Lecture: The Mind-Body Problem
Lecture script · about 15–20 minutes spoken · Week 6
Opening
Picture someone at a clinic telling a nurse that a healed burn still stings. The skin looks fine. A brain scan could show which areas are busy. But could any scan show what that sting feels like from the inside? Hold that question. This whole lecture grows out of it.
It isn't only a question for clinics. A nursing student asking how the pain a person reports relates to a scan is asking it. So is a computer science major wondering what a help-desk chatbot really understands. Each of them is asking what a mind is, and whether a machine could have one.
In Week 1 you learned to test a claim with a counterexample. In Week 2 you learned standard form, validity, and which forms are valid. Weeks 3, 4, and 5 asked what is real and what you can know. In this lecture I'll turn those questions on the knower. We'll meet three answers about what a mind is, and test each one with a famous case. Next week, in Week 7, we ask what makes you the same person over time, and whether you're free.
1. The mind-body problem
Think about the last time you wanted a snack. A wish showed up, and then your legs carried you to the kitchen. The wish seems to be in your head. Your legs are in the room. How does the first thing move the second?
That's the puzzle in its oldest form. The mind-body problem is the problem of how mental states, such as pains, thoughts, and wishes, are related to the physical body and brain. And a mental state is any state of mind a creature can be in, such as feeling a pain, believing a claim, or wanting something.
The course pins three standard answers. Substance dualism says mind and body are different kinds of thing. The identity theory says mental states are brain states. Functionalism says mental states are defined by their causal roles. I'll state each one the way its best defenders would, and only then raise the hardest objection.
Pause and try this. A friend says the mind-body problem means nobody has any idea what a mind is. Is that fair?
Think about what you heard a moment ago: three answers.
Here's the answer. It isn't fair. There are well-worked answers with real support; the problem is that each one faces a hard objection.
2. Substance dualism and the problem of interaction
Try this. Think of the number seven. Now, how big is that thought? How much does it weigh? The questions sound strange, because thoughts don't seem to have sizes or weights at all.
That's where substance dualism starts: the view that mind and body are different kinds of thing, so that a mind is not a physical object and is not made of the matter that makes up a body. In its strongest form, it points to two things. Thoughts lack size, shape, and weight. And you know your own pain from the inside in a way no scan can match.
Now the hardest objection. The problem of interaction is the objection to substance dualism that asks how a mind that is not physical could cause changes in a physical body, or be changed by it. You decide to wave, and your hand goes up. If the decision has no size, no place, and no push, how did it lift your hand?
The dualist has a reply. A puzzle about how a cause works isn't proof that it doesn't happen. The link between mind and body might be a basic fact with no further explanation.
Pause and try this. Does the problem of interaction show that substance dualism is false?
Separate "how does it work?" from "does it happen?"
Here's the answer. No, not by itself. It asks how a mind could move a body, and the dualist replies that not knowing how is not proof that it doesn't happen.
3. Physicalism and the identity theory
For most of human history, people saw lightning and had no idea what it was. Then it turned out to be an electrical discharge. Not a flash plus a discharge. One thing, described two ways.
Physicalism is the view that everything that exists is physical, so that minds, too, are made of nothing beyond the physical world. Its best-known form for the mind is the identity theory: the physicalist view that mental states are brain states, so that a pain and the brain process that goes with it are one thing described in two ways.
Here's its strongest case. A blow to the head, a drug, or a night without sleep changes how a person thinks and feels. That's what you'd expect if mental states are brain states. And there's no problem of interaction, because a brain state can send signals down nerves to a hand.
Pause and try this. Does the identity theory say pain isn't real?
Go back to lightning.
Here's the answer. No. Finding out that lightning is a discharge didn't make lightning unreal, and the identity theorist says pain is real and is a brain state.
4. An argument against the identity theory
Here's an argument from the course's argument set. It's illustrative: its premises were written for teaching. If the mind is the brain, then mental states have locations. Mental states have no locations. So the mind is not the brain.
