In a recent thread on BlueSky, a question was posed: “Do we experience the dreams as they unfold or do we only experience them as we remember?” Undoubtedly, the same question can be asked of our awake daily experiences. Do I experience the taste of a matcha latte as I drink, or do I only experience the taste as I actively attend to it and think about what I experience? Peter Carruthers would deny the former and the latter, while putting forward his own middle ground. I read his book “Consciousness: Essays from a Higher-Order Perspective”, and there are some controversial ideas.

The author advocates for a dispositional higher-order thought theory. 

Let’s first situate this theory with respect to others. This theory implies representationalism of some sort, meaning contents of perception are somehow represented in the brain (in the mind). Also, it is a theory of perceptual consciousness, of why it feels a certain way to perceive color or taste. Representational theories of phenomenal consciousness come in two varieties: first-order and higher-order. First-order theories suggest that only sensory processing of color is needed for conscious perception. In contrast, higher-order theories argue that an additional higher-order processing is required. Higher-order theories, in turn, can be divided into higher-order perception (HOP) theories and higher-order thought (HOT) theories. HOP theorists argue for the presence of an inner sense “organ” that receives and processes output from first-order systems and generates higher-order representations of “the feels”. In contrast, HOT theories posit that consciousness involves higher-order thought. HOT theories, in turn, can be subdivided into actualist and dispositionalist theories. Actualist HOT theorists would say that a first-order perceptual state and a higher-order thought should occur together. Dispositionalist HOT theorists would say that if a first-order state has a potential (or disposition) to evoke higher-order thought, then it is conscious. The author advocates for a dispositionalist HOT theory that he calls “the dual-content theory”.

The dual-content theory is named such because a conscious perceptual state contains two types of content. First-order content is ordinary perception. For example, “there is a red apple”. Higher-order content is experience. For example, “there is a visual experience of red” or “seems red”. The theory is dispositional because, for the author, a perceptual state should only be available for thought machinery. It can be conscious without machinery actually thinking about it. Perceptual consciousness, thus, is availability for higher-order representation.

The author claims that his theory answers important questions about conscious perception.

1. Why is there a subjective element in conscious perception: “It’s me who is perceiving”? Because some percepts have dual content.

2. Why is there an explanatory gap: “My perception of red feels like something”? Because some concepts are purely recognitional concepts. For visual perception of red, higher-order content “seems red” works as a pointer, as a purely recognitional concept “this experience”. These recognitional concepts create the semblance of an explanatory gap.

3. Why do phenomenal properties of experiences feel intrinsic? Why does it seem that there is an inner experience of redness: “The redness is just there, inside the experience itself”? It is due to the pointing nature of recognitional concepts. Purely recognitional concepts, such as “this” and “that”, don’t provide information about any relations. Consequently, the experience feels like a simple intrinsic property.

4. Why phenomenal conscious experience is indescribable or impossible to fully communicate: "I can tell you everything about red, but I can't tell you what it's like to see red”? Because purely recognitional concepts are not linked to other ordinary concepts. Therefore, we don’t have the language to describe them.

5. Why is phenomenal conscious experience private: “It happens in my head”? Because experience is represented there, in your head.

6. Why does it seem phenomenally conscious experiences are trustworthy: “I know what I have seen”? Because if a percept has the first-order content “red”, it also possesses the higher-order content “seems red” (dual content). Those two contents are bound together in a way that is reliable.

The author's version of the dispositionalist HOT also relies on consumer semantics being true. It is entailed that there are downstream systems that may use the content of the representation. These systems are belief-forming systems, reasoning systems, action-planning systems, mind-reading systems, etc. Because the same experience is available for different systems, it will possess dual content.

Among those systems, particular importance is taken up by the mind-reading system. This system, according to the author, is required to possess the capacity for HOT self-representation. I can only be conscious of red if my mind has a disposition or capability to represent the thought “I am experiencing red”. Because of this requirement, most of the animals are excluded, meaning that most animals don’t have conscious perception.

The last point is controversial, and intended to be so. As many of the other points in the book. Those controversial points seem counterintuitive, but the author’s rebuttal of intuitions is inspired by experiments where “the brain” seems to trick “the mind”. For instance, “the brain” may make “the mind” think it made a conscious, rational, logical decision, while in reality, “the brain” was relying on biases. This and similar findings are then extrapolated onto all intuition that we may have. From which follows that we should not take intuitions at face value.

There are many things in the book that I cannot agree with. 

The issue with non-conscious animals is particularly contentious. If the animal feels pain and acts on the environment to reduce the pain, the author would still claim that this behaviour can be explained without higher-order representations. Because for the author, no mind-reading = no conscious perception. Then, why and when do we need to apply higher-order representations to explain human behaviour? Also, do those mind-reading system capacities come in degrees? Is it possible to be a little bit conscious?

When discussing humans, the author gives an example of two visual streams: dorsal and ventral. The ventral stream (or “what” stream) sends information from the visual cortex to the temporal lobe. Information manipulation in the ventral stream has been shown to correlate with conscious object recognition, including colors, shapes, and faces. The dorsal stream (or “where” stream) sends information from the visual cortex to motor planning and motor execution areas. Information manipulation in the dorsal stream has been shown to correlate with motor action and movement estimation. Notably, the information inside the ventral stream can be described, while information in the dorsal stream is hard to describe. For example, try to describe the appearance of a mug versus try to describe exactly how your hand became configured to grasp the mug. The author explains this discrepancy with availability: contents in the ventral stream are available for higher-order thought, while contents of the dorsal stream are not. However, it is not obvious what the difference is between these two streams in terms of structure and/or electrophysiology. In other words, how do neurons and their connection define what is available and what is not?

Lastly, I didn’t fully understand the author’s metaphysical commitments. Is phenomenal consciousness = higher-order content? One option would be to answer with identity. Phenomenal consciousness is nothing over and above a first-order perceptual state plus a higher-order representation of that state. Another option would be to introduce some degree of epiphenomenalism, and that is what the author does. When talking about animal consciousness, he speculates that humans are not that far from animals after all, and “phenomenal consciousness might be almost epiphenomenal in its functioning within human cognition”. This, in turn, raises a question: if phenomenal consciousness is not identical to higher-order content, what is it? And what dual-content theory actually explains?

Overall, the book is fairly technical and philosophy-heavy. I admit that I might not have understood everything. However, I saw some inconsistencies in the theory that made me reluctant to agree.

Quote:

“phenomenal consciousness is constituted by dual-content perceptual states, wherever they might occur. To the best of our knowledge, such states are also always actually accessible to the reasoning processes and reporting systems of their subjects. But there is nothing in my account of phenomenal consciousness that logically requires it”

June, 2026