Possible Congruence of Harman's Withdrawal Thesis with Ecological Science?

For the longest time I’ve had trouble understanding Graham Harman‘s radical thesis of withdrawal. What does it mean for two entities to “withdraw” from each other as they interact? How can we reconcile this idea with complete causal closure? Doesn’t gravity, for example, affect an apple through-and-through? It wasn’t until I read the following passage from Varela and Maturana’s interesting little book, The Tree of Knowledge (1987), that I made sense of it as a general ontological requirement.

Ontogeny is the history of structural changes in a particular living being. In this history each living being begins with an initial structure. This structure conditions the course of its interactions and restricts the structural changes that the interactions may trigger in it. At the same time, it is born in a particular place, in a medium that constitutes the ambience in which it emerges and in which it interacts. This ambience appears to have a structural dynamics of its own, operationally distinct from the living being. This is a crucial point. As observers, we have distinguished the living system as a unity from its background and have characterized it as a definite organization. We have thus distinguished two structures that are going to be considered operationally independent of each other: living being and environment. Between them there is a necessary structural congruence (or the unity disappears). In the interactions between the living being and the environment within this structural congruence, the perturbations of the environment do not determine what happens to the living being; rather, it is the structure  of the living being that determines what change occurs in it. This interaction is not instructive, for it does not determine what its effects are going to be. Therefore, we have used the expression “to trigger” an effect. In this way we refer to the fact that the changes that result from the interaction are brought about by the disturbing agent but determined by the structure of the disturbed sysem. The same holds true for the environment: the living being is a source of perturbations and not of instructions.

Now, at this point the reader may be thinking that all this sounds too complicated and that it is unique to living beings. To be exact, as in the case of reproduction, this is not a phenomenon unique to living beings. It takes place in all interactions. And if we do not see it in all its generality, it becomes a source of confusion. (p. 96)

Basically, Varela and Maturana are saying that we can only think about objects in terms of unities that are structurally determined in their organization. All changes in a system are changes of structure, which can be the result of intrinsic structural dynamics or triggered by interactions with everything else. Moreover, Varela and Maturana claim that this feature of structural unification undergoing change applies to  the scientific description of all interactions because otherwise we couldn’t make sense of how and why object unities breakdown (a car is the most obvious example).

While I might be mistaken, this sounds remarkable like Harman’s thesis of objects withdrawing from each other as they interact. Harman seems to claim that every object-object interaction is characterized by each object retaining some “inner core” or “subterranean essence” while nevertheless interacting or “translating” on a “sensuous” level. I’m a visual thinker, so I always had trouble conceptualizing how this process of withdrawal works. But now I think my problem was with the specific term withdrawal, which is actually a misnomer if I am understanding Harman, Varela, and Maturana right. From what I gather, a better description of object-object interaction might go as follows: when an object interacts with another object, both objects need to retain their internal structural unity during the interaction in order for it to be considered an object-object perturbation or “translation”. If there was not a retention of unified structure, we wouldn’t be able to talk about the interaction in terms of two, separately organized unities. Without a separation of unitary structure (i.e without being “operationally distinct”), the structural changes become merely changes of the state of one system rather than changes as a result of the perturbation or translation between different objects. Varela and Maturana also point out that another possibility of object-object interaction is destruction, whereby one object destroys the organization of the unity of another.

One could say that my eyes have now been opened to the relevance of object-oriented philosophy to developments in ecologically oriented cognitive science. I still have trouble with how Harman argues for his thesis, but now that I have at least “translated” it into a familiar conceptual domain, I think I can finally accept some kind of generality of withdrawal for all possible object-object interactions. There’s gotta be a better metaphor than withdrawal though. It’s far too “spooky” for my likening. I’m also not convinced that this was Heidegger’s greatest lesson. But like I said, my eyes are now open to new possibilities of relevance.


In response to my comment about Heidegger’s greatest lesson, Harman asks

Well what on earth are the other candidates?

My reply:

One of Heidegger’s greatest lessons was his distinction between the phenomenon and the semblance. In my mind, this was a realist “upgrade” to Husserl’s transcendental reduction. In his surrounding lectures and in BT, Heidegger critiques Husserl’s transcendental reduction for missing the original phenomenon to be described: our experience as embodied entities living on a physical earth, but “worlded” in terms of the categorial (i.e. socially constructed) intuition. Husserl accepted the metaphysics of the “natural attitude” but bracketed realist questions from the ultimate reduction for sake of Cartesian “certainty” and wanting to achieve “apodicticity” . Heidegger recognized the irony (and ultimate futility) of using Cartesian standards of certainty for the investigation of something as concrete as lived experience. This is why he ultimately endorsed a hermeneutic phenomenology that started from within the messy circle of lived experience, rather than from eidetically purified descriptions of transcendental correlation. By introducing the concept of “semblance” into phenomenology, Heidegger provided a means to capture by formal indication the natural attitude’s acceptance of empirical realism without suffering from internal inconsistency (since he can show that his opponents’ positions stand upon the strength of unquestioned assumptions). Husserl would have never said things like:

With circumspective interpretation, the way in which the entity we are interpreting is to be conceived can be drawn from the entity itself, or the interpretation can force the entity into concepts to which it is opposed in its manner of being (SZ 150).

