Reading Thomas Metzinger

1.1 The Two Aspects of Transparency

Although Metzinger defines transparency precisely (constraint 3.2.7), I found the bundle of accompanying examples difficult to unify under a single concept. The notion spans two distinct aspects:

Phenomenal Simplicity: Experience is transparent when we cannot direct attention to any more details of the underlying process. There exists a strict, impenetrable attentional border—the aspects of perception that could become part of experience correspond to what we can potentially attend to. Large portions of the perceptual process remain permanently inaccessible to attention (except perhaps under psychedelic substances). This makes experience feel “substantial”—things appear to us as they fundamentally are, with no experiential clue that appearances might be abstractions or statistical inferences.

Phenomenal Givenness: In normal waking states (and vivid non-lucid dreams), we are predisposed toward naive realism. We experience immersion in a world as it independently is. Experience is transparent when its content is experienced as its only cause—as exclusively about the actual world. Under this aspect, thoughts, imaginations, and lucid dreams are phenomenally opaque: they are perceived as “representing,” as standing for something else.

These aspects relate through concepts like ineffability and immediacy, but they can dissociate. In vivid dreams we experience qualia (simplicity), yet the dream may lack givenness if we recognize it as dream. Is Metzinger treating these as two sides of one coin, or defining transparency as their conjunction?

1.2 Metzinger’s Definition and Its Interpretations

Metzinger’s formal definition: “For any phenomenal state, the degree of phenomenal transparency is inversely proportional to the introspective degree of attentional availability of earlier processing stages.”

Here “earlier processing stages” can mean temporally earlier (aspect of givenness) or constitutionally earlier (aspect of simplicity). Both are “internal causes” of experience. If we could attend to edge detectors in visual processing, Metzinger suggests, we would have to “will objects into existence” for them to appear—by analogy to imagination, where we must will imagined objects into being.

On homogeneity (constraint 3.2.10), Metzinger writes that without it, we could “introspectively penetrate into the processing stages underlying the activation of sensory content,” and “the multimodal, high-dimensional surface of our phenomenal world would start to dissolve. We would then phenomenally experience the model as an ongoing global simulation permanently generated from scratch.”

1.3 My Interpretation: Top-Down vs. Bottom-Up

Lucid dreams offer an interesting test case: we can affect scene construction by directing attention to top-level fragments of top-down information propagation, while bottom-up processing proceeds unaffected. I propose:

Givenness obtains when we do not clamp any fragment of the top-level layers for top-down information propagation—either because bottom-level layers are clamped by sensory input, or because we don’t realize we can affect them.

Simplicity obtains when bottom-up information propagation starts below the lowest attentionally available layer, rather than in the middle.

1.4 The Transparency Confusion

Later in the book, the transparency/opacity distinction becomes muddled. Sometimes opacity means “presenting misrepresentation” (as in opaque mental representations), other times “presenting something as representational” (correlated with different content). The later claim that “presentational content is always fully transparent” while “representational content” (cognitively available) tends toward opacity helps somewhat, but phrases like “fully opaque self-model” remain puzzling—this cannot mean presenting misrepresentation in any straightforward sense.

Metzinger’s eventual clarification: “A transparent representation is characterized by the fact that the only properties accessible to introspective attention are their content properties. It does not allow for the representation of a vehicle-content distinction using on-board resources… opaque experience is the experience of knowing while also (nonconceptually, attentionally) knowing that you may be wrong.”

2.1 The Puzzling Claim About Identity

In section 5.4 (“From Mental to Phenomenal Self-Presentation: Embodiment and Immediacy”), Metzinger writes that at some level of elementary bioregulation or molecular self-organization, the system is “forced—from a conceptual third-person perspective—to maintain the distinction between content and vehicle.” Yet he also claims philosophy’s task is to “demarcate a more fine-grained level of description on which it is plausible to assume a full match between content and causal role, that is, the identity of vehicle and content.”

This remains unclear to me. A side note: while Metzinger discusses the brain being insensitive to itself, he never addresses the phenomenology of headache—an interesting omission.

