Philosophical functionalism

Philosophical functionalism is often understood as applying to mental activity, such as thinking, or what might be considered biological activity, such as feeling pain. But functionalism can be understood more generally as a focus on actions or functions performed rather than on objects. A standard example is that of catching mice. One can build mouse traps having many different designs and being constructed of many different kinds of materials. They all perform the function of catching mice. A functionalist perspective focuses on the function or action performed rather than on the objects performing those functions or actions.

I had a discussion with Gemini (Google’s “AI Deep Dive” into a query about philosophical functionalism) attempting to clarify issues related to functionalism and emergence. I know that emergence is by now a pretty tired idea, but I think the result was worth the effort.

Gemini’s responses have been edited for clarity and brevity.

Me: Emergence is often applied to functionality. Discuss how functionalism relates to emergence.

Gemini: Functionalism and emergence are deeply intertwined concepts that both seek to explain how complex, high-level properties (like consciousness or social stability) relate to their simpler, underlying parts. Emergence is the ontological claim that a system has novel, irreducible properties. Functionalism is the methodological insight that these properties are defined by their causal roles rather than what they are made of.

A core pillar of functionalism is Multiple Realizability—the idea that the same function can be performed by different physical materials (e.g., a “clock” can be made of gears or silicon).

• Anti-Reductionism: Because a high-level function isn’t tied to one specific physical state, it can be viewed as “emerging” from the physical level rather than being identical to it.

• Projectibility: Functionalism argues that emergent properties are scientifically “real” because they follow their own laws (like the laws of psychology or economics) that remain true even if the underlying “hardware” changes.

[Me: Gemini then offered to discuss Functionalist Emergentist Materialism (FEM). I accepted its offer.]

Gemini: Functionalist Emergentist Materialism (FEM) is a specific synthesis in the philosophy of mind. It attempts to "have it all": it keeps the physical world as the only foundation (Materialism), acknowledges that the mind is a set of processes (Functionalism), and argues that these processes create something genuinely new and irreducible (Emergence).

Here is the breakdown of how these three pillars fit together:

1. The Materialist Foundation. … There are no “ghosts in the machine” or non-physical souls. …

2. The Functionalist “Middle-Man.” … FEM rejects the idea that a mental state (like “fear”) is identical to one specific brain state (like “neurons firing in the amygdala”). Instead, it uses Functionalism to argue that “fear” is a high-level organizational pattern. …

3. The Emergentist “Top.” … When physical matter is organized in a sufficiently complex functional network, emergent properties appear. You cannot understand “justice” or “longing” by looking at a single atom. These properties only exist at the level of the whole system. …

Me: Is FEM a version of Non-Reductive Physicalism?

Gemini: Yes, FEM is a classic version of Non-Reductive Physicalism (NRP). In addition, FEM makes an ontological claim that the high-level functional patterns actually emerge as real, new features of the universe.

Me: Since FEM is a theory of functionality, does it include some concept of atomic functions or atomic actions?

Gemini: Yes, but with a twist. The Tension: Emergence vs. Atomism. Atomic functions are just physics. They follow strict, boring laws. When you get enough of these atoms together in a specific functional architecture, the “whole” begins to exhibit properties that the “atoms” do not have.

Me: Do quantum transitions (or observations) count as primitive functionality or primitive actions?

Gemini: If you define the "atoms" of function too low (at the quantum level), you risk losing the "Functionalist" part of the theory, because the "role" becomes indistinguishable from the "physics".

Me: Defining quantum transitions as functional primitives still implies a functionalist perspective. The function of a quantum transition is to change the state of the elements that participated in the transition, thus preserving the functionalist perspective.

Gemini: You are making a sophisticated move here: you are pushing functionalism all the way down to the “floor” of reality. If a quantum transition is defined by its role in changing the state of a system, then you are treating the laws of physics themselves as a functional program. In this view, an electron isn’t just a “thing”; it is a functional operator that processes inputs (fields, energy) into outputs (new states). This suggests that "function" isn't something that only "emerges" in biological brains or computers, but is the fundamental nature of matter itself.

Your argument aligns closely with Informational Physics (the “It from Bit” doctrine of John Wheeler).

• If a quantum transition’s function is to “change the state of elements,” then every transition is essentially a logical gate in the universe’s computation.

• In this light, a “state change” is the most primitive form of information processing.