Adrian Ivakhiv and I have been having a terrific exchange on objects and relations which really goes to the heart of what I’m trying to think about (here, here, and here). Indeed, in an earlier draft of The Democracy of Objects I had pitched the project of the work as working out “the relation between relata (objects) and relations”. As I read Adrian’s remarks, I get the sense that he worries that object-oriented ontology might lead us to ignore relations. After all, OOO begins from the premise that objects are absolutely independent of one another. This could certainly cause a lot of worries for an ecological theorist, where relations and systems are so important.

I think this too quickly passes over, however, one of Harman’s most radical and original ontological claims; a claim that I have also made a center-piece of my own ontology. It is indeed the case that the ontological nominalist contingent of object-oriented ontology (Harman, Bogost, and myself), holds that objects are independent and autonomous with respect to one another. This is in contrast to the ontological relationist contingent (Whitehead, Latour), that holds that objects are constituted by their relations to everything else in the universe. However, this is not the whole story. One thesis that lies at the heart of the nominalistic variants of OOO pertains to mereology or part/whole relations.

read on!

As Harman puts it in Guerrilla Metaphysics,

If someone asks where substances are located in reality, it is impossible to single out an elite cadre of substances at the expense of all other entities. We find substance neither in the really, really tiny things, nor in the really, really natural things, nor in the really, really divine things. Substances are everywhere. What we have is not a universe split between aristocratic natural kinds and miserable, pauper-like accidents. Instead, we have a universe made up of objects wrapped in objects wrapped in objects wrapped in objects [my emphasis]. The reason we call these objects “substances” is not because they are ultimate or indestructible, but simply because none of them can be identified with any (or even all) of their relations with other entities… We never reach some final layer of tiny components that explains everything else, but enter instead into an indefinite regress of parts and wholes. Every object is both a substance and a complex of relations [my emphasis]. (85)

Harman then goes on to remark that,

…if every object can also be considered as a set of relations between its parts or qualities, it is equally true that any relation must count as a substance. When two objects enter into a genuine relation, even if they do not permanently fuse together, they generate a reality that has all of the features that we require of an object. Through their mere relation, they create something that has not existed before, and which is truly one… Granted, a relation between two objects may last only a brief while. But the same is true of objects that are obviously substances, such as mayflies or the fleeting chemical elements of californium. Durability is not a requirement for objecthood, just as being part of nature or having an exceptionally tiny size is not. Substances are filled with relations; relations become substances. (ibid.)

Inspired and radical claims which, I believe, bring a lot of clarity to a set of issues that have haunted Continental thought (especially its French variants) since the late 60s (more on this in a moment). For now, let’s recap:

1. Objects are autonomous from one another.

2. Objects are wrapped in other objects.

3. Scale– whether an object is large or small –is irrelevant to what constitutes an object.

4. Duration– how long an object lasts in time –is irrelevant to what constitutes an object. In other words, an object could be the smallest thinkable unit of time and be no less an object for that, and an object could exist for a near eternity and still be no less an object for that.

5. When objects enter into genuine relations (something that needs clarification, I think), we get a new object.

I think this army of claims is one of the most underdiscussed and ignored set of claims within OOO, and this because I believe that among the many claims of OOO, this one has the most far reaching consequences and implications for how we have hitherto posed a number of questions within Continental thought. Now notice the tension between claims 1, 3, and 5. In claiming that when objects enter into genuine relations they form a new object, Harman is not claiming that the objects making up the multiple-composition of the new object disappear or cease to exist. In this respect, there’s a very real sense in which Harman’s radical and admirably weird mereology could be characterized in Deleuzian terms as a thought of composition premised on “disjunctive synthesis”, so long as the term “synthesis” is emptied of its unfortunate cognitive connotations. Because objects are autonomous, because objects are wrapped in other objects, the subsets of the larger set persist as autonomous objects in their own right independent of the new set, but also as having entered into composition with this larger scale object or set. This is one reason I’ve elsewhere proposed that the proper being of objects has the status of a ghost or a poltergeist. The proper being of an object is not its parts (other objects), but is rather a ghostly endo-relational structure that cannot directly be perceived but only inferred.

