The theory of epiphenomensalism states that mental phenomena are causally inert by-products. Like the steam of a train whistle, mental phenomena are merely side-effects of an underlying physical/functional process. It seems to me that the question of epiphenomenalism is usually framed in terms of whether all mental phenomena are epiphenomenal. However, it has just struck me that this might be the wrong question. Maybe we should ask whether only some mental phenomena are causally inert by-products.
Let’s define pain as the brain’s nonconscious detection of cell damage signals. Let’s define the feeling of pain as the brain’s metacognitive detection of its nonconscious detection of cell damage signals. I assume, unlike David Rosenthal, that there is something-it-is-like for a creature’s brain to nonconsciously detect cell damage. I also think there is something it is like for the brain to metacognitively detect nonconscious detection of cell damage. I think that it is these metacognitive feelings of pain that are conscious. I want to say that these feelings of pain only (actually) exist in the functional sense. This functional sense runs “automatically”, just like the nonconscious detection of cell damage. However, feelings of pain also have a subjective sense. It is this sense that is our conscious experience, which we can introspect and report on. This is the feeling of pain as opposed to just the pain itself. It is the type of feeling that makes pain unbearable and unpleasant. The idea I am beginning to develop is that the subjective component of this metacognitive system is epiphenomenal, in the sense that it is a by-product.The subjective component of conscious feelings is what-it-is-like to be a metacognitive brain system.
This conscious/nonconscious distinction can be made for more modalities than pain. We can define “feelings of vision”, “feelings of touch”, “feelings of taste”, “feelings of body”, “feelings of smell”,and “feelings of hearing” all in the same way. Partial epiphenomenalism is the view that only some aspects of mentality are causally inert. Since the conscious feelings are realized by metacognitive neural assemblies, and all neural assembles deterministically “run” automatically, the conscious feelings themselves are epiphenomenal. Partial epiphenomenalism is partial because it says that these nonconscious first-order systems are genuinely mental and obviously any nonconscious system is causally efficacious. After all, rocks do not detect cellular damage at all, nor do they have to stay in homeostasis. And I have been arguing for awhile now that there is something-it-is-like for these nonconscious organisms to exist. But now take a nonconscious brain and provide it with a Global Neuronal Workspace that can metacognitively synthesize disparate specialized nonconscious processors and broadcast its information throughout the brain. There is going to be something-it-is-like for that GW to process its information. That something-it-is-likeness for the GW to run is what I mean by feelings of consciousness. Strictly speaking, the subjective component does nothing causally. So why is it there in the first place? Because it’s the inevitable result of what-it-seems-like to be an embodied organism. This might just be a brute fact of organic life. Either that, or panpsychism is true.
Regardless, why are only the metacognitive systems conscious? Because of how the GW is organized, it is tied into “executive” decision making, which is I think what ordinary people mean when they talk about “voluntary” actions, the actions we are responsible for because in some sense we consciously willed them. But I am in agreement with Libet and Dan Wegner: the feeling of conscious will is epiphenomenal. But it would be rash to therefore conclude that consciousness itself is causally inert. Because the feelings of will are realized by fully functional causal systems like the GW and these systems are very much causally active, it would be a mistake to conclude that consciousness itself is an illusion simply because the feelings of consciousness are illusions. The illusory feelings are the what-it-is-like for a GW to operate. Another way to say it is that consciousness is in the business of producing illusions.
Part and parcel of the content that guides the priors of the GW operations are an internalization of learned linguistic concepts. And as Julian Jaynes said:
Let no one think these are just word changes. Word changes are concept changes and concept changes are behavioral changes. The entire history of religions and of politics and even of science stands shrill witness to that. Without words like soul, liberty, or truth, the pageant of this human condition would have been filled with different roles, different climaxes. And so with the words we have designated as preconscious hypostases, which by the generating process of metaphor through these few centuries into unite into the operator of consciousness.
Compare two creatures, equally intelligent. One has been taught how to use psychological words like “mind”, “belief”, “desire”, “consciousness”, “will”, “reason”, and the other has not. Who is going to have the kind of intellectual life we most associate with conscious thinking adults? Obviously the one in possession of the right concepts. These concepts operate as the contextual priors of the metacognitive GW system. Micah Allen and I have called this “preloading”, with the idea being that socio-cultural factors “load up” the default mode network with social information and then this disseminates to the lower systems, influencing their processing, which in turn feeds back into the DMN for more processing. This loop is the continual loop of nonconscious states being operationalized into consciousness by metacognitive systems. This enables humans to do things like reminisce about a lover, to imagine what the future will be like, to imagine unseen vistas in your mind, to contemplate your past or future actions, to muse on an idea, to use inner speech as a cognitive aid, to project categorical structure onto the world, to narratize the world into stories, to think about your own life story, to tell stories, to articulate, to reason, to think about your own thinking, to reflect, deliberate, pontificate, to wonder, to gaze, to savor. You get the picture.
Of course what I have just said is not a full theory. It is only a sketch of a theory. But I think the overall picture is becoming clearer.