mattermorphosis

Graham Harman on Manuel DeLanda

Graham Harman’s critique of Manuel DeLanda’s (2006) ‘new philosophy of society’ emerged in 2008 in the form of an essay entitled DeLanda’s ontology: assemblage and realism. [Cont Philos Rev (2008) 41:367–383] While apparently celebrating DeLanda’s realism, Harman’s critique in a way ends up saying, ‘not bad, but not as good as mine.’ The problem to my mind, is that Harman seems to ignore or misunderstand DeLanda in a number of crucial areas.

1. He corrects DeLanda’s usage of the term ‘naïve realism,’ elsewhere suggesting that a more appropriate term is ‘narrow realism,’ presumably to protect his own particular brand of ‘expanded’ (actualist) realism, for DeLanda’s point about naïve realists is that they are naïve because they fail to recognize the virtual as central to reality. However, he does not expand on this by explaining how one can do away with the virtual, which DeLanda is at pains to establish by way of mathematics, science and metaphysics. Later in the essay, Harman goes on to suggest that the virtual may be done away with entirely, ‘But if we define objects as inherently deeper than all the states of affairs in which they ever become involved, then no recourse to a disembodied virtual is needed.’ The sticky point here is firstly that DeLanda does not in fact suggest a disembodied virtual; the whole point of assemblages is that these social wholes are virtual and actual composites. Secondly, if Harman’s ontology is an actualist one, however expanded, which the essay suggests, and if he excises the virtual, then he is left in the position of championing (presumably actual) objects which do not in fact exist in reality. [Admittedly, I am not an expert on the Harmanian school, so perhaps others may correct me here]

2. Harman also claims that ‘DeLanda draws an absolute separation between species and genus.’ Now, for anyone who has read all of DeLanda’s works, this is clearly a misreading for DeLanda replaces the categories of species and genus rather than working within that framework. Harman’s misreading becomes more obvious when he says of DeLanda’s use of species and genus, ‘they actually refer to two different ontological structures.’ The whole point of DeLanda’s new philosophy of society is to streamline Deleuze and Guattari’s work, and thus to posit a flat ontology, only assemblages may be considered real. DeLanda is thus not working with two different ontological structures, but one, and that one is the assemblage… the only entity he is prepared to defend as an ontological commitment. DeLanda (2010) makes all of this much clearer by spelling out the steps taken in parametrizing the concept of assemblages and doing away with the concept of strata, invoked by Deleuze and Guattari.

3. The most serious misunderstanding on Harman’s part concerns the status of the virtual in DeLanda. One wonders if he has just read the one book (A New Philosophy of Society: Assemblage theory and social complexity), ignoring for example, Intensive Science and Virtual Philosophy, or if it is a deliberate move. How for example does Harman come to the idea that DeLanda’s virtual is ‘unusually quantized,’ that is to say, ‘far more fragmented into discrete chunks than it is for many Deleuzians?’ In any case, if it was not clear in his preceding books, DeLanda (2010) spells it out quite clearly in Deleuze: History and Science, when he speaks of an ‘unsegmented or continuous virtual world’ in relation to the intensive and extensive.

Given that Harman is by no means a careless reader, it seems strange that his review of DeLanda’s ontology would contain these misunderstandings. Of course, I could go on to read Harman’s own work, and I have in fact browsed through some of it. But I admit, object oriented philosophy somehow leaves me cold. My interest here is to correct what I see to be a misleading review of DeLanda’s rigorous philosophical work, to correct Harman’s mischaracterization on a number of counts. It is not to enter into lengthy debates about the merits of object oriented philosophy or ontology, though I would be interested to hear an explanation of actual objects that are deeper than states of affairs, and what merits there are in conceptualizing processes as objects.