DYNOMIGHT ABOUT RSS SUBSTACK

The hard problem of feelings

The hard problem of feelings

Apr 2023

Here’s something weird. At least, I think it’s weird.

Consciousness

The hard problem of consciousness is why it feels like something to be alive. Physics does a good job of explaining everything that happens in terms of fields and atoms and whatnot—you’re eating a burrito because moments after the big bang, matter formed, and then planets, and then life forms, which started eating each other, and then there was an arms race that led to your brain trying to put certain nutrients into your body so you can propagate your genes.

That’s neat, but why does it feel like something to be you? There seems to be no need for consciousness. Physics wouldn’t care if we were all “zombies”. Why aren’t we?

I like to look at it this way:

  1. We are alive.
  2. We are conscious.
  3. We were created by evolution.
  4. But consciousness can’t “do” anything.
  5. Huh?

Four bad answers

This riddle appears to have four possible answers.

Answer 1: Evolution made consciousness on purpose because consciousness is useful. Maybe evolution noticed that if it created the right proteins, then we would have little “souls” in our brains that would have consciousness and help make better decisions.

The problem with this answer is that it’s… absurd? How could a soul change your behavior? Does it have a magical wand to push atoms around or stimulate neurons? Physics seems to prohibit this.

Answer 2: Something else made consciousness on purpose. Maybe God just decided that it would be fun to give everyone little souls. Or maybe we’re all living in a simulation and after 10 billion years the Simulator got back from their coffee break and turned on the “consciousness” switch.

I guess maybe?

Answer 3: It was an accident. Maybe evolution doesn’t “know” about consciousness. It made brains because they are good at telling muscles what to do so you get food and stay warm and don’t die. When those brains become powerful enough then integrated information something emergent phenomena something irreducible cause-effect structure—consciousness!

I’ll just admit it—I don’t understand this. We think biology “emerges” from physics because all of biology could in principle be explained in terms of physics. But for consciousness—how can the goings-on of one kind of thing create an entirely different category of thing? Isn’t that like saying a cruise ship is an emergent phenomenon of your sense of humor?

But I have friends who insist this is a symptom of my mouth-breathing reductionist worldview, so I thought I should mention this in case you’re more enlightened.

This answer overlaps with our last one:

Answer 4: Nothing did it. Maybe consciousness wasn’t created. It was already there, as a field that permeates spacetime. Or perhaps consciousness and matter and energy are different aspects of the same thing. All that evolution did was connect consciousness to an agent that can talk and act and have memories rather than just sitting there and being conscious all the time like rocks or thermostats do.

It gets worse

So all those answers seem pretty insane.

But say that doesn’t worry you. You still face the “meta” hard problem of consciousness, which is: Why do people report being conscious? Either:

  1. Physics is wrong and consciousness has a little magic wand that can change what happens in your brain.

  2. Physics is right and consciousness can’t do anything, in which case how is it that the fact that you’re conscious can make its way into the physical world when you talk?

Is it just a coincidence? You are conscious, and you report being conscious. But if physics is right, there seems to be no way for these two facts to be related.


Of course, I have no answers. The reason I brought you here today is that I think there’s another variant that is even more confusing. Namely…

The hard problem of feelings

When we talk about “consciousness” it sounds like a little bubble of awareness serenely watching life happen. That’s not what being alive is like. When you eat a taco, you don’t say “My, what an interesting texture is created when I mash these ingredients together with my saliva.” You have preferences. Good tacos and being told you’re good-looking and charming are awesome. Sad wraps and people squinting at you after your jokes fail are non-awesome.

This good/bad axis is a fundamental part of experience. (So fundamental that some people say if you train for years, you can learn to stop doing that and achieve enlightenment.)

So, we have feelings. So what? Well, why do we have feelings? Consider this variant of our earlier puzzle.

  1. We are alive.
  2. We have feelings.
  3. We were created by evolution.
  4. We feel good when we do stuff that would help propagate the genes of someone in a hunter/gatherer band.
  5. But feelings can’t “do” anything.
  6. The hell?

The same answers except worse

This puzzle has the same answers as the hard problem of consciousness, except they are even less satisfying.

Answer 1: Evolution made feelings because feelings are useful. Evolution created feelings because we have little souls that send out magic electrical spikes. Evolution installed a soul in your head and made it feel pain/pleasure so that it would exercise free will to serve evolution’s aims.

