At the heart of object-oriented ontology is something of a surprising paradox. Object-oriented ontology treats objects as black boxes. This is part of what it means to say that objects are withdrawn. To say that objects are black boxes is to say that neither we nor any other entity has any direct access to their inner world. That inner world is beyond representation. It is here that the paradox emerges. While object-oriented ontology treats objects as withdrawn from relations, it turns out that because of withdrawal knowledge of objects becomes thoroughly relational (is this really such a surprise?).
What we know in knowing is not the object per se, but rather the relation. Here knowledge shifts from the register of representation to performance. Because objects are black boxes they will always have the capacity to surprise. Representation is unable to predict the behavior of objects based on a representation of the object, but rather must set the object loose in the world to see what it does. As Pickering nicely describes it, “[t]his is the aspect of cybernetics that interests me most: the aspect that assumes an ontology of unknowability, as one might call it, and tries to address the problematic of getting along performatively with systems that can always surprise us…” (The Cybernetic Brain, 23). As Pickering goes on to remark, “[o]ne is more concerned with what entities do than what they are” (28).
Here then knowledge ought to be thought in relational terms rather than representational terms. Here we should think in terms of a simple feedback loop. Let the two circles in the diagram to the right above be two distinct entities. We can call the first entity labeled “input” the “knower” and the other entity labeled “output” the “to-be-known”.
What is known here is not the output but the dynamic relation between the input (knower) and output (to-be-known). Knowledge here is not knowledge of the output circle, but rather is performance, the dynamic interaction of the two. Knowledge here is performative. It is not a representing, but a doing. “Input” here should be thought as an act exercised on another entity. “Output” here should be thought as a response on the part of the entity that is the recipient of the act.
Here knowledge should not so much be thought as represented as revealed. In many respects, this is the entire point behind my distinction between local manifestation and virtual proper being. Local manifestation is something that takes place in an object. It is an actualization of potentials populating the virtual proper being of the object. While local manifestations aren’t entirely relational, a large subset of them are. These are events that take place within the object as a function of how it interacts with other entities in what I call “regimes of attraction”. Many of the qualities (local manifestations) of an object are thus relational in the way I outline above. They aren’t fixed in the object, but are emergences within certain fields of relations among objects. This entails that the qualities of an object are situated. Shift the situatedness of an object to another context, another regime of attraction, another set of relations, and new properties, qualities, or local manifestations take place. To think quality is thus not to think objects in isolation or as withdrawn, but to think the relational field as per the simple diagram above.
The object as such remains a black box. It’s internal universe is withdrawn from relation, and it is for this reason that it is capable of surprising us. This point can be illustrated with respect to Walter Grey’s “machina speculatrix or electronic tortoises depicted below:
The interesting feature of Grey’s machina speculatrix is simplicity. The electronic tortoise does not have a representation or internal map of the world, but rather has a touch sensor, a light sensor, two cells, and a design that tells it to seek light and move away from other objects it bumps into. What is interesting here is that such a simple machine develops complex and unpredictable behavior when set loose in its environment. Much of this behavior is depicted in this video below:
Based on the circuitry alone, the designers of the robot were unable to predict that it would do strange “mating dances” with its own image in the mirror or when encountering members of its own species. The behavior of the tortoise is only revealed as local manifestations when set loose in a regime of attraction or assemblage composed of a variety of other objects and light sources. Change these regimes and we might very well encounter other unpredictable behaviors as well.
This too is how it is with knowledge. Knowledge is not a representation of the object, but a bilateral performance where objects are acted upon and act in return, such that all the terms involved in the relation change and evolve over the course of this interaction. There’s an order of time here with an irreversible arrow. To know is to know not the entity taken in isolation, but rather the relation field of actions and responses. Shift these actions and fields of relations and new local manifestations might very well take place. As a consequence, knowledge takes on the characteristic of being situated. Our tendency is to reify this situatedness, treating it as an inherent feature of the entity in all circumstances, rather than understanding it as the feedback relation that it is.