As humans we have to start somewhere, and here is where scientific inquiry starts:
The scientific method agrees that, fundamentally, knowledge is a highly philosophical matter. Nothing can ever truly be known. We perceive the world through our senses, which are imperfect, which is then processed through a brain. The inner workings of the brain, neurally and philosophically, are unknown and what ends up in our conscious mind is never a true representation of our senses. By the time our consciousness is told by the brain what is going on, we have already interpreted our senses multiple times. We're basically just looking at a derivative.

Beyond that, we only perceive what our senses tell us. That information can be made up, and what they sense does not have to be real. We could be living in a simulation, our brain could be simulating a reality itself, maybe there is a god feeding us experiences directly into our brains. We might never know. The reality is, and the scientific method acknowledges this, that we can never make any substantive, definitive statements about reality. We do not experience reality, we experience the depiction of a depiction of a depiction of reality, if reality even exists at all.


However, there seems to be an awful lot of consistency. Not only do many of us seem to experience the same (depiction of a depiction of a depiction of a) reality, it is also that the reality itself seems to be somewhat consistent. Why exactly that is will never be a question for an empiricist - for all we know that might truly be the work of a God - but what we can do is to find out the rules and mechanics behind the things we experience (in our depiction of a depiction of a depiction of a reality - I will stop mentioning it here, but remember that this is always implied).

Secondly, the scientific method works closely off of Hanlon's razor. Whenever there are ideas for why something might happen, some are better than others.

something something

When we try to explain something, we try to describe the underlying mechanics. Ideally, each mechanic can be properly isolated and described directly. If your explanation requires more than one description, you are probably working around a mechanic you don't yet understand. That's why we want the number of factors and exceptions to be as low as possible.
Take, for example, gravity. At home, you will probably be accelerated towards the ground at the gravitational constant of 1g. If you have no clue about gravity, you could create an explanation that goes as follows: “Every physical object is gravitationally attracted to the nearest bigger object at 1g.”
This would already explain a lot, but you're going to run into problems fast: Some guy on Mount Everest just jumped, and they were accelerated back to earth at 0.98g. And yes, they double and triple checked, that's accurate. Well alright, how about
“Every object is gravitationally attracted to the nearest bigger object at 1g, except for some reason if you're nearer or farther from earth.”

This is where the problems begin - you have now made your explanation accurate enough to explain Mount Everest man, but clearly gravity just doesn't work the way you think and you're now patching your way around the problems that will arise with your explanation. There more that this happens, the less likely you are to actually be hitting the true essence of what gravity really is.
In the case of gravity, this would become obvious once you look at the stars. They are not attracted to their stars at 1g, or any alternative you could come up with if you think purely about “distance from earth”. Earth itself is pulled towards the sun at well over 1g, and trying to accomodate that into your explanation will make it ever and ever more problematic. You will have to fit in other planets, which will have different pulls even though they are closer or father from earth and the sun, you will have moons which you could try to accomodate and ultimately you could come out at a formula that works, but is absolutely horrible and just tries really hard to work its way around a problem it just doesn't actually know how to solve.


Here is a much better explanation: All mass attracts all other mass. The more mass and the closer you are, the more attractive you are. Make a formula out of this, put it into a computer and.. it works! You seem to have hit the essence of what gravity really is. This works for humans on Earth, it explains why Mount Everest man experienced different gravity, it explains other planets, moons, orbits and why Earth itself orbits the sun as well. To be fair, This formula presupposes Newtonian mechanics (ie. “objects in motion stay in motions, objects at rest stay at rest”) and ultimately we cannot truly know whether it's true - but it's a really, really good base. In fact, it's the best base we have so far. Sure, it could just be that anytime someone jumps god just deploys a magical hand to pull you back to Earth, but that doesn't explain anything. In fact, to try to explain the mechanics of how we fall back to Earth, “hand of god” doesn't explain anything at all. You are working your way around a mechanic you do not understand.

Now, why we fall back to Earth - ie. why there is gravity - cannot be known and might be a god, but the how doesn't make sense to be answered with “god” or anything other supernatural¹⁾.