The Heterophenomenological Fallacy, Part 4: Conclusion

(Sorry, Dear Reader, I couldn't come up with a good title for this post.)

I hope I have shown that Heterophenomenology is not taking us Philosophers of Mind one step closer to the Hard Scientists, but actually one step further away.  What could justify this regression?

I think that the major intuition that makes Heterophenomenology so appealing is that experiences are (supposed to be) Subjective, whereas the objects studied by physics, chemistry and biology are supposed to be Objective.  I did my basketball experiments alone.  No one else was watching, or at least, no one else was paying attention.  But *in principle* others *could have* been watching, and those others would have seen the *very same* basketball that I saw.  The entire Scientific Community could have been watching, in principle if not in practice.

I agree that this is a real difference between physics and phenomenology, a true difference between images of pencils and real basketballs.  However, I do not agree with the conclusions that Dennett and the Heterophenomenologists draw from this difference.

In practice, the entire Scientific Community could not have seen my basketball, and in fact, the Scientific Community has no interest in *my* basketball.  What the scientific community really cares about is the universal basketball.  The universal basketball is intersubjective in exactly the same way that seeing double is intersubjective.  The Law of Falling Basketballs is intersubjective in exactly the same way as The Law of Double Pencils is intersubjective.

The Heterophenomenological Fallacy, Part 3: Are the College Students Really Necessary?

To make my point clear, I will return to the story of the diverging pencils, and retell the story so that it is close as possible to the basketball story.  I will retell the basketball story, keeping all the details constant, swapping in the pencils experiment for the basketball experiment.

One day, I notice that pencils often produce diverging images when I hold them close to my eyes.  I decide to study the phenomenon more carefully.  I stand in front of a wall and hold a pencil vertically against the wall.  I focus on the wall and slowly move the pencil closer to my face, between my eyes.  Sure enough, the two pencil images diverge as the pencil moves closer to my face.

I repeat the experiment ten more times, and each time, the images of the pencil diverge.  I seem to have stumbled on a Law of Nature, “When a person brings a pencil to his nose, he sees two diverging pencil images.”  (This is not a fundamental law, of course, but it is a law.)  I publish a paper with my findings.

Other scientists read my paper and get interested.  But they don’t just take my findings on faith: they either see if they can reproduce my results, or if they are busy and lazy, they hope that others will try to reproduce the results.  Ten scientists repeat my experiment, and they all get the same results: the images diverge every time.  Satisfied, The Scientific Community arrives at Consensus: when a person brings a pencil close to his nose, he really, really sees two diverging images.

Interestingly, in the new story of the diverging images, the college students are nowhere to be found.  This is as it should be.  Real world experiments in hard science do not round up a simple random sample of college students to do the experiment; they simply employ how ever many experimentalists are practically necessary to carry out the experimental procedures and measurements.

As Dennett points out in "Intuition Pumps", when phenomenology was first proposed by Husserl and others, the college students were not part of the deal; the phenomenologists were writing about their own experiences.  In my opinion, this is how it should be.

The Heterophenomenological Fallacy, Part 2: Is Heterophenomenology Scientific?

At first, heterophenomenology sounds very clean and elegant, and most of all, scientific.  But is it really scientific?

For comparison, let's take a toy physics experiment that we can compare with our toy phenomenology experiment.  Let’s say I go outside with a basketball and throw it up in the air.  Before long, the basketball comes down.  Now I want to know: was the basketball’s return to earth some kind of happenstance?  Do basketballs always come back down, or do they sometimes just keep going up?

So I throw the basketball up ten more times, and each time, it comes back down.  I start to get excited; I seem to have stumbled on a Law of Nature, “When a person throws a basketball up, it always comes back down.”  (This is not a fundamental law, of course, but it is a law.)  I publish a paper with my findings.

