I’m not sure Scott isn’t just falling victim to the sorites paradox here. There are lots of macroscale definitions which seem to break down at their smallest application, and it’s not immediately obvious that consciousness couldn’t be one of them.
The question is whether to interpret such a falling apart of a definition (which I take to mean that related decision problems cannot be clearly answered anymore) as an inherent or even necessary attribute of concepts which ‘live’ at a macroscale, or as a weakness of said definition, as a sign that we’re mistaking a fuzzy word cloud for a precisely defined set.
Hmm. I see his point, I thinks, but I … think it does mean that, actually. Without fully understanding the definition, you should be less sure that a better understanding wouldn’t classify them differently.
Picture a slave-owner saying something similar about a slave, for instance. Slave-owners were even more confused than we are about personhood, and I think it’s clear that they weren’t “crystal clear on what [isn’t a person”, in retrospect.
Sure, there are debatable cases. But there are also clear-cut ones, like a bacteria, while alive, has no personhood”, and if your model predicts that it has more personhood than a human (as IIT does for consciousness for a certain 2D configuration), then you should not call whatever your model describes a bacteria has more of as “personhood”.
Scott Aaronson in reply to the statements like “A stone is conscious to the “inputs” of gravity and electrostatic repulsion”
I’m not sure Scott isn’t just falling victim to the sorites paradox here. There are lots of macroscale definitions which seem to break down at their smallest application, and it’s not immediately obvious that consciousness couldn’t be one of them.
The question is whether to interpret such a falling apart of a definition (which I take to mean that related decision problems cannot be clearly answered anymore) as an inherent or even necessary attribute of concepts which ‘live’ at a macroscale, or as a weakness of said definition, as a sign that we’re mistaking a fuzzy word cloud for a precisely defined set.
Hmm. I see his point, I thinks, but I … think it does mean that, actually. Without fully understanding the definition, you should be less sure that a better understanding wouldn’t classify them differently.
Picture a slave-owner saying something similar about a slave, for instance. Slave-owners were even more confused than we are about personhood, and I think it’s clear that they weren’t “crystal clear on what [isn’t a person”, in retrospect.
Sure, there are debatable cases. But there are also clear-cut ones, like a bacteria, while alive, has no personhood”, and if your model predicts that it has more personhood than a human (as IIT does for consciousness for a certain 2D configuration), then you should not call whatever your model describes a bacteria has more of as “personhood”.
It reminds me of Justice Potter Stewart: “I know it when I see it!”
Well, it’s the converse, which seems a lot more useful a criterion to me.