Consciousness explained by Dennett

Experience without a central meaner

Consciousness Explained (Dennett, 1991) sets out to demystify consciousness using empirical work in cognitive science and neuroscience.

Part I sets up the central questions and existing methods for studying the mind, and
challenges traditional notions and assumptions on which these methods rest. It then proposes the new approach of ‘heterophenomenology’, which treats a subject’s reported conscious experiences as data that must be interpreted, rather than reliable truths about an internal mental world.

Part II aims to dissolve the idea of a ‘Cartesian theatre’—a single inner place where everything is experienced—and to motivate and build an empirical model of the mind—the ‘multiple drafts model’—in which parallel, distributed processes in the brain continuously produce and revise the contents of experience.

Part III discusses the possibility of conscious robots, as well as argues that many familiar puzzles disappear once we give up the inner theatre picture.

Dennett condemns contemporary research on the mind for being fragmented into separate silos. Neuroscience, cognitive psychology, AI, anthropology, philosophy all study consciousness, but mostly in isolation. Each enterprise makes valuable contributions, yet on its own offers only a narrow and incomplete view. Dennett therefore stresses the need for an interdisciplinary project that attempts to combine the strengths of each discipline.

The current landscape is a mix of real progress and deep confusion: new models and experimental results coexist with overhyped ‘proofs’ and premature declarations that rival ideas have been refuted. In this context, Dennett advocates seeking independent lines of evidence that converge on a single hypothesis, rather than placing excessive weight on any one ingenious thought-experiment or odd experimental result.

I recommend this book for two main reasons. It offers a sustained critique of common, unquestioned intuitions on which much work on consciousness still heavily relies. Second, it advocates, and takes an important first step toward, a scientifically driven, interdisciplinary approach to the study of consciousness.

§ Phenomenology

In his Critique of Pure Reason, Kant (1781) drew a distinction between what appears to us—phenomena—and things as they are in themselves—noumena. Husserl (1913) sought a method for knowing things as they are in themselves and introduced introspection: the careful examination of one’s own conscious experience from the first-person point of view. However, it was soon realised that introspection is a theory-laden and fallible way of observing inner life. We are constantly interpreting what we experience, so introspection amounts to a kind of impromptu theorising, prone to bias and distortion.

Because introspective observations are inherently first-person, their claims cannot be independently checked. Introspection has therefore failed to provide a reliable, intersubjectively verifiably way of accessing experience.

Heterophenomenology—Dennett introduces ‘heterophenomenology’ as a method for taking introspective reports of experience seriously without treating them as infallible access to inner facts. First-person reports should be considered alongside behavioural and neural evidence and analysed as data about what a subject claims to experience. This shifts the focus of the study of the mind towards constructing testable, third-person models of what the subject experiences and believes. In this way, first-person reports remain central to the study of consciousness, but are now open to intersubjective, third-person assessment.

§ The illusion of a Cartesian theatre

In his Meditations on First Philosophy, Descartes (1641) famously separated the non-material mind from the material body. Although many philosophers have since rejected his dualistic view of the mind–body relation, his framework has had a lasting influence on how we think about consciousness. The common picture of the mind is that of a ‘Cartesian theatre’: there is a special place in the brain where all sensory inputs are brought together and observed by a privileged inner observer.

Dennett argues that consciousness is indeed manifested in the brain, but there is no reason to believe that the brain contains a deeper ‘headquarters’ whose activation is necessary or sufficient for conscious experience.

To illustrate this, he considers two explanations of error in experience and memory: imagine your brain is tampered with so that you now vividly ‘remember’ a woman in a hat at Sunday’s party, even though she was never there. When you recall the party, you confidently picture the woman and have no grounds for doubting your memory. There are two ways to interpret this:

• Orwellian explanation. You really experienced the party without her and only later had your memory rewritten.

• Stalinesque explanation. Your brain manufactured the whole scene from the start, so that it was as if you had always experienced her there.

