Object-Oriented Sci-Fi: Harman's Four Methods

17 April, 2012

The following is an excerpt from a talk by Graham Harman at the “Hello Everything” symposium. In it, Harman describes four methods for reversing common errors in failing to see objects. These methods are: counter-factuals, the hyperbolic method, simulation, and falsification. Each of them is an imaginative strategy for revealing the withdrawn core of objects, the aspect of them that makes them real for Harman’s Object-Oriented Ontology.

As philosophical techniques these four methods are quite striking. Together they constitute a kind of science fictional approach to philosophical thinking; each advocates imagining the world as different from reality in order to explore the limit and meaning of that reality.

I reproduce these methods here because I think they are promising ingredients in a recipe for something like an Object-Oriented Aesthetics or artistic methodology. Like much good SF I find them to be rich compost for my own imaginings, in this case of a set of procedures for generating multimedia art that inhabits an Object-Oriented perspective.

Here’s Harman:

"How do we reverse the error of seeing objects as events? We do that through counter-factuals. This is already a known method. You can imagine objects in different situations and imagine what the effects would be.[…]

"Imagining Lincoln in ancient Rome. How might he have played out there? Imagine a middle east with an Iranian atomic bomb or imagine an invaded Iraq instead. What are the possible things that would have happened in either of those cases. These help as allude to the thing as a style. Lincoln isn’t something that was confined to that historical period and that country but is something over and above that that could be translated.

"There are computers that do this. They take On Top of Old Smokey and turn it into a Bach fugue.

"Counter-factuals would be the first method for getting at the reality of things. The second would be what I call hyperbolic analysis, which I’ve used in three publications. This is reversing the error of impact. This is reversing the tendency to see things in terms of the effects they have. Instead of critique, also. I did this in the article on deLanda; I did this in the book on Latour; and I did this in the book on Meillassoux that hasn’t been published yet.

"In order to look at the impact of these philosophers what I did is not critique mistakes that they’ve made, but imagine that they have total success. Imagine that they become the dominant philosopher on the planet 20, 30 years from now. And then you imagine what would still be missing. What would still be missing if Meillassoux was the dominant world philosopher in 2050. Don’t fuss around with detailed mistakes that he makes but grant him everything and then see what’s still missing.

"If a philosophy can not survive the hyperbolic test then its less of a real philosophy, I would say. If you take some perfectly respectable minor article about some detailed point and then try to imagine that this is the most important philosophical text of the 21st century it can’t survive that test, obviously. It needs to be a work of a certain level, a certain comprehensiveness and that’s a more real philosophy. The more it can pass that sort of imaginative test the more real it is.

"The other two are a little harder. What we’re trying to do is talk about the mutual independence of a thing and its pieces where the thing is not reducible to its pieces and the pieces are not reducible to the thing. And we actually do this all the time: we call this simulation – where you’re removing a thing from its pieces and simply trying to treat it as a formal model. You’re testing the behavior of a tornado or the 1976 Cincinnati Reds – drawing on my sports writing career – without having to reassemble all the physical pieces that made them those things, of course. You’re simply testing them to see what will happen.

"And what I’ve realized while thinking about this is that paradoxically a thing is more real the more it can be simulated, the more it can be parodied. You can parody good poet better than bad ones, can’t you? If imitation is the sincerest form of flattery then simulation and parody are an even more sincere form. The less real something is the harder it is to simulate. It’s harder to simulate a bad writer, a bad philosopher than a good one.

"In other words the style of a thing is not just an aggregate of all of the deeds it has done. The style of a thing is something over and above those that can be simulated. And so here I would say, against some Luddite principles, if there were truly a computer that was able to write new Shakespeare plays I think that would be outstanding. I think this would be a tribute to Shakespeare, not some kind of cheapening of his greatness. It would show that the style there is perhaps something more real than the mass of works that one person wrote.

"And that leaves one last feature of pseudo-objects which is reducing them to sets, reducing them to pointing at an extensive number of things and saying that’s just a set it’s not a real thing with a unifying principle. We already saw that Rilke or earthquakes are substantial forms independent of their material components that can be removed and put on a computer and generate effects. What about the reverse? Is there a reverse situation where we can show those material components are real beneath all simulation?

"Actually yes. The answer to this is accidents: when things happen that weren’t expected. In what sense are accidents a method? Well, all the time. This is what falsification is about in science. You’re finding accidental things that happen to a theory that weren’t expected, things that point to the independence of the material components from the model that you had of them. So that would be the forth method to use.

“So now there are four methods to use: counter-factuals, the hyperbolic method, simulation, and falsification. And you could say that the humanities tend to benefit more from the first two and the sciences from the latter, but that’s not necessarily the case. There are significant exceptions. And what this suggests to me is that if this way of setting out the different methods is valid, the division between the human and natural sciences is actually an imperfect approximation to the real fissure running through human knowledge, which has to do with the kind of knowledge that shows the independence of a thing from its pieces and the kinds that show its difference from it outer effects, which are not strictly identifiable with either the sciences or the humanities.”