I’m not so sure we can dispense with #2. I think we think that because we abstract the rules of logic from the substance of speech and assume the former but not the latter is the real essence of thought.

But I think that’s backwards. I think that at least as far as organisms designed by natural selection go, every concept just is the feeling it gives us, every inference just is the feeling of adding, subtracting or modifying the cognitive architecture that makes concepts possible, and every complex idea just is the push and pull of feelings that move us to either embrace it or reject based on our needs and desires.

Those feelings were with us before language, and they’re still essential to understanding and motivation.

But they’re fuzzy. When it comes time to code rules of inference, we want formal structure and clearly defined terms.