ISTM that if you weren't consciously aware of being in control, the feeling that you were in control must be retrospective - you remember that you were driving.
This doesn't follow.
Most of the time when I walk, I'm not aware of it. On occasion, my leg falls asleep to the point that I lose sensation in the leg. When I try to walk, suddenly I'm very much aware of it. What's interesting is that there being feeling in the leg affects my ability to walk, even though I'm not consciously aware of the feeling (in fact I'm only consciously aware when something goes wrong when I try to walk--and I'm not even consciously aware of the goal of walking until that thing goes wrong).
This suggests that, in the nominal case, I feel my leg subconsciously. If that's the case, feeling isn't limited to consciousness--only the awareness of feeling. So it doesn't follow that I only retroactively "feel" when I reflect back on it. Only the tautology follows--that I'm only aware that I'm feeling when I reflect on it.
There is a linguistic difficulty here in the need to differentiate between 'self' as the whole person, the actor, and 'consciously aware self' that is the reflector.
My objection is that I think the self you're describing is a broken concept.
There are prerequisites for one to feel he is in control--and they correspond to particular types of sensations. There are various disorders of control and oddities that demonstrate that these are real sensations (alien hand syndrome, schizophrenia, intentional binding, involuntary movements such as tics, and legs that fell asleep).
This in mind, at the highest level, we have Joe. J is only the "consciously aware" part of Joe. J feels that J is in control. But really, Joe is in control. So J's feeling that J is in control is an illusion. But quite similarly to my leg asleep scenario, "Joe" feels "Joe" is in control anyway, when "he" is, and when "he" isn't, there's a conscious alarm (this much seems to fit some of Eagleman's descriptions).
But this doesn't work, because the feeling of control is produced by Joe. So, when J is feeling a sense of control, J is actually feeling Joe in control. Furthermore, J is only in conscious awareness when J happens to be self reflecting, which is rare. So in those other cases, Joe is feeling that Joe is in control (and when not, Joe is made consciously aware of it quite quickly).
Furthermore, awareness is the state of being aware of something--that is an exercise in conceiving, in information gathering. Conscious awareness is a mode of awareness. So it doesn't make sense to me to describe conscious awareness as
ever "doing" anything, where doing is understood in the agency sense of initiation of goal-based behaviors. It could cause something to be done, by virtue of communicating this concept, but it in itself is simply the conception.
I'm very suspicious that there's even a coherent entity to talk about that is the "conscious you", where "conscious" is used as a qualifier.
The consciously aware 'self' is informed and updated about the actions of the non(or less?)-conscious self.
I'm not sure these are two different pieces. I think consciousness is a mode of awareness, not a type of self. I don't see any coherent "type of self" consciousness to talk about, so I prefer dealing only with a "self" that has a "conscious mode" of awareness.
All of the
teleological actions are of the same sort. They deal with the same sort of models we talk about in language, and the same sort of goal-based actions, whether they are unconsciously performed or consciously performed. I don't think these concepts are made once for the "conscious self" and once for the "unconscious self", and communicated from the unconscious to the conscious when the need arises. I think they just exist singularly as concepts, and they're conscious when the spotlight hits them, unconscious otherwise.
Likewise, I don't think there's a separate coherent "conscious self"; any aspect of yourself that you can be conscious of, can involve itself in an unconscious activity. It doesn't make it a new aspect--it's the same thing, only without the spotlight on it.
I would also like to note that there's almost always a singularity of purpose in our actions, for normal subjects. Either I'm driving on autopilot, or I'm driving consciously; there's never a fight between the two. Since there's a lot of things that go on subconsciously, this suggests a collaboration. Rather than a separate internal mind discussion for every kind of potentially conflicting action between subconscious decision makers and conscious ones, a simple global teleological "agency", which simply has a spotlight that can shine here or there, is in my mind a more workable model than a "conscious self"/"unconscious self" one.
However, I suspect that the line between conscious and subconscious is pretty blurred. The 'CEO' is not the main actor, but an occasional adjudicator; present to direct general policy and resolve internal conflicts.
The concept of "CEO" conveys what I believe to be a false notion--that conscious awareness is itself an "agency", can itself initiate actions, has its own feelings/sensations (especially that there's no such thing as a feeling unless there is a conscious awareness of it), and so on. I think it's better to think of conscious awareness as a sort of highly focused attention, than a separate thing with a separate copy/control line to any idea or notion we're capable of being consciously aware of.
