I am more at ease with the direction Hofstadter is taking his argument about how the actual architecture of the brain and the actual molecular arrangements of proteins on neurons do not need to be fully explicated in order to at least think about thought processes. This is the same way how you don’t really need to know how a microprocessor actually works in order to program it in assembly.

This is, however, in contrast to the example of the C++ programmer. While it’s not necessary to understand what the compiler actually changes your code into, if you ever want to do optimization, it’s helpful.

Consciousness is yet another example of an emergent phenomenon—essentially a process that is not readily predictable from its component parts. In this way, he’s right, you can’t just look at individual neurons, or even encapsulated subsystems within the brain in order to understand consciousness.

A very common emergent phenomenon that most of us deal with almost daily is traffic. While clearly patterns are caused by socioeconomic trends, the price of real estate, the location of jobs, how well-repaired a particular road is, the particular way a segment of freeway curves, the price of gas, etc., etc., studying any of these things in isolation is not going to do much to inform you about where the slowdowns are going to be today. Even studying individual drivers and their cars will not yield very much.

Another example of emergence is weather: you can’t really look at individual molecules of nitrogen, oxygen, water, and carbon dioxide, and decide whether it’s going to rain or not.

My point, though, is that we don’t exactly know how many levels above the physical substrate consciousness arises from. While neuroanatomists tend to refer to particular subsystems in the brain as if you could disentangle them and look at them in a vacuum, any neurologist can tell you that there’s no way you’re going to hit just one subsystem with that blood clot or bleed. So we’re stuck with looking alternately between the somewhat abstract idea of cortical subsystems, and with the realization that we’re really just dealing with networks of neurons all connected in a particular, and remarkably reproducible way.

Which gets us to another point: how is my consciousness different from your consciousness? Clearly we have different genes. We’ve experienced different things. And yet, at least on a gross, subsystemic level, my neurons probably aren’t hooked up very differently from your neurons. At least, that’s what we assume (the thalamus is connected to the amygdala, the amygdala is connected to the hippocampus, the hippocampus is connected to the frontal lobe, etc., etc.) The individual connections probably differ minutely (no more than 1%, I bet), but is it enough to explain why I behave the way I do, and you behave the way you do?