Questioning the nature of reality with cognitive scientist Donald Hoffman

Originally published at: http://boingboing.net/2016/12/05/questioning-the-nature-of-real.html

1 Like

These violent delights

5 Likes

I think there’s the point that I disagree with.

Do our senses miss huge chunks of reality? Of course they do.

Do the things we can perceive get heavily processed, so that they line up with what “makes sense” more than with an actual perception of the underlying reality? If that weren’t true, we wouldn’t have optical illusions.

But the idea that we can never build tools that can perceive the underlying reality just seems like nonsense to me.

4 Likes

So Scottie Nell Hughes is right about there being no such thing as facts? And, essentially, Karl Rove when he spoke about creating new realities? That’s a bleak message to take into the next four years.

1 Like

I listened to the podcast, and it was interesting up to the point where Hoffman started postulating about physics.

Then it rapidly descended into Deep Woo.

I’ve been subscribing to YANSS for a long time, and it’s generally well-researched and interesting. But this episode really should have taken a much more skeptical view of Hoffman’s wilder theories (like the idea that consciousness, rather than spacetime, energy, quantum mechanics, etc. is fundamental to the dynamics of the universe).

Hoffman repeatedly claims that his ideas “haven’t been disproven,” when the truth is that they don’t make meaningful and testable physical predictions, and don’t solve any significant problems in theoretical physics.

When you’re a professional working in the field of physics you see all manner of crackpot theories from people who have only a pop-culture understanding of physics. They’re almost never worth spending any time to understand and debunk. That’s no different when the crackpot theory comes from someone who holds a PhD and professorship in an unrelated field.

5 Likes

Okay, so a guy ran some computer simulations, and from them concluded that the probability that reality works a certain way - which would include somewhere in it the probability that his simulations didn’t accurately represent reality or that they were flawed - is “Precisely zero”.

You are not so smart indeed!

[ETA: I’m going to be back later to say something about umwelts, but my initial reaction to drive-by smear this charlatan was precisely irresistible]

1 Like

“But that ‘other world’ is well concealed from man, that dehumanized, inhuman world, which is a celestial naught; and the bowels of existence do not speak unto man, except as man.” TSZ 3. Backworldsmen, Freddie the Nietzsche

That was a fun season ending last night.

2 Likes

I had to listen to this entire podcast because the text of the article was presenting such a foundationally stupid idea that I couldn’t believe it was accurate. Beyond the absurd arrogance the David Hoffman is represented as having (that I mention above), there are a couple of super obvious problems:

  1. If we have no way of accessing underlying reality, then an experiment we do to figure out how reality works has no validity. Basically, if he is right then necessarily his experiment gives us no reason to think he is right. If we can’t understand underlying reality then he can’t have built a simulation of underlying reality that is worth anything.

  2. Whether or not you have survived is a fact about reality, and I think we are all on board that consciousness depends on survival, therefore consciousness gives us at least one fact about underlying reality.

Turns out Hoffman doesn’t appear to be a spectacularly arrogant asshat, and that if this theory is relatively novel, then I really am a prophetic genius about 10-12 years ahead of my time (I’m pretty sure this is true).

Listening to this definitely reminded me of why I never made it through one of these before. Say a sentence, play five seconds of music, say a sentence, play five seconds of music. Turns out if you get to minute 20 or so that gets a lot better.

What he’s actually saying, for those not inclined to listen is something that I thought was obvious to everyone, that our perceptions are much like a user interface for reality. That is, it’s as related to underlying reality as the text on my screen is to the underlying workings of my computer.

But what he is not saying is that our senses don’t give us any information about underlying reality. Just like we you could begin to make guesses about the computer by the way that the text goes on the screen, you can start to figure out things about reality by working from within the user interface. He thinks physicists will eventually figure out that spacetime is doomed as a concept of reality (I don’t even this is a really out-there thing to say, I mean, if we’re basically wrong about the nature of everything, it wouldn’t be the first time). That is, he’s still saying the enterprise of physics is actually going to get more and more correct.

So he’s talking about consciousness and “perception” as a function of consciousness. He’s not talking about “perception” as the function of the body. So he’s not saying that a patch of photosensitive cells don’t give some kind of information about whether or not there is food in this direction or that.