2.2 Intentional vs. Phenomenal Content

The intentional (referential) content is straightforward: the representanda, the referents. Phenomenal content is “the way certain representational states feel from the first-person perspective.” In chapter 8, Metzinger says phenomenal content is “a special form of intentional content, namely, in satisfying the constraints developed in chapters 3 and 6.”

This cannot be quite right, since constraints are satisfied by intentional vehicles, not content. Perhaps the thesis is that phenomenal content is the semantic aspect of phenomenal experience—that how conscious processes “feel” is what they mean. An intriguing possibility.

2.3 On Double Dissociations

Metzinger argues double dissociations between phenomenal and intentional content do not exist. Unconscious intentional content abounds, but in standard situations no conscious state fails to be representational. Even diffuse emotions like jealousy have intentional objects—“The original representandum may be something that was only present in the world of our distant ancestors.” The geometric hallucinations under certain conditions represent a potential exception: “purely phenomenal content.”

3.1 The Role of the PSM

The Phenomenal Model of the Intentionality Relation (PMIR) appears to be a structure and process for managing attention. When an object is integrated into PMIR, directing attention toward it becomes phenomenally available—available for action control, concept formation, and “higher order” attention.

Integrating the Phenomenal Self-Model (PSM) into the world model provides relations between the PSM concept and “outside” object concepts. Such relations are the basis of goal-directed behavior: when a simulated relation differs from an actual relation, the difference drives action.

3.2 Object Integration and Attention

The PSM integrates the representation of any process that preattentively integrates features as an object. Only then does the object become attentionally available—this is more basic than PMIR, enabling all “higher order” relations. I’m uncertain whether Metzinger thinks objects are available for concept formation prior to PSM integration; given the lack of attention, this seems unlikely. What gets integrated is the distinct concept of perceiving the object. Now it becomes possible to have the goal of “taking a better look.”

This constrains my earlier understanding of attentional availability. Since global availability means integration into the world model, integrating it subjectively (into the PSM) shouldn’t require much additional machinery. Object formation already happens in both world-model and PSM; in this process, object-encoding processes become mental representations—potentially globally available.

3.3 Properties of the PSM

The PSM is distinguished in representational space by high invariability. Attention may be drawn to places where model output and sensory input differ most, but background self-awareness persists. All properties of the system are represented in one integrated data format—the PSM is “holistic.”

Metzinger writes: “The nervous system… is not so much a top-down controller, but more a system whose task consists in generating adequate patterns within thebody–nervoussystem–environment overall dynamics… if an internal self-model is to be a successful instrument in predicting the future behavioral dynamics of an individual system, it must in some way mirror or reflect important functional characteristics of precisely that part of its internal dynamics, which in the end will cause the overt actions in question.”

3.4 The Paradox of a Grounded Self-Model

A striking claim: “A fully grounded self-model would simply disappear. In principle, phenomenal selfhood emerges as long as there is a conflict or incoherence between bottom-up and top-down processes, between expectancy and actual perception.” The discrepancy draws attention to emulators, making them conscious—globally available for flexible reaction. “A certain level of autonomous, residual self-modeling is preserved.” Dynamics, including goal-directedness, is part of the self-model.

4.1 Content as Relation

Metzinger emphasizes that “the content of a perceptual state really is not a part of the environment, but a relation holding to this part… Full-blown, phenomenal self-consciousness always involves a relation between the self and an object component.”

4.2 Attentional Agency and Selectivity

Some stages of attentional agency are conscious, others unconscious. The process displays “an extremely high degree of flexibility and short-term adaptability, involving the explicit internal simulation of alternative objects for attentional processing. We like to call this ‘selectivity.’” What we experience as selectivity is “a globally available representation of the process in which different alternatives are matched against each other and the system settles on a single solution.”

4.3 Opaque Mental Representations

“Opaque mental representations” are those not perceived as standing for reality—like doubled vision when pressing one’s eyeballs. There is conscious meta-representation, but without dedicated channels. It is the knowledge that some experiences are not veridical, and the knowledge of how to use mental faculties to manipulate higher, more abstract layers of “modality stacks.” In volitional thought, “the object component is opaque. We know that we take a certain attitude toward a self-generated representation of a goal.”

