The mistake made here is mistaking a simulation for the thing simulated. We don't make that mistake for anything

else. I mean, nobody would mistake a kidney function simulation on a computer for a kidney producing urine. It seems

almost impossible to think, though, that we could ever reproduce or have a control group with consciousness, um, and

that I think leads many people to think about, uh, maybe philosophically entertaining, but ultimately, you know,

impractical sorts of experiments. I mean, are you aware of any, you know, true experiments that could shed light, uh,

no pun intended, on consciousness as a, uh, other than, you know, kind of metaphorically, the dashboard, the

instrument panel of yours, etc.? Are there only Gedanken experiments, or can we actually do real, honest-to-goodness

experiments in the physical sciences? I think we can. I think neuroscience has been showing that experimentation around awareness, phenomenality, consciousness

itself is possible and it's productive to a large degree. Um, the entire field of the neuroscience of consciousness, the

notion of the neural correlates of consciousness is based on measurement and subjective reports. Individuals who report what they are experiencing now,

are there problems with that? Are there limitations? Surely there are. Um, for instance, uh, when you are basing your

study on reportability, you do not distinguish between raw consciousness or phenomenal consciousness, as we say, state

on the one hand, and on the other hand, meta-consciousness or conscious metacognition, which sits on top of

awareness. It's meta-conscious consciousness. It's what you have when, in addition to experiencing, you know that

you experience. Now, to report an experience, you have to be meta-conscious of it; otherwise, you will not say

Well, I'm not having that experience; well, in fact, it may be an experience in itself, so there is that limitation, the

Neuroscience, neuroscientists today try to get around that with the so-called 'no-report' paradigms, but then you get

limited in all kinds of other ways. But I think it is productive to have some science around, uh, phenomenality, around

consciousness. What we will not succeed in—and unfortunately, that seems to be the goal for most sciences, uh, around

consciousness—is to reduce consciousness to physiological brain activity, to, to, to neurometabolism. The very

attempt to do that presupposes a certain metaphysical assumption, which is that, well, consciousness is reducible, and that

the correlation between brain activity and consciousness is a causal one, and, uh, that I think is hopeless. Um, and to

complement the neuroscience of consciousness, I think the way to go is to, um, to have introspective

investigation, which is not science. We should not add introspection to science because it will break the back of

science as a horse, and we will lose a very, very useful horse that we should keep and protect and maintain healthy. Um, but

in addition to science, I think it would be very productive, even for the scientists who practice the science of

consciousness, to develop a little more sophisticated, uh, introspective awareness; to know a little bit more firsthand what

it is that they are actually studying. Because the level of introspective naiveté amongst philosophers of mind and

and neuroscientists of consciousness is sometimes scary. It's, it's, it's, um, because why, why, why do you say

that? Well, because you, one is dedicating one's life to studying consciousness, which can only be actually known from

Introspection, a first-person perspective, uh, but one is extremely naive about what that word means. So how can one study it?

From the outside, if one doesn't actually know what one is studying. I mean, there are attempts, uh, made by famous

neuroscientists who say that, and I quote, "Consciousness doesn't happen. Consciousness doesn't exist." Um, and when

you read their material, what you realize is that what they call consciousness is a kind of ethereal sense

of individual subjectivity, a kind of individual eye within a kind of soul or spirit. And and that's what they explain away. Well, very

good. It's it's not like anybody was waiting for for that kind of proof of the obvious, but um, they confuse that

with what philosophers call phenomenal consciousness, which has nothing to do with individual identity or some kind of

individual soul. It's about experience itself, phenomenality. It's it's a type of existence that is defined in terms of

qualities as opposed to quantities. You cannot specify or fully describe, uh, uh, uh, experiences on on the basis of

physical quantities. What is the length in centimeters of a thought? What is the weight in kilograms of an emotion? What

is the color of of a memory? I mean, um, that's where things go wrong when a, know, otherwise famous

neuroscientist makes proclamations about the ability to reduce consciousness to nothing, while in fact what he's

referring to as consciousness is not it's not consciousness at all. I think that is that is unfortunate and a little

more introspective insight would would improve and help everybody to engage in a more productive dialogue. So, uh, I think it is uh it is

useful to kind of, maybe for the audience, to kind of state some of your positions, which I'll summarize at least as I

understand them, but please correct me if I'm wrong—in the likely scenario that I am wrong—but as far as I understand it,

you are not—although you are materialistic in some sense—you don't believe that computers will ever be able

to approach the, uh, conscious experience or even the thought process that humans can engage in. Is that

correct? Am I summarizing your position, or...

all. I don't think that the foundational level of nature is material or physical or describable in terms of

quantities at all. I don't think that is true. Um, I am very skeptical of the notion that one can build a silicon

computer and expect it to have private conscious experience the way you and I have. Because when you talk about

conscious computers, what we mean is more than just the fact that there is consciousness associated with it; what we

mean is that a computer would be conscious in the same way as you and I are. In other words, a computer would have

private conscious states that are accessible from the inside but not from the outside. Yes, that I think is a very

unproductive fantasy. It arises from some silly psychology because nature is telling us that what private conscious

experience looks like when observed from the outside is warm, moist neurochemistry. That's what it looks like.

So why would a silicon computer that, you know, doesn't metabolize, doesn't burn ATP, doesn't release neurotransmitter

molecules, uh, doesn't have action potentials—well, we can imitate action potentials, so that's not a good

example—but doesn't have any of the rest, why would that be what private conscious experience looks like? It's completely

Arbitrarily, the mistake made here is to mistake a simulation for the thing simulated. We don't make that mistake for

anything else. I mean, nobody would mistake a kidney function simulation on a computer for a kidney producing urine.

Nobody would expect a computer to pee on one's desk because one knows that a kidney simulation, even if accurate down

to the molecular level, is not the thing simulated. But when it comes to the patterns of mental activity in a

human being, we think the simulation is the thing simulated, and that's extraordinarily naive. No, I don't think any silicon computer will ever

Bernardo Kastrup: Proof of Consciousness

Bernardo Kastrup: Proof of Consciousness