Materialist Phenomenology

1. Phenomenology Needs Non-Reductive Materialism

We can affirm substance monism while retaining property dualism.

Bridging the divide. Traditionally, materialist and phenomenological philosophies have operated in isolation, with materialists often reducing the mind to the brain or dismissing it as causally inert. This book argues for a non-reductive materialism, a philosophical stance that acknowledges the physical basis of the mind while preserving its distinct mental properties and causal efficacy. This approach allows for a methodical analysis of phenomenal experience without importing idealist ontologies.

Three core ideas. Non-reductive materialism rests on three fundamental tenets:

• Mental properties are distinct from physical properties.
• Mental properties depend on physical properties for their existence.
• Mental properties can confer causal powers on mental events.
  This framework provides a robust alternative to substance dualism, which posits separate physical and spiritual materials for body and mind, and to eliminative materialism, which discards mental properties entirely.

Embodied cognition. The book specifically focuses on visual experience, asserting its objective existence independent of our conceptual frameworks. It emphasizes that cognition, particularly visual perception, is largely nonconceptual and deeply intertwined with our physical bodies. Perception, in this view, evolved as a guide for action, reminding us that the mind's operation inherently requires a body capable of causal interventions in the world, challenging fantasies like the "brain in a vat."

2. The World Actively Informs Our Senses

The light that conveys those signs to our eyes, in turn, should also be conceived as being structured by its interactions with the surfaces that bind objects.

Beyond mere stimulation. The world is not just a source of "raw sense data" or passive stimuli; it is actively populated by natural signs that biological evolution has opportunistically exploited. These signs are not arbitrary symbols but fall into two categories:

• Indices: Causal effects indicating presence (e.g., smoke for fire, footprints for prey).
• Icons: Signs that resemble their object or share similar relations (e.g., maps, drawings).
  These natural signs become "intentional signs" when a living creature treats them as such, assigning them a function in a task, a process often shaped by co-evolution.

Structured light and statistical regularities. Ambient light, referred to as the "optic array," is highly structured by its interactions with objects and surfaces. This structure carries rich information, such as solid angles indicating distance or optic flow patterns revealing self-motion and depth through occlusion. Crucially, the world exhibits statistical regularities—like the prevalence of solid objects or horizontal/vertical orientations—that have profoundly influenced the evolution of our sensory organs, allowing the brain to disambiguate ambiguous retinal projections.

Stability and recurrence. The persistence of objects and surfaces over time, explained by dynamical systems theory and the existence of "preferred states" (attractors), ensures stable and recurrent sources of natural signs. This objective structure of possibility spaces, where some states are far more probable than others, provides a continuous basis for tendencies and capacities that evolution can track. This dynamic view of the world's contribution moves beyond simplistic notions of direct causation, highlighting the intricate interplay between environmental structure and biological adaptation.

3. The Body Offloads Cognitive Work and Shapes Perception

Much of the computational work would still be offloaded to the dynamics of the robot’s legs interacting with the ground.

Beyond detailed commands. The brain's control over the body is not through specifying every anatomical detail but by manipulating parameters that select "preferred configurations" of muscles, joints, and tendons. This concept, drawn from robotics and dynamical systems, suggests that the body itself, with its inherent physical properties and dispositions, performs much of the "computational" work. For instance, a leg's pendular swing during walking is largely a spontaneous dynamic, requiring minimal neural control.

Affordances and effectivities. The interaction between an organism and its environment is characterized by "affordances" – the environment's capacities to offer opportunities or risks – and "effectivities" – the animal's abilities to act upon them. These dual dispositions are scale-dependent; a hole might afford shelter to a rabbit but not a fox. Animals can also actively modify their "effective environment" through constructions like dams or nests, further shaping the opportunities and risks they encounter.

Internal perception and self-awareness. The body continuously provides crucial information through:

• Proprioception: Informs the brain about posture, limb configuration, and movement via mechanoreceptors.
• Interoception: Monitors the visceral environment for metabolic balance via internal receptors.
  These internal perceptions contribute to our "body schema" (a first-person spatial awareness) and generate feelings of ownership and agency. Furthermore, the brain's ability to simulate actions, combining visual, somatic, and motor maps, allows us to experience the "accessibility" of unobservable object features, demonstrating how action non-instrumentally shapes perception.

4. The Brain Operates as a Distributed Multi-Agent System

We can recursively apply this part-to-whole relation between agents as far as needed until we reach a level populated by entirely mindless agents, avoiding the problem of infinite regress.