Heidegger is talking about situations like when we see a stick in the grass as a snake or make perceptual mistakes wherein we radically misinterpret the given phenomenon. Fool’s gold is good example. Such language was phenomenological heresy for Husserl, which is why he thought Heidegger was trying to naturalize consciousness like some sort of objective anthropologist.  But by allowing realist concepts like the semblance into his methodology, Heidegger was able to account for the full spectrum of human experience, such as  when “The present-at-hand, as Dasein encounters it, can, as it were, assault Dasein’s Being; natural events, for instance, can break in upon us and destroy us” (SZ 152). Our experience with earthquakes, volcanoes, animals, etc. shows us that there is indeed an objective reality “out there”, ready to stand in our way or assist us (as with the sun, the wind, and the sea). If some hapless fool interpreted an earthquake as a simulation or dream, Heidegger would say that he was experiencing a semblance of the earthquake, not the genuine phenomenon of the earthquake as it is “in itself”. Indeed,”the fact that Reality is ontologically grounded in the Being of Dasein, does not signify that only when Dasein exists and as long as Dasein exists, can the Real be as that which in itself it is” (SZ 212). Surely this is a vast improvement over Husserl’s phenomenological method and deserves attention as a philosophical methodology.



Filed under Philosophy

7 responses to “Possible Congruence of Harman's Withdrawal Thesis with Ecological Science?

  1. Pingback: another from Gary « Object-Oriented Philosophy

  2. Joseph C Goodson

    Gary, I have been enjoying this back and forth between you and Graham on these Heideggerian issues. Very interesting.

    Here is how I interpret, in a very elementary sense, Graham’s withdrawal thesis: knowledge is a kind of relation, a subset of relation. For relation to make sense, you have to have something to relate to, there has to be a difference between the relation to the object and the object itself. Knowledge is a relation to the object, but nowhere does knowledge merge with the object — we would cease to have a relationship and it would become an identity. Then, of course, you would have to accept a kind of ontological monism of some type in which relation is just an illusion or epiphenomenon which disguises the true fact of all objects’ identity with every other object (or at least the identity of many levels of objects into undifferentiated lumps of some kind). In a way, it is a very basic and concise idea: knowledge never becomes identity, relation never becomes identity. I think your use of the autopoietic structure (as Graham says, Levi has made very good use of this concept as well) is very enlightening and gets at the same idea: there is a consistency of any object’s particular window of relating that already includes and excludes another object’s qualities, and all according to the object’s own internal demands. Here, too, we can see why objects can never touch directly as they bring their own environment, their own horizon of relation, with them. The idea of an object’s withdrawal is one part of a constellation of ideas, each implying and reinforcing the other, which tries to solve the problem of an extreme monism on one end, and a world where nothing can touch at all, on the other.

    As for the word “withdraw,” itself, I think that your sensibility about it being “spooky” isn’t off the mark. Conversely, that is the reason I like the term, and I can see why it appeals so much to Graham, with the penchant for the weird and exotic that you can see at every moment in his philosophy. When I think of withdrawal, I can’t help but think of, say, the great Cthulhu, slumbering in the depths of the world for eons.

    In any case, very nice blog. I will follow it with interest.

  3. Gary,

    I’m glad to see someone else is picking up on this connection. In The Democracy of Objects I draw heavily on autopoietic theory in the development of my account of objects (especially in chapter 4 which is devoted to autopoietic theory, developmental systems theory, and Lacanian psychoanalysis). While I do not argue that all objects are autopoietic systems (I retain Varela and Maturana’s distinction between allopoietic and autopoietic systems), I do argue that all objects maintain only selective relations to the rest of the world and that these relations are constituted self-referentially (i.e., by the system itself). You can get a sneak peak at some of these lines of argument at the following two links:



  4. Pingback: Autopoiesis and Object-Oriented Ontology « Larval Subjects .

  5. Pingback: Withdrawal in Harman : Mormon Metaphysics

  6. Thanks for this. It made a lot of things click finally. I think the term “withdrawal” more so than anything else has confused me precisely because of the Heideggarian terminology transformed to this new context. I think I “get” OOP a bit more now, although I’m still pretty skeptical it is that novel.

  7. You might be interested in a recent post of mine where I compare Harman on with withdrawal to JJ Gibson on perceiving reality: “Heidegger/Harman and Gibson are considering the same situation, someone examining an object. But they frame it differently. Heidegger/Harman assume the object is real and note that it always outruns scrutiny. Gibson doesn’t know whether the object is real or not, but notes that, if it is real, it will outrun scrutiny. Though I’m not aware that Gibson knew of Heidegger’s ideas – he worked in a rather different intellectual tradition – his treatment of Descartes’ epistemological problem presupposes something rather like Heidegger/Harman’s analysis of real objects.”


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s