I've been discussing with Objectivists. I learned something new:
Lots of their thinking about epistemology presupposes an intelligent consciousness and proceeds from there.
They don't say this clearly. They claim to have answers to epistemological problems about how learning works (with perception, concept formation and induction). They claim to start at the beginning and work everything out.
Traditional approaches to induction try to say how intelligence works. They claim they solved the problem of induction. But they aren't actually focusing on the traditional problem. They aren't very clear to themselves about what problem each idea is meant to answer, and don't consistently stick to addressing the same problem.
Their approach to concept formation presupposes intelligence. How do you know which concepts to form? How do you know which similarities and differences are important? How do you decide which of the many patterns in the world to pay attention to? Use common sense. Use intelligent judgement. Think about it. Use your mind. Consider what you value and which patterns are relevant to pursuing your values. Consider your interests and which patterns are relevant to your interests. And, anyway, why do you want a mindless, mechanical answer someone could use without thinking, anyway?
So induction requires concept formation which requires being intelligent. Their take on induction presupposes, rather than explains, intelligence. It's kinda like saying, "You learn by using your intelligence to learn. It handles the learning, somehow. Now here are some tips on how to use your intelligence more effectively..."
They don't realize what's going on but this is a dirty trick. Induction doesn't work. How do you fix it? Well, induction plus intelligent thought is adequate to get intelligent answers. The intelligent thought does all the work! Any gaps in your theory of learning can be filled in if you presuppose an intelligence that is able to learn somehow.
One of the big points of epistemology is to figure out how intelligence learns without presupposing it works somehow. Yes it does work somehow, but let's figure out the details of the somehow!
I say new knowledge is created by evolution. They don't address the problem of how new knowledge can be created. Intelligence can do that, somehow. They don't know how. They seem to think they know how. They say intelligence creates new knowledge using perception, concept formation and induction. But then when you ask about the details of concept formation and induction, they presuppose intelligence...
Note: I do not blame Ayn Rand for this. I don't know how much of this is her fault. As far as I know from studying her writing, she didn't do this herself in her published works.