Graham Harman

For a Thought of Objects

Environmental awareness redefines our human position as part of a whole in a global and complex interrelationship. However, philosopher Graham Harman, the founder of object-oriented ontology (OOO), refutes the commonly accepted idea that everything in nature is connected. Certain things are intimately—and dangerously—interconnected, others almost not at all. Holism thus denies the “problematic” dimension of relationships between objects. Harman specifies the understanding of the object according to OOO—which cannot be reduced to its parts or its effects—but also contests the idea of a negative theology, an approach which gives a “glimpse” of objects even without formulating discursive knowledge. Though OOO advocates the theory of an ontological equality between all objects, it does not claim a political or moral equality. A man is not on the same level as dust, but at the same time, Harman sees no reason for philosophy to focus so much on the human.

Considering the “living” in the twenty-first century demands that we reconsider the relationship between subject/object and culture/nature, and ask ourselves new questions: will humanity follow the geo-engineering route? Will we find some intermediary relationship with the environment? What political engagement should we have as architects? What are the philosophical and technological tools that can make this engagement effective? My first question to you would then be, can Object-Oriented Ontology provide a framework to address these different issues?

Object-Oriented Ontology (OOO) takes objects to mean the fundamental unit of reality. But we mean “object” in a much broader sense than solid physical things. An object for OOO is simply anything that cannot be exhaustively reduced either downward or upward, to its parts or to its effects. Any sort of knowledge does one of those two things. Notice that there are really only two kinds of knowledge that we can have about anything: we can say what something is made of or we can say what it does, with all the subvariations of these two types. These are what I call undermining and overmining. Undermining is an attempt to say that a thing can be paraphrased in terms of the smaller pieces of which it is made. Overmining is the attempt to paraphrase an object in terms of what it does, what it shows us directly, what are the “events” in which it participates, what are the relational effects it has. And so, when people wave the Deleuzian flag and say they are more interested in what a thing can do than in what it is, this is just the opposite form of reductionism from the usual kind. They are merely going upward instead of downward. What we really need to get at is the object that exists in between its component and its effects: what I call “the third table.” This is neither the table’s pieces nor its effects in the practical world, but the table itself.

I think the arts deal with this better than philosophy, because the arts are fully aware that they are not primarily a form of knowledge. Neither is philosophy, yet many philosophers have convinced themselves otherwise. Art and philosophy are cognitive disciplines, but knowledge is not the only form of cognition. Knowledge is obviously important—our entire modern civilization is based on knowledge. We have so much knowledge that we don’t even know what to do with it anymore. And yet, knowledge is not the only form of cognition worth pursuing. We know from Socrates that philosophy is not a kind of knowledge.

One of the lessons taught by the Platonic dialogues is that we never really obtain a definition of anything. Socrates always asks what is virtue, love, friendship, justice, but he never gives an answer. He demolishes the answers of others, but never gets to one himself. He tells us that only a god can have this knowledge. The word philosophy itself—philosophia—means love of wisdom. It is not a wisdom in and of itself, but something you can never reach.

Critics of this model often say: “Oh, then you’re left with nothing but negative theology; you’re just saying what objects are not rather than what they are.” But this assumes an all or nothing result—that if you do not give us discursive knowledge stated in prose propositions, then you must be giving us nothing but vague mystical gesticulations. That’s not the case. Humans have a lot of knowledge that is metaphorical: that alludes to things rather than presenting them directly. We have hints, innuendos, threats. All of these acts of speech are not literal paraphrases of what the thing is, but somehow hint at what the things are. In the arts as well, if you’re able to reduce a particular artwork to a prose summary, then most likely it’s either not a very good artwork or not a very good summary. One of the two.

We know that there’s never going to be a final analysis of Hamlet or of the poetry of Baudelaire because these objects cannot be paraphrased. By contrast, the natural sciences are all about paraphrase, except perhaps in moments of scientific crisis. So, if you start with a concept like an electron, your job as a scientist is to discover new true attributes that belong to electrons. You’re doing the opposite of the arts and philosophy.

OOO is a way to deal with this in philosophical terms. Often people will say that this focus on what escapes discursive language isn’t new, because of Kant’s thing-in-itself or some other precursor. The problem is that even in Kant, the thing in itself is there and it’s something we can never know. We can think it but never know it: a tragic human burden. But OOO makes a more radical claim, which is that in every causal interaction there’s an unexpressed residue or surplus. OOO is really about looking at how this works on all levels, including the inanimate one.

