For: “Plain” Analytical Phenomenology Against: Dennett’s “ Heterophenomenology ”

For: “Plain” Analytical PhenomenologyAgainst: Dennett’s “Heterophenomenology”

The Viability of a Phenomenological Approach to Consciousness I (Lecture 1)

Charles SiewertRice University

Contemporary philosophy of consciousness

• Largely: defense or criticism of attempts to reductively explain consciousness

– Attempts, that is, to identify consciousness with, and explain it as, what is conceived of in terms that don’t presuppose it (e.g., in either information-processing, or physiological terms).

• Typical guiding concern: is the doctrine of physicalism true or false, and what are the implications of our answer?

Contemporary Philosophy of Consciousness—standard options

1. Explain consciousness reductively. Say how we might reductively explain it.) (Frequent strategy: reductive representationialism—“first order” (e.g., Dretske, Tye, Prinz, Hill); “higher-order” (e.g., Rosenthal, Lycan, Carruthers)

2. Marginalize the importance of consciousness.

– Can be combined with 1. (e.g., Geoff Lee)– Or: effort at threat containment. (e.g., Kim)– At the extreme: eliminate altogether the troublesome sense of ‘consciousness.’

(e.g., Dennett, Rey)

3. Show 1 and 2 don’t work. Persistence of “hard problem,” “gap,” “mystery.” (e.g., Chalmers, Levine, McGinn)

My project for the philosophy of consciousness

• Not antireductionist, but nonreductionist: my primary focus is neither on defending nor on criticizing attempts to reductively explain consciousness.

• Rather: initially bracketing such theories, first ask what can we say about consciousness that will help us understand its place in our knowledge, understanding, values and agency.

My research program I: substantive, but open identification of topic: consciousness

II: “phenomenal intentionality” (how experience is related to both sensory object perception and exercise of conceptual capacities)

III. epistemic significance of consciousness (its role in warranting both sense-perceptual, and introspective judgments)

IV: consciousness, mind and meaning

V:. consciousness, agency and selfhood

VI. consciousness and value

VII: examination of (causal, nomological, reductive) frameworks for explaining consciousness in light of I-VI.

Controversial questions we should initially leave open

• How RICH OR POOR in detail is sensory experience/consciousness?

• How is it related to ATTENTION?

• Is sensory experience “INTENTIONAL”? “REPRESENTATIONAL”?

• Does it “REPRESENT HIGHER-LEVEL PROPERTIES”? Confined to “lower” or “intermediate level representations”? How is what we experience through the senses related to our grasp of CONCEPTS?

• Is CONCEPTUAL THOUGHT PHENOMENAL in the same way as sense experience?

• Is introspection based on “INNER SENSE”?

• Is consciousness essentially “SELF-REPRESENTATIONAL”? Or “SELF-REFLEXIVE”?

• Is there experience/consciousness of AGENCY? Of SELF?

My approach: a kind of phenomenology

• Phenomenological Movement. What links together, as a matter of intellectual history: Brentano, Husserl, Stein, Scheler, Heidegger, Sartre, Gurwitsch, Merleau-Ponty, &?.

• My Version of “Analytic Phenomenology.”

A practice of self-critical first-person reflection used to clarify distinctions and show their relevance to theoretical and philosophical issues.

Influenced by and arguably continuous with the phenomenological movement, understood historically

Responsive to issues and arguments in the contemporary analytic tradition.

Contrasts with, e.g., “heterophenomenology,” and other approaches characterized by either a disregard for, or rather different use of “introspection”—or, as I prefer, first-person reflection.

An “underived” and “self-critical” use of “first-person reflection”

• FIRST-PERSON: my phenomenology relies on a way of judging about your own attitudes, thought and experience—and a type of warrant in so judging—distinctive of the first-person case, in consideration of real and hypothetical situations.

• UNDERIVED: It does not assume that the warrant we have for first-person claims about experience is limited to what they can derive from the fact that granting them accuracy best explains “third-person” observational data acquired without reliance on first-person reflection.

• SELF-CRITICAL: it acknowledges the fallibility of first-person judgment about experience, but affirms and tries to strengthen its powers of self-correction.

