Hacker News new | ask | show | jobs
by psychoslave 2456 days ago
Frankly, I don't understand how this autocentric posture can resist any serious enquiry. To my mind it seems alarmingly full of causal vicious circles, and lake any clear way to be invalidated experimentally.

In particular, taking the "number of possible universes" seems problematic to me, as we can only obverse the one in which we interact. And even taking that for granted, what is so astonishing that the actual one is the one that is compatible with our existence? I mean, if any possible universe could have happened, then the one we live in had actually an equal chance to exist with any other. Since this perspective assume that at least one universe must actualize, then pure randomness by drawing of lots is enough to explain the result. No need for a teleological purpose. Apply Law of Parsimony and you are done: Anthropic principle add needless ad hoc assumptions to reach its predetermined conclusions, especially in its strong version.

Also, the anthropic principle stands on many assumptions, including that the simulated models of tweaked universes do reflect other conceivable world in a fiddle manner would they be actual things.

It also neglects that among an infinity of variation, you surely should be able to come with totally different forms of "life". That is, for example it takes life as a carbon based structure, so you need universes with constant tweaked for enabling star to produce and disseminate carbon before they collapse. But on the counterpart, it can't pretend if you take it seriously, that other universes could not have produced very different form of "life". To my mind, we have no idea if our current actual universe is not also populated with non-carbon based life form.

Also this kind of theory emphases greatly on what the world did allowed with these tweaking. But, among an infinity of other combinations, who would seriously consider that nothing better could have been produced than a mankind tormented in wars and questions which are at best extremely hard to solve for its limited cognitive abilities, if not definitely out of its reach?

1 comments

> And even taking that for granted, what is so astonishing that the actual one is the one that is compatible with our existence?

You've just verbalized the weak anthropic principle. The entire point of it is that this is not astonishing (at least not without additional knowledge), but a necessity for the universe to be one compatible with our existence, irrespective of how rare the conditions are, given we're here to observe it. The prior probability of "us" arising might have been incredibly small (or not, we can't tell), but the posterior probability is 1 - we're here.

> I mean, if any possible universe could have happened, then the one we live in had actually an equal chance to exist with any other.

Yes, but that does not mean that the odds of any given configuration need to have an equal chance.

What GP is saying is if there is an infinite number of universes and if life is extremely unlikely, then the odds of any given universe with life only having one life form might be very high, even if there being even one life form might be extremely unlikely.

We know that the posterior probability of there being one suitable for us is 1, since we're here. But we don't know whether or not life is a rare event or not.

The antrophic principle explains why we "won the lottery" so to speak in being in one of the universes with at least one life form, but life being incredibly scarce might be one explanation for why we might be in a universe with only one life form.

Though of course we can not infer that life is rare across any given set of universes if we determine we're alone in this one, because we don't know how other universes would be configured.

> Though of course we can not infer that life is rare across any given set of universes if we determine we're alone in this one, because we don't know how other universes would be configured.

Not necessarily, you could assume a universal prior distribution over possible universes and count how many of those that have life have more than one.

https://en.wikipedia.org/wiki/Solomonoff%27s_theory_of_induc...

This might be a sort of a "last tool of science", a way to discover the laws of our universe without observing them.

That presupposes knowledge of more than one universe to even start to enumerate the possible parameters deciding the set, and without a sufficiently complete sequence we would be unable to determine all of the possible parameters, or how they can change.

E.g. even given that we know, say, gravity exists, we do not know if or how its strength varies in a set of infinite universes. Based on a sample of one universe, it could be constant or vary in an infinite number of ways.

> That presupposes knowledge of more than one universe to even start to enumerate the possible parameters deciding the set

I think you entirely missed my point (because you didn't read the link!). You could use the universal prior, that is, you could enumerate all possible computer programs (e.g. Turing machine programs) starting from the shortest and find the shortest one that generates our universe. Now, 2^L where L is the length of such program is the prior probability of our universe.

So to determine e.g. the average value of the gravitational constant across all universes, you would measure the gravitational constant in all universes and calculate a weighted sum where you weight the value of the constant by the prior probability of the corresponding universe. The same for the occurrence of life, etc.