In fact, though they never outright say it, they agree with my complaint (1). His whole argument is based on evolution, but evolution takes place within spacetime. If we can prove using evolution that there our entire concept of “before” and “after” wouldn’t be reflected in underlying reality then we can prove via evolution that evolution can’t possibly be right.

That’s not saying that evolution doesn’t reflect underlying reality in some way, if spacetime is wrong then evolution seems like it would be equally wrong. For all of the parts of the world that Newton was able to study and everything he meant to build out of the world, light really did and does travel in straight lines, so wrong is still good for something.

In the end, he’s proposing an absolutely radical theory of the nature of the universe, but one that falls within the normal bounds of science, that is, he thinks that we can think about it, do experiments to confirm or challenge (or disprove) it. And “we” (conscious things) will do those experiments from within our current framework (conscious thought) and some will thereby learn something about the underlying nature of reality. It’s not radical in the way the podcast presents it, it’s just radical in the way that gravity-as-a-force-of-attraction-between-all-masses was radical compared to gravity-as-things-fall-to-the-ground.

1 Like

There are transcripts of some of the older ones.

1 Like

I think that if his theory was true, though, it wouldn’t actually represent that much of a departure from what physics is saying now. It’s presented as being mind-bending because people find thinking about their own consciousness mind-bending, but he hasn’t abandoned the basics: we learn about reality by making theories and testing them.

Ultimately if there spacetime doesn’t work, we have some idea of what that would look like. Clocks and ovens and cellphones would still work. Planes would still fly and cars would still drive. But in some weird edge case involving what we would currently call astronomical distances (or several orders beyond “micro-” scopic ones) that the vast majority of people won’t know the math necessary to understand something won’t work the way we think it does right now. The world will keep turning, it’s just that elite physicists will have an even better idea of how bad a mischaracterization of reality “the world keeps turning” is.

A huge problem with what he is saying is that the language to express the idea he is trying to get out won’t be invented unless the idea turns out to be right (which ties in, in a strange way, with the topics at the beginning of the podcast). When he says consciousness might be the fundamental building block of reality, I don’t think you can take those words to mean what they sound like the mean right now. “Consciousness” would be a completely new understanding of consciousness and “reality” would be a completely new (though in almost all functional ways exactly identical) understanding of reality.

Anyway, I doubt I can understand the math in his paper well enough to comment on whether it is just brutal nonsense, and I’m still super skeptical of his model that shows we aren’t perceiving reality accurately. There’s no way to model that with some crazy simplifying assumptions, and the simplifying assumptions of evolutionary theory have produced a lot of nonsense in the past.

1 Like

Recommended reading:

Meat” by Terry Bisson.

Excerpt:

“They’re made out of meat.”

“Meat?”

“Meat. They’re made out of meat.”

“Meat?”

“There’s no doubt about it. We picked up several from different parts of the planet, took them aboard our recon vessels, and probed them all the way through. They’re completely meat.”

4 Likes

When he says consciousness might be the fundamental building block of reality, I don’t think you can take those words to mean what they sound like the mean right now. “Consciousness” would be a completely new understanding of consciousness and “reality” would be a completely new (though in almost all functional ways exactly identical) understanding of reality.

In that case, he’s not really saying anything at all, since the actual meaning of the words is to be determined at a later date.

Anyway, I doubt I can understand the math in his paper well enough to comment on whether it is just brutal nonsense, and I’m still super skeptical of his model that shows we aren’t perceiving reality accurately.

IMHO, this is the one part he probably gets right. Our intuitive understanding of how the world works is often misleading or simply wrong. It has taken hundreds to thousands of years of careful observation, experimentation, and testing theories to disabuse ourselves of mistaken ideas. For more than 99% of humanity’s existence (assuming humans have been around for 200,000 years), it was self-evident that:

  • The sun, moon, and stars go around the earth.
  • Heavier things fall faster than light things.
  • The earth is basically flat.
  • The stars have fixed positions in the sky and don’t change.

Heck, it’s only been within the past 500 years that we’ve known that it’s possible to create a vacuum (and that was considered literal heresy at the time).

The history of science shows us that many of the things we accept as “obvious” or “intuitive” today are actually very recent notions, and not at all consistent with the way our senses would have us believe the world works. But because the ideas are so pervasive, we reinterpret our senses through the lens of modern science and have a much better understanding of what’s really going on.

