Some thoughts on Graham Harman, "Lavalampy materialism", and Deleuzian "undermining"

Recently, Graham Harman and friends have been unfairly criticizing Deleuze for holding positions that he never held. In his recent article in The Speculative Turn, Harman cites Deleuze is as the arch-example of an “underminer” i.e. someone who eradicates or “undermines” objects in favor of some kind of primordial process or undifferentiated goo. The metaphor Harman is working with is that Deleuze goes “underneath” objects and says that they are “nothing more than the derivative actualization of a deeper reality—one that is more diverse than a lump, but also more continuous than specific horses, rocks, armies, and trees.” Accordingly, Harman says that

Undermining occurs if we say that ‘at bottom, all is one’ and that individual objects are derivative of this deeper primal whole. It happens if we say that the process of individuation matters more than the autonomy of fully formed individuals. It also happens when we say that the nature of reality is ‘becoming’ rather than being, with individuals just a transient consolidation of wilder energies that have already moved elsewhere as soon as we focus on specific entities. There is undermining if we appeal to a pre-objective topology deeper than actuality, or if we insist that the object is reducible to a long history that must be reconstructed from masses of archival documents.

This whole idea of undermining is thus leveled at Deleuze with the intention of making fun of his metaphysics of objects, of pointing out his apparently childish emphasis on flows and processes at the expense of stability. Harman and friends seem gratuitously adept at constructing strawmen of their philosophical opponents. Having just taken a grad seminar with John Protevi on A Thousand Plateaus, I can say confidently that Harman and friends are painting a ludicrously simplistic picture of an incredibly nuanced and sophisticated metaphysical position. To completely dismiss Deleuze for being “lavalampy materialism” is completely childish. Lavalamps? Seriously? That’s a pathetically immature way of characterizing a serious philosophical giant such as Deleuze. Harman and friends want you to think that Deleuze is all about flow flow flow. But If there is one thing I took away from Protevi’s class, it’s that this is  a completely wrong reading of Deleuze. Deleuze never reduced all of reality to pure process or flow. Not at all. His ontology is one of flows and breaks.

To think that for Deleuze an object is somehow “unreal” or “derivative” is to completely misunderstand his concept of stratification and destratification. As Deleuze says in “The Geology of Morals”, “Strata are Layers, Belts. They consist of giving form to matters, of imprisoning intensities or locking singularities into systems of resonance and redundancy, or producing upon the body of the earth molecules large and small and organizing them into molar aggregates” (40). And being a concrete thinker, Deleuze immediately gives an empirical example of stratification:

In a geological stratum, for example, the first articulation is the process of “sedimentation,” which deposits units of cyclic sediment according to a statistical order: flysch, with its succession of sandstone and schist. The second articulation is the “folding” that sets up a stable functional structure and effects the passage from sediment to sedimentary rock. (41)

Does that sound like lavalamps and wishy-washy goo? No! While, yes, Deleuze did emphasize flows and intensive processes, he never did this at the expense of stable resonances, stratifications, and actual organizations. Protevi emphasized over and over again that many Continental thinkers make the mistake of thinking Deleuze was all about flow flow flow. This is a shallow and quick reading. Deleuze always emphasized both flows and breaks, never one at the expense of the other. And it would also be a mistake to read Deleuze as saying that stratifications are somehow less real than the underlying intensive processes. Strata are fully real insofar as they have affects on other bodies. And strata have all the autonomy as Harman’s withdrawn objects. And, yes, in some sense Deleuze saw fully autonomous objects as being limit cases rather than full fledged realities. But this is obviously true and not at all incompatible with Harman’s position since even Harman agrees that objects aren’t eternal: they are routinely destroyed and come into and out of being, only being semiautonomous from the rest of reality. A rock, e.g., while seemingly stable to us humans, would look like a fluid flow to the eyes of a creature with a metabolism on the geological timescale.

But, in my opinion, Deleuzian metaphysics is superior to OOO in that it has more explanatory power. What does OOO explain? What phenomena does it make more clear? What data does it synthesize? What predictions does it make? What errors does it correct in previous systems? What grounds does it give for explaining the reality around us? In my opinion, the rich structure of Deleuzian metaphysics has far more explanatory power than OOO.

Take the example of crystallization. First, you have a supersaturated solution. Then you have the process of nucleation and the subsequent crystal growth which actualizes out of the potentiality of the supersaturated solution. We can explain this in terms of Deleuze’s metaphysics.* The supersaturated solution is undifferentiated yet its Virtual field contains the possibility of crystal actualization. When a singularity crosses a threshold, the Virtual possibility of crystallization actualizes and a process of stratification/actualization occurs wherein a line of flight is selected out of the virtual phasespace and novel strata/organizations are formed through immanent processes of organization.

This is a clearcut example of Deleuzian metaphyiscs at work. How would OOO make sense of the process of cystallization? Well, as I see it, it would be forced to say that, on some level, the supersaturated solution is itself a withdrawn object, or composed of withdrawn objects. And somehow the process of nucleation is a matter of a withdrawn solution-object transforming into the withdrawn object that is the newly formed crystal. Do you see the problem here? Because OOO is forced to say that “it is objects all the way down”, it is unable to account for the undifferentiated solution qua undifferentiated solution. This is why you need a process philosophy that includes the ontological register of intensive flow. The most parsimonious and scientifically respectable explanation of the process of crystallization must include the intensive level in addition to the Virtual realm, which accounts for the ready-possibility of the supersaturated field to nucleate. I am unclear on how OOO would explain this example. Saying it’s “objects all the way down” seems decidedly unexplanatory, especially in the context of “things” like solutions.

