** # I. Introspection: Preliminary Conceptual Demarcation ## A. Core Notion: The capacity, process, or act of attending to, examining, or becoming aware of one's own concurrent or immediately past [[mental states]] and processes. Often characterized as "looking inward." ### 1. Distinction from Exteroception: Contrasted with perception directed at the external world. - a. Directionality: Inward vs. Outward focus of attention. - i. Metaphorical Nature of "Inward": Critically examining the spatial metaphor. The mind is not literally inside in a spatial sense analogous to physical objects within a container. - (1) Non-Spatial Locus: [[Mental states]] lack precise spatial location in the way physical objects do. Where exactly is a belief located? - (a) Cartesian Heritage: The metaphor arguably inherits from substance dualism (mind as non-spatial substance). - i. Modern Physicalism: Within physicalism, [[mental states]] are brain states (or realized by them), which do have spatial location. - A. Problem Reframed: If [[mental states]] are brain states, is introspection then just attending to specific brain activity? - I. Neuroscientific Introspection: Could advanced brain imaging allow third-person access to the same information? - α. If Yes: Does this collapse the first-person/third-person distinction for [[mental states]]? - • Privilege Question: Would the subject's introspective access still be epistemically privileged over the neuroscientist's scan? - · Arguments for Privilege: Subjectivity, directness, acquaintance. - * Subjectivity: The irreducible "what it's like" aspect seemingly only accessible from the inside. - o Nature of Subjectivity Claim: Is it ontological (a non-[[physical property]]) or epistemic (a mode of access)? - ■ Ontological View: Qualia as intrinsic, non-physical properties (Property Dualism). - • Implication: Neuroscience can only ever access correlates, never the thing itself. Privilege is absolute. - ■ Epistemic View: Subjectivity arises from a unique first-person mode of presentation of physical states/properties. - • Implication: The information might be physically encoded, but only accessible as subjective via the first-person mechanism. Privilege relates to this mode of access. - ¤ Can this mode be replicated? Could an AI simulating the brain gain the same privileged access? Depends on whether the mode is functional or requires specific biological substrate. - * Directness: Introspective access often feels immediate, not mediated by instruments or complex inference (phenomenologically). - o Phenomenological Claim: Is this feeling of directness veridical or an illusion? - ■ Veridical View (e.g., Acquaintance): The phenomenology reflects a genuine unmediated access. Privilege stems from lack of intermediaries. - ■ Illusion View (e.g., Interpretive Models): The feeling of directness masks rapid, unconscious inference or construction. Privilege is weak or non-existent. - • Source of Illusion: Cognitive fluency? Automaticity of self-interpretation? - * Acquaintance: Potential direct awareness relation (Russellian) to [[the state]] itself. - o Scope of Acquaintance: Does it apply to all introspected states or only specific types (e.g., qualia)? - ■ Broad Scope: Acquainted with qualia, attitudes, etc. (Less common). - ■ Narrow Scope (Gertler): Acquainted primarily/only with phenomenal states. Privilege is domain-specific. - · Arguments Against Privilege: Fallibility, potential for neural data to correct first-person reports. - * Fallibility examples: Misidentification of emotions, confabulation, biases affecting self-assessment. - o Implication of Fallibility: Does fallibility undermine privilege completely, or just limit its scope? - ■ Undermines Completely: If introspection can be wrong, and external evidence can be right, external evidence can override. No absolute privilege. - ■ Limits Scope: Privilege might hold only for certain aspects (e.g., phenomenal presence) or provide only prima facie justification, defeasible by strong external evidence. - * Correction examples: Neural evidence suggesting a different emotional state than reported, or detecting precursors to decisions before reported awareness (Libet-style findings). - o Nature of "Correction": Is the neural data revealing the true state, or just a conflicting signal whose interpretation is also theory-laden? - ■ Strong Correction: Neural data provides ground truth. - ■ Weak Correction: Neural data provides evidence that challenges the first-person report, requiring reconciliation, but doesn't automatically override it. Maybe both reports are partial or misleading. - * Anosognosia Example: Cases where brain damage leads to lack of awareness of deficit, conflicting with external evidence. Demonstrates radical failure of self-monitoring accessible to third parties. - o Interpretation: Does anosognosia show failure of introspection itself, or failure of belief-formation based on (perhaps degraded) introspection or bodily feedback? - β. If No: What essential aspect does introspection capture that brain imaging misses? - • The Subjective Quality: The "what it's like" (qualia) aspect. - · Irreducibility Argument: Qualia seem irreducible to physical/functional properties detectable by scans. (Chalmers' Hard Problem). - * Basis of Irreducibility: Conceptual gap? Ontological gap? Explanatory gap? - • The First-Person Mode of Presentation: Accessing [[the state]] as mine. - · Indexicality: The essential "I" in "I am in pain" not captured by "Subject X's C-fibers are firing." - * Nature of Indexical Element: Is it purely linguistic, conceptual, or grounded in a pre-conceptual self-awareness? - o Linguistic View: "I" is just an indexical term like "here" or "now". - o Conceptual View: Requires deployment of a first-person concept. - o Phenomenological View: Grounded in pre-reflective self-awareness inherent in consciousness. - II. Access Differences: Introspection accesses brain states via their functional/phenomenal role, not their raw physical properties. - α. Analogy: Software vs. Hardware: We interact with software functionally, not by inspecting transistors. - • Limits of Analogy: Software is inspectable at the hardware level, potentially revealing hidden processes. Is the mind inspectable similarly? Does introspection access the "user interface" layer while neuroscience accesses the "code/hardware"? - · Disanalogy 1: [[Mental states]] might not be software on hardware; relation might be identity, realization, emergence. - · Disanalogy 2: Introspection might access phenomenal properties, which software lacks (unless panpsychism/functionalism about qualia holds). Accessing phenomenology vs function. - (2) Functional Interpretation: "Inward" might refer to accessing information proprietary to the system itself, not accessible via external senses. - (a) Information Domain: The "inner world" as a domain of information uniquely available to the system hosting it. - i. Types of Proprietary Information: Bodily sensations (interoception), cognitive processes (thoughts, plans), affective states (emotions, moods), executive commands (efference copies). - A. Why Proprietary? Is it a matter of fundamental architecture or contingent limitations of external sensors? - I. Architectural: The system's design inherently routes this information for internal consumption first. (e.g., Efference copies needed for motor control are inherently internal signals). - α Significance: Suggests first-person access is deeply integrated with basic functioning, [[not just]] a passive observation mechanism. - II. Contingent: Future technology (perfect brain reading) might overcome this limitation. - α. Conceptual Problem: Even with perfect brain reading, would it capture the subjectivity or just the correlates? Does reading the code equate to running the program? Does knowing the physical details give the first-person perspective? (Revisits I.A.1.a.i.(1)(a)i.A.I.α.) - (3) Phenomenological Interpretation: "Inward" might describe the subjective feeling of attending to thoughts/feelings as distinct from attending to sights/sounds. - (a) Felt Difference: A qualitative difference in [[the experience]] of attending. - i. Source of Felt Difference: Is it the act of attending or [[the object]] attended to? - A. Act-Based: A unique phenomenology of introspective attention itself (subtle, perhaps). A specific "flavor" to inner focus. - I. Nature of Attentional Phenomenology: Is it a distinct quale, or a [[structural feature]] of attention (e.g., focus, effort)? - B. Object-Based: The difference derives from the intrinsic feel of thoughts/feelings versus sights/sounds. The targets have different feels. - I. Nature of Object Phenomenology: Sensory qualia vs. cognitive qualia vs. affective qualia. - b. Sensory Modalities: Exteroception operates via distinct sensory modalities (sight, sound, touch, etc.). Does introspection have analogous modalities? - i. Lack of Dedicated Sensory Organs: There's no obvious "inner eye" or "inner ear" for introspection. - (1) Implication for Inner Sense Models: Creates immediate disanalogy with external perception. Weakens [[the analogy]]. - (a) Rebuttal Attempt: Perhaps the "organ" is a specific cognitive mechanism or neural circuit, not a peripheral sense organ. (e.g., Armstrong's "brain scanner"). - i. Identifying the Mechanism: Candidates include prefrontal cortex areas involved in self-monitoring, insula for interoception/affect. - A. Specificity Problem: Are these areas dedicated solely to introspection, or do they serve broader metacognitive/executive functions? Most evidence points towards broader functions (e.g., error monitoring, attention control, emotional processing). - I. Interpretation: Introspection might be one function implemented by these broader systems, not a dedicated "module". - (2) Unified vs. Diverse Access: Is introspection a single, unified mode of access, or does accessing a pain differ fundamentally in mode from accessing a belief? - (a) Potential Modal Differences: The way one is aware of a bodily sensation (e.g., proprioception, often considered quasi-introspective) feels different from awareness of a complex thought process. Does this reflect different introspective modalities or differences in [[the objects]] introspected? - i. Argument for Modal Diversity: The access to qualia (pain) feels direct/acquaintance-like, while access to belief often feels more conceptual/interpretive. Access to emotion involves feeling + appraisal. - A. Hybrid Theories: Suggests introspection might involve multiple distinct mechanisms depending on the target state (e.g., acquaintance for qualia, interpretation for attitudes, interoception for bodily states). - I. Plausibility: Captures phenomenological diversity. - II. Complexity: Less parsimonious than unified theories. Raises questions about coordination. - ii. Argument for Unified Mode, Diverse Objects: Introspection is a single attentional mechanism directed at different types of internal representations (phenomenal vs. propositional). The feel difference comes from the representation type being attended to. - A. Simplicity: More parsimonious than positing multiple introspective mechanisms. - B. Analogy: Attention can shift between visual and auditory stimuli (different objects, different feels) using a potentially unified attentional system. The attentional act might feel similar even if the objects feel different. ### 2. Distinction from Memory (Episodic): While introspection can target immediately past states, it's distinct from recalling autobiographical events. - a. Temporal Focus: Introspection