Dennett’s Consciousness Explained: The Multiple Drafts Model

Michael BrenndoerferUpdated December 2, 202514 min read

Daniel Dennett’s 1991 account of how perceptual processes shape experience without a single point of presentation.

PhilosophyAll chapters
Reading tools Notes & comments
Notes & comments
Part 7 of 12Mind, Self, and Experience
Chapter 85Explaining consciousness without a central observer › Where does a moment of experience happen?
Reading the full chapter · 9 sections · 14 min9 sections · 14 min

All 9 sections · 14 min read

Explaining consciousness without a central observer › Where does a moment of experience happen?Full chapter · Sections in this chapter1 / 9 ↓

Where does a moment of experience happen?

Imagine two small lights just a few centimeters apart. The first flashes red. The second flashes green a fraction of a second later. Under the right conditions, observers report seeing one moving spot that changes color somewhere between the two locations. There was no light in the middle, and the green flash came after the red one. How could the change of color appear to take place before the spot reached the second location?

Daniel Dennett approaches such puzzles by distinguishing when the brain processes information from when the event it represents seems to have happened. In Consciousness Explained (1991), he asks why perception should have a single internal point where everything is assembled into a show. He calls that picture the Cartesian Theater. His alternative, the multiple drafts model, allows different parts of the brain to make and revise discriminations that can have different effects over time. Dennett, Consciousness Explained, chapter 8, pp. 227–231.

The red and green lights give the reader somewhere to start. You can readily ask how the visual system detects the two locations, how it responds to the colors, and how it represents apparent motion. A separate question asks which version was first shown to consciousness. Dennett thinks the temptation to ask it comes from imagining consciousness as something in the brain that needs information delivered to it after the brain has already done the processing. Dennett and Kinsbourne, “Time and the Observer,” §§1.2.A and 2.2.

The Cartesian Theater

A theater is an intuitive picture of what it feels like to look around. You see the far wall, hear someone across the room, and notice the pressure of the chair under you. You can understand why a theorist might suppose that these sources have to send their results to a common destination where they become available to a single observer. That observer need not be an immaterial soul. It could be described as a physical part of the brain. Dennett's argument applies to a materialist who accepts that architecture as readily as to a dualist who assigns the job to an immaterial mind.

The objection is to what a destination by itself could explain. If an internal display contains both the wall and the sounds, someone or something still needs to distinguish the objects, combine the relevant information, and use it to recognize a situation. Giving those jobs to a little observer passes the original problem to a smaller version of the person. The model gets traction only when the jobs are assigned to mechanisms whose operation can be explained without assuming that each is itself conscious. Dennett, Consciousness Explained, chapter 8, §§1–2.

Dennett does allow that information gets integrated. An account of how someone turns in response to their name, for example, might show how auditory information connects to identity, attention, and the control of movement. A particular connection may have a measurable physical location. The claim about the theater is that these paths need not all lead to the same final place or that crossing such a location is what makes their contents conscious. A person can hear a sound, orient toward it, and describe it, with each effect relying on a different subset of what their nervous system was doing. The person performs the resulting actions as a whole.

What the theater assumes

The Cartesian Theater assumes a point of final presentation that marks which contents are conscious and when they became so. A time at which the brain responds to a stimulus, combines two signals, or prepares a response can be defined and measured for a specific task. That measurement does not by itself establish a finish line for experience.

What the multiple drafts model proposes

In the multiple drafts model, information about an event is processed across overlapping paths. A color can be discriminated at a particular place and time in the brain, influence a response, and subsequently have its effects overwritten or supplemented by other information. Several contents can have consequences at once. Their consequences can be detected in what someone says or does, remembered details, and changes in subsequent processing. The drafts metaphor describes processes with different histories. It does not require the brain to store several complete versions of the same scene as if they were files in a folder. Dennett and Kinsbourne, §§1.1 and 2.2.

Consider hearing someone say part of a sentence just as a loud vehicle passes. You may initially take their words as a question and then infer from the next sentence that they were making a suggestion. You might respond appropriately while remembering only your final interpretation. If the person stops halfway through to ask whether you heard, you might reveal the initial interpretation. A real hearing experiment needs to control the words and the noise. The example illustrates the model's proposal that what is available depends in part on which process is sampled and when.

Dennett and the psychologist Marcel Kinsbourne call this sampling a probe. Asking a question is a probe, as is asking someone to signal when they notice a target. Waiting can let relevant details fade or allow an interpretation to replace them. A very early question can interrupt the process the investigator intended to observe. These possibilities belong in a theory of how an experience is described. They do not make self-report dispensable. A report can ordinarily be very good evidence of experience, and the authors explicitly distinguish it from a requirement for having experienced anything. Dennett and Kinsbourne, §§1.1 and 2.2.

Play the same recording for two listeners. Before playback, ask one to signal the order of the singer and the drums as soon as both have entered. Ask the other about the order a minute after both entrances. The first listener may attend differently because of the advance question. The second may draw on what they know of the arrangement. The point is not that one answer will always be wrong. The two procedures may test different things. A method that compares them has more to tell us than a search for the one moment at which the order of entrance was shown on an internal screen.

