In Bryant's blog post "Critique as second-order observation" he once again brings up Spencer-Brown's law of form. Distinction is the condition for the possibility of observation, and with every distinction there is an unmarked space that is not visible to the distinction so made. And yet we can and do use second-order observation to "observe the observer" in this process of drawing distinctions, thus gaining a more comprehensive observation when we make visible the heretofore invisible assumptions upon which the first-order observation was made. While true that even the second-order observation also had to draw a distinction and thereby has its own invisible unmarked space, it nevertheless broadened the view of the previously hidden first-order invisible unmarked space.
So while there is no end in sight we still make progress in this process. And it it here that we need the spiral dynamics of evolving worldviews, both individual and collective, via developmental psychology. And it seems this sort of observational developmental process is missing from the likes of Bryant, even though it is incipiently implicit in this post. He grants a mereological progression of scale with his machines, but he is not familiar with Edwards' 3 types of holarchies. Bryant's might fit more into the ecological holarchy but the developmental holarchy is necessary to understand this re-iterative progression of second (and third, fourth etc.) observation, which only comes with human consciousness.
This is also missing in the work of Luhmann, on whom Bryant draws heavily. And the likes of Thompson and Varela, which point I was making in the last few posts in the Thompson thread. This sort of interative observational loop is the hallmark of reflective or narrative self consciousness, which is not the suobject of study with Thompson. He focuses on the present-centered core self and its benefits, not the benefits of the reflective observations of that developmental process which indeed provides the kind of 'integral' worldview context he intuits but does not explicitly state sans this modeling.