Main business today

• Examine a prominent, contrasting approach to the study of consciousness: Dennett’s third-personal “heterophenomenology”

• To clarify and make the case for my alternative “plain” (neither “third-personal” nor “pure”) self-critical phenomenology.

What is at stake in these methodological disputes:

How should we seek self-knowledge?

What is Dennett’s methodological view?

• What he calls “heterophenomenology”

• Grounded in criticism of the idea that introspection is infallibile

An alleged demonstration of introspection’s fallibility :

Dennett:

• “The visual field seems to naïve reflection to be uniformly detailed and focused from the center out to the boundaries…

• But a simple experiment show that this is not so.” (Consciousness Explained, p. 53)

The playing card surprise

Fixate straight ahead. Slowly bring the card towards the place you’re staring at.

“You will probably be surprised at how close to center you can move the card and still be unable to identify it.” (CE, 54)

The lesson of this surprise?

“Were you surprised…?

[If so] that must mean that had you held forth on the topic before the surprising demonstration, you would very likely have got it wrong.

People often claim a direct acquaintance with more content in their peripheral visual field than in fact they have.” (CE, 68)

Dennett: evidence of fallibility undermines reliance on introspection

• There is often a “first-person plural presumption” in discussion of what is found in our “streams of consciousness”.

• Problem: disagreement.

• Dennett’s diagnosis:

– “we are fooling ourselves about the high reliability of introspection.” – “the doctrine of infallibility is just a mistake” – We claim to be using powers of inner observation when really we are

theorizing.

What to do about introspective fallibility/disagreement? Shift to “third-person point of view”

• the first-person perspective: “a treacherous incubator of errors.”

• Mental events are “not among the data of science”

• “The challenge is to construct a theory of mental events, using the data that scientific method permits.

• Such a theory will have to be constructed from the third-person point of view since all science is constructed from that perspective.” (CE )

Dennett’s aim: a “neutral path”?• Supposed to provide a neutral description of the data that

does not prejudge the issue Nagel raises (whether we can give a complete account of the world in objective terms).(CE, 71)

• Supposed to be neutral regarding which entities have consciousness.(CE, 72)

• Supposed to be neutral about whether there could be “zombies” in the philosopher’s (Chalmers’ style “conceivability argument”) sense. (CE, 73)

To “heterophenomenology”• From mere sounds, to text, to “speech acts”. View subjects

via “the intentional stance” (CE, 76-77)

• Let subjects’ texts “constitute the subject’s heterophenomenological world”—comparable to the world created in fiction.

• Then we can determine (by theorizing from the third-person point of view) to what extent (if any) the “heterophenomenological world” of the subject corresponds to reality. (See CE, p. 98)

Dennett on “first-person authority”

• “If you want us to believe everything you say about your phenomenology…that is asking too much. You are not authoritative about what is happening in you but only about what seems to be happening in you, and we are giving you total, dictatorial authority over the account of how it seems to you, about what it is like to be you.” (CE, 96)

• “…[T]he facts about what people believe and report when they express their beliefs—are phenomena any scientific theory of mind must account for…The question whether items thus portrayed exist…is an empirical matter to investigate.

• “If suitable real candidates are uncovered, we can identify them as the long-sought referents of the subject’s terms; if not, we will have to explain why it seems to subjects that these items exist.” (CE, 98)

HP: “Heterophenomenology”: the basic idea

The “primary data” for the study of consciousness are subjects’ reports, interpreted as expressions of their beliefs about experience.

HP inquiry is “neutral” or “agnostic” about the truth of any of those beliefs. (pp.82-83)

Subjects are assumed to have “dictatorial authority” regarding what they believe about their experience.

But they are presumed to have no authority regarding the experience itself.

Dennett describes this as granting subjects authority regarding “how it seems” to them, and regarding “what it’s like to be them.” (p.84) Thus, in Dennett’s HP, what it’s like to be a subject is identified with what the subject believes to be true about his or her experience.

Dennett’s NO PRESUMED AUTHORITY thesis. The methodolgical lesson of “Feenoman”

• Heterophenomenology says: Treat our beliefs about our own experience as anthropological observers would treat tribe’s beliefs about their forest god “Feenoman.”

