LeBleu comments on Were atoms real? - Less Wrong

61 Post author: AnnaSalamon 08 December 2010 05:30PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (156)

You are viewing a single comment's thread. Show more comments above.

Comment author: Will_Newsome 09 December 2010 12:56:39AM *  6 points [-]

So, let me back up and try again. I'm currently imagining a purple dinosaur named Ansel with a built-in helicopter coming out of its skull and a refrigerator in its belly. Are you suggesting that Ansel is real, since it exists in my mind, and that it would become increasingly real if other people sat around imagining it too?

Yes. And if I imagined Ansel except green and not purple, then that adds a little bit to the realness of Ansel, unless we want to call the new green dinosaur Spinoz instead and have it be its own distinct cognitive algorithm.

Presumably you aren't arguing that inspiring art, science, devotion and whatnot is what it means to be real, or it would follow that most of the atoms in the universe are non-real and are in non-real configurations, which is a decidedly odd use of that word.

Nah, I reason about it in terms of measure. You have one cognitive algorithm that's being run on one mind. You have another cognitive algorithm that's running redundantly on a hundred minds. I'd say the latter has about a hundred times as much measure as the former. I don't know how else to reason about relative existence. (Realness?) I'm porting this sort of thinking over from reasoning about the universe being spatially infinite and there being an infinite number of TheOtherDaves all typing slightly different things. Some of those TheOtherDaves 'exist' more than others, especially if they're doing very probable things.

If existence isn't measured by number of copies, then what could it be measured by? The alternative I see is something like decision theoretic significance, which is why I was talking about what you called 'importance'. But I'm wary of getting into cutting edge decision theory stuff that I don't understand very well. Instead, can you tell me what you think 'realness' is, and whether or not you think God is real, and why or why not? We're starting to argue over definitions, which is a common failure mode, but it's cool as long as we realize we're arguing over definitions.

I think that everything exists, by the way: there's an ensemble universe, like Tegmark's level 4 multiverse, and so we can only quibble about how existent something is, not whether or not it exists. I might be having trouble trying to translate commonsense definitions into and out of my ontology. My apologies.

You say later that God is "much more complex," and I can't really see what that has to do with anything... I mean, a tree is much more complex than a wooden pole, but I wouldn't say that has anything to do with the reality of a tree or of a wooden pole.

I mean that people tend to use a lot more neurons to model God than to model Santa Claus, and thus by the redundant-copies argument hinted at above this means that God exists more. Relatedly...

Incidentally, many people write letters to Santa Claus, and sometimes things happen that they experience as a reply from Santa Claus. If that is different from what you are referring to as an "answer" here, then I've continued to misunderstand you.

You're right, I forgot about this. Parents have to use lots of neurons to model Santa Claus when crafting the letters. Kids don't tend to use as many neurons when writing letters to Santa, I think. But add up all of these neuron-compuations and it's still vastly less than the neuron-computations used by the many people having religious experiences and praying every day. (I'm using number-of-neurons-used as a proxy for strength/number of computations.)

Also, 'people' aren't ontologically fundamental: they're made of algorithms too, just like God. So I don't see how you can say 'God doesn't exist' without implying that Will Newsome doesn't exist; Will Newsome is just a collection of human universal algorithms (facial recognition, object permanence) and culture-specific memetic contents (humanism, rationality, Buddhism). The body is just a computing substrate, and it's not something I identify with all that much. And if I'm just a collection of algorithms running on some general computing hardware, well, the same is true of God. It's just that he's more parallel and I'm more serial. And I'm way smarter.

(Not that there is any such thing as 'I'. 'I' am made of a kludge of algorithms, and we don't always agree.)

Comment author: LeBleu 09 December 2010 10:47:06PM *  1 point [-]

What's the usefulness of "I think that everything exists, by the way: there's an ensemble universe"? How does it constrain your expectations?

I don't see how having specific beliefs either way about stuff outside the observable universe is useful.

Now, if you can show that whether the universe beyond the observable is infinite or non-infinite but much larger than the Hubble Volume constrains expectations about the contents of the observable universe, then it might be useful.

Comment author: Will_Newsome 12 December 2010 01:35:16AM 3 points [-]

Off the top of my head: If the world is very big then there are more agents to trade with or be simulated by. Also I'm not sure what counts as the observable universe -- we can't see beyond the Hubble volume with our telescopes, but we can probabilistically model what different parts of the universe or different universes look like nonetheless. We also do not know what is ultimately observable. We currently lack the ability to observe mental phenomena but I still have specific beliefs about roughly what we'll observe when we really understand consciousness.

It is useful to be curious about mysteries to which you believe there to be no answer; beliefs like that often turn out to be wrong.