Beyond the homunculus. To avoid the infinite regress of a "little person" in the head interpreting signs, the book proposes a multi-homuncular model. This hierarchy consists of populations of "mindless cognitive agents" (like neural circuits or cortical areas) that perform specific tasks, consuming and producing hybrid iconic-indexical signs. These agents are defined by their architecture and emergent cognitive capacities, not by conscious interpretation.

Coordination through dynamics. The brain's distributed processing, where different properties (shape, color, motion) are handled by specialized agents (e.g., V1-V6 visual areas), is coordinated through synchronized oscillatory activity. Neurons, modeled as relaxation oscillators, tend to "entrain" or synchronize their firing, forming transient coalitions of active populations that bind disparate information into coherent percepts. This dynamic mechanism allows for rapid, flexible coordination without a central command.

Attention as a top-down modulator. Voluntary attention acts as a crucial top-down influence, cascading down the processing hierarchy to:

• Increase the coding quality of signs for attended objects.
• Decrease noise in signal transmission.
• Bias competition among rival neural coalitions.
  Attention also provides the "indexical" component of internal signs, effectively tagging specific parts of retinotopic maps with "Your attention is here," thereby informing agents about the subject's current focus and valuation.

5. Perception is a Multi-Stage Construction of Reality

The process of recovery proceeds one step at a time, extracting line segments from an image and grouping these into one dimensional entities to represent boundaries of different types (the primal sketch); selecting and grouping those lines to create two-dimensional contours representing surfaces, while at the same time incorporating information about their distance and orientation from the observer (the 21/2D sketch); and finally, assembling those surfaces into a solid shape using a simple stick figure to define the articulated segments and the spatial distribution of the surfaces (the 3D model).

From raw input to object models. Visual perception is a complex, multi-stage process that progressively constructs our experience of objects. Starting from the retina's initial analysis of light intensities, the brain builds up geometric representations:

• Primal Sketch: Extracts oriented line segments from luminance discontinuities (retinocentric coordinates).
• 2.5D Sketch: Groups lines into contours, adding information about surface depth and slant (egocentric coordinates).
• 3D Model: Assembles surfaces into solid shapes, identifying object axes (allocentric coordinates).
  This hierarchical construction, though simplified, illustrates how "mindless agents" transform natural signs into increasingly sophisticated intentional signs.

Separating appearance from reality. A key function of object perception is to distinguish the "perspectival" (how objects appear from a given viewpoint) from the "factual" (their objective properties). This is evident in "constancy effects," where an object's size, shape, or color appears stable despite drastic changes in its retinal projection. Mechanisms like detecting ratios of apparent size to distance, or comparing reflectances across a scene, allow the brain to achieve this separation.

Coordinate transformations and binding. To link visual input to bodily action, the brain performs a series of coordinate transformations: from eye-centered to head-centered, then to body-centered, and finally to object-centered coordinates. These transformations, potentially mediated by "gain fields," integrate information from visual, proprioceptive, and motor maps. The final coherent perception of an object, with all its properties bound together, is achieved through the synchronization of oscillatory activity across distributed brain areas.

6. Biological Values Guide Our Cognitive Choices

The central place that evaluations have in an emotional response points to the main evolutionary function of emotions: to encode values that guide successful action in the world.

Beyond pure cognition. The brain's cognitive agents don't operate in a vacuum; their decisions, especially in action planning, are guided by a sophisticated biological value system. This system, implemented through neuromodulators and neurohormones, reflects basic homeostatic needs and emotional responses (e.g., fear, reward). When multiple cognitive "coalitions" compete for control, these values help select the most appropriate response.

Subcortical appraisal. Key "relay stations" in the brain, such as the basal ganglia, cerebellum, and hippocampus, play crucial roles:

• Basal Ganglia: Evaluates saliency, reward probability, and aids in action selection and habit formation.
• Cerebellum: Modulates motor control and correlates sensorimotor acts with mood states.
• Hippocampus: Creates spatial maps and attaches affective values to locations and objects.
  The amygdala, in particular, performs rapid and slower appraisals of situations, assessing risks and opportunities to guide motivational states.

Feelings as learning tools. While emotional responses involve complex endocrinal, neural, and behavioral components, the subjective "feeling" part is not a mere epiphenomenon. Feelings, like pleasure or displeasure, are indispensable for learning, particularly in instrumental conditioning. They provide the necessary feedback for an organism to associate actions with positive or negative outcomes, thereby shaping future behavior and refining the value system over a lifetime.

7. Consciousness Emerges in Graded Layers of Agency

The overall effect of many protoselves consuming internal signs as feelings yields a subjective state we may refer to as the feeling of being alive.