How the spot changes color halfway across

Dennett and Kinsbourne discuss the color phi finding reported by Paul Kolers and Michael von Grünau. When separated lights of different colors flash in rapid succession, some observers report apparent movement along a path with a relatively abrupt color change between the lights. There is an objective order of physical stimuli and a perceived sequence that includes an event at a location where neither stimulus appeared. A response can represent a color change as taking place along the path once both signals have arrived. No backwards signal from the second flash to the first is needed. Dennett and Kinsbourne, §§1.2.A and 2.2.

The investigators can be interested in how many milliseconds separate the two flashes and how many milliseconds later the observer can use the difference in color. Those are questions about the order and time of physical processes. They can also be interested in where the change appeared to occur along the motion. That is a question about what the observer perceived. Confusing the time a representation is formed with the time represented by it makes the result look mysterious. A newspaper printed today can describe a meeting that happened yesterday. A representation constructed in the brain can likewise organize content into its own temporal sequence.

Now suppose there really is a theater. One story says the brain waits for the second flash, works out the movement and the color change, and displays the edited version. Dennett and Kinsbourne call this Stalinesque revision. Another says the first flash is displayed separately, then the second flash arrives and alters the observer's memory. The final report describes motion and a color change, while the observer has lost access to the original experience. They call this Orwellian revision. The labels designate where the editing falls relative to the supposed moment of presentation. Dennett and Kinsbourne, §2.2.

The two stories can lead to the same answers about what was seen. If an earlier probe reveals that a discrimination was already made, that result tells us something about processing. What it establishes about presentation still depends on how the theater boundary is defined. The multiple drafts model keeps any measurable difference in the processing and drops the requirement for a further verdict about which of two versions was presented first when nothing distinguishes them in the relevant respects. The experiment supplies a case in which an assumption about the format of consciousness makes an otherwise intelligible event seem inexplicable.

What a report can tell us

A person says a wine tastes a little like grapefruit. This gives a scientist something to explain. The vocabulary might depend on earlier tasting experience, the comparison the person was just asked to make, or a distinction they had learned to notice. Dennett's method of heterophenomenology takes the person's description seriously as evidence of how things seem to them. It leaves open the theoretical interpretation they might give of what produced that impression. In chapter 8 the fictional Otto thinks he compares several words to an experience internally before choosing the best one. Dennett asks how an account of speech could explain this feeling of selection without assuming an inner speaker who has already composed the words. Dennett, Consciousness Explained, chapter 8, pp. 229–231.

An important distinction here is between acknowledging the report and treating the person's account of its own origins as settled. Otto might say that he considered the grapefruit and berry descriptions and found one more apt. A theory can acknowledge that account of how the comparison seemed and still investigate what influenced the choice. Speech errors give the inquiry something concrete to work with. Dennett discusses familiar slips in which sounds exchange positions or two candidate words blend. Such errors may disclose processes that organized the utterance without having been available for the speaker to describe in advance. Dennett, Consciousness Explained, chapter 8, §2.

A critic might think this approach offers tools for examining reports while leaving their felt character unaccounted for. Dennett's reply is to ask how the critic is identifying the felt character and what explanatory work remains after the full account of experience is given. That disagreement should survive a fair description of the method. Collecting reports and investigating their causes is already possible without deciding the metaphysical question. The point where the theories diverge is what an account of those processes could leave out, even if it were complete by its own standards.

Qualia and the flavor of coffee

Dennett takes up a particularly strong conception of qualia in his companion essay “Quining Qualia.” On this conception, an experience has properties that are intrinsic, wholly private, ineffable, and immediately known to the subject. Dennett questions whether the cluster of requirements holds together. Disputants sometimes use qualia more broadly for the qualities experience seems to have. Under that broader use, asking whether you can sense bitterness is a different question from asking whether bitterness as experienced must be a wholly private property, independently of how it affects your judgment and comparisons. Dennett, “Quining Qualia,” §§2 and 4.

His two imaginary coffee tasters, Chase and Sanborn, used to like the same coffee and now dislike it. Chase thinks its flavor has stayed the same while his taste in coffee has changed. Sanborn thinks his preferences are unchanged and the coffee tastes different to him. Both offer recognizable descriptions of their lives. Dennett asks what would establish the contrast each draws. They each have to compare what the coffee tastes like today with a memory and with judgments learned over years. Those comparisons can be affected by several changes at once. The felt certainty each taster expresses is evidence of how things strike them, but it does not by itself specify which mechanism changed. Dennett, “Quining Qualia,” §4, intuition pump 7.

There is room for a reply. A defender of phenomenal qualities can agree that memory and comparisons are fallible and still hold that the coffee has a distinct character right now. The existence of such a character need not depend on infallible access. Dennett's challenge becomes sharper if that character is then defined as something wholly independent of anything the taster can notice, learn, remember, or do. Which part of the taster's situation would motivate a theorist to include that property? A critic of Dennett can press an equally direct question. Does having an account of when a person can discriminate flavors and what they can say about them tell us why the taste feels as it does?