• Scientists would accept without question the tribe members’ reports of what they believe. But the anthropologists would not presume these beliefs are justified to any degree.

• They would only accept what they need to explain the beliefs/practices from a third-person observer’s point of view.

Dennett’s NO PRESUMED AUTHORITY thesis. The methodolgical lesson of “Feenoman”

• In studying consciousness we should proceed with respect to our beliefs about our own experience in much the same way anthropologists would proceed with respect to the tribe’s “Feenoman” beliefs.

• “Only when such a theory [one that explains why we believe what we do about experience] is in place will be able, for the first time, to know what we mean when we talk about ‘the experiences themselves’ as distinct from what we, each subjectively, take our experiences to be.” (CE p. 90)

Crucial thesis of HP: no presumed first-person authority about experience

In HP, subjects are assumed to have “DICTATORIAL AUTHORITY” regarding what they BELIEVE about their experience.

But they are presumed to have NO AUTHORITY regarding the EXPERIENCE itself.

Why accept this methodological claim? Dennett’s argument seems to be:

1. It is the only good way we can recognize and correct introspective fallibility.

2. It is presupposed in scientific practice.

We need to distinguish three theses re FP judgment about experience

• INFALLIBILSM: FP judgments about experience (or some important subclass of these) are such that, when you make them, you can’t be wrong.

• NO PRESUMED FP AUTHORITY (Dennett’s HP): FP reports/beliefs about experience are presumed to have no authority—unless and until their accuracy is confirmed by third-person theorizing.

• DEFEASIBLE FP AUTHORITY (e.g., Siewert): the fact that someone makes a sincere FP judgment about experience gives one some reason to accept its truth, but doesn’t guarantee this: the judgment is still open to rational correction.

Dennett seems to neglect the third

• DEFEASIBLE FP AUTHORITY (e.g., Siewert): the fact that someone makes a sincere FP judgment about experience gives one some reason to accept its truth, but doesn’t guarantee this.

• Compare our treatment of memory: we are entitled to rely on it without deriving its authority in some “memory-free” way—but still we recognize memories are fallible, and correctible by other memories.

Why think that HP is presupposed by science? Dennett’s evidence (“Who’s on First?” (WF)

1. Imagery Research2. Visual Illusion Research3. Research using “masking”

Here: Comment on (2) and (3)

Dennett’s evidence that HP is presupposed by science

In research on visual illusions researchers only assume as data what subjects believe to be the case:

“…[O]f course experimenters on illusion rely on subjects’ introspective beliefs (as expressed in their judgments) about how it seems to them, but that is the agnosticism of heterophenomenology;

to go beyond it would be, for instance, to assume that in size illusions there really were visual images of different sizes somewhere in subjects’ brains (or minds), which of course no researcher would dream of doing.” (WF p.87)

Dennett’s evidence that HP is presupposed by science

3. “Masked priming”—scientific researchers should not (and would not) just assume that the subjects’ were not conscious of the “masked” stimulus, when they deny being conscious of it.

All that should be accepted is that subjects believe that they were not conscious of the stimulus.

One needs to seek evidence as to whether subjects were not conscious of the stimulus, or were conscious of it for an instant but had no memory of it (because of the “masking” stimulus). (WF, p. 89)

Critique of Dennett’s HP: Preview

1. Dennett’s “surprise” argument fails doubly: (a) it fails to show we actually have the false introspective

beliefs he attributes to us, and

(b) it fails to show us how to justify his key methodological principle (“no presumed FP authority”) on the basis of recognizing FP fallibility.

2. His examples from experimental psychology don’t show that scientific practice does or needs to presuppose the “no presumed FP authority” principle essential to HP. Actually, there is reason to believe the opposite.

Critique of Dennett’s HP: Preview

3. The “no presumed FP authority” principle of HP, implies we can grasp the concepts we use to form FP beliefs about experience well enough to seek TP evidence to justify them, without assuming we can accurately apply these concepts in the FP case. If this is false, the practice of HP undermines our grasp of the very concepts it needs to employ.