4.4 The First-Person Perspective as Dynamic Subject-Object Relations

Metzinger: “Please note how a phenomenal first-person perspective now reveals itself as the ongoing conscious representation of dynamic subject-object relations: to see an object, to feel a pain, to selectively ‘make a thought your own,’ to chose a potential action goal, or, to be certain of oneself, as currently existing.”

This leads me to conclude that we have something like a self-perception modality. The subject–object character of experience comes from cross-modal binding with this modality. But Metzinger complicates this: “Cognitive self-reference, therefore, on the phenomenal level is necessarily experienced as direct and immediate, because it is not mediated through any sensory channel (it takes place in a supramodal format).” The supramodal aspect arises from cross-modality binding in higher layers.

4.5 Structure of the PMIR

“Phenomenal models of the intentionality relation consist of a transparent subject component and varying object components, which can be transparent as well as opaque, transiently being integrated into an overarching, comprehensive representation of the system as standing in a specific relation to a certain part of the world.”

Episodic memory involves reconstructing a PMIR—memory retrieval requires associating a past-event simulation with a self-representation. “Reactivating a PMIR inevitably means reactivating a PSM.” Patients lacking PMIR display akinetic mutism: wakefulness without speech, emotional expression, or movement. They can track objects or say their name, but show “complete absence of any willed action or communicative intention.”

4.6 Agency and Ownership

“The experience of agency seems to be the ongoing representational dynamics collapsing a phenomenal model of the practical intentionality relationship into a new transparent self-model.” Two kinds of ownership are involved: ownership for body movements (PSM), and ownership for the corresponding volitional act (PMIR)—“for example, the conscious representation of the selection process preceding the actual behavior.”

4.7 My Assessment: Is PMIR a Distinct Module?

PMIR seems to describe various binding processes where one component is the transparent part of the PSM. I’m not convinced these form a distinctive separate module. Given the structure of the PSM (transparent lower layer, more abstract higher layers) and binding processes in general, there doesn’t seem much to add specific to PMIR. Perhaps Metzinger’s thesis is that there is something special about this subclass of integrative processes.

4.8 Additional Observations

PMIR may be key to intersubjectivity: its representation can be its object. Goal representations (via goal-encoded objects) exist in non-egocentric frames of reference; they can be integrated into PSM for actual behavior, into “allocentric frames” as simulations of others, or activated unconsciously by mirror neurons. Mirror neurons first emulate low-level, non-goal movements; low-level and high-level resonance mechanisms do not coincide.

Linguistic concepts are much more than simple concepts (non-homomorphic representations gaining recombinability through activation links). They are goal-encoded objects, having qualia as all objects do—here, the words or other symbols themselves.

5.1 The Weakness of Metzinger’s Eliminativism

Metzinger’s metaphysical argument against the existence of mental notions such as the self is weak. It is mostly a proof by assumption of the thesis, starting from materialist grounds. His position resembles eliminativist materialism as in Richard Rorty’s Philosophy and the Mirror of Nature (Chapter 2). Metzinger uses his developed theory to provide what Rorty calls “criteria for the identity of reference.”

5.2 Rorty’s Move Beyond Eliminativism

Rorty moves past eliminativist materialism: “But since I think that the reductive and eliminative versions of the identity theory are both merely awkward attempts to throw into current philosophical jargon our natural reaction… I do not think that the difference between the two should be pressed. Rather, they should both be abandoned, and with them the notion of ‘mind-body identity.’ The proper reaction… is to adopt a materialism which is not an identity theory in any sense, and which thus avoids the artificial notion that we must wait upon ‘an adequate theory of meaning (or reference)’ before deciding issues in the philosophy of mind.”

5.3 The Antipodeans and Phenomenal Life

Consider Rorty’s Antipodeans: people just like humans who never developed mental terms, instead using brain-state descriptions from an early point in their cultural development. For Rorty, there’s no real difference between Antipodeans and humans outside their “folk psychology.” For Metzinger, by contrast, the scenario is implausible—Antipodean terminology would indicate vastly different phenomenal lives. Antipodeans are probably only “system-conscious.” Such phenomenal differences would project onto their entire culture.

