I think this has to be the longest-running Asylum conversation ever, but what the hell, here's to spawning a new thread of it...
On the subject of a real, physical world, and the mental perception of it, I am of the persuasion that humans are rather complex systems of matter and energy, which happen to have the unique trait of being able to model other systems, existant or non-existant, within itself.
Other, less complex systems - the rest of the animal kingdom, and pretty much anything with any kind of sensors - can also do this to a limited degree. Humans are fairly unique, as far as we can tell, in the ability to model non-sensory information within ourselves. We can imagine things that don't exist, or things which we think may exist but can't directly perceive ourselves.
We also use a nifty little modelling process we call math to quantify certain aspects of the systems we are modelling, and thus more accurately model these other systems, or model systems that could not be usefully modelled any other way. But most of the time, since we are designed to model things in formats akin to the information we get from our senses, we ultimately translate the results of our models back into a "physical" meaning of them. Math or "concepts" don't ultimately matter - math is just a concept, that we use to model certain systems that cannot be effectively modelled otherwise. But then math itself is rather difficult for us to render in our minds as meaningfull information, so we ultimately translate it back to more "abstract" concepts.
There is a "physical", distinct universe out there, but we cannot seeing it. Or rather, we are seeing it, and hearing it, and otherwise perceiving it, and that's the problem - the nature of existance is not a subset of our system, a memory or model in our minds, and so we can't simply KNOW what is. The only way we could would be if it were a smaller part of us, as though we were God.
So we must perceive it, or cause other things, scientific instruments and such, to perceive it and then incorporate that information by observing them.
The problem is, other things can also effect our models. Innacurate math in models where that is being used, flawed translation back into "tangible" conceptual models, imagination, drug, or other mental modelling induced by non-sensory information... or, if we want to get sci-fi, someone feeding inaccurate information into our senses or directly into our brains, ala The Matrix and all its kin.
So for the most part, we CAN try to "perceive" the physical universe, if we can rule out bad equations, sloppy transliteration of them to the perceptable systems in question, people's wild imaginations, drug-induced hallucinations, or the mind-slavery of a race of malicious artificial intelligences.
Most of the time, as far as we know, we're not hallucinating, dreaming, or having false senses fed to us (though you can never actually tell, kinda by definition, but then it just comes down to having to accept that seeing, within reason, is believing). If we can control our imaginations, as most professionals can, then we can rule out that we "made it up". Transliteration to sensory formats of data is useful for humans to understand what it is exactly that they're predicting, but there is no actual "right" or "wrong" model, so long as it maps accurately enough to the variables used in the equations.
For most of modern physics, we've got all of the above hammered out and straight, occasionally creating new perceptual models to fit the new quantitive, mathematic models. And now it's mostly just down to getting the values right, figuring out how much of x is added to attribute y of system z when ABC happens.
So, to coin a phrase: "Nobody KNOWS anything. We're just getting better and better at guessing."
(To Archer, and his dropped-ball experiment: no, you don't know what is going to happen to the ball. You don't even KNOW what the ball is. You have a model in your mind of the ball and other sensory data, and can model a 99.99_% accurate simulation of the results that will come to be when you perform the action of dropping it. I leave that infinitesimal chance that your model of the ball falling is wrong because you're not modelling the entire universe, and blue-sky lightning could come down and destroy your bedroom for all you know. Besides, 99.99_% is still giving you a lot, because you're not modelling EXACTLY what will happen down to the finest level, and when it gets down to the quantum level, you couldn't gather enough data to get 99.99% accurrate even with the most advanced scientific instruments).
