Chapter 23

Consciousness and the Brain — The Hard Problem

Take a slice of liver. Drop it in warm fluid. Give it the right nutrients, and it will start producing bile. That is what livers do.

Now take a slice of brain. Drop it in the same dish. Hook it up to careful instruments. The neurons will fire. Tiny chemical messengers will spill out. You can measure all of it.

But the brain slice will not feel anything. It will not see colors. It will not hear an oboe play. It will not smell fresh bread. It will not be aware — of the dish, of itself, of anything at all.

That little thought experiment is not mine. It belongs to Michael Marsh.¹ He poses it on page 99 of his book. And then he asks the question that has haunted every honest thinker about the mind for at least a century: are we really nothing more than “a pack of neurons,” as Francis Crick once put it?²

Marsh answers, in his way, no — or at least not so simply. The conscious mind is harder to explain than a bile-making liver. Much harder. He admits, quite plainly, that there is “no explanation as to how, or why, consciousness should necessarily arise” from the firing of nerve cells.³

And then — this is the strange part — he goes right on arguing as if it does.

That is the puzzle this chapter is about. It is called the hard problem of consciousness. Once you see it, you cannot un-see it. And whatever else NDEs may or may not prove, they place this problem squarely in front of us and ask us to look.

A. The Critic Acknowledges the Mystery

To Marsh's credit, he does not run from the hard problem. He names it. He attributes the term to David Chalmers, the Australian philosopher who coined it in the mid-1990s.⁴ On page 100, Marsh writes that this is what Chalmers “famously termed the ‘hard problem’ in our attempts to grapple with the nature of conscious-awareness.”⁵

What is the hard problem? Here is the simplest way to say it. Brains process information. We can study how. Brains store memories. We can study how. Brains help bodies move. We can study how. These are what Chalmers calls the easy problems. They are not actually easy in any practical sense — they involve enormous scientific effort. But they are the kind of problem science knows how to chase.

The hard problem is different. Why is there anything it is like to be a conscious being? Why does the firing of neurons produce — or accompany, or coincide with — the felt experience of redness, the taste of coffee, the ache of grief? These felt qualities are called qualia (KWAH-lee-uh). They are the raw “feel” of conscious life.⁶

A thermostat measures temperature. It does not feel hot. A camera processes light. It does not see. A computer runs through chess positions. It does not enjoy the game. Brain cells fire and release chemicals. Why, then, should there be a somebody home — a felt point of view that sees the chess board, tastes the coffee, knows what it is to be alive?

Think of it this way. Even if a brain scientist could one day describe, in perfect detail, every neuron that fires when you bite into a strawberry — every chemical, every electrical signal, every cascade of activation across the cortex — that complete description would still leave one thing unexplained. It would not tell you what the taste of the strawberry is like. The sweetness. The redness in the mind's eye. The little bright pleasure. That felt quality — the qualia — is what the easy problems leave behind. And it is what the hard problem is all about.

Nobody knows. Not after a century of brain science. Not after billions of dollars in research. Not after fMRI machines and electrode arrays and supercomputers running staggering models. We still do not know.

Marsh knows this. He says so on page 99. And he says even more. A petri dish full of cultured neurons, he writes, would not feel cozy when the room is warm or annoyed when disturbed. Neurons in a dish do not have illusions. He concedes that the move from neural firing to conscious experience is a deep mystery.⁷

And yet — here is where it gets strange — Marsh's whole argument against NDEs depends on assuming that consciousness is produced by the brain. Every objection he raises, every neurological mechanism he proposes, every “this lobe explains that experience” argument, rests on that quiet premise. The brain makes the mind. The mind cannot exist apart from the brain. So when patients report consciousness during cardiac arrest — well, they must be wrong about the timing, or wrong about the content, or remembering it badly later. They cannot really be conscious without a working brain.

Notice what just happened. A premise that Marsh himself admits has no explanation just did all the heavy lifting in his argument.

B. Where the Argument Breaks Down

Three weaknesses are worth naming.

The first is the move from we don't know how to but it surely does. Marsh treats the hard problem as a debt that physicalism — the view that everything real is physical or made of physical stuff⁸ — will eventually pay. Like an IOU. Give us another fifty years, give us another scanner, give us another model of neural networks, and we will close the gap. Maybe. But that is not an argument; it is a promise. And the gap is not narrowing. After thirty years of intensive work since Chalmers named the problem, no one is closer to a real explanation of why neurons should produce experience. The IOU keeps getting renewed.