Other scientists read my paper and get interested.  But they don’t just take my findings on faith: they either see if they can reproduce my results, or if they are busy and lazy, they hope that others will try to reproduce the results.  Ten scientists repeat my experiment, and they all get the same results: the basketball comes down every time.  Satisfied, The Scientific Community arrives at Consensus: when a person throws a basketball up, it really really must come back down.

Let’s look at the basketball story from the point of view of philosophy of science.  What is the explanandum?  Is the explanandum (A) “Ten experimentalists threw basketballs up in the air ten times, and each time the basketball came down” or (B) “Ten experimentalists *reported that* they threw basketballs up in the air ten times, and each time the basketballs came back down”?  I submit, Dear Reader, that it is A, not B.  The explanandum is the event, not the experimentalist’s report of the event.

What is true of basketballs is true of all real experiments in physics, chemistry and biology, those that are being performed today and all those that have been performed since the invention of Real Science.  Theorists try to explain the physical outcomes of experiments, not the reports of the experimentalists.

The Heterophenomenological Fallacy Part 1: Heterophenomenology Explained

I've written a lot about heterophenomenology (also known as "the third person perspective") and I'm going to write some more, so I decided to dedicate a post to explaining heterophenomenology in my own words.

Heterophenomenology is kind of like experimental psychology.  You get a simple random sample of college students (of course a simple random sample of human beings would be better, but more expensive) put them in some well-defined situation, and ask them what they are experiencing.

As a toy example, take the following experiment.  You tell the student to stand in front of a wall and hold a pencil vertically against the wall.  Tell the student to focus on the wall and slowly move the pencil closer to his face, between his eyes.  Ask him to describe what happens to the image of the pencil.  (Hint: in this situation, the student should be seeing double, and the two pencil images should diverge as the pencil moves closer to his face.)

So far this is all very straightforward.  The novelty of heterophenomenology is in how the theorist interprets the data gathered by the experiment.  According to heterophenomenology, the data gathered by the experiment is the report of the experience, not the experience itself.

In the case of the pencil, the data collected by the experiment is the fact that the student *reported* seeing double, not that the student *actually* saw double.  About whether the subject actually experienced what he reported experiencing, the theorist is meant to be agnostic, at least initially.  A successful explanation must account for the fact that the subject reported the experience, but not the fact that the subject actually had the experience.  In theory, the theorist is allowed to conclude that the experience report was accurate, if such a conclusion is warranted; but the theorist is not allowed to assume that the report was accurate.

Dennett compares heterophenomenology to anthropology.  The theorist is like an anthropologist interviewing a tribe of hunter gatherers about their system of mythology.  The anthropologist’s data is not the ghosts and goblins themselves, but rather the reports of the hunter gatherers about those ghosts and goblins.

In philosophical terms, the report of the experience is an explanandum, but the experience itself is not an explanandum.

Shabbos Menucha: A Hasidic Parable by MC Complete

Shalom and Menucha were a newlywed Hasidic couple.  They were very happy.  But there was one problem.  Menucha worked very hard to make delicious Shabbos dinners.  But Shalom was so exhausted on Friday night from working and learning Torah all week, that he said Shalom Alechem, Eshses Hayil, Kiddush, Hamotzi, Birkat Hamazon, and promptly collapsed in bed.

This made Menucha so sad that she couldn’t eat any of the food.  She sadly put it all in the refrigerator and served it as leftovers during the week.

It went on like this for a few weeks.  Then one day, Shalom came home with his face glowing.  “My dear wife, you won’t believe it,” he said.

“My dear husband, what is it?” Menucha asked.

“The Rebbe has invited me to learn with him bechevrusa!”

“My dear husband, I am so happy for you!”

When the highly anticipated hour arrived, Shalom was bouncing off the walls with excitement.  The Rebbe greeted Shalom warmly, took two Gemaras off the shelf, opened them to the page, and launched into the most brilliant shiur that Shalom had ever heard.

