Loading field map
The philosophy of mind’s treatment of self knowledge and subjectivity addresses two intertwined puzzles: how a subject can know its own mental states, and what it means for mental life to be had from a first-person perspective. These are not merely technical problems for epistemologists; they sit at the intersection of metaphysics, philosophy of language, and empirical psychology. The subfield asks what the self is, whether it is a thing or an activity, and how the peculiar authority we seem to have over our own thoughts, feelings, and intentions is possible—or whether that authority is an illusion.
Self knowledge refers to the distinctive way a person knows their own mental states—that one is in pain, believes it will rain, or intends to leave. This knowledge appears different from knowledge of the external world: it is typically immediate (not based on evidence or inference), first-personal (available only from the inside), and authoritative (we usually trust a person’s report of their own experience over third-party observation). The philosophical problem is to explain these features without making the mind a private theater that only its owner can inspect.
Subjectivity names the fact that mental states are for someone. A pain is not just a physical event; it is felt by a subject. This “what it is like” dimension—sometimes called phenomenal consciousness—raises the question of whether subjective experience can be captured in objective, third-person terms. Subjectivity also concerns the structure of experience: every thought seems to be my thought, every perception seems to be from my point of view. The self is not merely an object among objects but the standpoint from which objects appear.
The two notions are linked. If self knowledge is immediate and authoritative, that suggests the subject has a special access to its own mental life. But if subjectivity means that mental life is essentially first-personal, then the question becomes whether that first-personal access is a form of knowledge at all, or something more primitive—like an acquaintance or a performance.
The modern subfield crystallized in the twentieth century, but its questions have deep precursors. Descartes’s cogito made self knowledge the foundation of certainty: the one thing I cannot doubt is that I am thinking. Yet Descartes also treated the mind as a substance whose contents are transparent to itself—a view later philosophers would challenge. Locke’s account of personal identity tied the self to memory and consciousness, raising the question of whether self knowledge is a matter of inner observation or of narrative construction. Kant argued that every experience carries an “I think” that must be able to accompany it, making the subject a formal condition of experience rather than an object of it. These precursors did not use the modern vocabulary of “self knowledge” as a distinct research program; they were concerned with the nature of the soul, the conditions of experience, and the limits of knowledge.
The decisive shift came in the twentieth century, when philosophers began to treat self knowledge and subjectivity as problems in their own right, often in dialogue with psychology and neuroscience. Three broad approaches now organize the field: the Cartesian or inner-sense tradition, the expressivist or constitutivist tradition, and the anti-authority or deflationary tradition. These are not a simple succession; they coexist and cross-fertilize, and many philosophers combine elements of more than one.
The oldest and most intuitive approach holds that self knowledge is a form of perception or introspection. Just as I see the tree, I “see” my own mental states through an inner sense. This view—often associated with Descartes and later with Locke—treats the mind as a domain of objects that the subject can inspect from within. Introspection is then a quasi-perceptual faculty, and its deliverances are reliable because the mind is transparent to itself.
The inner-sense tradition faces a well-known set of problems. First, it struggles to explain the authority of self knowledge: if introspection is a perceptual faculty, it could err, and we would need evidence to correct it—but we rarely treat our own reports as fallible in the same way we treat perception of external objects. Second, it seems to require a homunculus—an inner observer watching the mind—which leads to an infinite regress. Third, it cannot account for the fact that many mental states are not objects of introspection at all. I do not introspect my belief that $2+2=4$; I simply have it. The belief is not an inner object I observe but a way I am oriented toward the world.
Despite these problems, the inner-sense view retains influence in two forms. Some philosophers defend a moderate introspectionism, holding that while not all mental states are introspected, some (like pains and visual experiences) are given through a distinct inner awareness. Others, particularly in the empirical tradition, treat introspection as a fallible but useful method for accessing mental content—a method that can be studied experimentally, as in the “think-aloud” protocols of cognitive psychology. The philosophical version, however, has largely moved away from the claim that introspection is a perception of inner objects.
