Consciousness as Having Input
Re-reading Chalmers's 'Moving Forward on the Problem of Consciousness' (JCS, 1997, 4, 1, 3-46) I remember that he argues that consciousness cannot be explained by materialism in its present form because materialism only deals with structure and function. Consciousness is not structure or function. But is that all that materialism deals with? Phenomenal consciousness is having sensa, being aware of something, i.e. having an input. Materialism is quite comfortable with the idea that things have inputs. It is just that input tends always to be described in terms of an input-output relationship because things other than us can only be tracked by their outputs.
Phenomenal consciousness does not include output. Something that is aware can only be aware of its input. Whatever in our heads is aware tends to hold the belief that its output is responsible for certain aspects of the subsequent input. However, it cannot have direct evidence of this. Moreover, there is every reason to think that most of what alters input to awareness from within the brain is generated outside awareness - in 'the unconscious'. If we assume that consciousness is a property of some defined physical structure then we cannot include output in consciousness because we cannot know what the output of something with a particular input is. All we can know is the aggregate output of all the things with inputs, and there are lots of those in the brain. Theories of consciousness which describe it as 'motor', 'active' or 'enactive' are not theories of consciousness but theories of function of part or all of the brain.
The input of our phenomenal consciousness is rich and it is true that physics relatively rarely seems to deal with rich inputs. Meteorology would like to, but in practice probably has to rely on empirical prediction from past patterns. Bridge design famously fails to consider complex inputs, as in the case of the footfalls on the London Millenium Bridge. Rich inputs are familiar when it comes to lenses, pinhole cameras and magnetic resonance imaging machines but each component of the input tends to be dealt with as interacting locally with a separate part of the receiving structure. Holograms may be closer to what we want - a wave interacts with a mass of incoming information. Beyond that there is probably considerable scope for studying rich inputs but for outside objects there is probably usually no great pay off. Thus it is not that materialism cannot deal with inputs like consciousness, it may simply be a low priority in most cases.
For consciousness that priority changes. Part of the problem we have in responding to this different requirement may be that our language is often sloppy with regard to distinguishing inputs and outputs. A 'firing' seems to be an output and a 'stimulation' seems to be an input, but it is not entirely explicit. I suspect that to build a clear account of human experience we will need to tidy up our use of language about inputs.