4. By denying any presumption that we can accurately apply concepts of experience in the FP case, HP would, without justification, de-legitimize critical self-examination of beliefs about experience.

Dennett’s failed argument from surprise

What exactly does Dennett think surprise in the playing card demo shows we would have got wrong had we “held forth” on it? Two possible mistakes:

a) False claims about how well we see shape and colors outside of the focal point when staring fixedly.

b) False claims about how much “content” we have in “peripheral visual field” in normal (active, saccadic) vision.

It seems that Dennett wants to say he shows we make mistake (b) (“people often claim more direct acquaintance with content… than they have”).

Dennett’s failed argument from surprise

a) False claims about how well we see shape and colors outside of the focal point when staring fixedly.

b) False claims about how much “content” we have in “peripheral visual field” in normal (active, saccadic) vision.

It seems that Dennett wants to say he shows we make mistake (b).

But surprise at how poor our peripheral vision is when staring fixedly does not show we have mistaken views about how much we see in normal active vision.

It’s unclear surprise even shows we previously had mistaken views about (a). To be surprised that p does not necessarily require that you previously expected that not p. It might be enough that you did not expect that p.

Also notice: to disconfirm the alleged error, we rely on introspection, and make no apparent commitment to the “no presumed FP authority” principle.

So: Criticism #1 of Dennett• His “simple demonstration” (supposedly) showing the fallibility

of introspection, attributes beliefs to us he offers no good reason to think we actually have. ((“The visual field seems to naïve reflection to be uniformly detailed and focused….”)

• Anyway, he doesn’t show to get from demonstrating the fallibility of introspection, to justifying the “no presumed authority” principle. This is a big gap in the argument.

• The kind of “correction” he illustrates seems to presume the reliability of introspection.

Dennett’s failed argument from Visual illusion Research

When subjects’ say: “That circle looks bigger to me than that one,” don’t researchers accept both

(i) that the subjects believe that one circle looks bigger than the other, and

(ii) that it actually does look bigger to them than the other?

What is supposed to show researchers accept only (i) (the belief report), and then (“heterophenomenologically”) seek out third-person evidence of the accuracy of the subjects’ beliefs (about experience)?

Dennett suggests accepting (ii) would commit researchers to the existence of variously sized circular items in the subjects’ brains (or minds). But this is simply mistaken.

Dennett’s failed argument from Visual illusion Research

Further, suppose researchers don’t accept (ii)—subjects’ reports that one figure looks bigger to them than the other.

Then they simply have no reason for regarding this as a case of visual illusion (rather than introspective error).

But if they do accept (ii) on the authority of subjects’ reports, then they aren’t heterophenomenologists.

D’s failed argument from the “masking” example.

Granted: researchers do not (should not) assume that subjects would be correct in denying they were very briefly conscious of a stimulus.

But this can easily be justified without HP, as follows.

Why we shouldn’t just accept subjects’ denial of disk experience

When there is no masking stimulus, subjects do report seeing the priming stimulus, and we give prima facie credence to their reports then—so there is at least some reason to expect they would see the same stimulus briefly under masking conditions.

Now the subjects’ denial of seeing the stimulus is based entirely on a lack of memory of having done so—but memory is fallible, and the conditions for forming reliable memories are especially poor here. For if there is an experience of the disk, it is very brief and it is quickly replaced by something else that captures attention.

This is reason to be cautious about accepting the truth of subjects’ reports in these particular circumstances. But this reason appeals to the presumption of first-person authority that HP denies.

So: Criticism #2 of Dennett

• HP is not in fact, presupposed in the conduct of experimental research on vision.

• HP’s “no presumed first-person authority” plays no discernible role in the research.

• On the contrary, a presumption of first-person authority seems to be playing a crucial role in his examples (much as Goldman suggests).

How Dennett Might Reply

Even if it’s unclear HP is required for legitimate research about consciousness…

What have we got to lose, if we go with it? What harm would it do, to accept HP? Why should we grant introspection some “prima facie” authority?