The second weakness is the silent leap from correlation to causation. Brain scans show that certain experiences correlate with activity in certain regions. True. Damage to the visual cortex usually impairs sight. True. Drugs alter consciousness. True. From all these correlations, Marsh concludes that the brain produces consciousness. But that conclusion does not follow from the data alone. It is added on. Correlations show that consciousness and brain activity are linked. They do not show which way the linkage runs, or what kind of linkage it is. We will return to this in section C, because it is here that the older filter model enters the conversation — and changes the picture.

The third weakness is more subtle. Marsh's chapter 5 wants to do two things at once. It wants to admit that consciousness is a mystery, and it wants to proceed with confidence as if the mystery were already solved. Those two moves do not go well together. When you have a mystery this deep at the center of your science, you owe it more than a paragraph or two before pivoting back to physiology. You owe it real weight. Marsh nods at it and moves on. The whole rest of the book is built on what he has just admitted he cannot explain.

C. Following the Hard Problem Where It Leads

Here is the heart of the matter. Once you take the hard problem seriously — really take it seriously, not as a footnote — the NDE evidence stops being a curiosity and starts looking like a clue.

For most of the twentieth century, the dominant assumption in brain science has been what is called the production model. The brain makes consciousness, the way the liver makes bile or the kidneys make urine. Mind is a product of meat. Damage the meat, and the mind frays. End the meat, and the mind ends.

The production model has the appeal of being simple. It also has one enormous problem. Nobody has ever observed it actually happen. Nobody has ever watched a neuron fire and seen consciousness come out. We see the firing. We see the chemical signal. We see the report of an experience. What we never see — what no instrument has ever detected — is the moment where physical activity is transformed into felt awareness. As William James put it more than a century ago, the production of consciousness by the brain, if it occurs, would be as great a miracle as if thought were spontaneously generated out of nothing.⁹

James saw something else, too. He pointed out that there are other ways to think about how mind and brain are related. The brain might not be a producer. It might be a transmitter — like a radio, or a television set, or a prism. The light is not made by the prism. It passes through. The keys of an organ do not make the air; they let the air through, in shaped form.¹⁰

This is the filter/transmission model. Brain damage impairs the transmission. Drugs modify the transmission. Death — perhaps — ends the transmission, but not the consciousness on the other side. The brain is the receiver. The mind is the signal.

The everyday analogy is the television. If your TV picks up nothing but static, the trouble could be in the set. The picture tube might be cracked. The circuits might be fried. But the trouble could just as easily be in the antenna, or in the cable, or in the way the signal reaches the receiver. Take the back off the TV and study the circuits all you want — you will not find the football game living inside the wires. The game is happening somewhere else. The TV only catches it. Damage the TV, and the game does not stop. Only your reception of it stops.

The filter model has been around a long time. The Oxford philosopher Ferdinand Schiller laid it out in 1891, anonymously, under the pen name “Troglodyte.” Henri Bergson developed it. William James preached it.¹¹ In our own day it is defended in serious detail by Chris Carter,¹² by the Kelly group at the University of Virginia in Irreducible Mind,¹³ and by the philosopher Bernardo Kastrup, among others.¹⁴ It does not require that the brain be unimportant — far from it. It requires only that the brain not be the source.

Now here is the part that matters for our book. The two models — production and filter — predict different things about what should happen when the brain is severely impaired during a near-death experience.

Production model: when the brain is impaired, consciousness should be impaired. Severely impaired. When the brain stops working, consciousness should stop. End of story.

Filter model: when the filter is partially damaged, consciousness should appear distorted. But when the filter is deeply impaired — when the brain that normally restricts and shapes conscious awareness is no longer doing the restricting — the signal might come through differently. Perhaps more freely. Perhaps even more clearly than before.

What does the data show?

Not what the production model predicts. Not even close.

When NDErs are asked what their thinking was like during the experience, around eighty percent describe their consciousness as “clearer than usual” or at least “as clear as usual.”¹⁵ Bruce Greyson and his colleagues have shown that this enhanced clarity is reported more often by people who were physiologically closer to death — more often, not less.¹⁶ The closer to flatline, the sharper the reports.

That is not what a dying brain should produce. That is what a freed signal might look like.

