The problem of ontological depth: « If entities are defined by their relations, this deprives them of any depth beyond their current deployment in the world, and there is no reason that anything would ever change » – Graham Harman
This provocative affirmation is far from obvious and calls for deeper analysis.
First remark does Graham Harman mean « defined » or constructed? Harman is using a personal definition that is semantically loaded in favour of his conclusion, but it still does not work. This formulation is pure assertion, often repeated, and easily shown to be wrong.
For example, in physics force implies acceleration, and thus change in the deployment of velocities and positions. This means that relations between forces define a configuration that deploys change.
In fact, the same argument applies to the relations between velocities, as velocity is itself a rate of change. An assemblage defined in terms of multiple velocities is constantly changing its « current deployment ».
In Deleuzian terms, the kinetics (latitude, velocities and positions) define the current deployment and the dynamics (longitude, forces and power to affect and to be affected) of a configuration define the « depth » or potential modifications of that actual deployment.
Thus the notion of a static « current deployment » conceals a radical incomprehension, it involves the reduction of configurations to non-kinetic properties and non-dynamic properties.
Thus relational ontologies can easily accommodate depth and change in a unified account, whereas objectal ontologies cannot.
Harman’s OOO splits the object between depth and change, between deep object (real but unchanging) and changing object (sensual, but superficial).
On this point see HARMAN’S TEMPORAL DENIALISM.
This whole argument is part of Harman’s rhetoric of projection: isolate a flaw in his system (in this case the inability to handle time, change, and becoming) and then project it onto rival ontologies. The actual argument does not matter, what counts is bare assertion, bold projection.