Bundling and Unbundling I


“The identity of any real entity must be accounted for by a process, the process that produced that entity […] When it comes to social science the idea is the same: families, institutional organizations, cities, nation states are all real entities that are the product of specific historical processes which created them and those that maintain them.” -DeLanda

One high school summer vacation, I lost a few hot Florida afternoons devouring a copy of Daniel Dennett’s Darwin’s Dangerous Idea. He called this “dangerous idea” a universal acid that would burn through any container imposed on it and would destroy and reform everything in its wake. The container of biology would not hold this fundamental idea- that there is no need for essences; that simple mechanisms can create evident complexity; that the category does not necessarily precede the instance.

That book began to formalize an idea that I was holding in separate silos:

  • The broadly understood Turing thesis that a machine could calculate without “knowing how to calculate.” (Infamous example: cellular automata)
  • Herbert A. Simon was quickly becoming a major influence in my view of organizations, wherein artifacts and interfaces were crystallized habits/memory. (Influenced my decision to go to Carnegie Mellon, and what I studied when I got there).
  • My readings into Pierce and the Pragmatists in college, and their epistemological methods.
  • The idea of the artificial as a subset of (instead of antithesis to) the natural.
  • The Dennett/Hofstadter “Mind’s I” view of the self as a “narrative center of gravity” (that is, as a fiction that is immensely useful). Heterophenomenology.
  • The general idea of Wittgenstein’s ladder, the useful fiction/abstraction, etc.
  • The flavors of agency and intelligence of people, machines, animals, and organizations
  • Later: the idea of an “ecological perspective” of a system
  • Later: DeLanda’s perspective; Bogost’s critical perspective “Unit Operations”

It was not necessarily a deliberate project for me (or else maybe I’d have worked it all out much sooner, I’d think). I sketched out my trajectory from 10,000 feet here six months ago.

I’ve almost been writing here for a year. Isn’t that odd.



I’ve been writing about particular instances of this broad perspective over at Sweet Talk:

In “Only the Variation Is Real“, I used the concept of speciation (and the organization of the United States) to begin speaking about operational definitions. A cut from DeLanda that I didn’t end up including but roughly captures the idea:

In this ontology all that exists in the actual world is singular individual entities (individual atoms, cells, organisms, persons, organizations, cities, and so on) whose main difference from each other is spatio-temporal scale. There are no totalities, such as “society as a whole”, but a nested set of singular (unique, historically contingent) beings nested within one another like a Russian Doll Between one entity and the larger one is the relationship of parts to whole (but not one of membership in a general category).

Being a Good Person” is probably my most-viewed post anywhere in a single day thanks to Reddit- (with only one other contender, a summary post that Cory Doctorow retweeted). In it, I tied in Kevin Simler’s timely definition of Personhood into the Sweet Talk discussion. The same motif- I enjoyed the new Personhood definition for its transactive, nonessentialist conceptualization.



I’ve long been fascinated by hybrid systems of humans, organizations, and machines. It’s the cross-section that I always sought in school and it’s what occupies my thoughts while I dabble with building things. That’s often why certain stories and concepts tend to blur together on this blog: for example, Cory Doctorow’s War on Generalized Computing and Daniel Estrada’s Organisms of the Future. One is about legal and procedural challenges about computer ownership, around problems that form when computers are not distinguishable from our traditionally “dumb” tools, from the environment, from our own organs. The other is about a definition of organism that is useful but that is also broader than our typical biological category “organism”, allowing for humans to be part of organisms in tandem with nonhuman components, or for machine systems to be organisms or parts of organisms (or both concurrently).

There are, I think, pragmatic implications for this kind of rhetorical and ontological nitpicking. It’s not just window-dressing, and I hope to make a forceful case for that sometime.



Chris Dixon and Benedict Evans have written some great stuff about the aggregation and disaggregation of services and functions in certain consumer markets over time. Newspapers, for instance, came to fulfill several functions: as a platform for the distribution of weather forecasts, local and national news, wanted ads, op-eds, advertisements, and other forms of curation. There was no guarantee that newspapers ought to do all of these things, or that the technologies that would replace newspapers would continue all of these tasks. Instead, Dixon argues technologies we might now recognize as successors have “unbundled” and tackled smaller sets of functions: Craigslist, the Huffington Post, and Twitter all eat at parts of the Newspaper without being a newspaper itself. Likewise, I’d argue that the car is not just an iron horse- new functions related to car culture and expected car behaviors have aggregated over time. What is a car [in the first world] without an audio player? It’s plausible that the entertainment system will survive the steering wheel as an essential part of the personal vehicle package.

Leave a Reply

Your email address will not be published. Required fields are marked *