Does OOO relate specifically to the new anthropocentric condition? Could this idea have existed fifty years ago? Most importantly, how does OOO help us today in this Anthropocene era? Is it helping us to reconsider the relationship between object and subject? In a moment where all we learn from ecology is that everything is related to everything else, how can OOO justify its focus on the individual autonomous agency of things? Why at this moment are we looking at an object when we should be looking at networks, systems, structures, and ecologies?

I think it is wrong to hold that ecology tells us that everything is related to everything else. What it tells us, instead, is that some things are crucially interrelated in ways that can kill us all, while other things are barely related to each other. To say that “everything is interrelated” is to skip the hard work of showing how certain things come to be mutually dependent even though others are not. To show this requires that we acknowledge that relations are problematic. It does not go without saying that one entity should be able to affect another. Why aren’t entities just mutually impenetrable, walled off from any influence from other things? Yes, we can see empirically that certain things do affect other things, but there is still a philosophical problem here. When I saw James Lovelock’s terrifying Dublin lecture in April 2009, his point was not that everything affects the world climate in cataclysmic fashion. Rather, he emphasized the dangerous positive feedback loop made up of just three specific factors: the eventual death of the rainforests, death of the algae, and the melting of the Canadian and Russian permafrost. Holism is actually a lazy philosophy.

You ask an interesting question as to whether OOO could have existed fifty years ago. Not only would I say yes—OOO arises from a reading of Heidegger that, in principle, could have been carried out immediately after the publication of Sein und Zeit—but there is a sense in which OOO could have existed immediately after Kant. It seems to me that the post-Kantian period of German Idealism, from the 1790s forward, is one of the great counterfactual crossroads in the history of philosophy. They got rid of Kant’s thing-in-itself by saying, “To think a thing outside thought is already a thought, and therefore the thing-in-itself is a contradiction in terms.” They then found ways to immanentize the Ding an sich and not allow it to occupy some impossible Beyond closed off from all human access. But a different turn was possible after Kant. Instead, it could have been argued that Kant is right about the thing-in-itself, and merely wrong to think that only humans and other “rational beings” are unable to reach the noumena. OOO’s claim is that the noumena are not that which exceed human perception and cognition, but that which exceed any relation whatsoever. Whether or not a stone is “conscious” (and who really knows what “conscious” means anymore?), it is unable to make direct contact with a cliff, a river, a sheet of paper, or the hand of a boy who throws it. If post-Kantian philosophy had taken this path instead of the German Idealist one, the last 200 years of philosophy would look entirely different. OOO can be seen as going back in time and making a course correction.

To be sure, this argument by OOO has led to a number of foolish responses. One of them runs as follows: “Since OOO wants to treat humans and non-humans on the same footing without granting ontological privilege to humans, then humans are on the same footing as a garbage dump. Therefore, OOO is saying that humans are no better than garbage.” This is an obvious equivocation. The OOO approach to flat ontology means that all objects are equally objects, not that all objects are equally dignified or valuable. Ontological equality does not mean political or moral equality. 

Notice that even before bringing humans into the picture, objects themselves are not morally equal. We view certain objects as treasures, others as garbage. And they’re not treasures just because we’re “fetishizing” them. Some objects are legitimately treasures because of their rare beauty or because of some important utilitarian function they might have. So, there’s already a hierarchy of objects, and there’s no reason why humans couldn’t be the most important object in our hierarchy of objects. But what I reject is the central modern idea that humans and non-humans are so different in kind that humans must represent 50 percent of philosophy and absolutely everything else in the cosmos is the other 50 percent. That’s so unlikely, so implausible. H.P. Lovecraft helps teach us otherwise—that we are not so important cosmically however important we are to ourselves. Our intelligence may be fairly puny compared to the other intelligences that may exist now or in the very near future.

One place where a subject does play a role is in the arts—a work of art needs a beholder. I do not believe that there is such a thing as art without humans, or at least there can be no human art without humans. Perhaps dolphins and parrots have a sophisticated form of aesthetic experience, but what we call art needs humans as a catalyst. The artwork itself is always deeper than whatever the beholder or the spectator sees of it. In the end, an artwork is not a physical thing lying outside of us, but a compound entity made up of the artwork and us.

