Awareness

On Simulating Awareness


There's a body of thought which says that you can program a computer to be self-aware. You can simulate consciousness and you'll have consciousness.





Here's a little animation of an electric motor. It gradually speeds up until it's going too fast for the refresh rate on your monitor to display. Then it stops dead.
This is a proof of concept, or more accurately a proof of no concept. Call it a proof of no idea.
When the motor stops, all the rotational energy has to go somewhere. The motor is bolted to your monitor - note the bolts in the base - so it can't move. It's a very powerful motor. This is a detailed simulation.
The energy has to go somewhere, so we'd assume that your monitor at the very least jerks up on the right side and slams down. Probably it cartwheels across the room and sticks in a wall.
It doesn't do that. It has to, but it doesn't. There's a flaw in our thinking.
Anyone who can see where this is going says, "Of course it doesn't do that. It's just a simulation!"
Got it. A simulation is not the thing it simulates. It looks like the real thing, but it isn't real. That's completely obvious.

For the dedicated programmed mind enthusiast it's time to backtrack.
There's nothing in the world a simulation of which will do anything, with one amazing exception: a simulation of a mind will be somehow different.
You can't simulate a motor and get torque. You can simulate a mind and it'll be a mind. A simulation of a mind is a mind.

This is based on the bizarre idea that computers are already in some completely undefined way minds, so your simulation isn't really a simulation.
A picture of a bear isn't a bear, an animated motor can't twist your screen.
If you believe you can program a mind, you've found the one simulation that is real. You've drawn a picture of a bear and any idiot can see that we're not going to get mauled. We already know about the motor. A simulated mind is a mind.

That's a problem, because nobody knows how a mind works.
Every wacked theory I've ever seen says that when you get enough connections, enough complexity, enough processing power, you're conscious. That's science!
"What does it take to create a mind?" Answer: "A lot."
Maybe it's even more mystical. Adding two and two on paper does nothing, adding two and two in a computer makes a mind form somewhere for just a second and think four, before it's gone. That only works on a computer, because computers are magic. We don't know how a mind works but computers do stuff you need a mind to look at, so a computer is a mind!

People spend years and generations prying up tiny pieces of the truth to assemble into a pattern. If you don't have any pieces you start looking for them. That's how we learn. It's very slow. Sometimes you don't make any progress for years, or forever, but it's how science advances.

Consciousness is apparently different. If you don't have time or a clue you make up some utterly embarrassing bullshit and wave your hands. You claim the absurd is obvious, that magic is science, that something that's not even a vague or coherent idea is a theory so good it doesn't have to be tested. You brush aside the whole concept of testing, and with it science and logic. A computer can generate a mind because someone quoted you on their web page. Probably that web page is self-aware, if you're right.

You're not. In what bizarre witless world is a simulation of a thing the thing? A simulation is a description that you can watch and make predictions from.
A picture of a motor is a simulation.
Can you get torque from it?
Has there ever been a simulation that held more than was put into it, that accidentally had an entirely different and more complex set of properties in addition to the ones you gave it? Can a description be more complicated than the real thing?
Can you simulate something you don't understand and have a computer magically add the properties you missed and do something you wanted but didn't understand enough to simulate it, using unknown properties of your computer, a complex system that wasn't designed to have them?
How witless do you have to be to consider that?
People are saying it. It's ridiculous, but in this one special case it's going to work.
According to this idea it must already be happening. If a complex computer can be conscious then a simpler one must be a bit conscious. It doesn't behave any differently, so this is the first theory that claims that the behaviour of a system will be exactly the same when the most important thing about it has changed.
It's a kind of anti-science - it can't be observed or tested because it changes everything and yet nothing.

We could go further and enter the psychotic domain of people who claim that consciousness doesn't really exist, but I refuse.
James Jeans said, "The universe begins to look more like a great thought than a machine."
The view that people are mindless machines, and I suppose by extension that there's really no such thing as thought, is so ludicrous as to defy comprehension, assuming you have thoughts and can comprehend stuff.
(It's tempting to state that people who claim consciousness doesn't exist may be actually, through some defect, mindless automatons, but I think the evidence we have indicates that you can't build a functioning mind without consciousness, so they're just denying their experience because they can't think of a theory to explain it.)
I've said this before: it'll take a very dedicated genius to figure this out.
It appears to be better to deny the absolutely obvious than to admit that the world needs another Isaac Newton and you're not him.