I think this is thoroughly expected. Imagine you have a set of legal words (legalese) and then you have a set of scientific words or statistical words (all can

I think this is thoroughly expected. Imagine you have a set of legal words (legalese) and then you have a set of scientific words or statistical words (all cants) and you show the sets to those not associated with those sets… we’d see similar results. It’s neat to find these oddities but that’s all it is, linguistically.