You talked about the relationship between OOO and art, but how does it work with architecture when there is clearly a difference in the balance between function and form.

Architecture is already a more complicated situation than the arts. OOO’s main theme, of course, is the autonomy of objects. This is clear in a visual arts context even if humans are, I hold, necessarily one of the ingredients of the artwork. But humans are an ingredient of architecture in a different and more profound sense: namely, if you didn’t have any relation to human purposes then it wouldn’t be architecture, but an artwork. Conversely, if you were only concerned with the function, then it wouldn’t be architecture, but engineering. Patrick Schumacher, in his big two-volume book, The Autopoiesis of Architecture, writes that the lead distinction in architecture is the one between form and function. He says that you need both. He criticizes Peter Eisenman here at Yale for only focusing on the form, and he criticizes some engineering people for only focusing on the function. Patrick’s argument against the latter group is that engineering problems always underdetermine design decisions. Now, I happen to love the Eiffel Tower. I don’t know what French people think about it these days—Paul Verlaine is said to have hated it so much that he changed his walking routes around Paris in order to avoid having to look it—but I love it. And, of course, that’s an engineering tour de force by Gustave Eiffel. But I heard a lot of architects at our conference criticizing it yesterday in design terms and saying it’s a terrible piece of architecture. If they’re right, then the grounds for their making that criticism would be to say that the Eiffel Tower is purely an engineering achievement and the design decisions do not follow logically from the engineering. That’s what Schumacher would say if he were anti-Eiffel Tower (I have no idea whether he is). And there’s something to that. But there’s also a certain aesthetic grace to sleekly executed engineering projects.

Leading on from that, in this scenario or situation of the Anthropocene we really have something that calls for universal engagement, and yet we can see clearly that something is going wrong. What has happened over the past ten years is that we have seen many solitary objects of iconic architecture, which are just illustrative of “la societé du spectacle.” This is a tendency that I find dangerous. Not to take away the aesthetic and artistic dimensions of architecture, but I’m struggling to see in what terms you could rethink this relationship between function and form in the shadow of this question of the Anthropocene. Is that in your thoughts?

Let’s start with form, which must not be overidentified with overt visual appearance. The medievals already knew there were two kinds of forms. There’s the substantial form and the accidental form. The latter is the one we can identify with how a thing appears, how it looks to someone or relates to someone. I would also argue that the mathematical formalization of a building is also an accidental form, though I’m aware that there are trends afoot in architecture, opposed to OOO, that want to place algorithms at the center of architecture. Anyway, there’s also the substantial form, which is the one you can never see entirely (much like Kant’s thing-in-itself) but which structures the thing. It’s what used to be called the “substance” of the thing. 

Now, I wonder, couldn’t the same split be made between two kinds of architectural function? That is to say, we usually think of function in terms of what I would call accidental function. I don’t mean the unanticipated side effects of a building, though these are not unimportant. Instead, I mean the overt, explicit program of a building. Even if a building comes about in order to serve a precise program, there is something accidental or superficial about this, since it defines the building purely in terms of its relations to other things. For OOO this is never enough, of course. A building could (and usually does) end up serving many other purposes from the ones demanded in the competition brief. The “substantial function” of a building would be inaccessible just like its substantial form, but would be something pertaining solely to architecture (which necessarily has function concerns) and not to the visual arts (which do not). On this note, there is an idea I call “zero form, zero function.” It has to do with elements of form or function that are there without being deployed.

Like the hammer.

Yes, the form of the hammer is both visible and not visible. There are two different kinds of forms. And the function of the hammer is both visible and invisible. So, there’s substantial function just as well as substantial form. And so we come to the Anthropocene.

Yes, what are your thoughts about that?

The Anthropocene… in a way it’s not surprising that Bruno Latour took it as his next major topic, since he always seems to be nearby there when important intellectual matters arise. He has that public side to him where he doesn’t want his work to be detached from contemporary topics. The Anthropocene, among other things, creates a scientific predicament in which we can’t look into a microscope and see with 100 percent certainty that global warming is happening. You have to assemble all these mediators and chains of reference in order to show that it is happening. That’s a very Latourian theme, of course. He would say that this has been happening all along in science, but there’s no question that the indirect access to scientific truth becomes more evident than usual in the case of global warming. Hence the relative ease of denying its existence. 

