I agree that the first part was the strong part of the paper (and was the reason I decided to link it on HN), and e.g. statistics about differing word usage in

I agree that the first part was the strong part of the paper (and was the reason I decided to link it on HN), and e.g. statistics about differing word usage in different fields isn't what I'd call conclusive evidence. But are you postulating that all arguments (for a weaker version of) Scientism (either all arguments there are or even only the all arguments the author presents), are silly? That's something I'd disagree with. For example, while science isn't the perfect process like in the textbooks portrayals, a model that produces testable predictions (and thus can be evaluated) seems the very good way to characterize how we get new knowledge about various phenomena of the world and can convince others that it is knowledge and not just an esteemed opinion. Also the bit "vicious circularity" resonated with me: in my opinion, (looking at what we know of the evolutionary history our species[1]) it's quite reasonable hypothesis that humans could develop logic and mathematics in the first place because of the physical world we observe is regular and predictive enough so that human cognition had a chance to develop. [1] Not in the evo psych sense, just that we descend from the same great-great-*-parents as our cousins (the other) apes, and during the history we ended up being the ones with wider cognitive capabilities. Even though our cousins are relatively bright.