I think it's a little bit of both. Scientific-type thinking is not restricted to modern times and a lot can be deduced simply from observation, given a large en

I think it's a little bit of both. Scientific-type thinking is not restricted to modern times and a lot can be deduced simply from observation, given a large enough sample size. On the other hand, a broken clock is right twice a day.