primarily targets currently active or very recently active states, whereas episodic memory targets events further in the past. - i. The "Specious Present": How immediate must a state be to count as an object of introspection rather than short-term memory? Where is the boundary? (cf. William James). - (1) Functional Definition: Perhaps the boundary lies where active processing/phenomenal presence ceases. - (a) Defining "Active Processing": Online manipulation vs. passive maintenance in working memory vs. offline storage. - (b) Defining "Phenomenal Presence": Currently experienced vs. recently experienced but still "echoing" vs. recalled representation of past experience. Is there a phenomenal continuity that defines the specious present? - (2) Epistemic Definition: Perhaps the boundary lies where the presumed special epistemic security of introspection gives way to the known fallibility of memory reconstruction. - (a) Gradual Decline?: Is the epistemic security a sharp drop-off or a gradual decline as the state recedes in time? - i. Implication for Reliability: Introspection of just past states might already involve reconstruction, making even immediate retrospection fallible. (Dennett's "Orwellian vs. Stalinesque" revisions - does consciousness edit the past immediately, or does memory reconstruct it later?). - A. Orwellian Model: Subsequent processing revises the contents of consciousness itself; later reports reflect the revision. Memory is accurate about the (revised) conscious state. - B. Stalinesque Model: Consciousness is as it was, but memory access involves reconstruction/confabulation. Report reflects faulty memory. - C. Empirical Distinction: Difficult to distinguish experimentally. - b. Nature of Content: Introspection accesses the state itself (or purports to), while memory accesses a representation of a past event/state, often reconstructed. - i. Accessing the State "Itself": What does this mean? - (1) Constitutive View: Awareness is part of the state. (e.g., Self-representationalism). To introspect is just for the state to occur consciously. - (2) Direct Realism (Internal): Unmediated awareness of the state. (e.g., Acquaintance). No intervening representation. - (3) Representationalism: Introspection accesses a representation, but one generated concurrently with the target state, unlike memory's later representation. (e.g., Higher-Order Perception). - (a) Introspective Representation: What is its format? How does it guarantee accuracy about the concurrent state? Is the representation itself the locus of introspection? Does the target state cause the introspective representation reliably? ### 3. Distinction from Self-Knowledge (General): Introspection is often considered a primary source or method for gaining self-knowledge, but isn't identical to it. Self-knowledge encompasses dispositional states, character traits, personal history, etc., which may not be directly introspectible. - a. Scope: Introspection targets occurrent/conscious states; self-knowledge is broader. - i. Dispositional States: Beliefs, desires, traits existing even when not currently active. - (1) Accessing Dispositions: How do we know our dispositions? - (a) Inference from Occurrent States: Noticing patterns in introspected thoughts/feelings. ("I often feel anxious in crowds, so I must be somewhat agoraphobic"). Requires reliable introspection + reliable inference. - (b) Inference from Behavior: Observing our own actions. ("I always choose vanilla, so I must prefer it"). Links to Self-Perception Theory (Bem). Often unconscious inference? - (c) Inference from Others' Testimony: What others tell us about ourselves. ("People say I'm generous"). Requires trust and accurate perception by others. - (d) Direct Introspection? Some argue for introspecting dispositions directly (controversial). Perhaps a felt sense of a disposition? A subtle affective signature? (e.g., feeling inclined towards P even when not actively thinking it). - b. Source: Self-knowledge can also derive from inference, observation of one's behavior, testimony from others, etc. - i. Integration of Sources: How are these diverse sources integrated to form a coherent self-concept? (Narrative self-construction, weighting evidence based on context/motivation). - ii. Conflicts Between Sources: What happens when introspection conflicts with behavioral evidence or others' views? Which source is typically privileged? (Often introspection, but perhaps mistakenly - see biases, self-deception). # II. The Nature of Introspective Access: Core Models and Debates ## A. Metaphorical Models and Their Discontents: ### 1. Perceptual / "Inner Sense" Models (Locke, Armstrong): Introspection is analogous to perception – an inner sense scans internal states. - a. Key Features: - i. Mechanism: Posits a dedicated cognitive mechanism for internal monitoring. - (1) Evolutionary Plausibility: What selection pressures could lead to such a mechanism? Monitoring internal states for behavioral control? Social cognition (predicting others by analogy)? - (a) Behavioral Control: Detecting errors, monitoring goals. (Metacognition link). Needed for flexible, goal-directed behavior. - (b) Social Cognition (Simulation Theory): Understanding others by simulating their states and introspecting the result. - i. Requires Accurate Self-Monitoring: Simulation is useful only if introspection of the simulated state is reliable. Potential co-evolution? - (2) Neurological Correlates: Attempts to identify brain regions associated with self-awareness/introspection (e.g., insula, anterior cingulate cortex, medial prefrontal cortex). Are these evidence of a dedicated mechanism or just neural activity correlated with the process? - (a) Correlation vs. Causation: Neural activity might be necessary but not sufficient, or might be a consequence rather than the cause of introspection. Activity could reflect the target state, not the introspective act. - (b) Domain Generality: These regions are involved in many functions beyond introspection (error detection, emotion, attention). Does this argue against a dedicated introspective mechanism? - i. Counter-Argument: Key perceptual areas (e.g., V1) also interact widely; domain-generality doesn't preclude specific function. Perhaps introspection is a specific mode of operation of these general metacognitive circuits (e.g., attending to internal representations generated by these areas). - ii. Quasi-Sensory Data: Introspection delivers information about mental states, analogous to how vision delivers information about external objects. - (1) Information Format: What is the nature of this introspective "data"? Propositional? Phenomenal? Conceptual? Depends on the target state? - iii. Detachment: The introspective mechanism is distinct from the states it observes (contra constitutive models). - (1) Higher-Order Representation: Implies a second-order state representing the first-order state. - (a) Regress Problem? Does the second-order state need to be introspected by a third-order state to be conscious? (Avoided by arguing the second-order state doesn't need to be conscious itself, or that it makes the first-order state conscious - Rosenthal's HOT). - b. Challenges and Criticisms: - i. Lack of Phenomenal Qualities: Introspecting a belief doesn't seem to have a distinctive sensory quality like seeing red or hearing a C#. If it's perception-like, where are the "introspective qualia"? - (1) The Core Objection: Inner sensing lacks the qualitative feel of outer sensing. (See Deep Dive Below - Section II.A.1.b.i) - (a) Response 1: Maybe there are subtle introspective qualia we overlook (e.g., cognitive phenomenology - feeling of belief, desire). Requires defending existence of cognitive qualia. - (b) Response 2: Maybe the analogy is structural/functional, not phenomenological. Introspection provides information like perception (monitoring mechanism producing representations), but doesn't need to feel like it. Analogy focuses on causal role. - (c) Response 3: Maybe the targets (e.g., beliefs) lack qualia, so the introspection of them also lacks qualia. The mechanism isn't the source of qualia, it just transmits information about the target. Introspection of qualia does have qualia (the target qualia). - ii. Target State Specificity: How does the "inner sense" differentiate between closely related states (e.g., belief vs. assumption vs. supposition)? Perception has mechanisms for feature detection; what are the introspective analogues? - (1) Introspective Feature Detectors? What features define "belief-ness" or "desire-ness" for the inner sense? - (a) Functional Role: Perhaps it detects the state's characteristic functional role (connections to inference, behavior). How is this "detection" achieved mechanistically? Requires mechanism sensitive to complex relational properties. - (b) Associated Phenomenology: Perhaps subtle feelings or cognitive phenomenology differentiate attitudes. (See Cognitive Phenomenology debate). Requires reliable link between attitude type and specific feel. - (c) Conceptual Overlay: Perhaps the differentiation relies on applying concepts, not raw perception. Blurs line with interpretive models. Inner sense provides raw data, concepts categorize it? - iii. Transparency Problem (Evans, Moran): Often, to know if you believe P, you attend outward to the world regarding P, not inward to a supposed state of "belief." (See Section IV.A.2.a.i for detailed Transparency discussion) - (1) Inner Sense Response: Maybe looking outward is the way the inner sense gathers evidence about belief states (indirectly). Seems counter-intuitive to the "inner looking" metaphor and less like perception. Requires beliefs to be tightly coupled to world-assessment mechanisms. - iv. Calibration Problem: How could such an inner sense be calibrated for accuracy? External perception is calibrated by interaction with the world (feedback, error correction). What calibrates the inner sense? - (1) Internal Coherence? Calibration via checking consistency with other internal states? (Seems circular if other states also accessed via potentially uncalibrated inner sense). - (2) Behavioral Feedback? Calibration via observing if actions based on introspected states succeed? (Indirect, noisy signal). - (3) Innate Calibration? Perhaps the mechanism is innately reliable? (Requires evolutionary justification - why would evolution select for accurate self-monitoring?). - (4) Lack of Calibration? Perhaps it isn't well-calibrated, explaining introspective fallibility. Maybe "good enough" for basic functions, but not highly accurate. ### 2. Self-Interpretation / Attributional / Theoretical Models (Ryle, Sellars, Bem, Gopnik): Introspection is not direct detection but rather a process of interpreting available cues (behavior, context, internal cues) using a folk-psychological theory to attribute mental states to oneself, much like we attribute them to others. - a. Key Features: - i. Theory-Mediated: Access is indirect, filtered through a (possibly tacit) theory of mind (folk psychology). - (1) Nature of Theory: Is it explicit propositional knowledge or implicit procedural skill? Embodied know-how? Set of causal generalizations? - (2) Acquisition: How is this self-theory acquired? (Learned concurrently with theory for others?). Innate constraints (core knowledge)? Cultural influence? - ii. Evidential Basis: Self-attributions are based on evidence, which may include internal cues but also behavior and context. - (1) Parity Assumption: The process of self-attribution is fundamentally similar to other-attribution, though the