I think this thesis is incredibly rich in implications for both philosophy and social and political theory. Within the framework of onticology I’ve tried to capture this thesis in the distinction between endo-relations (which I often refer to as “endo-structure” or “endo-consistency”) and exo-relations. Endo-relations are the ghostly internal structure of an object which make up its proper being or its status as a substance. It is that assemblage of powers, attractors, or singularities (tendencies presiding over the entity as act) that make up its proper being, no matter how brief the duration of that being might be. By contrast, exo-relations are local and temporary interactions between autonomous and independent entities. Now, what this distinction allows us to think is interactions among objects at different levels of scale, even if one of the objects in the interaction happens to be a part of the object that it is interacting with. The weird conclusion is that the smaller scale object is both a part and necessary condition for the smaller scale object and an autonomous object in its own right. My cells are literally other than me, though fortunately they generally cooperate and condescend to assist me in the continuing adventure through time of my body.

This point can be illustrated in terms of ecological theory. An ecological system is one object. It has a unique endo-relational structure that makes it act as one. But this ecological system both contains other ecological systems (for example, the ecology of a single tree in a rain forest), and contains other objects (frogs, trees, insects, soil, droplets of water, bacteria, birds, etc., etc., etc.). These other objects both rely on that eco-system for their own continued existence in a particular way, and are independent in their own right. If I emphasize reference to the particular way in which a subset or smaller scale object enjoys its existence within an eco-system, then this is to underline that the way an object exists within a particular texture of relations is a local actualization that does not exhaust the virtual potentiality or excess of the object. The object, in principle, can be detached from the ecosystem (though this might bring about a nil actualization: death), and this detachment would generate other actualizations or local manifestations that would never appear within that particular contexture of relations (here we might think of the sad fate of many animals in earlier zoos that began exhibiting markedly different behaviors in captivity). Without this sort of meriology we’re unable to account for these sorts of possibilities.

The point can also be made in terms of social theory. In my view, Althusser is essentially correct when he claims that a society is not its members, nor groups, but relations. The study of a society is essentially the study of relational structures and how they are organized, such that the terms within that structure are mutually referring. For example, identities like “professor” are not atoms, but are rather uniquely relational structures that necessarily refer or relate to other things such as students, educational institutions, systems of production, government, etc., etc., etc.

Where Althusser goes wrong, however, is in his conception of other entities as being illusions of social structure. For example, his famous rejection of the subject in his characterization of the social as process without a subject. What we have here, at the ontological level, is bad meriology. Rather than the social exhausting the field of entities, the social is instead one object that is internally differentiated and composed of its own endo-relational structure, that contains many other autonomous objects (humans, tools, materials, groups, etc., etc.) that have their own autonomy and endo-relational structure. The absence of this sort of meriology in Althusser’s thought had a cascade of consequences that have determined the destiny and nature of the questions posed by French inflected political thought indebted to Althusser down through Ranciere, Laclau, Balibar, Badiou, Zizek, and so on. In other words, because the autonomy of objects was not affirmed, because the thesis was not adopted that you can both have objects wrapped in objects and have these objects autonomous from one another, the question that came to haunt this variant of structuralist political theory was how it was possible to locate a void or point of in-determination within structure from which change not determined by the regulative structure could become possible. And it is here that we get all the worries about the subject and the act.

Had a meriology such as the one I’m trying to outline here been adopted in these early moment of the development of French structuralist political thought, the subsequent questions would have been very different. With Althusser we would still, of course, seek to investigate the endo-relational structure, the virtuality, of those objects called “societies”. However, no longer would we search for the “Christ-point” in these endo-relational structures of the “empty square”, “degree zero”, “void”, or “subject”. Rather, the issue would become that of investigating how objects can act on one another, how they interact, and how it is possible for objects that are parts of another object to change the object of which they are a part. For example, how is it possible for a group– what Guattari, following Sartre, called a “subject-group” –to act on and change the endo-relational structure of the social object in which its enmeshed but from which it is nonetheless autonomous. And here one of the central questions would be one of how to navigate the autopoietic feedback loops that characterize the endo-relational structure of the social object as it strives to regulate the smaller scale objects of which it is composed, but from which it is nonetheless distinct. These feedback loops are points in phase space, limiting the basins of attraction of the social system. The issue here becomes that of how to avoid being enmeshed in these tendencies, feedback loops, or attractors such that ones actions simply end up re-enforcing the endo-relational of the structure, and how it is possible to push social systems into new basins of attraction without generating catastrophe. The point, however, is that there is already an autonomy within these systems. The question is that of how it is possible to act on the endo-relational organization of these systems.