But there’s another Latourian theme that I don’t like quite as much, which is the idea he shares in his Gifford lectures—that Gaia has not yet been fabricated, and thus in a way has not come into existence. I think that is simply wrong. It’s true that we can’t split so cleanly between nature and culture, because we can point to all sorts of “hybrids,” like the ozone hole, or like garbage dumps that become ecological preserves. With We Have Never Been Modern, Latour has written the most important philosophy book in the last thirty years, I think. And we haven’t begun to discover the implications of this book yet. But then he ends up slipping tacitly into a position where everything has to be a hybrid. The human always has to be somewhere on the scene for something to be registered in reality. Ramses II cannot have died of tuberculosis, because it wasn’t discovered yet; microbes did not really exist before Pasteur, but only pre-existed Pasteur “for Pasteur.” And now, Gaia doesn’t really exist until we assemble it. That’s certainly not Lovelock’s position: his position is that Gaia exists, and it may destroy us all very soon whether we like it or not. It’s a classical scientific realist position, whereas Latour politicizes reality, albeit in a much broader sense of “political” than used by anyone else.

Latour is trying to communicate this idea of an Anthropocene through emotions. He made several attempts with art—his exhibition in Toulouse was almost like a theater performance, with play and dance. So, in the field of architecture is it possible to be simultaneously motivated by a global issue and an aesthetic proposal? If I’m going to design this building, it is not an object—it’s more like a living element. It’s going to consume things, it’s going to die, it’s going to disappear. It is very technical and data driven, but at the same time it has to be sublime, beautiful, or distinctive. Do you think there is that idea of aesthetic activism?

That term suggests that architecture should be activist in spirit and should be trying to improve society, which is a provocative proposition. I’m instinctively sympathetic to formalism, and want to preserve every discipline’s autonomy and prevent it from being reduced to the handmaid of another one—even if all disciplines are supposed to become servants of a purpose as noble as saving the climate, saving the proletariat, or saving the world. And so, I always tended to resist the idea of the moral superiority of politically active art. 

Yet one has to admit there are numerous artworks whose full impact cannot be felt apart from the political aspect: Picasso’s Guernica, Stowe’s Uncle Tom’s Cabin. It’s impossible say in principle that all works are cut off from their context—this is what high formalism wants to do. It has to be realized that there are gateways in artworks that allow certain influences in, but not others. Stephen Greenblatt, the anti-formalist par excellence in literary criticism, says that Shakespeare’s plays are one set of Elizabethan texts among others—even he wouldn’t reduce Shakespeare to all of his influences. He’s choosing selectively which were the important cultural features that were integrated into his plays. So, you have to be selective, and if you’re selective, that means the artwork has firewalls that only let some influences in and not others. Again, holism gets us nowhere, in the arts or anywhere else.

Do you see this aesthetic activism evident explicitly or implicitly in any architectural examples?

The obvious, most moving example in Washington is the Lincoln Memorial. The Washington Monument is a nice smooth obelisk commemorating the great President and father of his country, but Lincoln’s is the more moving memorial. The man gave his life, was on the right side of history, and is still probably our greatest President. It’s hard not to be moved, as a US citizen, by the larger-than-life statue of Lincoln inside the memorial. In some way, it’s edifying. It’s not popular to say that art can be edifying, and yet somehow it makes you a better person to encounter Lincoln in statuesque form.

In terms of non-monumental buildings, what would do that? And what could do it beyond the sphere of historical significance? I suppose you could say that certain kinds of buildings inherently suggest certain political orderings. The skyscraper—we like to suggest that it hints at American enterprise, but also at hierarchy and complexity. We know that not all hierarchy is good, but it’s also the case that not all complexity is good. One of the problems I have with the rise of 3-D printing and with parametricism is that yes, you can generate all sorts of complex, varied forms with incredibly precise cuts and contours, but these minute variations are not all memorable. There is such a proliferation of complex 3-D-printed forms generated by so many different algorithms, yet most are so unmemorable that they may as well be clones. Whereas, consider something like the Sydney Opera House. What’s amazing to me about this public favorite is that it’s a fairly basic, if unusual, geometric configuration. But you remember it.