evidence base differs (privileged access to internal cues). - (a) What are "Internal Cues"? Sensory imagery? Bodily sensations (interoception)? Sub-vocalizations? Affective twinges? Cognitive phenomenology (if it exists)? Proto-thoughts? - i. Access to Cues: How are these cues themselves accessed? Does this access require an inner sense, leading back to the previous model? - A. Potential Regress: If accessing the cues for interpretation requires introspection, the model becomes complex or circular. Needs non-interpretive access to cues. - B. Non-Introspective Cue Access: Perhaps cues are accessed non-consciously or via bodily awareness (interoception), not inner sensing of mental states per se. Cues might be non-mental or proto-mental phenomena that correlate with mental states. - I. Interoception as Cue Source: Awareness of physiological state (heart rate, gut feelings) interpreted via theory. - II. Imagery/Subvocalization: Directly accessible phenomenal contents used as evidence for underlying thoughts/beliefs. - (2) Weighting of Evidence: How is evidence from different sources (internal cues, behavior, context) weighted in self-attribution? Is internal evidence always primary? - (a) Nisbett & Wilson Implications: Suggest internal cues are often ignored or misinterpreted in favor of plausible causal theories derived from context or folk psychology. ("I must like this person because I'm helping them"). - (b) Motivational Weighting: Do we weigh evidence differently based on self-serving biases? (e.g., attributing success internally, failure externally). Biased interpretation. - iii. Fallibility: Explains introspective errors easily, as interpretations can be wrong or based on faulty theories/evidence. - (1) Sources of Interpretive Error: Faulty theory (misunderstanding mental states), misinterpretation of cues (mistaking anxiety for excitement), ignoring evidence (motivated reasoning), motivational biases, conceptual limitations. - iv. Developmental Trajectory: Aligns with developmental psychology findings suggesting children learn to attribute mental states to self and others concurrently using a developing theory of mind. Self-attribution follows similar error patterns as other-attribution. - b. Challenges and Criticisms: - i. Phenomenological Inadequacy: Doesn't seem to capture the immediacy and directness often associated with introspecting simple states like pain or basic thoughts. It often feels like we just know, without interpretation. - (1) Counter-Argument: The interpretation process might be rapid, automatic, and unconscious (System 1 thinking). Expertise makes it feel direct. - (a) Analogy: Visual Perception: Visual object recognition also feels immediate but involves complex unconscious inference/interpretation based on cues and stored knowledge. - (b) Disanalogy?: We still experience visual qualia directly, which seems missing in introspecting beliefs per this model (unless qualia are also interpreted constructs). The "immediacy" of pain feels different from the "immediacy" of recognizing a cat or inferring one's belief. Pain seems less interpretative. - (2) Rejoinder: This pushes the problem back – how do we access the cues upon which the unconscious interpretation operates? Is that access direct? (See II.A.2.a.ii.(1)(a) above). If cue access is direct/acquaintance-like, the model becomes hybrid. The immediacy might come from the cue access, not the interpretation itself. - ii. First-Person Authority: Struggles to explain the apparent special epistemic status (authority, non-inferentiality) often granted to first-person reports. Why should my theory-based interpretation of myself be more authoritative than my interpretation of you? (Privileged evidence helps, but is it enough?). - (1) Role of Privileged Evidence: Access to internal cues (thoughts, imagery, sensations, interoceptive data) provides data others lack. Quantity and quality of data differs. - (a) Is this data sufficient? Can behavior or context sometimes provide stronger evidence, even to the subject? (e.g., realizing one's anger from behavior despite feeling calm; implicit bias tests revealing attitudes contradicting self-reports). - (2) Self-Fulfilling Interpretations? Perhaps self-interpretations can sometimes shape the very states they attribute (e.g., interpreting ambiguous feelings as anger might amplify anger). Does this create authority, or just complicate the picture? Might apply only to some states. - iii. Access to Qualia: How can interpreting behavioral/contextual cues give access to the qualitative character of experience (e.g., the redness of red)? Theory-theory seems better suited to propositional attitudes than phenomenal states. - (1) Qualia as Interpretive Constructs?: Dennett's view leans this way – qualia aren't intrinsic properties but part of the story we tell ourselves based on discriminatory capacities and dispositions. Highly controversial. Seems to deny the phenomenon. - (2) Qualia as Cues: Perhaps qualia are among the "internal cues" the theory interprets, but the theory doesn't explain qualia themselves or our access to them. - (a) Requires Separate Account of Qualia Access: Needs supplementation from another model (e.g., acquaintance, direct perception) for qualia. Leads to hybrid model: interpretation for attitudes, acquaintance for qualia? ### 3. Acquaintance Models (Russell, Gertler): We have direct, unmediated acquaintance with some of our mental states, particularly phenomenal experiences (qualia). - a. Key Features: - i. Directness: No mediating mechanism or interpretation; awareness is constituted by the state itself being present to consciousness. A fundamental subject-object relation. - (1) Metaphysical Relation: Acquaintance as a fundamental, irreducible relation between subject and state/property. - ii. Non-Propositional: The fundamental relation is acquaintance with the state/property, not formation of a belief about the state (though beliefs can be formed based on acquaintance). - (1) Contrast with "Knowledge By Description": Direct grasp vs. knowing via properties. Knowing redness vs knowing facts about redness. - iii. Focus on Qualia: Primarily motivated by and applied to phenomenal consciousness. - (1) Why Qualia? Their intrinsic, seemingly self-presenting nature fits the model well. They seem to be given to consciousness directly. - b. Challenges and Criticisms: - i. Scope Limitation: Seems less applicable to propositional attitudes (beliefs, desires). Are we "acquainted" with our beliefs in the same way we are with pain? - (1) Acquaintance with Attitudes?: What would it mean to be acquainted with the "belief-ness" itself? Seems phenomenologically thin or non-existent for many. Cognitive phenomenology advocates might disagree. - (2) Acquaintance with Content?: Are we acquainted with the propositional content? How? Requires account of content acquaintance, potentially linking to theories of concepts/propositions. - ii. Explaining Introspective Judgments: How does direct acquaintance lead to beliefs/judgments about the state, which are necessary for reporting or reasoning? What is the mechanism connecting non-conceptual acquaintance to conceptualization? - (1) Concept Application Problem: How does non-conceptual acquaintance ground the application of concepts (like "pain" or "red")? How does the raw feel connect to the word/concept? - (a) Bridging Mechanism Needed: Requires an account of how acquaintance interfaces with cognitive systems for judgment. Gertler suggests acquaintance grounds non-inferential justification for applying concepts. The acquaintance relation provides evidence for the concept's application. - i. Nature of Grounding: Causal? Constitutive? Does acquaintance cause the judgment, or does the judgment refer to the acquainted state? - iii. Nature of "Acquaintance": The notion of acquaintance itself can be metaphysically obscure. What is this relation? - (1) Risk of Primitivism: Defining it as simply "direct awareness" risks explaining little, positing it as a brute fact. - (2) Comparison to Perception: How does it differ from direct realist theories of perception? Is it perception of internal states/properties? If so, is it a form of inner sense after all, just characterized metaphysically differently? ### 4. Deliberative / Rational Agency / Constitutive Models (Moran): For certain states (especially attitudes like belief), introspective knowledge arises not from inner observation but from exercising rational agency. To know what you believe about P, you deliberate about whether P is true; your committed answer constitutes your belief. - a. Key Features: - i. Transparency: Explains the "transparency" phenomenon (looking outward to determine one's belief). We answer the question "Do I believe P?" by answering the question "Is P true?". (See IV.A.2.a.i for detailed Transparency discussion). - (1) Rational Constraint: This method works because beliefs rationally aim at truth. My committed stance on P's truth is my belief. Assumes rationality and commitment link. - ii. Agency and Commitment: Emphasizes the role of the subject as an agent who forms and endorses attitudes, rather than a passive observer of internal states. Self-knowledge is linked to self-constitution. - (1) Making Up One's Mind: Introspection isn't discovering a pre-existing state but actively endorsing one. Links self-knowledge to commitment and responsibility for one's attitudes. - iii. First-Person Authority: Grounded in the agent's authority to make up their mind and commit to reasons, rather than privileged epistemic access to inner facts. - (1) Authority over Content: I have authority because I decide what to believe based on reasons relevant to the truth of P. - b. Challenges and Criticisms: - i. Scope Limitation: Seems primarily applicable to attitudes amenable to rational deliberation (beliefs, intentions). Less clear how it applies to sensations, emotions, moods, or passively occurring thoughts. Can you deliberate yourself into feeling pain? - (1) Non-Deliberative States: How do we know our pains, tickles, fleeting thoughts, unwanted emotions? Requires a different model for these. Hybrid theory needed? Moran acknowledges limits; focus is on attitudes where we take responsibility. - ii. Unconscious/Recalcitrant States: Doesn't easily account for knowledge of attitudes that are unconscious, irrational, or resistant to deliberative change (e.g., implicit biases, phobias). We can know we have a belief even if we don't endorse it upon deliberation. - (1) Knowing vs. Endorsing: Moran distinguishes knowing what one does believe (potentially discovered observationally, falling outside the deliberative model's scope of agential self-knowledge) from knowing what one should believe or commits to believing (via deliberation). Does this concede ground to observational models for non-ideal cases or mere discovery? Yes, acknowledges a role for empirical self-discovery. - (2) Implicit Attitudes: How does this model handle states detectable via implicit association tests, etc., which might not be accessible via deliberation? These seem like states we discover (often observationally or via testing) rather than constitute through deliberation. Fits the non-agential category. - iii. Descriptive vs. Ascriptive: Is it describing how we do introspect, or providing a normative account of how rational self-ascription should work? Sometimes we do seem to look inward even for beliefs (e.g., "What do I really believe about this, deep down?"). - (1) Phenomenological Counter-Evidence: Sometimes self-discovery feels observational, even for