When I say that intentionality is not definitory of subjectivity but rather a secondary formation, I am thinking of Deleuze’s idea that what is primary are affective states, qualitative apprehensions that create a moment of indetermination in the input-output causal chain governing action. All this is material, I am certainly not claiming that the mind is immaterial. This corresponds to what C.S.Peirce calls “Firstness”. For Deleuze we are in the first instance “affected” by our interactions with our environment and this leads to a possibility of novelty in our reactions that are not constrained by a fixed sensori-motor schema, i.e. this affective capacity is a key part of why we are not mindless robots, but capable of new reactions.
I do not deny the existence of internal representations, but I do not give them explanatory centrality. I would not say that we only “need” to be embedded, but that this embedding is constitutive of what we are. Bernard Stiegler makes a difference between cognition and knowledge in the sense that knowledge is constituted within and dependent on this primary sociality and technicity of humans, and it acts back on our individual cognition forming it, correcting it, transforming it.
A key example he gives is learning to read. Contrary to learning to speak a language, which is hardwired into our brains, there are initially no neurones of reading. Reading is a cognitive skill that is mastered from a massive rerouting and readapting of brain modules that initially had nothing to do with such a skill. It is a materially inscribed apprenticeship, that is not to be explained in Darwinian terms, except that natural selection has given us enough plasticity to be capable of such neural (and “cognitive”) re-assignment. Reading implies the technicity of inscription and the sociality of shared meaning. As cognition it is of a different order than successful or failed perception, accurate or erroneous memories. As dependent on and impacting the cognitive resources of the brain, it involves the mobilisation of pre-existing brain mechanisms, and their re-grouping and re-assigning to other functions,
Thus although I think that there is no incompatibility between the Deleuze-Guattari-Foucault-Stiegler account and the reliance on internal representations, I do not think that representations are primary in the explanation of subjectivity, nor are they sufficient to account for the interaction between cognition and knowledge that constantly re-works and re-grounds our cognition. I don’t think representation can even get going without social interaction and technical interfacing, in the very general sense where language itself is technical.
This does not reduce subjectivity to a mere epiphenomenon, as in my account subjectivity is distributed more extensively than the limits of the brain and acts back on and reinforces, validates, corrects, and transforms the brain’s cognitions. This psychic, social, and technical grounding of knowledge is not just an external yet necessary framework, but constitutive of subjectivity and cognition. The further question of the presence and the localisation of internal representations and their degree of causal efficacy is an empirical question that I have no desire to legislate on.
For Stiegler, who is a materialist, all language is inscribed, it is marks and inscriptions, that nonetheless instantiate and facilitate processes of categorisation. Stiegler insists that the representations, based perhaps on mnesic traces and their assemblage and transformation, that he calls secondary retentions, and that can be individual or collective, must be discernabilised, discretised, stabilised, and synchronised by collectively and technically inscribed tertiary retentions. This is a post-Husserlian re-working of the impossibility of a private language.