I am guilty of proposing an alternative form of ”material agency” called polyagency which, in my dissertation thesis (2006:156), is defined as a “collective term for intensive processes between actualized entities whose virtualities generate a multitude of transformations. Polyagency consists of four interrelated concepts that describe inseparable phases of becoming: the in-between, individuation, stratification and the time-shelter. These intensive processes also occur in the formation of actual entities where there is no human relation. Polyagency is used to explain how matter and materiality changes in encounters”.
This was clearly influenced by Deleuze and Bergson. However, the very term agency does not come from them. It comes from the anthropologist Alfred Gell and the sociologist Anthony Giddens which were great inspirations in my earlier licentiate thesis. In my subsequently published articles on causeway assemblages I ditched polyagency and polyagents as terms since some people confused it with Latour’s actants (although Latour emphasizes a completely actualized perspective where processes of emergence are lacking).
I have decided to publish at least on more article based on my original concept. After that I will not only drop polyagency but any form of “agency” in my terminology. It is because it is a superfluous concept in object-oriented ontology. Human agency is dependent on the subject-object distinction which is erased in OOO where only objects exist. According to Ingold, agency is usually described as something that is acquired because otherwise the human would only be a thing and therefore agency must also be bestowed on things so that they can act like people. Hence, any form of material or object agency gives objects the capacities of humans and therefore maintains the subject-object distinction. OOO works the other way by showing that the human is a variety of the “inanimate”, only different in degree, not in kind. Although my late “Deleuzean” definition of polyagency actually has dropped agency as a crucial component (as seen in my definition in the beginning of this post), the term itself may be confusing. Therefore I have decided to not use it anymore after the article in question has been published.