beliefs, especially uncertain or conflicted ones. The "making up one's mind" phenomenology isn't always present. - (2) Normative Idealization: Perhaps transparency describes an ideal of rational agency and integrated selfhood, not always the empirical reality. Actual introspection might often be messier, involving both deliberation and observation. ### 5. Expressivist Models (Wittgensteinian influences, Bar-On): Introspective reports (avowals like "I am in pain") are not descriptions of inner states based on epistemic access, but rather sophisticated expressions or showings of those states, akin to natural expressions like wincing or crying. - a. Key Features: - i. Non-Epistemic: Avowals are not based on evidence or inner observation; their security stems from their expressive function, not epistemic privilege. They don't report knowledge, they manifest the state. - (1) Bypassing Justification: No need for inner sense, interpretation, etc., to justify the avowal. It's not a knowledge claim in the standard sense, hence doesn't need standard justification. - ii. Expressive Function: They serve to communicate/show the state directly. - (1) Learned Extension: Verbal expressions as learned extensions of natural expressive behaviors (e.g., learning to say "Ouch!" instead of just crying). Language refines expression. - iii. First-Person Authority: Explains authority by denying the standard epistemic model; the expression is authoritative because it's a direct manifestation, not because it's a well-justified report. Authority of source, not report. - (1) Security, Not Infallibility: Expressions can be suppressed or misleading, but when sincere, they directly show the state. The authority lies in the direct link between state and expression, not impossibility of error or deception. - b. Challenges and Criticisms: - i. Cognitive Sophistication: Seems more plausible for simple states (pain) than complex ones (believing that economic sanctions are counterproductive). Avowals of complex thoughts seem more descriptive than purely expressive. - (1) Expressing Complex States: How can a belief with complex propositional content be expressed rather than reported? What natural behavior does "I believe P" extend? Perhaps assertion itself? - (a) Neo-Expressivism (Bar-On): Argues avowals express the state while also being truth-apt judgments, attempting to bridge the gap. They have a dual function: expressive (explaining security) and descriptive/reportive (explaining truth-aptness and role in reasoning). Preserves security while allowing for judgment/belief aspect. - ii. Introspective Beliefs vs. Reports: While it might explain verbal reports, what about the underlying introspective awareness or belief one has before speaking? Does expressivism deny such awareness exists, or just redescribe its nature? - (1) Denies Intermediary Belief?: Some versions might deny a separate introspective belief/awareness preceding the avowal. The state itself + expressive disposition is sufficient. No cognitive intermediary needed. - (2) Reinterprets Awareness?: Perhaps the "awareness" is the disposition to avow/express sincerely. Having the awareness is being poised to express. - iii. Possibility of Error/Insincerity: How does expressivism account for insincere avowals or cases where someone seems genuinely mistaken about their own state (if avowals are just expressions)? - (1) Insincerity: Explained as misuse of the expressive behavior (like feigning a wince). Not a genuine expression, breaking the direct link. - (2) Error: Harder for pure expressivism. If it's just expression, error seems impossible (unless one expresses the 'wrong' state?). Neo-expressivism allows for error as the judgment aspect can be mistaken (e.g., mischaracterizing the expressed state due to conceptual confusion) even if the expression is linked to some state. # III. Epistemology of Introspection: Justification, Reliability, and Self-Knowledge ## A. Epistemic Status of Introspective Beliefs: ### 1. Infallibility Thesis: The claim that introspection cannot be mistaken. If one introspectively believes one is in mental state M, then one is in mental state M. - a. Strong Infallibility: Introspection is immune from any error regarding the presence, nature, and properties of the mental state. (Historically attributed to Descartes, though debated). - i. Arguments For: Conceptual arguments (e.g., being in pain is feeling pain; the awareness is constitutive); arguments from immediacy/directness. - (1) Constitutive Argument Detail: If awareness is part of the state (e.g., self-representationalism), error seems impossible. The state wouldn't be that state without the awareness. To believe one is in state M is (part of) being in state M. - ii. Arguments Against: Empirical evidence of introspective errors (e.g., Nisbett & Wilson studies on misattribution); philosophical arguments about misidentification or mischaracterization (e.g., confusing jealousy with indignation); possibility of conceptual error (misapplying a mental state concept). - (1) Misidentification: Mistaking state A for state B (e.g., anxiety for excitement). Requires states have sufficiently similar phenomenal or functional profiles to allow confusion. - (2) Mischaracterization: Correctly identifying state A but misjudging its properties (e.g., intensity, cause, object). Believing pain is in left foot when it's referred pain from back. Judging mild annoyance as rage. - (3) Conceptual Error: Lacking or misapplying the correct concept (e.g., not knowing what "ennui" is, mislabeling schadenfreude as simple malice). Introspection might be accurate at pre-conceptual level, but conceptual report is wrong. - b. Weak Infallibility: Restricted claims, e.g., infallibility only about the existence of a conscious state, not its precise properties, or only for phenomenal states, not attitudes. - i. Example: One cannot be mistaken that one is experiencing phenomenal consciousness (e.g., feeling something), even if one misidentifies what it is. (Chalmers' "directly acquainted with consciousness itself"). Or infallibility about the phenomenal character itself, if not its label or properties beyond the core feel. - ii. Plausibility: More plausible than strong infallibility, but still faces challenges regarding the precise boundaries of states and the role of concepts in identifying them. - (1) Boundary Problem: When does one quale shade into another? Can we be infallible about vague boundaries (e.g., warmth turning into pain)? If states are vague, infallible access seems problematic. - (2) Concept Requirement: Does even existential infallibility ("I feel something") require some minimal conceptual grasp ("something is happening," "I feel")? Or is there purely non-conceptual infallible awareness? ### 2. Indubitability Thesis: The claim that introspective beliefs cannot be rationally doubted. Even if error is metaphysically possible, the subject has grounds for certainty that preclude rational doubt. - a. Relationship to Infallibility: Often linked, but distinct. One might argue that even if fallible, the nature of introspective access provides a unique form of certainty. A belief can be fallible but indubitable for the subject. - i. Basis for Certainty: Immediacy? Self-presentation? Lack of conceivable counter-evidence from the first-person perspective? Constitutive link? Psychological certainty vs. epistemic certainty. - b. Challenges: Can one not rationally doubt an introspective belief if confronted with strong contradictory evidence (e.g., neurological evidence showing no pain signals, behavioral patterns suggesting denial)? What if one suspects a cognitive bias is influencing the introspective judgment? Rationality seems to demand considering all evidence. ### 3. Self-Intimation ("Omniscience") Thesis: The claim that if one is in mental state M, one necessarily knows (or is in a position to know) introspectively that one is in M. Mental states automatically reveal themselves to introspection. - a. Strong Self-Intimation: Applies to all mental states. - i. Arguments Against: Existence of unconscious mental states (Freudian unconscious, cognitive unconscious); states one fails to attend to; states one lacks concepts for. Clearly false. - (1) Unconscious States: Well-established in cognitive science (e.g., implicit biases, automatic processing). - (2) Attentional Requirement: Many states seem to require attention to become introspectively known (e.g., background mood, peripheral sensations). They don't force themselves upon us. - b. Weak Self-Intimation: Restricted to conscious states, especially phenomenal states. If one is consciously feeling pain, one knows one is feeling pain. - i. Plausibility: Seems highly plausible for phenomenal states – what would it be to be consciously in pain but not know it? Seems contradictory for phenomenal consciousness. - (1) Constitutive Link?: Perhaps being conscious of pain is part of being consciously in pain. The consciousness is the knowing (at least implicitly). (Higher-order or self-representational views might support this). The state includes its own appearance. - ii. Challenges: Fringe cases? States of marginal consciousness? Attentional requirements (does the state reveal itself even if unattended?). What about moods that color experience without being explicitly noticed? Can one be consciously sad without knowing one is sad until attention is drawn to it? - (1) Marginal Consciousness: States barely above the threshold of awareness. Are we omniscient about these fleeting, barely-there states? - (2) Unattended Qualia: E.g., the feeling of clothes on skin, present but unnoticed until attended. Was it phenomenally conscious before attention? - (a) If Yes: Weak self-intimation fails (conscious but not known). Implies overflow (more phenomenal consciousness than accessible/reported consciousness). - (b) If No: Then consciousness requires attention (or access consciousness). Self-intimation might hold for access conscious states. (Block's distinction: Phenomenal vs Access consciousness). ### 4. Incorrigibility Thesis: The claim that introspective beliefs cannot be corrected by others; the first-person perspective has final authority. - a. Distinction from Infallibility: One might be mistaken (fallible) but still immune from correction by others (incorrigible) if others lack any epistemic standing to challenge the first-person report. Authority vs accuracy. - i. Basis of Incorrigibility: Privacy? Unique access? Non-inferential nature? Expressive function (if avowals aren't reports subject to correction)? - b. Challenges: Therapy often involves challenging a person's self-understanding. Behavioral and neurological evidence can conflict with self-reports. Is incorrigibility a plausible claim? Perhaps only for certain aspects of certain states (e.g., the raw feel of pain - "It feels like this")? - i. Domain Restriction: Maybe incorrigible only for occurrent qualia reports ("It feels like this to me now"), not attitudes or causes or characterizations using public concepts (which others can assess consistency of use). ### 5. Fallibilism (Dominant Contemporary View): Introspection is fallible, dubitable, not self-intimating (except perhaps for conscious qualia), and corrigible. It is a source of knowledge, but imperfect and subject to various forms of error. - a. Sources of Error (Revisited): Misidentification, mischaracterization, conceptual inadequacy, attentional lapses, bias (confirmation, hindsight, self-serving), confabulation, theoretical overlay, misattribution of causes/objects (e.g., emotion attribution). Memory