I've always called those words 'soft categories'; they don't refer to the state of reality, but to the models we use to explore it. (And there are as many model

I've always called those words 'soft categories'; they don't refer to the state of reality, but to the models we use to explore it. (And there are as many models as there are brains to hold them.) They are prone to making people agree with each other without requiring that they understand each other or how reality actually behaves. They're sort of like low order approximations: if we all say "π is about 3," nobody disagrees. (Or if someone does, they may risk rejection for it.) Either way, it doesn't help you solve complex problems. The solution I usually propose is to not use the term at all: to avoid it in its entirety.