A second major approach denies that self knowledge is a form of perception or inference. Instead, it holds that first-person authority is not a matter of knowing a fact but of expressing a state. When I say “I am in pain,” I am not reporting an inner observation; I am giving voice to the pain itself, much as a groan expresses pain without being a description of it. This view, associated with Wittgenstein and later developed by philosophers like Donald Davidson and Richard Moran, treats self knowledge as a constitutive feature of mindedness: to be a subject is to be able to say what one thinks, and this ability is not grounded in evidence but in the very activity of thinking.
Wittgenstein’s contribution was to dissolve the picture of the mind as a private theater. He argued that the grammar of first-person statements (“I am in pain”) differs from that of third-person statements (“He is in pain”): the former is not a description of an inner state but a criterion for the state’s presence. This does not mean we are infallible about our own minds; it means that the standard for what counts as a mental state is partly set by the subject’s sincere avowal. If I sincerely say I am in pain, there is no further fact of the matter that could override my report—unless I am lying or using words incorrectly.
Richard Moran developed this into a distinction between the theoretical and the deliberative stance. When I ask “What do I believe?” I can answer either by observing my behavior and inner promptings (theoretical) or by making up my mind (deliberative). Self knowledge is primarily deliberative: I know my beliefs by forming them, not by inspecting them. This is why I can be authoritative about my own intentions in a way I cannot be about someone else’s: I am the author of my intentions, not their spectator.
The expressivist tradition has been criticized for being too deflationary. If self knowledge is just expression, then it seems to lose its cognitive content—it is not really knowledge but a performance. Critics also point out that we are often wrong about our own motives, and that psychoanalysis and experimental psychology have shown that self-reports can be systematically biased. The expressivist response is that these errors are not errors of introspection but errors of self-deception or interpretation: I can be wrong about why I acted, but I cannot be wrong about whether I intend to act, because intending is partly a matter of avowing.
A third approach challenges the very premise that self knowledge is special. This tradition—associated with behaviorism, some forms of functionalism, and more recently with “predictive processing” and “self-model” theories—holds that first-person authority is either an illusion or a useful fiction. On this view, we know our own minds the same way we know other minds: by observing behavior, inferring from context, and constructing a model of ourselves. The apparent immediacy of self knowledge is a byproduct of the fact that we have more data about ourselves (including internal bodily signals) and that we are motivated to maintain a coherent self-narrative.
The most radical version is eliminativism about introspection: there is no such faculty, and the very idea of a “self” is a grammatical illusion. This view has roots in Hume’s claim that when he looks inward he finds only perceptions, no self. Modern defenders, like Daniel Dennett, argue that the self is a “center of narrative gravity”—a useful abstraction, not a real entity. On this account, self knowledge is not a special kind of knowledge but a skill of self-description, and it can be mistaken in the same ways as any other description.
A more moderate version, sometimes called inferentialism, holds that we do have access to our own minds, but that access is inferential and fallible. We learn what we believe by observing our own behavior, our emotional reactions, and our patterns of thought. This view is supported by empirical findings: people are often wrong about their own attitudes (as in implicit bias research), and they can be induced to confabulate reasons for their actions. The inferentialist concludes that first-person authority is a social convention, not a metaphysical fact.
The deflationary tradition has been criticized for failing to explain the phenomenology of self knowledge. When I know I am in pain, it does not feel like an inference; it feels like direct awareness. Moreover, the deflationary view struggles to account for the normative role of self knowledge: we hold people responsible for their intentions, and we treat their sincere avowals as binding in a way that third-party reports are not. If self knowledge were just inference, it is unclear why we would grant it this special status.
A fourth approach, often overlooked in analytic philosophy but central to the continental tradition, treats subjectivity not as a problem to be solved but as the starting point of philosophy. Phenomenology, founded by Edmund Husserl, begins with the first-person standpoint and asks what structures of experience make self knowledge possible. Husserl argued that every act of consciousness has a structure of intentionality—it is directed at something—and that this directedness always includes a sense of the self as the one who is directed. The self is not an object within experience but the subject-pole of experience.