distortion even for immediate retrospection. Expectation effects. - b. Reliability: If fallible, how reliable is it? Is it reliable enough to ground self-knowledge or be used as scientific data (see V.A)? - i. Domain Specificity: Reliability might vary significantly across different types of mental states (e.g., more reliable for current sensations than for complex motivations or character traits). Higher for qualia presence than attitude content? Higher for occurrent thoughts than standing beliefs? Higher for focus of attention vs periphery? Higher for valence than specific emotion type? - ii. Methodological Concerns: How can reliability be assessed without circularity (i.e., without relying on introspection to check introspection)? Use of converging evidence (behavior, neural correlates, predictive success, triangulation with other reports). - (1) Converging Evidence Issues: Behavior can be misleading (acting); neural correlates are often correlational, not definitive; prediction based on reports tests action-link, not report accuracy itself; triangulation assumes others' reports are reliable. No independent gold standard. ## B. Introspection and Justification: How do introspective beliefs acquire epistemic justification? ### 1. Foundationalism: Introspective beliefs (especially about current conscious states) are non-inferentially justified, serving as a foundation for other knowledge (particularly self-knowledge). - a. Basis of Justification: Direct awareness/acquaintance; constitution (the awareness is the state); reliability of the introspective mechanism (if reliable enough to confer non-inferential justification). - i. Which Basis?: Different models of introspection suggest different bases for foundational justification. Acquaintance model fits naturally. Inner sense could if reliable. Constitutive models build justification into the state itself. Self-presenting states. ### 2. Coherentism: Introspective beliefs are justified by their coherence with other beliefs (including other introspective beliefs, beliefs about behavior, context, etc.). - a. Challenge to Immediacy: Conflicts with the phenomenology of introspection often feeling direct and non-inferential. - i. Coherentist Response: Phenomenology might be misleading; justification structure is distinct from subjective feeling. Justification emerges from the web of beliefs, even if the belief feels immediate. - b. Explaining Error: Coherence can shift, allowing for the revision of introspective beliefs deemed incoherent with the larger set. Accommodates fallibilism well. Justification is provisional. ### 3. Reliabilism: Introspective beliefs are justified if produced by a reliable belief-forming process (whether foundational or not). - a. Empirical Question: Justification depends on the actual reliability of the underlying introspective mechanism(s), which is an empirical matter. How reliable does it need to be? - b. Problem of Access: Requires knowing that the process is reliable, which might itself involve introspection (potential circularity). Or requires externalist justification (belief is justified if reliable, whether subject knows it or not). # IV. The Objects of Introspection: What Can We Introspect? ## A. Mental States: ### 1. Phenomenal States (Qualia): Sensations (pain, tickles), perceptual experiences (seeing red), moods, emotions (the feeling aspect). - a. Introspective Character: Often considered the most amenable to direct introspection, possibly via acquaintance. Focus on the qualitative "what it's like." - b. Debates: Existence/nature of qualia; possibility of absent/inverted qualia; completeness of introspective access (can we fully grasp the essence of "redness"? Can we introspect all aspects of a complex quale?). - i. Introspective Completeness: Can we attend to all features of our visual field simultaneously? Seems unlikely. Introspection might be selective even for qualia. Attention limits apply internally too. Granularity limits? - ii. Introspection and the Structure of Qualia: Can introspection reveal the intrinsic nature or structure of qualia (e.g., are color qualia simple or complex? structured?)? Or does it only reveal their presence and comparative relations (e.g., this red is brighter than that red)? ### 2. Propositional Attitudes: Beliefs, desires, intentions, hopes, fears (states with propositional content). - a. Introspective Character: Access seems different from phenomenal states. Do we introspect the content, the attitude, or both? - i. Transparency Phenomenon (Revisited - EXTREME DEEP DIVE) : Often, to know if one believes P, one considers whether P is true (world-directed), not an internal state. (Evans, Gallois, Moran). - (1) Scope: Does transparency apply equally to all attitudes (e.g., desires, intentions)? - (a) Attitude Differences: Arguably less transparent for desires (attend to feelings/needs?), intentions (attend to commitment/plan?), emotions (attend to feelings/appraisals?). Transparency seems strongest for belief. - i. Intentions Example: To know if I intend to go, I might check my plan/commitment (less transparent). To know what I intend, I focus on the intended action/state-of-affairs (more transparent regarding content). Difference between attitude presence and attitude content. - ii. Desire Example: "Do I want X?" Might involve checking feelings towards X (less transparent) or assessing X's properties (more transparent about why I might want it, perhaps indirectly revealing desire). - iii. Emotion Example: "Am I angry?" Often involves assessing feelings and the situation ("Does this situation warrant anger?"). Mixture of opaque and transparent elements. - (2) Limits of Transparency: When does it fail or become opaque (requiring attention to the mental state itself)? - (a) Irrationality/Uncertainty: When beliefs are irrational, uncertain, conflicted ("Do I really believe this conspiracy theory I find compelling?"). Requires looking inward at the state's properties (strength, origin, coherence, affective pull). - (b) Second-Order Questions: When considering the source, strength, justification, or phenomenology of the belief, rather than just its content. E.g., "Why do I believe this?" or "How confident am I?" requires attending to the state, not just the world. Metacognitive judgments are inherently opaque. - (c) Cognitive Phenomenology Link: If attitudes have a distinctive phenomenal character ("cognitive qualia" - the feeling of believing, desiring, judging), then introspection could target this feeling, making the state opaque (accessible as a state). (Pitt, Strawson). - i. Nature of Cognitive Qualia: Is it proprietary (unique to the attitude type, e.g., a "belief-feel") or common (general feeling of conviction, desire, effort)? - A. Proprietary View: Specific feels for believing, desiring, hoping etc. Makes attitudes potentially opaque objects. - B. Common View: General feels (conviction, effort) accompany different attitudes. Less support for full opacity of attitude type. - ii. Introspectability: If it exists, is it easily introspectible or subtle/elusive? Requires training? Easily confused? - (d) Emotional/Affective Attitudes: Attitudes like hopes, fears seem inherently less transparent. Knowing if I hope for P requires attending to feelings/dispositions, not just whether P is likely/desirable. The affective component demands opacity. - (e) Comparative Judgments: Comparing one belief to another ("Is my belief in X stronger than my belief in Y?") requires attending to properties of the beliefs. - (f) Pathological Cases: Delusions, thought insertion - transparency seems to break down dramatically. Subject may query the source or nature of the thought, not just its content's truth. - (3) Transparency and Opacity Debate: Is experience fundamentally transparent or can it be opaque? - (a) Strong Transparency Thesis (Tye, Harman, early Dretske): Phenomenal character is exhausted by the represented properties of external objects (or bodily properties for interoception). Introspection "looks through" the experience to the world (or body). We are only aware of what is represented (content), not the intrinsic qualities of the representing experience itself (vehicle). - i. Analogy: Looking through a clean window pane reveals the scene outside, not the glass itself. The properties experienced (color, shape) are attributed to the object, not the experience. - ii. Implication: There are no qualia intrinsic to the experience (vehicle qualia), only represented properties (content qualia, which are properties of the object). Introspection of experience reveals only the world (or apparent world). Pain represents bodily damage. Redness represents surface reflectance. - iii. Motivation: Physicalist compatibility. Avoids positing intrinsic, non-physical qualia. Explains focus on world in perception. - (b) Challenges to Strong Transparency: Evidence suggesting awareness of experiential features (vehicle awareness). - i. Blurred Vision / Phosphenes / Afterimages: When perception malfunctions (blurriness, double vision, phosphenes, afterimages), we seem aware of features of the experience itself, not just misrepresented external objects. Where is the blurriness located if not in the experience? Seems intrinsic to the visual field. - A. Transparency Response: These are still represented properties, just unusual ones (e.g., representing indeterminacy, or properties of the medium being represented as if external, or representing states of the sensory organ itself). Content, not vehicle, is accessed. Blur represents lack of determinate spatial properties. Afterimage represents adaptation state of retina projected onto world. - I. Counter-Response: This seems strained. The phenomenology feels like a property of the seeing, not a property of the seen. Explaining away the phenomenology? - ii. Attention to Experience: Can't we deliberately attend to the qualitative character of experience itself? E.g., attending to the specific hue of red, its saturation, its grain, not just the object being red. Bracketing the object. - A. Transparency Response: This is just attending more closely to the represented properties of the object, perhaps very fine-grained ones. Still world-directed attention. - I. Counter-Response: Phenomenologically distinct? Attending to how the object is presented vs what is presented. Attending to paint vs attending to painted object. - iii. Moods / Bodily Sensations: These seem less world-directed. Anxiety or pain seem to be properties of our conscious state, not properties of external objects represented by that state. - A. Transparency Response: Bodily sensations represent properties of the body (interoceptive content). Moods represent complex relational properties or global states of the organism/world interaction (e.g., representing the world as threatening/draining). Still representational content. - I. Counter-Response: Does the representational content exhaust the phenomenal feel? The feeling of dread seems distinct from representing the world as dangerous. Is the quale identical to the content? - iv. Cognitive Phenomenology: If thoughts have feel, this feel doesn't seem to represent external properties. Seems like an intrinsic feature of the thought process/state. What does the "feeling of knowing" represent externally? - A. Transparency Response: Cognitive phenomenology represents properties related to cognitive processing (fluency, confidence, coherence). Still representational content about internal cognitive states/processes. - v. Structural Features / Vehicle Awareness (Soteriou, Richardson): Perhaps