“Why not live by the heterophenomenological rules?...Nobody has yet pointed to any variety of data that are inaccessible to heterophenomenology.” (WF, p.92)

Criticism #3: HP could keep us from grasping concepts of experience. Thus it could deny us access to all the data of

consciousness.

Note HP identifies “what it’s like to have an experience” with what a subject believes about her own experience.

But there is a difference between, e.g., the way an itch feels (= what it’s like for one to feel it) and what you believe about itches. The former is the “phenomenology” (phenomenal character) of an itch.

If you don’t get this distinction, you simply don’t get the very idea of consciousness. But HP explicitly obliterates the distinction.

Criticism #3: HP could keep us from grasping concepts of experience (consciousness)

If you don’t get the distinction between, e.g., the way an itch feels (= what it’s like for one to feel it) and what you believe about itches, you

simply don’t get the very idea of consciousness. But HP explicitly obliterates the distinction.

This criticism presupposes I am generally competent in saying “I feel an itch” when I do, and that I can tell my own itches apart from tickles or toothaches and other states that aren’t itches (like beliefs about itches).

Admittedly, HP asks us not to presuppose this competence.

But then it prevents our understanding the very notions whose accuracy it pretends to investigate.

HP risks destroying our grasp of the beliefs it claims to investigate

This example points to a deeper more general problem with HP.

HP presupposes, without justification, that we can understand the concepts we use in forming FP beliefs about experience well enough to seek TP evidence for these beliefs…

without assuming we can accurately apply these concepts in the FP case. But to start, it is strongly counterintuitive to suppose that I could

understand well enough what I meant by ‘itch’ though I was quite incompetent at correctly telling when I feel an itch.

HP risks destroying our grasp of the beliefs it claims to investigate

Suppose: our grasp of the concepts we use in forming FP beliefs about experience depends on our being able to correctly apply these concepts in such beliefs.

Then we cannot understand them well enough for inquiry into the truth of such beliefs, unless we assume those beliefs are not massively false.

But HP forbids us from making that assumption. So HP is self-defeating. Our grasp of these concepts does not depend on the accuracy of these beliefs, only if our

concepts of experience are pure “functional role” concepts.

But such “conceptual functionalism” is, and should be, extremely controversial. It should not be assumed.

And notice: to assume it would be to abandon the “neutrality” Dennett claims for HP (about, e.g., “zombies”). For we would in effect presuppose the inconceivability of zombies.

Criticism #4: HP robs us of the ability to improve and correct our introspection by critical reflection.

Consider the claim Dennett thinks we erroneously hold: “The visual field [is] uniformly detailed and focused from the center out to the boundaries.”

How am I supposed to correct such an error (by, e.g., Dennett’s playing card demonstration) without presuming myself entitled to at least some of my introspective judgments about experience?

Mustn’t I presumed I’m warranted in saying, e.g., “I am now looking straight ahead, not at the card”—”The card now looks nearer to the spot I’m looking at”, etc.)

There is no way to correct alleged errors in some of our first-person judgments about experience through critical self-examination without reliance on other first-person judgments about experience.

It seems HP, strictly followed, would rob us of critical self-examination of FP beliefs about experience.

Heterophenomenology: unnecessary and harmful

• UNNEEDED AS RESPONSE TO FALLIBILITY. First, the argument for HP rests on dubious and unjustified claims about what introspection erroneously tells us. Second, to determine whether we really do believe what Dennett says we do, and to correct the errors we make, we can and do rely on first-person reflection without any HP third-person derivation.

• UNNEEDED TO ACCOUNT FOR EXPERIMENTAL PRACTICE. Experimental study of consciousness does not require HP’S implicit “no presumed first-person authority”—such a claim plays no essential role in the experimental reasoning.

• POSITIVELY HARMFUL. HP (consistently pursued) could harm the study of consciousness by preventing: our understanding the concepts of experience that figure in our first-person beliefs about it; and critical first-person reflection to correct errors in those beliefs.

Points the way to an alternative

Plain, Self-Critical Phenomenology

Rely on first-person judgment about experience in response to “Socratic” questioning,

While granting that such judgment is fallible,

And seeking rational self-correction,

But without pretending to derive its warrant entirely from observational, third-person evidence.