Harman and friends will likely respond by saying that the supersaturated solution is itself composed on many tiny withdrawn objects, since they put forward an infinite regress. But at some point, you lose the explanatory power of the term object when you apply it to everything. This is why OOO is only partially complete as an ontology. I agree that on some level objects must be considered stable and semiautonomous. But this stability needs to be understood at the proper scale, spatially and temporally. Which is why process philosophies are so helpful, since they can capture both change and stability, flows and breaks, intensive processes and stable object-resonances.

I hope this post has cleared up some misconceptions about Deleuzian metaphysics. Harman and friends would be well-off if they stopped their ridiculous discussion of lavalumps and wishy-washy goo. A careful reading of Deleuze obviates any such misguided reading of flow at the expense of stability. As Deleuze says, “Saying stratified is not the worst that can happen.”

*EDIT: when I say that Deleuzian metaphysics can explain phenomena like crystallization, I do not mean “scientific explanation”. What I mean is that Deleuzian metaphysics can give an account of the conditions for the possibility of the phenomena, which is a metaphysical explanation rather than a scientific explanation. My answer to Morton’s question of “…what does the Deleuzian description add that I can’t simply see with my own eyes, aided by a decent chemistry textbook?…” is as follows: The Deleuzian account is not meant to supercede or replace scientific explanation. It is meant to give science an underlying metaphysics to account for the conditions of possibility for the phenomena it studies that does not invoke higher beings or transcendental forms. It does this not in terms of a Hylomorphic or Spiritualistic metaphysics, but rather, a metaphysics of immanence compatible with naturalistic monism (the idea that there are no supernatural events or processes). Metaphysics, as I understand it then, it meant to give ontological flesh to the scientific models such as dynamic systems models that actually do talk about phasespaces and singularities. What is the ontological register of the phasespace? Does it “really” exist? Deleuze can help answer these questions.



Filed under Philosophy

14 responses to “Some thoughts on Graham Harman, "Lavalampy materialism", and Deleuzian "undermining"

  1. Bits of lava in a lava lamp occasionally break off from other bits while they flow around.

    • This is the sort of de-concepted contempt for argument that gives OOO its bad name. Gary actually engages with the content of Deleuze’s philosophy to dispel a rumour concocted to replace that philosophy with a snide deceptive slogan.


    “…what does the Deleuzian description add that I can’t simply see with my own eyes, aided by a decent chemistry textbook?…”

  3. Pingback: Phasespaces, Normal Science, and Goo: On Tim’s reply to Gary « Networkologies

  4. Good stuff this. I suspect the OOO proponent has a job even explaining how she can be ontologically committed to crystals (or clowns) given that we never access them.

  5. Mark Crosby

    Gary, as a long time, silent admirer of your efforts to bring Phenomenology up to speed with with 4EA, I must say that IF a certain head-scratching post, made today, was aimed at your post here, then I have to say that You-Know-Who has really taken over the Ministry of Magic (or thinks he has 😉 Thanks, Mark

  6. Pingback: A Challenge to OOO: Prove me wrong, or, write Summa Contra Processis-Relationalis! « Networkologies

  7. Pingback: The Twilight of Becoming (and Process) « Naught Thought

  8. I think your remarks about explanatory power than the incredibly vague and all-encompassing terms of “undermining” and, even worse, “overmining”, which are just superficial appropriations of “reduction”. I think that much of object-oriented “ontology” is in fact a debased and disguised form of epistemology.
    I argue this point here:
    and in my series of posts on Harman’s THE THIRD TABLE beginning here:
    I think this little booklet is a must read for seeing the Klingon cloaking machine at work masking epistemology into deep space ontology. Harman can even go on condemning others as guilty of “epistemism” ( while blithely pursuing a rather naïve epistemological project.

  9. Reblogged this on AGENT SWARM and commented:
    Gary Williams has given the decisive reply to the lavalampy legend that OOOxians have been trying to stick to Deleuze’s name. He also calls attention to the biggest thing that is present in Deleuze’s philosophy but missing in OOO: explanatory power. For all its talk against reductionism in the form of under-mining and over-mining OOO is guilty of the worst reduction of all: meso-mining, or the reduction to objects. Harman’s “third table” is the meso-object, “deeper” than the scientific and the common sense table because shorn of all sensible and knowable predicates. For details see my review of THE THIRD TABLE here:, and its sequel explicating some positions with more explanatory power (Latour, Laruelle, Feyerabend) here:

  10. You are a very persuasive writer. I can see this in your article.
    You have a way of writing compelling information that sparks much interest.

  11. Bill Schaffer

    I think is exactly correct: “on some level objects must be considered stable and semiautonomous. But this stability needs to be understood at the proper scale, spatially and temporally.” The fact that things – all things – lose their integrity and/or isolation at the right scale of time and space, does not imply that said ‘things’ are less than real at the level of their constitutive dynamics, or reduced to mere flow. I am not through exploring OOO – I may still be missing something – but it seems to amount to a rhetorical hypostasis of the concept of ‘object’ and nothing else. Objects all the way down. (This may be a failure of my imagination, so I continue to explore at this point). What I am completely sure of is that the account of Deleuzean virtuality as ‘lavalampy’ is a caricature of criticism and deserves to be laughed at.

  12. I was on Digg looking for bulk headwear groups, when I found a link to this blog, glad I stopped by!!! Thanks!!!!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s