we can become aware not just of the content but the structural features or vehicle of experience itself. Opacity is possible through attending to the medium. - A. Soteriou's Focus: Temporal structure of experience. Argues we can be aware of the passage or flow of experience itself, not just the changing properties of objects represented in time. Awareness of experiential succession. - I. Introspecting Temporal Flow: Is the feeling of time passing an awareness of change in the world or awareness of the experience unfolding? Soteriou argues for the latter being possible via attending to succession within consciousness. - α. Example: Attending to the sequence of notes in a melody as experienced, not just the notes themselves. - II. Opacity Implication: If we can attend to the temporal structure of experience, then experience is not fully transparent; some of its own features (its temporal properties) are accessible. This is awareness of the vehicle. - α. Relation to Specious Present: This might be awareness of the specious present structure itself. - B. Richardson's Focus: Awareness of the sensory field/medium itself. Can we be aware of, say, the visual field as a field (its boundaries, its spatiality, its 'presence'), distinct from the objects within it? - I. Field Awareness vs Object Awareness: Attending to the boundaries, grain, or presence of the visual field itself. E.g., noticing the edge of vision, or the "screen" on which images appear (phenomenologically). - α. Analogy: Awareness of the cinema screen itself, not just the movie playing on it. - II. Opacity Implication: Awareness of the field is awareness of a feature of the experience (its structural framework/medium), making it opaque. Vehicle awareness. - C. Relation to Standard Transparency: These views suggest transparency might be the default mode for object-focused perception, but opacity (awareness of experiential features/vehicle) is possible, perhaps requiring specific forms of attention (reflective, structural attention). Ordinary introspection might be transparent; philosophical/meditative introspection might achieve opacity. - (c) Meditation and Transparency/Opacity: Recent studies and philosophical arguments suggest meditative practices might cultivate a form of awareness directed at the experience itself, challenging strong transparency. - i. Phenomenological Claims: Experienced meditators report capacity to observe thoughts/sensations as events in consciousness, distinct from their content or external referent. Awareness of impermanence relates to observing the arising/passing of experience. Seeing thoughts as "just thoughts". Awareness of awareness itself. - A. Type of Awareness Cultivated: Mindfulness (bare attention), concentration (focused attention), insight (vipassanā), non-dual awareness. Different practices might target different aspects. - B. Target of Awareness: Is it the content (observed non-reactively) or the vehicle/structure (observing the arising/passing of the state itself)? Or awareness itself? - ii. Scientific Studies (e.g., Dor-Ziderman et al. on 'Awareness of Awareness'): Some research attempts to correlate specific meditative states (like non-dual awareness) with neural signatures (e.g., gamma synchrony, default mode network changes), suggesting a distinct mode of awareness potentially focused on consciousness itself rather than objects. - A. Methodological Hurdles: Correlating subjective reports with neural data is complex. Defining and measuring "awareness of awareness" operationally. Risk of conceptual confusion between subject reports and scientific interpretation. Need for validated first-person methods (e.g., micro-phenomenology). - I. Subject Training Effects: Does training create genuine insight or impose theoretical framework on experience? - II. Neural Correlates: Are correlates specific to the claimed state (e.g., awareness of awareness) or reflect general relaxation/attention effects? - B. Plausibility: Does current neuroscience support the possibility of brain attending to its own global states or structural features in a way corresponding to these reports? Some models (e.g., Global Workspace Theory variants) might allow this. - iii. Philosophical Interpretations: - A. Refutation of Transparency?: If meditators genuinely access features of the experience vehicle (e.g., its spatiality, temporality, or 'consciousness-ness', its arising/passing), then strong transparency is false. (Kind, Bayne argue this direction). Experience has accessible intrinsic features. - I. Implication: Supports views where qualia are intrinsic properties of experience (vehicle qualia exist). - B. Alternative Mode of Awareness?: Perhaps meditation cultivates a different type of attention or awareness (meta-awareness, non-dual awareness) that can target experience, while ordinary awareness remains transparent. Transparency holds for standard introspection, fails for specialized forms. - I. Standard vs Expert Introspection: Analogy to expert perception (wine tasters detecting notes novices miss). Introspective skill development. - C. Reinterpretation within Transparency?: Could meditative insights be re-described as awareness of very subtle contents (e.g., representing the boundaries of the sensory field, representing the rapid change of states)? Defending transparency by re-describing the content as highly refined or unusual representational content. - I. Example: "Awareness of thought as thought" is representing the thought as having the property of being a thought (representational content), not accessing the vehicle. "Awareness of impermanence" is representing the high rate of change of represented properties. - II. Plausibility: Can all meditative reports be plausibly re-described this way without distortion? Seems challenging for reports of "pure awareness". - D. Skill/Expertise Model: Perhaps introspection of experience itself is a skill that can be developed, making opacity achievable but not typical. Ordinary introspection is transparent; expert introspection can be opaque. Links to B. - E. Conceptual Refinement?: Maybe meditation leads to developing new concepts for describing experience, allowing finer-grained reports about subtle contents, without necessarily changing the fundamental nature of introspective access (still transparent access, just better described). Learning a new vocabulary for experience. - ii. Accessing Attitude vs. Content: Can one introspect the "belief-ness" of a belief, distinct from its content? Or is the attitude only known via its functional role or associated feelings? - (1) Functional Clues: Knowing it's a belief because it guides inference and assertion in a belief-like way. Requires meta-cognition to monitor functional role? - (2) Affective Clues: Feelings of conviction, doubt, uncertainty associated with the content. Are these part of the attitude or consequences? - (3) Cognitive Phenomenological Clues: A specific (perhaps non-sensory) "feel" of believing, judging, desiring. (See IV.A.2.a.i.(2)(c) above). Makes the attitude itself potentially opaque. - b. Standing vs. Occurrent Attitudes: Can we introspect standing beliefs/desires (dispositions not currently active in consciousness)? Or only occurrent thoughts/feelings related to them? - i. Inferential Access: Knowledge of standing states might typically be inferential (based on past occurrent states, behavior) rather than directly introspective. E.g., "I guess I believe X, because I consistently argue for it." Links to theory-theory/interpretation. Self-knowledge via self-interpretation. - ii. Dispositional Introspection?: Can one directly "sense" the disposition itself? Seems unlikely for most theories, but perhaps a felt sense of inclination or readiness? Subtle occurrent manifestation of the disposition? ### 3. Emotions: Complex states involving phenomenal feelings, physiological changes, cognitive appraisals, and behavioral dispositions. - a. Introspecting Components: Can we introspect the feeling aspect (valence, arousal)? The cognitive appraisal ("This is dangerous")? The physiological change (heart racing)? The disposition (urge to flee)? All simultaneously or selectively? - i. Fragmented Access: Introspection might only grasp certain components easily (e.g., feelings, perhaps appraisal) while others require inference or interoception (e.g., precise physiological state, underlying disposition). Access depends on component type. - b. Misattribution: Prone to misattribution errors (e.g., Schachter-Singer experiment suggesting undifferentiated arousal is interpreted based on context). Challenges the reliability of introspecting emotions accurately. - i. Implications: Suggests introspection of emotion involves significant interpretation, fitting attributional models. We introspect arousal + context, then apply label (theory-laden). ### 4. Intentions: States related to planning and initiating action. - a. Introspective Character: Awareness of deciding, forming an intention, initiating action (cf. Libet experiments and debates about conscious will). - i. Libet Findings: Readiness potential precedes conscious awareness of intention to act. Challenges simple view of conscious intention initiating action. - (1) Interpretations: Veto power (consciousness monitors/stops actions)? Illusion of conscious will (Wegner - intention inferred post-hoc)? Flawed experiment (measurement issues)? Consciousness arriving late but still relevant for learning/adjustment? Intention is complex process, Libet captures only late stage awareness of urge? - b. Prior Intentions vs. Intentions-in-Action: Can we introspect both long-range plans (prior intention: "I intend to go to Rome next week") and the immediate intention guiding current movement (intention-in-action: the intention controlling finger movement now)? Access might differ. Prior intentions more like belief introspection (commitment); intentions-in-action perhaps linked to phenomenal feel of agency. ## B. Mental Processes: Reasoning, judging, deciding, remembering, attending. ### 1. Introspective Access: Can we directly introspect the process itself, or only its inputs, outputs, and perhaps some phenomenal correlates (e.g., feelings of effort, fluency, uncertainty)? - a. Nisbett & Wilson Critique: Argued that we often lack introspective access to the causes and processes of our judgments and behaviors (higher-order cognitive processes), instead confabulating plausible reasons based on implicit theories. We access the result, then invent the process story. Process opacity. - i. Confabulation Evidence: Studies on choice blindness, split-brain patients, justification of hypnotically induced actions, misattribution of causes for preferences. People readily provide causal explanations that are demonstrably false. - b. Process vs. Product: We might have good access to the products of processing (e.g., the resulting belief, the retrieved memory) but poor access to the process itself (the deductive steps, the memory search algorithm). - i. Analogy: We see the output on the computer screen, not the CPU operations or specific lines of code executing. - ii. Phenomenal Correlates as Process Clues?: Do feelings of fluency, effort, familiarity, "tip-of-the-tongue" provide indirect introspection of processing? (Metacognitive feelings). These feelings might track process features without revealing the process steps themselves. ## C. The Self: Can we introspect the subject of experience, the "I"? ### 1. Humean Skepticism: Hume famously reported finding only a bundle of perceptions, no distinct impression of a unified self when