Of Lumps, Lava, and Firehoses
Some notes on process philosophy contra object-oriented ontology
August 18, 2011

Prompted by Ben Woodard, there's been a recent flurry of posts in the philosophy blogosphere about the differences between process philosophy and object-oriented ontology. Specifically, Ben argues that thinkers of process are stuck "in the twilight of becoming" and content to allow "becoming to be utilized as an escape hatch in argumentation."

There have been several replies to Ben's charge, but I'm interested in responding to Steven Shaviro's in particular.

Shaviro offers mild objection to the characterization of lumpy or lava lampy materialism, while admitting that advocates of process philosophy would do well to clarify their positions more clearly. He adds my accusation that process philosophy is firehose metaphysics to the lumpy/lava mix, and here I should clarify that my reasons for using that term have more to do with my dissatisfaction with flow and continuance as metaphysical fundamentals than they do with my desire to accuse process philosophy of indistinction. It seems to me that the indistinction is meant to be a feature of process thinking, not a bug.

But as Levi explains, the object-oriented ontology's dispute with process philosophy is a philosophical one rather than a rhetorical one: "Whitehead undermines objects by treating "actual occasions" as the ontological foundation of being." That's right.

In my short paper about process and procedure that birthed the term "firehose metaphysics," I offered a distinction between process and procedure. Despite Shaviro's argument that Whiteheadian process is that of composition rather than flow, it seems quite clear to me that there's no denying the fundamental processuality of process philosophy, particularly the sort derived directly from Whitehead, in which actual entities are processes proceeding from from phase to phase. Instead of processuality, and borrowing from my interests in processes in the algorithmic or computational sense, I'm more interested in procedurality, or the logic by which something works. Levi and I sometimes talk about objects as systems or machines, and we do so because procedurally is of greater interest to us than processuality.

Thus it's no surprise to hear Steve say, "I'm inclined to think that all procedures are in fact processes, contra Bogost's opposition between them; but that not all processes are procedures." As a process philosopher, of course Shaviro sees the processes as the firmament that underwrites procedures.

To respond in a sideways way to Woodard's original question: on the one hand, more and better clarifications of the fuzziness sometimes present among process philosophers would be a welcome source of better discussion among all of us, particularly since so many such responses seem to amount to incantations of Whitehead. But on the other hand, the fundamental dispute between OOO and process philosophy is a legitimate philosophical disagreement, not just a failure to communicate or understand.

Comments (11)
Want to comment on my site? Use the comment form below.

Pardon me if I'm missing the point here. I'm certainly very far from a professional philosopher.

This distinction between process and procedure seems very curious to me, since I encountered some of (Alfred North) Whitehead's ideas from a mathematical perspective, particularly his idea of a point-free geometry and its philosophical connections with mereology. To summarize my perspective on the topic briefly, consider the problem of describing a shape mathematically (in 2 or 3 dimensions). If we follow the idea of set theory, then we say that the shape is composed of (uncountably) infinitely many points. We do this because we are trying to atomize the shape. However, the perspective is largely absurd and impossible to work with. We have to work rather hard to reintroduce the concept of continuity. In practice, we're therefore much more likely to work with a topological approach and describe the shape as an aggregate (formally set union) of non-atomic blobs/primitive shapes. But if we take this perspective from the beginning, using it as the basis for the construction of shapes, then we can avoid the introduction of points altogether and declare them an irrelevant fiction. For a discrete space like the set of natural numbers, this distinction is just a matter of perspective, but for continuous spaces we can arrive at a fundamentally different notions of what a shape is.

So, to get back to process vs. procedure, I wonder if the distinction parallels or perhaps is even deeply related to the distinction between these atomic and non-atomic ways of describing shapes/objects. In a process, it may not make sense to cleanly separate one stage from the next or last stage; they bleed into one another and overlap in a fundamentally continuous and topological manner. By contrast, the essence of a procedure (particularly in the algorithmic sense) would be to break steps down and separate them into atomic steps that are then sequenced. So in some sense this might also be seen as a continuous vs. discrete distinction; the difference between a system of differential equations specifying the continuous evolution of a physical system and a program specifying the discrete evolution of an algorithm.

Putting my ‘aesthetic’ hat on, I see this as an issue about context. I think it’s worth emphasising that on evidence of Unit Operations, you have a vastly different take on systems as reciprocal structures of regulation (different from Galloway’s Protocol for example). Hence the importance of procedure as a unit operation: a finite, discrete, step by step transformation.

I think the main difference between OOO and process philosophy is how one accounts for the discrete. It seems to me that process philosophers, like Shaviro, often separate Whitehead from other process philosophers by formulating occasions as the discrete mechanism of novelty. But often, I see no legitimate acknowledgement of discrete, in so far as, an actual discrete entity, is completely separate and unconnected with its surroundings. A thinker like Simondon would find this quite alien, as would Grant and Ben Woodard perhaps, because everything must have a genesis from which it occurs as grounded from ungrounded, hence a context must exist that causes the thing to emerge, and indeed alongside and after.

