> But if they answer no, it's easy to show that they're mistaken, and that in practice such labels are applied to statements regardless of their truth or falsit

> But if they answer no, it's easy to show that they're mistaken, and that in practice such labels are applied to statements regardless of their truth or falsity. That seems like a potentially hazardous undertaking.