[moring]

> In science nothing can be proven.

Asking here because it is mostly on-topic: This phrase is repeated often, but shouldn't it actually be, "In science, a hypothesis is either fundamentally verifiable or fundamentally falsifiable, but never both"? The two simply being the logical negation of each other.

"All swans are white" is fundamentally falsifiable (by seeing a black swan) but not verifiable, as you described.

"Black swans exist" is fundamentally verifiable (by seeing a black swan), but not falsifiable.

[RandomThoughts3]

I think both are wrong.

Science has a word encompasses a broad variety of definitions and practices. The philosophical question of the relationship of empiricism to truth (because fundamentally the question has more to do with the relationship between observations and deductions than science in itself) is a complex one and overall of little interest to actual practitioners of science. It can’t easily be subsumed in broad aphorisms.

I would hazard that most modern sciences - and physics especially - are nowadays mostly concerned about building models and the resulting predictions. The question then often becomes: does the tool I have built allow me to correctly predict how observations will unfold? The answer to which is often more shades of grey than simply yes or no.

[greiskul]

Maybe you think you saw a black swan around the orbit of Jupiter with your new fancy telescope, but later you figure out that the telescope was falty and the swan you saw was actually white.

Because science depends on the external world, and there might always be hidden variables that we are not considering, it can only give us extremely high confidence after we make repeated observations using different methods, but it cannot give us the absolute confidence that Math can give us.

Also, this example of faulty telescopes is actually really close to what actually happened in the history of astronomy. Stars were thought to have a visible radius that made them appear to be much larger than the Sun, which was considered evidence that the Sun was not a regular star.

[ninetyninenine]

> it can only give us extremely high confidence

Here's an interesting thing. Even high confidence can't be verified.

Let's say you examined 10 million swans and you think you observed all possible swans. But there's no way you can know whether or not the actual population is 10 billion trillion swans or a google swans.

If you observed 10 million swans and they are all white but those swans could only represent 1/99999999999999 of all possible swans. Then that means your observation is low confidence and there's no way whether we can verify what fraction of the population our sample size represents.

So actually high confidence is just an assumption. At a very technical level the confidence that science brings to the table is very very weak. We are making tons of assumptions and jumping to conclusions all the time.

[Koshkin]

> So actually high confidence is just an assumption.

This one is actually a funny statement. Not all assumptions are born equal :)

[ninetyninenine]

Yeah, we actually can't technically have any confidence for anything. Yet somehow we are. It's a bit paradoxical to our daily experience.

[epgui]

Science (as in the “natural sciences”, ie.: everything that is not philosophy or mathematics) is not about “proving” anything at all.

Science is about measuring, quantifying and qualifying uncertainty.

[ninetyninenine]

I answer it here:

https://news.ycombinator.com/item?id=41851265

If you want to learn more look up Karl popper. It’s his philosophy.

In response to the other sibling replies I’m basically talking about science as in the scientific method. What are the limits of the scientific method in terms of determining truth?

One thing that isn’t addressed is fundamentally limited accuracy and precision of observational instruments. This actually makes things not falsifiable as well but I don’t talk about this because it complicates everything.