> This is why it is imperative to build your model up from first principles.
The main problem with this approach (I'll call it the Cartesian approach because it was most famously used by Descartes) is that human beings are less than 100% reliable at logical reasoning. If you make an error anywhere in your chain of reasoning, your conclusions are going to be off and there's not going to be any way to check them. It's just like writing 10,000 lines of code without ever actually compiling it, let alone testing it. You also develop a sort of foolish confidence about the correctness of your own beliefs, which makes it even easier to be wrong. If you make enough wrong turns, you become Ayn Rand.
That's why empiricism is so good. It's not that empiricists don't make mistakes too, but when they do, they find that they are surprised by concrete facts that they observe, and know when to go back and reevaluate.
Another helpful trick is to understand that there are degrees between 0% and 100% confidence. I can entertain an proposition as being possible or likely rather than simply true or false based on the recognition that I have incomplete information. If you tried to take this approach you couldn't derive anything logically because you would just have a multitude of possibilities in front of you. Formal logic only works with statements that are 100% true. Otherwise you're stuck with Bayesian reasoning, which is even more mentally taxing to derive information from.
(Or, as an alternative response)
Please derive for me, from "first principles", why it is imperative to build one's mental model from first principles.
The main problem with this approach (I'll call it the Cartesian approach because it was most famously used by Descartes) is that human beings are less than 100% reliable at logical reasoning. If you make an error anywhere in your chain of reasoning, your conclusions are going to be off and there's not going to be any way to check them. It's just like writing 10,000 lines of code without ever actually compiling it, let alone testing it. You also develop a sort of foolish confidence about the correctness of your own beliefs, which makes it even easier to be wrong. If you make enough wrong turns, you become Ayn Rand.
That's why empiricism is so good. It's not that empiricists don't make mistakes too, but when they do, they find that they are surprised by concrete facts that they observe, and know when to go back and reevaluate.
Another helpful trick is to understand that there are degrees between 0% and 100% confidence. I can entertain an proposition as being possible or likely rather than simply true or false based on the recognition that I have incomplete information. If you tried to take this approach you couldn't derive anything logically because you would just have a multitude of possibilities in front of you. Formal logic only works with statements that are 100% true. Otherwise you're stuck with Bayesian reasoning, which is even more mentally taxing to derive information from.
(Or, as an alternative response)
Please derive for me, from "first principles", why it is imperative to build one's mental model from first principles.