Gisteron wrote: Since all of the difficulties you are naming here, Senan, also apply to medicine, would you call that a social science, too, with similarly questionable scientific rigor? Controlling variables doesn't mean eliminating them, it means being mindful of them. If you account for all the potential sources of error, your total "margins" of error might get enormous, but at least the results your reviewers would get might just be within those same margins. Intellectual honesty and transparency are quintessential hallmarks of science. Undue confidence is not. If you as a researcher either ignore important variables knowingly, or fail to mention them in your paper by mistake, that's what makes the results irreproducible. If the margins are understated, to get a result that deviates well beyond them is trivially easy, and that is so with history to no smaller extent than with chemistry.
Science is not a subject, it is an approach, a way of figuring things out. Its domain is all of our shared reality, no matter how obscure the causes of the final observables are.
But yes, I agree. Science is difficult. To account for variables one left uncontrolled retrospectively is boring and taxing, and there are plenty very difficult ones to account for particularly when dealing with humans, as you said. There is of course an easy, lazy way out by just not accounting for them at all. It is however, as I said, a way out.
First, I am not a scientist and my knowledge and practice of it has been limited to what I received in California public schools and universities, so I'm likely wrong most of the time. I count on you mainly, but others as well to correct my ignorance, Gisteron, so "thank you."
I guess what I was trying to get at is when you mix two chemicals under similar conditions (similar amounts, temperature, container, etc), you tend to get a very predictable and repeatable result because the chemicals don't change their minds or their behavior on a whim. They have no free will. They will generally do the same thing every time unless there is a new variable introduced or a condition is changed. Chemical reactions are not motivated by money or looking to impress anyone. A medication to treat cancer doesn't do it because it "hates" cancer. It just does what chemistry and biology dictate.
It is the scientist and their methods that become the less controllable variables because they are human beings with motivations and desires that may motivate them to achieve a certain result even if it isn't the provable one. Thankfully other scientists can usually prove their errors or omissions pretty easily when the results are not the expected ones and can't be repeated. Or the resulting drug to treat your itchy foot ends up giving you brain bleeding and warts on your eyeballs that the greedy scientist forgot to mention in his paper and the free market and legal system fix the problem instead.
Inserting people into science creates higher margins for error, and even more so when your subject is also human. Pretty soon the margin for error outweighs any sense of certainty about the results, which is why right now phenomenon like ESP are extremely difficult to "prove".