The difference between these explanations is empirically unresolvable: they both yield the same present, vivid but false recollection.

From this, Dennett concludes that the distinction is meaningless and that the idea of a single, privileged moment of conscious experience should be abandoned. This is what he means by his often-misunderstood claim that ‘consciousness is an illusion’: not that experiences aren’t real, but that the picture of consciousness as a special inner show is mistaken.

§ From brains to minds

To build a scientifically grounded theory of the mind, Dennett begins by exploring how our brain’s parallel specialist subsystems evolved from spotting predators and grabbing objects to supporting distinctively human activities, such as thinking ahead and exercising higher-level control over behaviour. These subsystems were not designed for such activities; rather, their plasticity allowed them to be reused for new functions.

As a result of this evolution, the operation of these specialists is organised by several layers of design: some organisation is innate and shared with other animals, some arises from individual habits of thought, and a large part is shaped by culture—memes of language, images, narratives, and social practices that are encoded in, and processed by, the brain.

§ The multiple drafts model

Pandemonium of specialists—Drawing on what modern neuroscience tells us about the brain’s massively parallel architecture, Dennett replaces the Cartesian theatre with the idea of multiple parallel, semi-independent processes distributed across various brain areas at various times. Each process is responsible for specific tasks, such as recognising words or monitoring bodily signals.

Content-fixation—These specialist processes continually fix and revise the content of perception, thought, and memory, composing ‘drafts’ that offer different, sometimes competing, narratives and interpretations of experience. Most of these drafts play short-lived roles in the modulation of current activity, but some are processed, discriminated, and stabilised in ways that allow them to guide behaviour or be reported verbally.

On this picture, there is no single final version; consciousness is simply the subset of drafts that become accessible to our higher cognitive systems—those that shape action, reasoning, and self-report—and this accessibility is what gives rise to the familiar sense of ‘what it is like’ to have an experience.

§ Conscious AI

Thus, according to Dennett, consciousness exists at a higher level of abstraction than individual brain processes. It is the information-processing trajectory that begins with sensory input, undergoes a series of transformations, leaves traces in memory, shapes verbal report and action, and competes with alternative trajectories in guiding behaviour.

The brain as a virtual machine—A virtual machine is a software system that runs on top of physical hardware and can be described in its own terms. Just as a virtual machine is a higher-level pattern in the activity of the computer hardware, Dennett thinks of the mind as a kind of virtual machine implemented on the biological hardware of the brain.

Thus, what matters for consciousness is not the specific biological material, but the functional organisation. It follows from this that, in principle, a suitably ‘programmed’ robot with a silicon-based ‘brain’ could also be conscious.

§ My nitpicks

This is probably the closest I’ve come to being convinced by functionalism, but I still have a few reservations. Dennett tells a compelling story about why we do not need a Cartesian theatre, but does not explain why it nevertheless seems to me as though there is such a theatre. Moreover, brain processes can be described at different levels of abstraction. All these levels have an evolutionary explanation, but not all of them are substrate-independent. Dennett does not really motivate why the information-theoretic level should be privileged as the level at which consciousness is fully captured, rather than one that depends on biological details. Instead, he largely assumes from the outset that consciousness is nothing over and above the activities of higher cognitive systems characterised at that level.

Overall, Dennett's writing is hard to connect to the rest of the discourse on the subject. He relies heavily on analogies and metaphors and, although his arguments have an analytic structure, his writing style often does more to persuade through vivid imagery. I therefore suspect this book will be most useful to readers who already have some background in philosophy of mind, metaphysics, and philosophy of science.

Bibliography

- Dennett, D. C. (1991). Consciousness explained. Little, Brown and Co.

- Descartes, R. (1641). Meditations on First Philosophy.

- Husserl, E. (1913). Ideas: General Introduction to Pure Phenomenology.

- Kant, I. (1781). Critique of Pure Reason.