looking inward. - a. Interpretation: Was Hume looking for the wrong thing? Expected a Cartesian ego (a simple, enduring substance)? Or correctly reported the lack of a substantial self within experience? - i. Looking for Object vs. Subject: Was he looking for the self as an object of introspection, rather than considering the subjective pole of experience itself? The self that experiences vs a self that is experienced. - ii. Bundle Theory Consequence: If correct, the "self" is a construct (perhaps narrative or grammatical), not a directly introspected entity. Self-knowledge is knowledge of the bundle/pattern. ### 2. Kantian Transcendental Ego: The self as a necessary condition for the possibility of experience (unity of apperception), but not directly accessible as an object within experience. Known through its function (unifying experience), not introspection. Cannot be an object for itself. ### 3. Phenomenological Views: Some traditions describe an awareness of self-presence or "mineness" accompanying experience. Is this introspective access to the self? - a. Pre-reflective Self-Awareness: A primitive, non-observational, non-introspective awareness of oneself as the subject of experience, inherent in the experience itself (Zahavi, Husserl, Sartre). Distinct from deliberate introspection of the self as an object. - i. Nature: Tacit, implicit, structural feature of consciousness. The "for-me-ness" of experience. Not something we find by looking, but the way experiences are had. Pervasive but non-objectifying. - ii. Relation to Introspection: Introspection (reflective self-awareness) might take pre-reflective awareness as its (implicit) object, but they are distinct levels. Reflection objectifies the pre-reflective subjectivity. # V. Methodological Role and Limitations ## A. Introspection in Science (Psychology): ### 1. Early Psychology (Wundt, Titchener): Structuralism relied heavily on trained introspection to analyze conscious experience into basic elements (sensations, feelings, images). - a. Criticisms: Subjectivity (private data), lack of replicability (different labs, different results), disputes between labs (e.g., imageless thought debate), restriction to conscious processes (ignoring unconscious), possibility of training influencing reports (demand characteristics, imposing theory). ### 2. Behaviorist Rejection: Watson and Skinner rejected introspection as unscientific (private, unverifiable), focusing solely on observable behavior. Shift away from consciousness as topic. ### 3. Cognitive Revolution: Re-emergence of interest in internal mental states, but often using indirect methods (reaction times, error patterns, computational modeling) rather than relying solely on first-person reports. Mental states inferred, not directly observed via introspection. ### 4. Contemporary Use: Introspective reports (verbal protocols, questionnaires, experience sampling) are used, but often cautiously, triangulated with behavioral and neural data, and acknowledging potential biases and limitations. - a. Think-Aloud Protocols: Used to study problem-solving processes. Assume report reflects concurrent thought, but verbalization might alter thought (reactivity). Might reflect rationalization rather than process. - b. Experience Sampling Methods (ESM): Prompting subjects to report current states/experiences at random intervals in natural settings. Aims for ecological validity. Still relies on self-report accuracy at the moment, subject to biases and conceptual limits. - c. Neurophenomenology: Attempts to combine detailed first-person reports (using trained subjects, often from meditative traditions) with simultaneous neuroimaging (Varela). Aims to bridge subjective-objective gap by finding mutual constraints. - i. Methodological Challenges: Training effects (does training improve accuracy or instill theory?), conceptual alignment between subjective reports (often using folk or specialized vocabulary) and neural data (mapping complex states to brain activity), interpreting correlations (causality unclear). Need for sophisticated interview techniques (e.g., micro-phenomenology) to elicit detailed, less theory-laden reports. Difficulty validating first-person data. ## B. Introspection in Philosophy: ### 1. Foundational Role: Used by Descartes (Cogito), Locke (ideas), Hume (impressions) as a primary source of evidence for theories of mind and knowledge. Method of intuition. ### 2. Armchair Philosophy: Continued reliance on introspectively derived intuitions and thought experiments, particularly in philosophy of mind and epistemology. Testing conceptual possibilities via introspection. ### 3. Critiques: Concerns about reliability, potential for bias, limitations exposed by empirical psychology. Can armchair introspection provide reliable data about the mind, or just about folk-psychological concepts/intuitions? (Experimental Philosophy - X-Phi - challenges intuitions by testing them empirically across demographics, finding variation). Is philosophical introspection just exploring culturally specific folk theories? ## C. Limitations and Pathologies: ### 1. Cognitive Biases: Systematic errors affecting self-assessment (e.g., confirmation bias, hindsight bias applied to one's own reasoning, Dunning-Kruger effect, self-serving bias, projection bias). Introspection inherits biases of general cognition. ### 2. Confabulation: Generating plausible but inaccurate explanations for one's own behavior or choices, often without awareness of the inaccuracy (common in everyday life, not just pathology - Nisbett & Wilson). Introspection accesses result, rationalizes cause. ### 3. Introspective Illusions: Misjudgments about one's own states (e.g., illusion of conscious will (Libet, Wegner), misattribution of emotion, intensity biases, duration neglect). Systematic errors in judging properties of one's own experience. ### 4. Limits of Attention/Awareness: We cannot introspect states we are not attending to or that are unconscious. Richness of peripheral awareness debate (do we have rich unattended qualia - supporting overflow, or is periphery sparse - supporting attention gating consciousness?). Introspection depends on attentional allocation. ### 5. Conceptual Limits: Introspection is constrained by the concepts available to the introspector. Can't report a state one lacks a concept for, or might misreport using wrong concept. Finer concepts might allow finer "introspective" distinctions, but may also impose theory. Language shapes report. ### 6. Pathologies: Conditions affecting introspective capacity (e.g., anosognosia - lack of awareness of deficit; certain forms of schizophrenia affecting self-monitoring/agency, thought insertion; alexithymia - difficulty identifying and describing emotions; some aspects of autism spectrum disorder related to identifying own internal states). Provide insight into mechanisms by seeing how they break down. # VI. Metaphysical Implications ## A. Consciousness: Introspection seems intimately linked to conscious awareness. Is introspection required for consciousness (higher-order theories), or vice versa (first-order theories)? Does introspection grant access to phenomenal consciousness (qualia)? ### 1. Higher-Order Theories (HOT): Consciousness arises when mental states are targeted by higher-order states (thoughts or perceptions/introspection). Implies introspection (or a HO state) is key to consciousness. Consciousness is introspectedness (or potential introspectedness). *   a. HOP (Higher-Order Perception): Consciousness via inner sense scanning first-order states. Armstrong, Lycan. *   b. HORT (Higher-Order Thought): Consciousness via non-inferential thought about first-order states. Rosenthal. HO thought need not be conscious itself. *   c. Challenges: Problem of the phenomenal feel (how does HO state confer qualia? seems mysterious); rock problem (why doesn't HO thought about rock make it conscious?); targetless HO states (HO thought without F-O state - misrepresentation). ### 2. First-Order Theories (FOT): Consciousness is an intrinsic property of certain first-order states (e.g., due to their representational content or functional role). Introspection is a separate process that accesses already conscious states. *   a. Examples: Biological Naturalism (Searle - specific brain powers), Representationalism (Tye, Dretske - specific representational content), Integrated Information Theory (Tononi - causal structure), Self-Representationalism (Kriegel - see below). *   b. Challenges: Explaining the link between the first-order property (representational, functional, biological) and subjectivity (the Hard Problem). Why *these* properties generate qualia. ### 3. Self-Representational Theories (SRT): Consciousness arises from states representing themselves (Kriegel, Williford). Blurs first-order/higher-order distinction. State has F-O content and HO content (representing itself). Intrinsic awareness. Introspection might be a further layer or mode of accessing these already self-aware states more explicitly. ## B. First-Person Perspective: Introspection is the paradigm of the first-person perspective. Understanding introspection is key to understanding subjectivity and the difference between first-person and third-person knowledge of minds. ### 1. Reducibility: Can the knowledge gained via introspection be fully reduced to or explained by third-person scientific knowledge? (Linked to qualia debate and the explanatory gap). *   a. If introspection reveals irreducible qualia or subjectivity, then reduction fails. Explanatory gap remains. *   b. If introspective knowledge is ultimately about functional roles or representational content describable physically (and qualia are identical to these), reduction may succeed in principle. Explanatory gap might be dissolved or shown to be epistemic limitation. ## C. Mind-Body Problem: The nature of introspective access (especially to qualia) informs debates about dualism vs. materialism. If introspection reveals irreducible subjective qualities, does this support dualism or non-reductive physicalism? If introspection is merely self-interpretation based on physical cues, does this support reductive physicalism? ### 1. Qualia Realism: If introspection reveals genuine, irreducible qualia, poses challenge for reductive physicalism. Supports property dualism (Chalmers) or non-reductive physicalism (some forms) or perhaps even substance dualism (historical). ### 2. Qualia Eliminativism/Illusionism: If introspection is unreliable regarding qualia (Dennett - qualia are theoretical posits or part of a self-narrative; Frankish - illusionism), removes this challenge. Supports reductive physicalism or functionalism. Introspection tells us about our beliefs or judgments about experience (which are physical/functional), not about experience's supposed intrinsic non-physical properties. The target of introspection is reconceived. --- ## Deep Dive Reference: Section II.A.1.b.i. (Extended from original position for clarity) ### II.A.1.b.i. Deep Dive: Lack of Phenomenal Qualities (Revisited & Extended) - i. Lack of Phenomenal Qualities: Introspection seems devoid of distinctive sensory qualia. - (1) The Core Objection: Inner sensing lacks the qualitative feel of outer sensing. - (a) External Perception's Benchmark: Outer senses possess specific qualitative characters. - i. Visual Qualia Example: Seeing involves subjective qualities like color and brightness. - A. Nature of Visual Qualia: These qualities seem intrinsic to the visual experience itself. - I. Subjectivity: The "redness" of red is a