> You seem to be engaging in difficult philosophical concepts and jumping over important steps (like rigorous definitions or showing why something “must be”).
Yeah, sorry, I bashed that out too quickly. I'll try to slow down a bit, though I can't promise a full treatise :)
For a thought to be about reality, there must be something about that thought that makes it about reality. How this might be the case is one thing, but the fact that it is the case is nonetheless necessary. And for the thought to be about the reality, there must in one way or another be similarity between thought and reality. Again, the 'how' may be extremely complex, but the existence of the similarity seems necessary. To put it in opposite terms, if a thought is not similar to reality, then it is not about reality, and therefore can't constitute knowledge about reality. Do you agree thus far? If so, let that be premise 1.
I'm obviously talking about the concepts we use to think about reality, not about the words we use to describe it.
I will assume that we agree that thoughts can be about reality (even though they're obviously sometimes not) as premise 2.
So, given our two premises, suppose Bob is thinking about a tree. With materialism, where is the likeness between Bob's mind and the tree? It cannot be the neurons or the physical structure of his brain or body, because these things don't have similarity with a tree. The atomic structure of a brain, in any state, is different from a tree's. That much is clear. And likewise, a tree doesn't somehow appear in Bob's brain. Given our premises, this is a problem for materialism. If the likeness is not physical, and it clearly isn't, then what is it?
We can discuss the idea of a tree as 'emergent' from the matter&state. But honestly, I struggle with this concept. Every time I've heard the word used, it seems to be a way of restating the problem rather than solving it. Perhaps you could explain in a bit more detail how this overcomes the 'similarity' problem I've outlined above? ISTM you still get the problem that a bunch of neurons is not a tree, regardless of their arrangements. Either the neurons in some sense 'cause' you to think about a tree, in which case you need to ascribe to the neurons an ability to cause immaterial effects, which again seems a problem; or it is simply a way of saying "we don't know how it works". Again, so ISTM.
On computers: very briefly, a computer doesn't 'know' anything about a tree, or anything else. A 'tree' on a computer is a bunch of magnetic states on a hard drive, or transistors in particular states, or what have you. The meaning of a computer's output (that is, the link between the computer and reality) exists only in the mind of a human being. An LLM's text about a tree is no different in principle from the binary for the word 'tree' on an hdd, or a jpeg, or whatever. An LLM gives the illusion of knowing, but no more than that. Again, I realize this is a pretty summary treatment, but I think the tendency to describe the mind as a computer is misguided for this reason. You face exactly the same problem with the computer as you do with the mind. For my part, I simply reject premise 2 as far as a computer is concerned.
Anyway, very simply, if materialism is true, a thought cannot be about reality, and it cannot have a similarity with reality. Therefore, we can't know anything about reality. But we can know about reality (premise 2), therefore materialism is false.