Object-Oriented Ontology

I find the most interesting philosophers are ones that make me rethink my previous ideas while still not quite sitting well enough with me to fully sign on to their philosophy, thus forcing me to develop my own ideas in response.  Such is the case with Graham Harman and his object-oriented ontology.

I've long been fascinated by Alfred North Whitehead's process philosophy(a fascination that runs in the family).  Actually, it would better be described as "process-relational philosophy," as Whitehead, like Wittgenstein and a few other philosophers of his day, took relations as the fundamental building blocks of reality.  Reality, for him, is constructed of single "actual occasions" which are relational in nature, and which form complex occasions called "societies" or "actual entities."  Some of these actual entities form what are called "enduring objects," which is pretty much exactly what it sounds like.  But the enduring objects have to recreate themselves at every moment to maintain their identity.  Thus, instead of a continuity of becoming, what you have is a becoming of continuity.  I won't yet go into the role that God plays in Whitehead's system, because that's not really important for the point at hand.  The point is that in his system, objects are defined by their relations, and maintain their identity through a process of inheritance from the past.

Harman, starting from a Heideggerian perspective comes to a very different view.  Heidegger famously spoke about the use of a hammer.  While you're using a hammer, its status as an object is invisible to you.  It exists purely as a tool that is "ready-to-hand."  We can use it with theorizing about it.  Only when it breaks does it become "present-at-hand" -- that is to say, present as a real object.  But whereas this analysis has traditionally been interpreted from a phenomenological angle, in terms of what tools are "for us," Harman offers the more radical suggestion that whether he knew it or not, Heidegger is actually giving us an ontology of objects.  Using the example of a hammer is a bit deceiving, since hammers have to be wielded by the person using them.  But the ready-to-hand can be found all around us at any moment.  As I speak, there are wood beams holding up the roof over my head, which in turn are held together by nails.  Solar radiation is penetrating the Earth's atmosphere and warming up the planet to a temperature at which I am able to survive.  Plants are producing oxygen I require to breathe.  Electricity is being supplied to my house, thus allowing me to type this sentence on the keyboard and have it show up on the internet.  Objects all around me are exerting a real force upon the world to form a background of order that is invisible because it functions so perfectly.  Only when these things break down do they become present to me.  These objects don't exist the way they do for me.  They exist in their own right, and they exert a real force upon the world because of some inner quality that gives them the capacity to relate to other objects.


Objects are not, therefore, reducible to their relations.  Rather, says Harman, they have an essence which gives them the capacity to relate to other objects in any number of ways.  Salt relates to me in a different way when poured on the mashed potatoes I'm eating than it does to a snail when poured on its skin.  These two contexts have an entirely different set of relations, but they both involve the same substance.  The salt did not creatively manifest a new set of relations in either of those situations.  Rather, related to the two contexts in ways that are consistent with its essence.  The essence of an object cannot be exhausted by any of its relations or even the sum total of them.  Objects relate to other objects in the particular ways they do because they have a particular essence that gives them the ability to relate in these ways.

What's more, no object ever encounters another object directly.  Every encounter with an object is an encounter with a caricature of it.  It is only one mode of the object's capacity to relate, and cannot ever touch the essence that the object that gives it this capacity.  Because it can never be exhausted by its relations, the real object always withdraws from view, hiding its essence behind any given relation.  Harman therefore calls for a kind of occasionalism, in which objects only relate to each other from within a larger object, namely their relationship.  If a piece of wood is set on fire, the wood and the fire become part of a larger wood-fire object that transforms both of them from the inside.

One objection I have to this philosophy may seem rather trivial, but I find the imagery and terminology a little troubling.  Harman defines "object" as any entity that can be taken as a single whole.  That's easy to imagine in the case of something you can pick up or separate from its environment, such as my phone or wallet, but some things seem to be more a part of a continuum.  A mushroom may appear as an object to us, but it is simply the reproductive organ of the mycelium underground.  You can't really point to a single mycelium.  Rather, there are mycelial networks that comprise a fungal system.  I'm sure Harman would consider that network to be an object, and the mushrooms it produces to be objects as well.