At first, Shalom sat riveted to his seat, drinking in the words of Torah.  But after a few minutes, something very strange happened.  Shalom began to feel sleepy.  His eyelids started to droop, his head started to nod.  Shalom didn’t understand.  The shiur was so fascinating; why was he struggling to stay awake?

Then Shalom picked up his head and looked at the Rebbe.  He realized that he had fallen asleep with his head in his arms on the Gemara.  He was mortified.   He didn’t know what rock to crawl under.

“What’s the matter, Shalom?” the Rebbe asked.  “Are you embarrassed that you fell asleep in the middle of our chevrusa?”

“Yes,” Shalom admitted.

“If you felt embarrassed to fall asleep while I was feeding you Torah,” the Rebbe said, “how much more so should you feel embarrassed to fall asleep while your wife is feeding you dinner.”

Why I am not a Third Person, Part 2: Papal Infallibility

In Chapter 4, Dennett presents an interesting objection to the first person perspective.  Doesn’t the first person perspective award “papal infallibility” to peoples’ reports of their phenomenology?  Since “introspection” would not seem to be infallible, shouldn’t we avoid it, and put our conclusions on firmer footing?

To answer this objection, first of all, I would distinguish between infallibility and reliability.  It’s possible for a reporter, or a spy, or a research assistant to be extremely reliable but never, of course, infallible.

We treat the testimony of a reliable source differently from the testimony of an unreliable source.  We rely on the testimony of a reliable source, though we are always prepared (in the back of our minds) to reject that testimony in the face of overwhelming evidence to the contrary.  We don’t ignore the testimony of an unreliable source -- the source may be unreliable, but he did say what he said -- instead, we treat it as data that can be accounted for or explained.  Dennett is asking us to treat peoples’ reports of their experience as data, which means that he is asking us to treat people as unreliable witnesses of their own states of mind.  But is that really a good idea?

I think a good analogy here would be memory.  Memory is not infallible, but if we were not to rely on our memories, we would not be able to do anything; neither philosophy, or science, or surviving for more than a few hours.  We know that memory is not infallible, but we assume it is reliable, and we rely on it.

Why I am not a Third Person, Part 1: Loading the Dice

Chapter 4 of Consciousness Explained by Daniel Dennett is called “A Method for Phenomenology”.  The essence of Dennett’s method is “the third-person perspective”.

The “third-person perspective” forbids the philosopher from drawing any conclusions from his own subjective experience.  The idea is to study phenomenology by listening to people’s reports of their experience, and treat those reports as data to be accounted for, or explained.

Dennett’s third person perspective sounds clean and disciplined, but there’s something funny about it.  Third person philosophy takes people’s accounts of their phenomenology and asks the following question: “Are the subjects actually experiencing the states of mind that they report, or are they storytelling machines who are designed to generate stories about states of mind?”

Once you ask the question that way, the answer is obvious.  Option 2  is quite plausible (as long as you ignore the fact that you are experiencing similar states of mind).  It’s consistent with our ideas about physics and computer science.  It’s tidy and coherent.  Option 1, on the other hand, seems to raise troubling questions.

As Dennett puts it in Section 4.6, “Isn’t it unimaginable that scientists might discover neurophysiological phenomena that just were the items celebrated by subjects in their heterophenomenologies?”  This is a somewhat roundabout way of saying that materialism is incompatible with the idea that people are actually experiencing what they say they are experiencing; which is equivalent to saying that materialism is incompatible with the idea that I am actually experiencing what I “seem” to be experiencing.

Once you have chosen option 2, all that is left is to haggle over the price, that is, to explain why the designer (e.g. God and/or natural selection), would have designed people to appear conscious when they are actually not conscious.  This is an inherently speculative task, and it’s not surprising that an imaginative person can come up with a compelling answer that’s hard to argue with.

If the materialist can convince the dualist to take the third-person perspective, the dualist has not just agreed to play in the materialist’s court, he has essentially conceded defeat.  The third person perspective is not just a method, it’s a solution.  It solves the problem simply by refusing to consider the problem.