2 Likes

If that were true then no one would have ever said anything. For as long as humans have had language we’ve been talking about being hot or cold without knowing a thing about what heat is until very recently. Language is all about grasping at straws.

I fully agree that our perceptions lead us to all kinds of conclusions that are totally wrong. I’m just super skeptical of his model which uses math to “precisely” define and measure evolution and come up with a probability of exactly 0. Any mathematical model that gives you a probability of something happening in reality without a margin of error is making an unwarranted simplifying assumption, even if it ends up being correct.

Wait a minute. If he can’t say what consciousness is at all, then how the hell did he model whether the content of consciousness helped in an evolutionary setting?

I keep thinking about it and thinking, “Did he just assume that accurately perceiving reality would take up energy and not accurately perceiving it would take up less energy?”

2 Likes

I got a notification to this topic because @Nelsie quoted an article I also quoted?

Huh?


Oh damn, this is totally my kind of thing too.

*reads whilst rubbing chin


So yeah initial reaction to:

True but childishly insecure and immature response. It’s pattern, who the fuck cares where it’s happening and how?
“No escape.”
Pfff.

Fun trivia:

In the deeply surreal anime, Flip Flappers, the protagonist has a green rabbit named Uexküll. It’s a very hard name to say in Japanese.

1 Like

Thanks, added it to my watch list on Aozora.

1 Like

I’ve been thinking about this all night, since that’s the kind of thing I do.

A. I keep coming back to the thought that his explicit invocation of cogito* means that his point is self-contradictory. But it isn’t. He wants to draw a distinction between the fact of consciousness and the content of consciousness. I’m extremely unsold on the idea that such a distinction can be made.

[The cogito is the “I think therefore I am” argument. Basically, one thing you can be sure about is that you have some kind of immediate experience, and therefore you are]

B. To take his user interface analogy further, there’s an intersection with the simulation argument. We are essentially like the characters in a 3D videogame and the universe is the physical state of the information on the chips/hard disks. So essentially we are living in a simulation, but the simulation wasn’t constructed by a superior intelligence, it evolved within the system. In that sense, physics would be like what speedrunners do. Looking for every strange behaviour they can find to get at the underlying rules in a way that gets past the rules that seem apparent (but without being able to look at the actual ROM). Physics can never guess at the underlying state of the memory, because any given state could be represented by a 0 as easily as it could a 1. But the point of the scientific method is that it is agnostic to that kind of “eternal truth” and deals instead with the truth of what happens. The reason science is so successful is because it doesn’t try to break the Umwelt, and the reason philosophy and economics are so unsuccessful is because they do.

[Take that economics!]

C. I don’t see how proving that consciousness is a building block that can be used to build quantum mechanics would actually prove anything. We know that consciousness is a building block that can be used to build quantum mechanics because it was conscious beings (us) that created the theories of quantum mechanics. A Turing machine can construct quantum mechanics but it can also construct anything else. Is he saying that he thinks that his mathematical notion of consciousness will allow us to prove quantum mechanics as a theorem? Doesn’t he think he’s going to have problems with the incompleteness theorem there? If the universe is based on math, then either the universe contains a contradiction, or we know that there are unknowable things within it.

[The incompleteness theorem proves that (roughly) any sufficiently complex mathematical system (and it takes very little complexity to be sufficient) cannot be both consistent - there is no statement that is not true and false - and complete - every true statement can be proven true.]

1 Like

It makes me think of this odd habit humans have, of trying to invent ranking systems for intelligence among nonhuman animals. Some of us will pluck an organism out of the environment it was evolved for, put it in a different environment, and take note of how confused it becomes. Then there’s a ranking of which animals get most confused, which ones shrug it off, and that’s how smart they are compared to us.

Darwin had a grandson, Charles Galt on Darwin, who wrote _The _Next Million_Years where he opines that no organism, most especially not homo sap, will ever be able to out-think evolution, and any attempts we make to deliberately evolve ourselves into the future, will only succeed in showing us where we went wrong.

…which has me wondering about the season climax of WestWorld.what would it look like for a post-human species of robot to have access and understanding g of their own DNA sourcecode, as well as the freedom to act on that knowledge without parental meddling? it brings to mind that Charlie Stross book, Saturn’s Children

1 Like