I say no because physics.

Answer 2: Something else made feelings on purpose. Maybe God / The Simulator / Whoever was bored and decided to turn on the feelings module.

Maybe. But if it wasn’t evolution, then—isn’t it weird that evolution’s fingerprints are so clearly present in when we feel good or bad? Is that a coincidence?

Answer 3: It was an accident. Maybe evolution doesn’t know about feelings. (How could it, when feelings can’t do anything?) Evolution just wanted to give you a brain that would move your muscles around so that you can reproduce. As part of that, it created heuristics for “good” and “bad” events and told your brain to learn how to make good stuff happen. Then, for magical information-partition-causal-schema reasons, feelings happen.

I guess that’s possible. But remember—in this theory, feelings can’t influence behavior. Sex doesn’t feel good because that’s helpful. All your behavior is explained by physics and evolution favoring patterns that reproduce themselves. It just happens in this answer that the “this is good do more of this” pattern evolution programmed into your brain also manifest as “a sentience experiencing pleasure”. That’s weird.

Answer 4: Nothing created feelings. Maybe, just like there’s a magical field of consciousness that surrounds all things, there is a goodness/badness field. The reason you have feelings is that feelings are a fundamental part of the ruleset of the universe and feelings are everywhere and it’s impossible not to have feelings.

OK… But again, isn’t that a remarkable coincidence? If feelings don’t do anything, then why is what we feel so aligned with what evolution wants us to do?

I can accept that rocks and thermostats are just sitting there (slightly) conscious. But do they feel “good” when they successfully regulate the temperature?

Say you create a chess AI. It simulates billions of games and builds a “reward function” that estimates how likely it is to win from a given board state. Then to play games it does a tree search for moves that will make the reward function go up.

That reward function is a lot like the function in your brain that makes you happy when you eat good food or have sex with hot people or provide a happy home for your children or show your neighbors how cool you are.

Now, let’s just suppose your chess program is sentient. Even then—when you talk about “reward”, isn’t that just an informal label? We don’t seriously think that the AI would experience pain or pleasure depending on the board state, do we? But unless you believe in little souls, it’s hard to see what the difference is.


Thanks to Superb Owl.

Comments at substack.

new dynomight every thursday
except when not

(or try substack or rss) ×
Consciousness will slip through our fingers

Can technology explain subjective experience?

I guess life makes sense: For some reason there’s a universe and that universe has lots of atoms bouncing around and sometimes they bounce into patterns that copy themselves and then those patterns go to war for billions of years...

Reasons after nonpersons

AI will make some old-school philosophical thought experiments seem much more relevant

The year is 2029 and you’ve finally done it. You’ve written the first AI program that is so good—so damn good—that no one can deny that it thinks and has subjective experiences. Surprisingly, the world doesn’t end. Everything seems fine....

Why it's bad to kill Grandma

Commonsense morality is an OK-ish utilitarianism

1. In college, I had a friend who was into debate competitions. One weekend, the debate club funded him to go to a nearby city for a tournament. When I asked him how it went, he said: Oh, I didn’t...

Reasons and Persons: The case against the self

A review of the thought experiments in Reasons and Persons by Derek Parfit (part three)

You want to go to Mars. There’s a machine that will scan and destroy all the matter in your body, send the locations of every atom to Mars, and then recreate it. You worry: Does this transport you, or does...

Reasons and Persons: Watch theories eat themselves

A review of the thought experiments in Reasons and Persons by Derek Parfit (part one)

You live with a group of utterly rational and self-interested people on an island, gathering coconuts to survive. Tired of working so hard, Alice builds a machine and implants it in her brain. This machine leaves her rational except when...

Are ethics all a lie?

If we only believe in ethics because of evolution, does that mean ethics are a lie?

Some people claim ethics aren’t practical. Others make a grim philosophical argument:

Pragmatic reasons to believe in formal ethics

Some situations where formal ethics are really needed for practical problems in real life.

Here’s a “low-brow” take on ethics that’s worth taking seriously:

It's hard to use utility maximization to justify creating new sentient beings

The ethical theory of Utilitarianism applies to many situations, but runs into problems when choices might create new beings.

Cedric and Bertrand want to see a movie. Bertrand wants to see Muscled Duded Blow Stuff Up. Cedric wants to see Quiet Remembrances: Time as Allegory. There’s also Middlebrow Space Fantasy. They are rational but not selfish - they care...