subjective feeling. - α. Apparent Irreducibility: This feeling seems resistant to purely objective, third-person physical or functional description. - • Knowledge Argument Relevance: Could one know all physical facts about color vision but not know "what it's like" to see red? (Chalmers, Jackson). Challenges physicalism if knowledge is gained. - · Jackson's Mary: This thought experiment probes this very possibility. Assumes Mary learns all physical facts from black-and-white room. - o If Mary learns something new upon seeing red, qualia seem non-physical (or represent non-physical facts, or provide a new mode of access to physical facts). - ** Property Dualism suggested. Qualia are fundamental non-physical properties of brain states. - † Physicalist Responses: Argue Mary gains know-how, or new concepts/acquaintance related to existing physical knowledge, not new factual knowledge about non-physical properties. The knowledge gained is non-propositional, or knowledge of old facts under new modes of presentation. - § Ability Hypothesis (Lewis/Nemirow): Mary gains abilities (to imagine, recognize, remember red experiences), not propositional knowledge that P. The learning is practical know-how, not theoretical know-that. - % Does this adequately capture the apparent epistemic gain? The feeling seems like learning a fact about reality (what red is like), not just gaining a skill. Seems like she learns that red feels like this. - = Critics argue the phenomenal feel is informational, providing factual knowledge about the nature of experience or the world. The ability is grounded in new information/acquaintance. Ability doesn't seem to exhaust the learning. - ^ Nature of this Information: Is it factual, indexical, constitutional, or sui generis? - ¬ Factual Information: If factual, is it about the object, the state, or the relation? - ÷ Information about the Object: E.g., "This object has this specific spectral reflectance property (presented phenomenally)." Links qualia to representing external properties. (Dretske, Tye). Transparency view. Qualia are just the way the world looks/feels. - ÷ Information about the State: E.g., "My visual system is currently in this specific neural state (presented phenomenally)." Links qualia to representing internal states (Sense-Datum theory echoes? Lycan?). Less transparent view. Qualia reveal brain state properties. - ÷ Information about the Relation: E.g., "I am currently related to the object in this specific sensory way (presented phenomenally)." Relational view. Qualia are properties of the subject-object relation. - ÷ Information about the Experience Itself: "Having this kind of experience." Seems most direct capture, but risks circularity if "this kind" is defined by the quale. Supports opacity. Qualia are intrinsic properties of experience. - ¬ Indexical Information: Information tied essentially to the first-person, present-tense perspective ("this quality, now"). Cannot be fully captured by objective description. (Perry). Knowledge is essentially perspectival. - ÷ Connection to Demonstrative Concepts: Requires a way to 'pick out' the quale directly via experience ("that feeling"). Acquaintance needed? Relies on context of awareness. - ¥ Is this 'picking out' itself dependent on the quale? (Potential circularity: need the quale to demonstrate it). Solvable if demonstration is constitutive or based on unique functional role allowing direct reference. - ¬ Sui Generis Phenomenal Information: A type of information distinct from physical/functional facts (Chalmers' Type-F Monism or Property Dualism). Information about intrinsic phenomenal properties. These properties are fundamental. - ÷ Metaphysical Burden: Requires positing fundamental phenomenal properties or protophenomenal properties. - ¥ Ontological Commitment: Adds complexity (qualia, phenomenal concepts) to one's ontology vs. parsimony. Is the complexity warranted by the explanatory gain? - ¥ Integration Problem: How do phenomenal properties relate to physical properties (causal interaction, emergence, realization)? Epiphenomenalism? Interactionism? Panpsychism? Russellian Monism (fundamental properties are proto-physical/proto-phenomenal)? - § Phenomenal Concept Strategy (PCS) (EXTREME DEEP DIVE) : Mary acquires new concepts (phenomenal concepts) for physical properties/states she already knew under a different (physical/functional) description. The fact known is physical, the mode of knowing (via the phenomenal concept) is new. (Loar, Papineau, Carruthers, Balog, Levine). A physicalist strategy acknowledging the epistemic gap. - % Does this explain the qualitative difference in the feeling itself, or just redescribe the epistemic situation? Does possessing a new concept explain why red feels the way it does? Seems to relocate the mystery into the nature of phenomenal concepts. How can a concept explain a feel? - = The explanatory power regarding the feeling's intrinsic nature is debated. - ^ Critics' View: PCS explains our judgments or beliefs about qualia (why they seem special, why we talk about them this way), but not qualia themselves. It explains the intuition of an explanatory gap without closing it. It explains the conceivability of zombies. It explains why we think Mary learns something factual, but doesn't explain the phenomenal feel itself. - ¬ Focus on Epistemology: PCS is primarily an epistemological strategy, explaining our epistemic situation regarding qualia, not a metaphysical account of qualia themselves (though it assumes qualia = physical states). - ¬ The Hard Problem Remains?: Does explaining why qualia seem special actually explain qualia? The "what it's like" might remain untouched by an analysis of concepts about it. - ^ Proponents' View: Phenomenal concepts are intimately linked to the phenomenal states themselves, perhaps constitutively. Deploying the concept involves the phenomenal state in some way, explaining the special character of the knowledge. The concept provides a unique cognitive access route to the physical state. - ¬ Constitutional Link: The concept might be partly constituted by the phenomenal state/neural basis (Papineau, Balog). Using the concept involves redeploying the state. - ¬ Unique Cognitive Role: Phenomenal concepts have a unique functional role tied directly to the experience, different from theoretical/physical concepts of the same state. They are used for fine-grained discrimination, memory, imagination based on the experience itself. - ^ Nature of Phenomenal Concepts: How do these concepts work? What makes them "phenomenal"? Several proposals: - ¬ Indexical Concepts: Refer to the quale via its role in the current experience ("this quality"). Tied to the context of experience. (Perry-like). Concept refers via context. - ÷ Dependence on Experience: Can only be formed/deployed when having the experience (or recreating it in memory/imagination). - ¥ Explains Mary's Limitation: She lacked the experience needed to form the concept/deploy it indexically from the experiential context. Seems plausible epistemic story. - ¥ Problem: Does indexicality capture the richness? "This quality" seems too thin. Does it explain why the concept feels non-physical? - ¬ Demonstrative Concepts: Function like "that thing", picking out the quale directly via attention/acquaintance. (Similar to indexical, focus on demonstration). - ¬ Constitutional Concepts: The concept includes or is partly constituted by the phenomenal quality itself, or rather its neural basis. (Papineau). Using the concept involves activating (part of) the neural state underlying the quale. - ÷ Blurs Concept/Referent: Concept is not fully distinct from its referent's physical basis. - ÷ Explains Special Feel?: The concept feels special because it runs the relevant neural process. Knowledge feels direct because the vehicle of thought includes the state itself. - ÷ Problem: How can a neural state be part of a concept? What is the metaphysics of concepts here? How does compositionality work? - ¬ Quotational Concepts: Concepts that effectively 'quote' the experience. Refer to the physical state by deploying part of the state itself in the concept, used quotationally. (Balog). Similar to constitutional, emphasis on quotation role. - ÷ Analogy: Like quoting a word " 'red' ". Phenomenal concept refers to neural state N by embedding N within a conceptual structure: [N]phen. - ÷ Problem: Same issues as constitutional view. How does neural state embedding work? - ¬ Recognitional Concepts: Concepts deployed based on the ability to recognize the experience type when it occurs, without needing descriptive knowledge. (Loar). Based on discriminatory capacity. - ÷ Problem: Seems too ability-focused, maybe susceptible to Ability Hypothesis counter-arguments. Doesn't fully capture the qualitative aspect? - ¬ Irreducibility/Primitiveness of PCs: Are phenomenal concepts sui generis cognitive kinds, irreducible to other concepts, whose deployment constitutes phenomenal knowledge? They have unique conceptual role, unique inferential connections (or lack thereof to physical concepts). (Levine?) - ÷ Explains Epistemic Gap: The gap arises because PCs are conceptually isolated from physical/functional concepts. No a priori deduction possible between them. - ÷ Problem: Why should such irreducible concepts exist for physical states? Why this cognitive architecture? Ad hoc? - ¬ Common Structure: Most PCS views agree PCs are distinct from 3rd-person physical/functional concepts, refer to physical states, are acquired via experiencing the state, and explain the epistemic gap/intuitions (Mary, zombies). Disagreement is on the precise mechanism. - ii. Introspection's Target (Revisited): So, if external perception involves these rich qualia (or phenomenal concepts/information), why does introspection of, say, a belief seem to lack analogous qualia? - A. Target Difference Hypothesis: Perhaps beliefs, unlike perceptions, simply lack intrinsic phenomenal qualities. Introspection can't reveal what isn't there. Qualia are tied to sensory/perceptual processing (or perhaps affect). Propositional attitudes are non-phenomenal by nature. - B. Cognitive Phenomenology Hypothesis: Perhaps beliefs do have qualia ("cognitive qualia"), but they are much more subtle, less salient, or different in kind from sensory qualia. Introspection does access them, but we often overlook them or lack concepts for them. (Pitt, Strawson, Montague). - I. Examples: The feeling of "knowing," the feeling of "doubt," the specific feel of entertaining a particular proposition (e.g., difference between thinking 2+2=4 and thinking about Paris). - II. Challenge: Are these genuine qualia (intrinsic, non-representational feels), or just combinations of sensory imagery, affect, and metacognitive feelings (like fluency, effort)? Debate over existence and nature. Reducible to sensory/affective halo? - C. Functional Difference Hypothesis: Perhaps the function of introspection differs from perception. Perception aims to represent the world qualitatively. Introspection aims to access information for self-monitoring and control, which might not require generating distinct "introspective qualia." Information format might be propositional or conceptual primarily for attitudes. - D. Transparency Argument Applied Internally?: Maybe even when introspecting belief, we focus on its content or functional role, not the 'feel' of introspecting or the 'feel' of the belief state itself (even if subtle cognitive qualia exist). Default mode is transparent even for inner world? Less plausible if target is a quale itself (like pain). **