> Do children "play" with dolls? I don't think that is a slam-dunk. Seriously? > As far as we know, children may be mindless automata Yes, that's possible.

> Do children "play" with dolls? I don't think that is a slam-dunk. Seriously? > As far as we know, children may be mindless automata Yes, that's possible. You can't prove that humans in general are not philosophical zombies. But I have memories of being self-aware around age 4 or so, so from my own personal experience I put that as an absolute upper bound on where humans might not be persons yet. On the other hand, I also remember exhibiting I/O behavior that would cause an outside observer to conclude that I was experiencing much more intense pain than in retrospect I was actually experiencing. > Facial Action Coding System Of course it is possible to create a taxonomy of facial expressions and how those correlate to physical stimulus. What you can't do is know the subjective experiences that those expressions correspond to because no adult knows what it is like to be a baby. If anyone actually knew that, we would not be having this discussion at all. > We can draw robust, valuable conclusions about subjective states without enjoying deductive certainty. I don't deny that. What I'm saying is that you cannot draw those conclusions simply by observing a system's I/O behavior. You have to take the underlying mechanism into account. If you doubt this, look at this photo: https://images-na.ssl-images-amazon.com/images/I/91YBjzSRmnL.jpg https://images-na.ssl-images-amazon.com/images/I/91YBjzSRmnL... The subject of that photo appears to be in distress, but it is not in fact in distress. The way I know this is that the subject of that photo is actually a doll (look at the eyebrows), and dolls cannot be in distress. So there is an existence proof that merely because a physical system appears to be in distress that is not enough evidence to conclude that it is in fact in distress. You need something else.