Let's check the form. The first premise says if P then Q. The second says not Q. The conclusion says not P. That's denying the consequent, from Week 2, and it's valid. If both premises were true, the conclusion couldn't be false.
But is it sound? That turns on the second premise. An identity theorist says a pain is located where its brain state is. A dualist says it makes no sense to put a thought two inches behind your eye. That's an open question, so the course keys this argument for validity, and never for soundness.
Pause and try this. Someone says, "That argument is valid, so the identity theory has been refuted." What's the mistake?
Ask Week 2's two questions, one at a time.
Here's the answer. Valid only means the conclusion would follow if the premises were true. The identity theorist denies the second premise, so nothing is settled yet.
5. Nagel's bat
Picture bats pouring out of a barn at dusk, swerving past the beams in the dark. Many bats find their way by sending out sounds and sensing the echoes. Now try to imagine what that's like for the bat. Not for you hanging upside down. For the bat.
That's the question in the title of Thomas Nagel's paper, "What Is It Like to Be a Bat?", from 1974. Nagel argues that physical accounts of the mind leave out the subjective character of experience: what an experience is like for the creature having it, from its own point of view.
Here's why it presses the identity theory. Suppose you learned every fact about a bat's brain. That list is written from the outside. It still doesn't seem to tell you what the echoes are like from the inside. If mental states are brain states, a full brain account should leave nothing out.
The physicalist replies that this shows a limit of our point of view, not a gap in the world. We can't take up a bat's perspective, because we don't have a bat's brain.
Pause and try this. Is Nagel's point that bats need more study?
Ask whether more outside facts would close the gap.
Here's the answer. No. The point would hold for any creature, even another person, because more study from the outside doesn't seem to give you the view from the inside.
6. Functionalism
Think about a goalkeeper. A goalkeeper can be tall or short, young or old. What makes someone the goalkeeper is the job: stand in the goal and stop shots. The role is defined by the job, not by the body filling it.
Functionalism is the view that mental states are defined by their causal roles, by what they do rather than by what they are made of, so that anything playing the right role, brain or machine, could be in the state. A causal role is the pattern of causes and effects that marks out a mental state: what typically brings it about, and what behavior and other states it typically leads to.
Take pain. Bodily damage causes it. It causes distress, the belief that something's wrong, and the urge to protect the hurt spot. On functionalism, pain is whatever plays that role. In a human, a brain process plays it. In an octopus, or a well-built machine, something else might.
Pause and try this. A classmate says, "Functionalism is another way of saying the mind is the brain." Is that right?
Think about the goalkeeper again.
Here's the answer. No. Functionalism says what matters is the role, and a brain is only one thing that could fill it; a machine or a very different creature could fill it too.
7. The Turing test
Imagine a game played by text messages. You chat with two hidden partners. One's a person. One's a machine. Your job is to tell which is which.
That's the shape of Alan Turing's imitation game, now called the Turing test, which I'm describing in outline: a questioner trades typed messages with a hidden machine and a hidden person and tries to tell which is which. Turing proposed it as a way to ask whether a machine can think.
It fits functionalism well. If a mind is defined by what it does, then a machine that does what a thinking person does in conversation looks like a strong candidate. But notice what the test checks. It checks typed replies. That's all.
Pause and try this. If a machine passed the Turing test, would that show it's conscious?
Think about what the questioner actually gets to see.
Here's the answer. Not by itself. The test checks conversation, not consciousness, so passing it wouldn't show the machine feels anything, which is Nagel's question again.
8. The Chinese room
Now the case at the center of this week. John Searle, in a 1980 paper called "Minds, Brains, and Programs", describes the Chinese room: a person who knows no Chinese follows rule-books for matching Chinese symbols and passes out fluent answers. Questions come in through a slot. The person looks up the symbols, follows the rules, and passes answers back out. Outside, the room seems to understand Chinese. Inside, the person understands none of it.
The case asks: is following a program the same as understanding? Here's the course's reconstruction. First premise: the person runs a program for answering questions in Chinese. Second: the person doesn't understand Chinese. Third, a premise we add: if running the right program were enough for understanding, anyone who runs it would understand. Conclusion: so running the right program isn't enough for understanding. That's denying the consequent, so it's valid.