There is more. NDErs almost never describe the experience as foggy or dreamlike. The opposite. Again and again, in study after study, they describe it as “more real than real.” Sharper than waking life. Brighter colors. A sense of perception that does not have to grope for the right word or strain to remember a face. If consciousness were a dimming candle being snuffed out by a failing brain, this is the last description anyone would expect to find. But if consciousness is normally narrowed by the brain — trimmed to fit the body's daily survival needs — and if that narrowing briefly relaxes when the body is shutting down, then “more real than real” is exactly the language we should expect to hear.

Pim van Lommel, the Dutch cardiologist whose 2001 Lancet study brought NDE research firmly into mainstream medicine, has been blunt about this. He argues that the evidence requires us to abandon the production model and adopt a non-local account of consciousness.¹⁷ He is not a Christian theologian arguing from prior commitments. He is a hospital cardiologist who watched the data force his hand.

The veridical NDE cases push the same way. We treated them in earlier chapters — the dentures case (Chapter 4), Pam Reynolds (Chapter 5), Vicki Umipeg and the other blind NDErs (Chapter 6). Each adds an experimental check that no production-model explanation has ever managed to absorb. People who should not be conscious are conscious. People who should not perceive perceive accurately. The brain that is supposed to be producing the experience is, by every measurable standard, in no shape to produce anything.

Now — this matters — none of this proves the filter model. The filter model could still be wrong. NDEs could still turn out to be some strange brain artifact we have not yet figured out. The epistemology here is honest. What I am saying is more modest: the filter model fits the data better than the production model does. And it fits in a way that gets stronger, not weaker, as the evidence accumulates.

Chris Carter puts it bluntly. To keep the production model afloat, the materialist must discount large portions of the data — especially the cases of accurate perception during cardiac arrest.¹⁸ He is right. The data fits the filter model. The filter model would be the obvious choice — were it not for the prior commitment to physicalism that Chapter 25 will examine in more detail.

One more thread is worth pulling. The Christian tradition has never required the production model. From the earliest creeds onward, believers have confessed a conscious intermediate state — the soul living between death and resurrection.¹⁹ The biblical case comes in Chapter 26. For now the simple point is this. If the filter model is right, NDEs are not an embarrassment to Christian faith. They are exactly the kind of evidence one would expect to see if Scripture is right about human nature.

D. Two Counter-Objections

A skeptic might press back two ways.

First. “If the brain is just a filter, why does brain damage usually diminish consciousness rather than enhance it? Strokes, dementia, tumors — these reduce mental functioning, not expand it. Doesn't that count strongly against your model?”

Fair question. And the filter model has an answer. A damaged radio does not play better music. It produces static. So we should expect, on the filter model, that ordinary brain damage will normally distort and reduce conscious life — because it disrupts the receiver. The filter model predicts impairment in most cases. What it also predicts is that under certain unusual conditions — when the brain is so deeply impaired that its filtering function is, as it were, suspended — something else might appear. A signal coming through differently. And this is exactly where two phenomena come in: NDEs (the focus of this book) and terminal lucidity, where deeply demented patients regain crystalline clarity hours before death (treated fully in Chapter 9). Both are predicted by the filter model. Neither is predicted by production.

Second. “The hard problem is real, but it is just a problem awaiting a solution. Science has cracked harder ones.”

I respect the optimism. But this is a promise, not an argument. Promises must be paid eventually. After more than three decades since Chalmers framed the problem — and a century since James saw it coming — we are not closer to closing the explanatory gap. We are arguably further from it. New work in integrated information theory, global workspace theory, predictive processing, and so on, has produced richer descriptions of what the brain does. None has explained why there is anything it is like to be the brain doing it.²⁰ At some point the IOU has to be cashed. Or quietly retired.

Where does this leave us?

The hard problem alone does not prove dualism. It does not prove that the soul exists. It does not prove that NDEs are visits to another world. What it does is more limited — and, in another way, more important. It shows that physicalism is not a solved scientific result. It is a working assumption. And like every working assumption, it can be overturned by sufficient evidence to the contrary.

The veridical NDE evidence is exactly that kind of evidence. Combined with the hard problem, it builds a cumulative case the next chapter (Chapter 24) will frame as a positive argument for substance dualism.

For now, hold this in mind. Marsh admits the mystery on page 99. He cannot explain it. Neither can his fellow physicalists. And the data we are studying — the consciousness that flares brightest just where his model says it cannot flare at all — is exactly what the rival model would expect.

Sometimes the simplest thing to do, with a stubborn pattern of evidence, is to follow it.