Yet, to advocate a discrete entity or ‘unit’, (and this is where aesthetic formalism kicks in for me) is to advocate something necessary by way of an intrinsic mode of generation. A unit is an entity of substantial form that holds it’s form despite it’s context. You don’t necessarily get that in Simondon nor Whitehead presumably.

yes, and in some ways (as Levi pointed out in his post on transference) points to a limit of conversation/debate/commenting, as D&G, Rorty (who could never quite take his own advice/insight), and Wittgenstein have said when people come from such differing orientations that they cannot see/accept differences than it is eventually counterproductive (one only has so much time and energy) to keep repeating oneself (there is perhaps a Hegelian spirit of Progress haunting much of this whole 'field' so I'm looking forward to seeing your pragmatist take on it).

*I have discussed a lot of the details of what I say below on my blog.


Having a degree in mathematics and being fond of algebra, I love your example/suggestion and think it quite appropriate. How we articulate a position really matters.


I agree that it appears that how one accounts for the discrete differs. Harman appears to want absolute discreta, where my conversations with Levi indicate that his view is more flexible as discussed in his forthcoming The Democracy of Objects.

For those who follow Peirce, and Whitehead is such, nothing can be said of absolute discreta, because they would be absolutely unrelated and could therefore not come into relation. This is a basic binary and is a fundamental axiom that likely differs between "process" and "OOO." (I use scare quotes because one must be careful about the generalizations.) However, consider Shaviro's repeated point "(As mentioned in my previous posting — for Whitehead “there is a becoming of continuity, but no continuity of becoming”)." The point that I would make is that mere relativity implies little as it is a logical requirement per a choice of axioms that distinguishes the process perspective from classical substance perspectives. The real question is about powers/activity/existence, the engines of the becoming, of relating in *significant* and existential ways rather than merely logical ones.

Said another way, those critical of process on this point may be mistaking a basic logical requirement with some substantive or existential claim. Some may follow, of course, but the critiques do not usually go that far an balk at the axiom.

As for advocating for a discrete unit, I do. Three. Possibility, Activity/Existence, Law or Habit. (Fellows Peirceans may now chuckle.)

I like the distinction between process and procedure, though it does leave me curious as to whether there are processes that aren't procedures. I would like to understand better, however, the essence of procedure (which is still debated in philosophy of computation circles), and what it means for objects to contain an internal procedure. The natural computational movement, of course, claims that ultimate being is fundamentally a computation, but this is a claim that OOO seems to not want to make. But then the danger is that "procedure" becomes mere metaphor (internally, objects withhold a procedure-like essence, whatever that means). This is exactly what I didn't like about Unit Operations. By the end of the book, a unit operation had dissolved into something so vaguely metaphoric that it wasn't clear anything was being explained by invoking them.

Michael Mateas on August 22, 2011 1:06 PM

Michael, to some extent that's the trouble with first principles philosophy. It's meant to be vague and up in the clouds, because it's trying to describe something at a higher level.

But that said, the question of whether there are processes that are not procedures, or vice versa, is a legitimate metaphysical question whose answer involves not merely an answer, but perhaps a fairly substantial philosophical position.

As for the natural computation movement, you're right that OOO would not agree that being is fundamentally computation. Thus, when I use computational procedurality I use it as an example of a more fundamental kind of procedurality, which leads us back to your discomfort with abstractions like "unit operation."

What made me uncomfortable with Unit Operations was the position that Unit Operations are somehow broader than a computational notion of a unit operation, and that in fact computational operations are special cases of Unit Operations. But all of the concreteness and denotative meaning of the term is only grounded in computation. You couldn't have written the book before the advent of computational manipulation of media. It's not like Unit Operations as a concept were sitting around before computation, and, historically, computation arose as a special case. Rather, the historical fact of computation gave rise to the notion of a unit operation - but then somehow the metaphor turns around and swallows the concept that begat it. I feel like a similar thing is happening in the notion of the withheld procedure at the heart of an object.

Michael Mateas on August 23, 2011 3:32 PM

That strikes a weird thing to bother you Michael. It's a book mostly about computation, thus the application thereto. This seems like an aesthetic objection.

In any case, my next book (Alien Phenomenology) returns to the unit operation without the contextual baggage of computation, so maybe it will sit better then.

Ah, the objection isn't that I'm bothered you focused too much on computation, rather that it wasn't clear the generalization of unit operations beyond computation carried any water. There wasn't enough theoretical material defining unit operations in the abstract (independent of computation), though you were claiming that they were independent of and more abstract than computation. Sounds like your next book addresses this dead on.

Michael Mateas on August 23, 2011 8:32 PM

Yeah, I think it'll do what you're hoping for.

Post a Comment

If you haven't left a comment before, your comment will be held for approval. Thanks for waiting.

  Remember me for next time?
To prove you are a human, type the word (loading...) below.
recent stuff