I’ve thought about this some, and I think I see your point now. I would phrase it this way: It’s possible for a “Level 3 savant” to exist. A Level 3 savant, let’s posit, has a very deeply connected model of reality, and their excellent truth-detecting procedure allows them to internally repair loss of knowledge (perhaps below the level of their conscious awareness).
Like an expert (under the popular definition), and like a Level 1 savant, they perform well within their field. But this person differs in that they can also peform well in tracing out where its grounding assumptions go wrong—except that they “just have all the answers” but can’t explain, and don’t know, where the answers came from.
So here’s what it would look like: Any problem you pose in the field (like an anomalous result), they immediately say, “look at factor X”, and it’s usually correct. They even tell you to check critical aspects of sensors, or identify circularity in the literature that grounds the field (i.e. sources which generate false knowledge by excessively citing each other), even though most in the field might not even think about or know how all those sensors work.
All they can tell you is, “I don’t know, you told me X, and I immediately figured it had to be a problem with Y misinterpreting Z. I don’t know how Z relates to W, or if W directly relates to X, I just know that Y and Z were the problem.”
I would agree that there’s no contradiction in the existence of such a person. I would just say that in order to get this level of skill you have to accomplish so many subgoals that it’s very unlikely, just as it’s hard to make something act and look like a human without also making it conscious. (Obvious disclaimer: I don’t think my case is as solid as the one against P-zombies.)
I’ve thought about this some, and I think I see your point now. I would phrase it this way: It’s possible for a “Level 3 savant” to exist. A Level 3 savant, let’s posit, has a very deeply connected model of reality, and their excellent truth-detecting procedure allows them to internally repair loss of knowledge (perhaps below the level of their conscious awareness).
Like an expert (under the popular definition), and like a Level 1 savant, they perform well within their field. But this person differs in that they can also peform well in tracing out where its grounding assumptions go wrong—except that they “just have all the answers” but can’t explain, and don’t know, where the answers came from.
So here’s what it would look like: Any problem you pose in the field (like an anomalous result), they immediately say, “look at factor X”, and it’s usually correct. They even tell you to check critical aspects of sensors, or identify circularity in the literature that grounds the field (i.e. sources which generate false knowledge by excessively citing each other), even though most in the field might not even think about or know how all those sensors work.
All they can tell you is, “I don’t know, you told me X, and I immediately figured it had to be a problem with Y misinterpreting Z. I don’t know how Z relates to W, or if W directly relates to X, I just know that Y and Z were the problem.”
I would agree that there’s no contradiction in the existence of such a person. I would just say that in order to get this level of skill you have to accomplish so many subgoals that it’s very unlikely, just as it’s hard to make something act and look like a human without also making it conscious. (Obvious disclaimer: I don’t think my case is as solid as the one against P-zombies.)