A hierarchy of conscious agents. The "hard problem" of consciousness is tackled by rejecting two-level models (brain vs. mind) and proposing multiple intermediate levels of agency. Consciousness is not a single, monolithic entity but emerges gradually through the sign-consuming activities of progressively more sophisticated agents:

• Protoselves: The simplest conscious agents, consuming signs for properties (e.g., glucose concentration, reflectances) as raw, lived feelings (e.g., hunger, colors). They operate at the temporal scale of tens of milliseconds, integrating stimuli into moments of simultaneity or succession.
• Core Selves: Composed of interacting protoselves, these agents consume signs for objects and their affordances as feelings of anticipation, dread, ownership, and agency. They live in a "specious present" lasting a few seconds, integrating information into a coherent, immediate experience.
• Autobiographical Selves: Formed from core selves, these agents consume signs for complex situations, integrating them with long-term memories and future plans. They generate feelings of involvement and are capable of creating coherent narratives of their past and future.

Consumption, not representation. Crucially, these upper-level agents are conceived as pure consumers of signs, not producers of further representations. Phenomenal experience, therefore, is identified with the consumption of signs, eliminating the idea of the visual field as a "veil" separating us from reality. The "transparency" of perception is explained by our natural tendency to attend to factual content, though we can train ourselves to notice perspectival appearances.

Graded intentionality and attention. Each level of agency possesses distinct forms of intentionality and attention:

• Protoselves: Exhibit basic arousal and vigilance, sensitive to changes and intensity gradients.
• Core Selves: Engage in "flow," a highly concentrated attention during skillful interaction with objects, characterized by effortlessness and immediate feedback.
• Autobiographical Selves: Employ "selective awareness," directing attention to relevant details in complex situations, informed by past learning and future contingencies.

8. Nonconceptual Cognition Underpins Our Perceptual World

Human beings can make sense of what they see without the help from any conceptual machinery.

Beyond concepts and beliefs. A central tenet is that much of our visual perception, particularly of properties and objects, is nonconceptual. We can be aware of a colored object without possessing the concepts to identify its type or color, or to form verbal beliefs about it. This challenges views that equate visual consciousness with the ability to form beliefs or make explicit judgments.

Perceiving properties and objects. The perception of isolated properties, like color, is akin to a measuring operation, where the brain maps ranges of objective properties (e.g., surface reflectances) onto ranges of internal states, yielding intrinsic phenomenal qualities. Object perception then builds on this, separating perspectival appearances from factual properties through mechanisms like constancy effects and the detection of higher-order relations (ratios, gradients). These processes do not require concepts but rely on the brain's inherent capacity to process information and its evolutionary history.

Demonstrative understanding of situations. Even the understanding of simple situations can be nonconceptual. Animals, like vervet monkeys, can recognize predator types and respond appropriately without possessing "sortal concepts." Humans also engage in "demonstrative understanding," using nonlinguistic equivalents of indexicals like "this," "that," "here," and "now," fixed by attention and a practical grasp of action possibilities. This form of understanding relies on statistical prototypes extracted by neural mechanisms, rather than abstract universals.

9. The Unified Self is a Narrative, Not a Primal Given

Its unifying narrative is the way it explains to itself all the sensations and feelings in which it is immersed and the sense of agency and ownership it gets from its body in action.

Challenging introspective unity. While introspection often suggests a single, unified conscious self, the book argues that this unity is largely a narrative construction by the autobiographical self. Evidence from phenomena like conscious confabulation (fabricating explanations for actions without conscious access to their true causes) and split-brain patients (where hemispheres operate with dissociated awareness) indicates that our sense of a singular self is often an interpretive act.

Multiple subjectivities. Split-brain studies reveal that the left hemisphere, with its linguistic abilities, may construct a coherent story even when it lacks direct access to experiences processed by the right hemisphere. This suggests the possibility of multiple "autobiographical selves," one capable of descriptive understanding (language-dependent) and another of demonstrative understanding (shared with other primates). The apparent unity in daily life is fostered by shared bodily experiences, unified subcortical structures, and competition for limited attentional resources.

Survival, not strict identity. Personal identity is not an absolute, strictly identical state but a matter of "survival" across stages, varying by degree. This survival depends on the broad similarity and causal connection between past, present, and future versions of oneself. The autobiographical self, through its capacity for episodic memory and future planning, weaves these stages into a continuous life story. However, this narrative is a dynamic, often retrospective, construction that explains the myriad sensations and feelings experienced by lower-level agents, rather than a direct reflection of an inherently unified consciousness.

Last updated: March 16, 2026