5.4 The Non Sequitur

Rorty’s point about eliminativism: “Some statements of the form ‘I just had a sensation of pain’ are as properly taken as true as ‘The sky is overcast’ and ‘The sun is rising,’ but none of them is true.” Metzinger’s identity-of-reference criteria function like a theory explaining how “The sky is a blue dome” is false, by explaining what the sentence speaks about and how facts differ from postulated facts.

But once Metzinger finishes building the referents of “pain,” instead of saying “and here is how talk about sensations and selves gets things wrong,” he claims “and since we can now see that sensations and selves are complex processes rather than simple substances, they do not exist”—wait, what? This is a non sequitur. Complex processes can exist just as readily as simple substances.

6.1 Agnosia (4.2.1)

Metzinger claims (pp. 220–221) that a patient using chromatic information for shape formation and motion detection—but lacking color experience (seeing in shades of gray)—has color “cognitively but not attentionally available.” This seems wrong to me: that chromatic distinctions feed into shape formation has nothing to do with recognizing color conceptually. Chromatic vision feeds into concept formation here, but not into color concepts.

6.2 Hemineglect (4.2.2)

Hemineglect is undoubtedly an attentional deficit (and a deficit of the “model of intentionality relation”), but likely the deficit exists simply because lesions leave nothing to attend to and model. Metzinger doesn’t claim otherwise; he analyzes minimal conditions that could generate hemineglect.

6.3 Anton’s Syndrome

The discussion focuses on the self-modeling deficit without mentioning whether offline phenomenal experience (nocturnal dream-like, but top-down modulated) is also absent. I suspect it is.

6.4 Charles-Bonnet Syndrome (4.2.4)

Metzinger writes that “percepts are missing characteristic features and are simply superimposed on, but not semantically embedded in, the phenomenal model of external reality.” This is contradicted by the following example: the patient reports pragmatic and (slight) phenomenal abnormalities as distinguishing hallucinated content; semantically it seems to be OK.

6.5 Dreams (4.2.5)

Metzinger claims “phenomenal dream content is not attentionally available” and “All there is is salience-driven, low-level attention.” This conflates two concepts: attentional salience and volitional attention. Phenomena must be attentionally available (in the salience aspect) to be minimally conscious.

This connects to the four forms of introspection from Chapter 2 (p. 36): (1) external attention, (2) consciously experienced cognitive reference, (3) inward attention/inner perception, and (4) consciously experienced cognitive self-reference. Forms 1 and 3 are subsymbolic re-representation (toward world-experience and self-model, respectively); forms 2 and 4 are symbolic variants.

Metzinger says “introspection 3 is almost impossible in a dream state, because high-level attention is absent.” Most dreams feature a phenomenal self, only not a volitional self due to lack of deliberation. In dreams, attention cannot rest on the self-model; it only integrates features selected by the generated world-model. “You cannot introspectively attend even to your most simple sensory perceptions in the dream state, because you are not an attentional subject.”

Anecdotally, I have had dreams without a phenomenal self—dreamt from third-person perspective, remembered as movies. I have certainly had dreams with a phenomenal self that was not my actual (waking) self-model. Normally my dreams are remembered in roughly the same format as normal memories.

6.6 On “Earlier Processing Stages”

I doubt “earlier processing stages” can become directly attentionally available due to architectural limitations. More likely they become available indirectly by “polluting” the bottom-up signal with “vehicle properties”—the maximum a posteriori distribution puts too much weight on lower-layer consistency, fixing them before they propagate information upward. This might be an important (perhaps primary) mechanism of attentional availability. For Metzinger, attentional availability is “subsymbolic re-representation”: additional neurodynamical structure correlating with and propagating information about the original phenomenon.

How does this differ from cognitive availability (“symbolic re-representation”)? The attentional structure is neurodynamically homomorphic with the original phenomenon and cannot be reassembled in arbitrary contexts, while symbolic structure is only activationally linked.