A Few Remarks on Agency

Over at Circling Squares and Struggle Forever, Philip and Jeremy are having a nice discussion of agency.  Since flat ontology is mentioned, I thought I’d pipe in with my two cents.  When I use the term “flat ontology”, I don’t use it to suggest that everything has agency.  I think agency is something pretty specific and isn’t common to all entities.  For me, flat ontology just means that everything is material (even incorporeal entities!) and that there is nothing outside of the world that conditions or overcodes everything else such as Platonic forms, God, and so on.  Flat ontology just means that there’s only the world.  It’s a synonym for immanence.  Why do I generally use the term “flat ontology” rather than “immanence”?  Because a number of people heavily influenced by phenomenology and the mind/body dualism seem to confuse immanence with what is in the mind rather than what is in the world.  For example, folks say that sensations are immanent to the mind while the object producing sensations is transcendent to the mind.  This is not how I use the distinction.  Immanence just means something is in the world.

On the topic of agency, I think that an entity must be minimally self-directing in order to be an agent.  That said, following Daniel Dennett in Freedom Explained, I also believe there are degrees of agency.  A bacteria has agency, but a very low degree of agency, because it is able to move towards and away from certain chemical gradients in a fluid or can move away from lie or darkness.  A cat has more agency than a bacteria or pine tree.  An adult human has more agency than a child, and so on.  If a cat has agency whereas a rock does not, then this is because a rock can only move when subjected to another force, whereas in one way or another all of these entities can initiate movement on their own (I sidestep the metaphysical question of freedom here as I just don’t know how to solve it).

Here are a few additional points:  First, the problem with humanist conceptions of agency is that they implicitly assume an idealized model of humans that says we’re all the same (usually that of white, middle class dudes).  Following Andy Clark and Deleuze and Guattari, however, I think this is a bad assumption.  A knight on horseback is a different agent, than a naked person and a person with a smart phone is a different agent than an aborigine.  We need to distinguish agents by what they can do, by their powers or capacities, rather than by resemblances.  I devote a lot of digital to ink to this thesis in Onto-Cartography, so I won’t rehearse it now.

Second, we need to recognize that there are agencies beyond human individuals.  Here I’m not simply referring to dog, blue whales, trees, and bacteria, but also corporations, governments, revolutionary political parties, and so on.  These entities– or machines as I now call them –both are self-directing in their own ways, and have forms of agency that can’t be found among the parts that compose them.  A revolutionary political party is able to do things that the individual person’s that compose them cannot.

Third, with Philip, I do not accept the structure/agency couplet.  I think the structure/agency couplet poorly describes how worlds are actually organized.  For me there are only machines and relations/interactions between machines.  What we call a “structure” is just the manner in which one machine, captures another machine in its “gravity” (a metaphor) or power (I try to abandon the term power because it’s too anthropocentric).  The moon moves about the earth because the earth bends time-space in such a way that this is the only path along which the moon can move.  A person that is paid by a company in “scrip” is trapped where they live because scrip can only be redeemed at the company store and cannot be exchanged for federal tender.  They are caught in the “gravity” of the company and scrip.  It is not that there is something called “structure” that is other than agents, but that one agent is caught in the orbit of another agent in a way that limits their movement, fixes their local manifestations, and that influences their becomings.

Finally, fourth, we should not assume that agents and subjects are the same thing.  Following Michel Serres, I argue that a subject is a machine that organizes the movements, local manifestations, and becomings of other entities.  Serres gives the example of the game of rugby to illustrate this point.  In rugby, he claims, it is the ball, not the players that are the subject.  The players are subjected to the ball.  The ball brings the players together in certain relations that are constantly shifting.  However, it also modifies their status throughout the course of the game.  The person who has the ball is now pursued.  When another person gets the ball he now either becomes a pursuer or a defender.  It is the ball that “assigns” the roles and that therefore functions as subject.  Clearly the ball’s status as subject is temporary.  When the game is over and it’s tossed in the back of a car, it ceases to be subject.  Anything can serve as a subject, even where the entity that takes on the function of subject is not an agent in the sense outlined above.  Enough for now.