A person without a central spectator

The attack on a central observer need not erase the person whose experiences matter. Dennett develops that distinction with his comparison between a self and a center of gravity. Mechanics describes the behavior of a complicated moving object by using its changing center of gravity. That point can help explain and predict what the object does without adding a new moving part to the object. Treating a life as a history of someone's actions, commitments, stories, and changes in outlook can likewise explain patterns that a list of individual movements would leave obscure. A center of narrative gravity is the common subject under which those patterns make sense. Dennett, “The Self as a Center of Narrative Gravity,” opening discussion.

Having a point of view then involves activities that can be studied, including remembering that something happened to you, picking it out when asked, and using it to decide what to do. The character of a life can organize those activities without someone sitting apart from them receiving the product. The center-of-gravity comparison describes the explanatory role of the self. It does not explain every way a person's memory or practical reasoning works. A theory still has to identify the processes that allow those patterns to persist, break down, and develop.

Where the argument leaves experience

Thomas Nagel's question about what it is like to be an animal can be asked without imagining a theater. Likewise, David Chalmers's later distinction between functional problems and the hard problem does not require him to posit a central inner spectator. For Chalmers, explaining discrimination, information use, and report still leaves the question of why such activity is accompanied by experience. Dennett invites us to see whether a careful account of the varied things organisms perceive, say, and do gives us reason to think there is that further explanatory target. His reply requires more than pointing out the absurdity of a spectator inside a head. Chalmers, “Facing Up to the Problem of Consciousness,” §§2–3.

The strongest place to test the disagreement is a detailed case. In color phi, varying how and when people are asked about the flashes can reveal real differences in how information is used. Researchers can look for processes that register a color without affecting a later response, or for information that persists despite a competing signal. These are sensible questions on Dennett's view. A critic may agree that a single finish line contributes nothing to the experiment and maintain that the appearance of motion still needs a phenomenal explanation. Distinguishing what an observer sees from how it came to be represented clarifies the point at which the two accounts disagree.

Further reading

Join the discussionAsk a question or share what you learned.

Comments

Add to the conversation

A question, a correction, or an example from your own work.

No comments yet. Be the first to share your thoughts!

A different angle
All writing
Philosophy

Aristotle's De Anima: Soul, Perception, and Intellect

Aristotle treats soul as the form of a living body and asks how living things perceive, move, and think.

14 min read
Philosophy

Husserl's Cartesian Meditations: Transcendental Phenomenology

Edmund Husserl's five meditations on transcendental phenomenology, from the epoché to the experience of other minds and a shared world.

14 min read
Philosophy

Husserl's Ideas I: The Epoché and Transcendental Phenomenology

Edmund Husserl's 1913 systematic presentation of transcendental phenomenology, introducing the epoché (bracketing) and eidetic reduction.

13 min read

About the author

Reference

Citation details

Cite or share this article.

BIBTEXAcademic
@misc{brenndoerfer2025dennetts, author = {Michael Brenndoerfer}, title = {Dennett’s Consciousness Explained: The Multiple Drafts Model}, year = {2025}, url = {https://mbrenndoerfer.com/writing/dennett-consciousness-explained-eliminativism-multiple-drafts-guide}, organization = {mbrenndoerfer.com}, note = {Accessed: 2026-11-19} }
APAAcademic
Michael Brenndoerfer (2025). Dennett’s Consciousness Explained: The Multiple Drafts Model. Retrieved from https://mbrenndoerfer.com/writing/dennett-consciousness-explained-eliminativism-multiple-drafts-guide
MLAAcademic
Michael Brenndoerfer. "Dennett’s Consciousness Explained: The Multiple Drafts Model." 2026. Web. November 19, 2026. <https://mbrenndoerfer.com/writing/dennett-consciousness-explained-eliminativism-multiple-drafts-guide>.
CHICAGOAcademic
Michael Brenndoerfer. "Dennett’s Consciousness Explained: The Multiple Drafts Model." Accessed November 19, 2026. https://mbrenndoerfer.com/writing/dennett-consciousness-explained-eliminativism-multiple-drafts-guide.
HARVARDAcademic
Michael Brenndoerfer (2025) 'Dennett’s Consciousness Explained: The Multiple Drafts Model'. Available at: https://mbrenndoerfer.com/writing/dennett-consciousness-explained-eliminativism-multiple-drafts-guide (Accessed: November 19, 2026).
SimpleBasic
Michael Brenndoerfer (2025). Dennett’s Consciousness Explained: The Multiple Drafts Model. https://mbrenndoerfer.com/writing/dennett-consciousness-explained-eliminativism-multiple-drafts-guide
Newsletter

Stay up to date

Get new articles, book updates, and notes on AI and technology.

Unsubscribe at any time. Privacy policy