[18] Thompson, Waking, Dreaming, Being, 322.

[19] Thompson, Waking, Dreaming, Being, 322.

[20] See Thompson, Waking, Dreaming, Being and Mind in Life.

[21] Thompson, Waking, Dreaming, Being, 330–331.

[22] Thompson, Waking, Dreaming, Being, 332.

[23] Zahavi, Subjectivity and Selfhood, 128.

[24] Zahavi, Subjectivity and Selfhood, 128.

[25] Here’s Zahavi from his essay “Being Someone” (a retort to Metzinger’s own Being No One) on where Metzinger overlaps with but then ultimately departs from his phenomenologist peers: “There is, superficially at least, a rather striking overlap between Metzinger’s description and the account favored by numerous phenomenologists. That is also where the agreement ends. The phenomenologists would argue that the self is real if it has experiential reality, and that the validity of our account of the self is to be measured by its ability to be faithful to experience, by its ability to capture and articulate (invariant) experiential structures. By contrast, Metzinger argues that it would be a fallacy (what he calls the error of phenomenological reification) to conclude from the content and structure of phenomenal self-experience to the literal properties of an internal and non- physical object, which is what Metzinger takes the self to be” (11).

[26] Zahavi, Subjectivity and Selfhood, 106.

[27] Zahavi, Subjectivity and Selfhood, 132.

[28] Zahavi, Subjectivity and Selfhood, 123.

[29] Zahavi, Subjectivity and Selfhood, 118.

[30] Zahavi, Subjectivity and Selfhood, 122.

[31] Zahavi, Subjectivity and Selfhood, 126.

[32] Zahavi, Subjectivity and Selfhood, 129.

[33] Thompson, “Contemplative Neuroscience as an Approach to Volitional Consciousness.”

[34] Thompson, “Contemplative Neuroscience as an Approach to Volitional Consciousness,” 187–188.

[35] Thompson, “Contemplative Neuroscience as an Approach to Volitional Consciousness,” 194–195.

[36] Thompson, “Neurophenomenology and Contemplative Practice,” 227.

[37] Thompson, Waking, Dreaming, Being, 325.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade