Hacker News Reader

Comment by ISL | Hacker News Reader

It's questions all the way down.

Even if we knew the complete theory of everything, the immediate question would become, "Why is the theory of everything the exact theory of everything?"

View on HN

Not only that, but empirical truths are always subject to measurement error -- we will never be completely certain that there is not a more accurate model of reality, and in all likelihood we will always continue to refine our models. Also consider the ever-increasing costs of this minute examination -- probing higher energy physics is only going to become more expensive. There will always be limits to what we know and what can be known about the Universe -- it's an inherent quality of empiricism.

I think of it in terms of a network of beliefs. In the network of beliefs about physics, and complete theory of everything would make all of the constituent beliefs mesh together and support each other perfectly.

For sure, there would still be questions, but the network would have been tidied up.

Chemistry is already (I think?) in this position. There are no contradictions in the network of fundamental chemistry beliefs, even if there are some unknows in terms of practical issues.

Because of the Anthropic principle !

Conceivably, you could have empirical evidence for multiple universes, at which point the anthropic principle becomes a normal explanation. Until then, obvious it remains a bit of a non-explanation.

I hate the Anthropic principle.. it appears to be able to justify anything

You should hate anything that isn't testable and falsifiable, that's good hatred.

Thanks!

Why? What fundamental reason is there to reject the untestable and the unfalsifiable? At worst, these claims fall into an epistemological netherworld where they can neither be ascribed truth nor falsity -- at best they may be useful tools.

So your idea of R&D is essentially random, but because you'll occasionally happen upon something useful, don't knock it?

Truly, the philosophy of science is dead.

That's not at all what I'm suggesting. I am not advocating for the arbitrary assignment of confidence intervals, expected values, or whatever your chosen epistemological metric is -- I am advocating for restraint in assigning falsity to what we merely cannot assign truth.

Logical positivism is not boundary of philosophy of science.

There's a difference between, "Can't assign truth" and "Not testable or falsifiable." A difference I hope not to have to explain.

Science can't tell us whether something will ever be testable or falsifiable; we just know what we know now.

So there's no scientific reason to hate any particular idea. Rather, the concepts direct the priority of effort--let's not spend time on things that are not currently testable or falsifiable.

The "hatred" aspect comes from somewhere else besides science, and honestly I think it's harmful more often than it is helpful.

Science can't tell us whether something will ever be testable or falsifiable; we just know what we know now.

That's... almost literally the opposite of reality.

I phrased it poorly.

What I meant is that, given a hypothesis that is not currently testable or falsifiable, we cannot use current scientific knowledge to prove that it will never be testable or falsifiable.

To your point, the way I wrote it above, it sounds like I believe nothing can ever be tested or falsified. Obviously we have tested and falsified many hypotheses.

I'm glad you clarified, because I was getting distressed at the idea of someone so eloquent thinking something so crazy.

To clarify a bit myself though, remember that we're talking about something for which the standards of "testable" and "falsifiable" often just mean, "even in some thought experiment." For example people predicting collisions with another universe have a pretty wild hypothesis, but they proposed some ways to explore that idea involving studying the CMB background.

When we're talking about the anthropic principle as a hypothetical foundation for a complete understanding of physics though... there's really no room even in theory for testing or falsifying that notion. That's what was being hated, and the hate of which I'm in support.

I can think of two hypotheses that can't currently be ruled out:

1) We're living in a computer simulation of a universe, which was built and operated for our benefit by some unknown intelligence(s).

2) Our universe is just one of an infinitely many universes, each of which has slightly different values for the physical constants. Our universe seems well-suited to our life because of selection bias--in the universes not suited to life, there is no life to observe that fact.

These both relate to various formulations of the anthropic principle. The principle itself can't be used to prove the correctness of one or the other. But it's possible that we will find some experiment in the future that will allow us to test one or the other.

I think that's why those concepts are entertained by a large number of people in the scientific community, over drinks at least. Of course, if someone tried to get serious funding to study it, right now, there might be a different reaction.

Reply on news.ycombinator.com