I suppose there's nothing logically wrong with this, but the terminology just seems misleading.  Sure, a system or network may be considered an object, but that seems to obscure what we mean by those terms.  A system is supposed to connect different objects together, whereas the defining feature of Harman's objects is the fact that they always withdraw from one another.  I suppose the mushroom withdraws from the mycelium in a way that the mycelium never touches the full essence of the mushroom.  But the mushroom is also part of the mycelial network, and is produced by that network in the first place.

I guess what I'm saying is that while Harman is rightly critiquing a tendency in philosophy to portray the world as a seamless continuum of relations, he seems to go too far in the opposite direction, giving us world that is clunky and non-relational.  Not that Harman rejects relations outright, but he has to appeal to larger objects in order to explain those relations, without fully explaining how these objects come into being in the first place.

Plenty of philosophers have attempted to describe a unit of which reality is composed, and many of them are simply different ways of viewing reality.  Whitehead described an ontology of process, and certainly Harman's objects can be described as processes, even if they can't be defined in the purely relational manner than Whitehead uses.  They could also be described as holons, as in Arthur Koestler's philosophy.  A holon is something that is simultaneously a whole and a part of a greater whole.  Koestler's ontology describes a cosmos of holons all the way up and all the way down.  But none of these descriptions seems to capture the heterogeneity of what is being described.

That is why I prefer to think of objects, networks, systems, and holons in terms of the Deleuzian concept of assemblage.  An assemblage is a system of heterogeneous components externally related to one another.  The idea of external relations is crucial here, in contrast to the internal relations emphasized by Whitehead.  An internal relation means the relationship defines the thing it is relating.  In the case of exterior relations, by contrast, the relations exist between entities that have their own identity, and would maintain their identity if taken out of that context.  An assemblage exerts a retroactive influence on its components, which is another ways of saying that it has emergent properties.  An assemblage must be constructed of different parts initially, but it also has the ability to construct new parts.  An assemblage has both material and expressive components, the expressive allowing for relation with other assemblages.  Finally, an assemblage is composed of other assemblages, and can combine to form new assemblages.

I'm not sure to what extent I'm actually disagreeing with Harman here, but I find this way of reframing these ideas to be useful.  I would here like to address some criticisms others have made about object-oriented ontology.  Matt Segall points out that withdrawal exists in Whitehead's system too, but in this case it involves withdrawal of an occasion from its relations just before passing from the subject to the superject to be prehended in the consequent nature of God.  He then goes on to state that it is a matter of the kind of withdrawal one prefers:  one that brings us back to a substance ontology, or one that brings us into theology.  As someone of a theistic persuasion, I can respect what he is trying to do here, but I don't really think this kind of withdrawal satisfies the problem that Harman is posing, nor do I think there is any need for such false dichotomy.  The withdrawal being emphasized by Harman is not something that happens at a given point in a process, but rather a fundamental condition of objects themselves.

Steven Shaviro points out that object-oriented ontology doesn't account for how objects come into existence in the first place.  I agree that this needs to be fleshed out, but I don't find Whitehead all that helpful in this regard either.  I think a Deleuzian idea of the virtual can be useful here, but I would also like to introduce into the discussion Rupert Sheldrake's theory of morphic resonance.  Sheldrake offers a theory of habit, whereby the more a phenomenon is repeated, the more it becomes a habit.  A phenomenon subject to morphic resonance can be a behavior or a relation, or more significantly, it can be a form.  Object-oriented ontology is essentially a theory of forms, and when we explore the idea of the virtual development of form through non-local morphic fields, we have the prospect for an exciting re-emergence of formal causation.

Of course, Sheldrake himself suggests that not everything has a morphic field.  Things that grow on their own do, but artifices created by people do not.  I would suggest that he is being overly cautious here.  Since morphic resonance applies to behaviors, we can apply morphic fields to human technologies such as writing, which seems to have emerged independently in at least three different areas of the world in what is virtually the blink of an eye in terms of human history.  Civilization itself emerged rather suddenly in a similar non-local manner.  So we can say that artificial objects still have a morphic field, even if one of the ingredients required for their actualization is human labor.

I think such a theory of virtual, non-local causation can help bring the sense of continuity missing from object-oriented ontology, as well as helping to reconcile it with some variety of process philosophy.  We would then have a theory of assemblages connected to one another not so much in the material realm, but rather in a virtual realm of habit and creativity.  This gives us a kind of mystical unity in something that approximates Whitehead's idea of God, while still giving us receding, self-contained objects in the realm of manifestation.