Maurice Merleau-Ponty extended this to the body: subjectivity is not a disembodied mind but an embodied being-in-the-world. Self knowledge, on this view, is not a matter of introspection but of practical understanding—I know my body not by observing it but by using it. Jean-Paul Sartre added that the self is not a fixed substance but a project: I am what I make of myself through my choices, and self knowledge is always retrospective and incomplete.
The phenomenological tradition has influenced analytic philosophy of mind through the work of philosophers like Hubert Dreyfus and, more recently, through the “enactivist” movement, which holds that cognition is not a matter of representing the world but of interacting with it. On this view, subjectivity is not a private inner realm but a mode of engagement with the world, and self knowledge is a form of practical know-how rather than propositional knowledge.
Since the late twentieth century, the subfield has become increasingly interdisciplinary. Cognitive science, neuroscience, and psychiatry have provided new data about self knowledge and subjectivity, and philosophers have responded in two ways. Some have used empirical findings to constrain philosophical theories: if introspection is shown to be unreliable in certain conditions, then the inner-sense view must be revised. Others have argued that empirical methods cannot capture subjectivity at all, because they necessarily adopt a third-person perspective.
This latter position is associated with the “hard problem of consciousness,” a term coined by David Chalmers to distinguish the easy problems of explaining cognitive functions from the hard problem of explaining why there is something it is like to be a conscious subject. The hard problem is not a problem about self knowledge per se, but it bears directly on subjectivity: if subjective experience cannot be reduced to physical processes, then self knowledge cannot be reduced to information processing. This has led to a revival of dualism in a modest form—not the substance dualism of Descartes, but a property dualism that holds that subjective experience is a real feature of the world that cannot be captured in objective terms.
The empirical turn has also produced a rich literature on self-disorders in psychiatry. Conditions like schizophrenia, depersonalization, and autism have been interpreted as disturbances of the sense of self or of the ability to know one’s own mind. These clinical phenomena provide a natural laboratory for testing theories of self knowledge: if a patient reports that their thoughts are not their own, is that a failure of self knowledge, a different kind of self knowledge, or a breakdown of the very distinction between self and other? Philosophers have used such cases to argue that self knowledge is not a single capacity but a cluster of abilities that can come apart.
The field today is characterized by a plurality of approaches that do not reduce to a single paradigm. The inner-sense tradition survives in a weakened form, as a claim about a specific class of mental states (sensations, emotions) rather than about all mental life. The expressivist tradition is influential in the philosophy of action and in debates about moral responsibility, where first-person authority plays a normative role. The deflationary tradition is strong in empirically oriented philosophy, where it aligns with cognitive science’s findings about self-deception and confabulation. The phenomenological tradition continues to inform work on embodiment, emotion, and psychopathology.
The most productive recent work has been integrative: it tries to explain how the different features of self knowledge—immediacy, authority, fallibility—can coexist. One influential proposal is the two-factor view: self knowledge has a constitutive component (I know my intentions by forming them) and an empirical component (I know my character by observing my behavior). Another is the self-model view: the brain constructs a model of the self, and this model is what gives rise to the sense of subjectivity; self knowledge is then a matter of the model being accurate or inaccurate. These proposals are not mutually exclusive, and many philosophers now treat the field as a set of overlapping problems rather than a single question with competing answers.
A central unresolved issue is whether self knowledge is unique or continuous with other kinds of knowledge. If it is unique, then there is a genuine mystery about how a subject can know its own mind. If it is continuous, then the mystery dissolves, but we must explain why it seems unique. This question is unlikely to be settled by philosophy alone; it will depend on how empirical research on introspection, self-monitoring, and the neural correlates of self-reference develops. What is clear is that self knowledge and subjectivity are not optional topics in philosophy of mind. They are the point where the mind’s objectivity—its place in the natural world—meets its irreducibly first-person character. Any adequate theory of mind must say something about both, and the tension between them is the engine of the subfield.