The functionalist answers with the systems reply: the person is one part of a larger system of person, rule-books, and room, and the whole system, not the person alone, understands. A brain understands, though no single cell does.
Pause and try this. Does the Chinese room show that machines can never be intelligent?
Read the conclusion word for word.
Here's the answer. No. It says running the right program isn't enough, on its own, for understanding; the room answers very well, so it says nothing against machines answering well.
9. The help-desk chatbot
Back to the chatbot from the opening. Suppose it answers in Chinese for a customer who writes in Chinese. Here's another argument from the course's set, also illustrative. If a machine understands Chinese, it can answer questions in Chinese. This machine answers questions in Chinese. So it understands Chinese.
Check the form. If P then Q. Q. So P. That runs the conditional backward, from the consequent to the antecedent, and it's invalid. Try the same shape with new content. If a person is a chef, they can boil an egg. This person can boil an egg. So this person is a chef. A kid making breakfast makes both premises true and the conclusion false.
Pause and try this. Since this argument is invalid, does that prove the chatbot doesn't understand Chinese?
Remember what "invalid" tells you, and what it doesn't.
Here's the answer. No. An invalid argument fails to support its conclusion, but the conclusion could still be true or false; this argument doesn't tell you which.
10. Jackson's Mary
Picture a scientist named Mary. She knows every physical fact about color vision. But she's lived her whole life in a black-and-white room. One day the door opens, and she sees a red apple.
That's the case from Frank Jackson's paper "Epiphenomenal Qualia", from 1982. It asks whether Mary learns something new when she first sees red. It's about qualia: the felt qualities of experience, such as how red looks or how a sting feels to the one who has it.
Many people feel sure she learns what red looks like. But she already knew every physical fact. If she learns a new fact, then some fact about experience isn't physical. That presses the identity theory and functionalism alike.
The physicalist replies that Mary gains a new ability, to recognize and imagine red, or a new way of knowing a fact she already knew. Think of knowing a city from a map, and then walking its streets.
Pause and try this. Is the point of the case that Mary didn't study hard enough?
Look at what the setup says she knows.
Here's the answer. No. By the setup, Mary knows every physical fact, and the case asks whether that could still leave something out.
11. The three positions side by side
Let's line them up. Substance dualism says mind and body are different kinds of thing, and the problem of interaction presses it. The identity theory says mental states are brain states, and Nagel's bat and Jackson's Mary press it. Functionalism says mental states are defined by their causal roles, and Searle's Chinese room and Jackson's Mary press it.
Count the cases pressing them: one for dualism, two for the identity theory, two for functionalism. That's five entries, with Mary counted twice. And the dates sit close: Nagel in 1974, Searle 6 years later in 1980, Jackson 2 years after that in 1982, 8 years in all.
Each view also has a reply. Here's where they disagree, in one plain sentence: is the mind a separate kind of thing, a particular kind of brain process, or any system that plays the right roles? The course doesn't key a verdict.
Pause and try this. Does the view with the most pressing cases have to be the weakest one?
Ask what a count can and can't tell you.
Here's the answer. No. The count says how many famous cases a view faces, not how good its replies are, and every position here has a serious reply.
Closing
Let me pull it together. The mind-body problem asks how mental states relate to the body and brain. Substance dualism says they're different kinds of thing, and it faces the problem of interaction. The identity theory says mental states are brain states, and the argument about locations is valid but not settled.
Nagel's bat asks whether a physical account leaves out what experience is like from the inside. Functionalism defines a mind by what it does. The Turing test checks conversation, not consciousness. The Chinese room asks whether following a program is understanding, and the systems reply answers back. The chatbot argument runs the conditional backward. And Mary asks whether knowing every physical fact could still leave something out.
Next in this module come the slide deck, the Readings and Resources page, the lecture tutorial, the practice set, the quiz, the discussion, and the assignment. Then, in Week 7, Personal Identity and Free Will, we ask what makes you the same person over time.
Built with The Course Maker · samples.thecoursemaker.com