big

NOOP.NL | The Creative Networker

Self-Organization vs. Anarchy

19/10/2009

Anarchy-eflon-3261945045 Jim Highsmith and David Snowden think that self-organization is different from anarchy. They say that self-organization (in a social context) needs some form of management/leadership, and that it otherwise degenerates into anarchy. I disagree.

The origin of the word “anarchy” is: anarchia, from Greek, and from anarchos, which means “having no ruler”. Various dictionaries list two meanings for anarchy:

a) absence of order (or presence of disorder)

b) absence or denial of any authority or established order

In system dynamics, this means either of two things: a) chaos (no order), or b) complexity (order, but not imposed by an authority). This is depicted in the following picture. Governance stretches from order into complexity. And anarchy, the absence of governance, stretches from complexity to chaos.

Governance-vs-anarchy 

Anarchy has a bad name, which is undeserved. In the minds of most people, anarchy is equal to chaos. It is a misconception, and probably the main reason why some experts don't like associating self-organization with anarchy. Galaxies behave in an anarchistic manner, and yet they are not chaotic. Ecosystems are anarchistic, but they are also not chaotic. And countries without (working) governments are anarchies, but are also not necessarily chaotic.

A self-organizing system is the complex variant of anarchy. This is true in physics, in chemistry, in biology, and in sociology. There are many definitions of self-organization, and none of them require leadership, management, or authority. In my opinion, it makes no sense to change the meaning of self-organization when applied in a social context.

I think the real issue that some people have with anarchy, is that such unmanaged systems can behave in a way that the stakeholders don't value. When my children are playing a game, running around me and yelling in my ears, I would eagerly agree that this is anarchy. But the children are self-organizing. It just means their way of self-organizing is not very much appreciated by me as their primary stakeholder. And yes, then I prefer to enforce some governance. As Dave Snowden said: “Then you draw a line on the floor, and you tell those kids: if you cross this line, you’re dead.”

(picture by eflon)

Twitter TwitterRss SubscribeEmail NewsletterDelicious Bookmarks

Latest, greatest and favoritest posts:
Managing Leaky Organizations
Empowered, Whether You Like It or Not
Eat Chaos, Poop Order

This article is written by on in Empowerment & Delegation. Jurgen Appelo is at Happy Melly. Connect with Jurgen Appelo on .

This article was posted in:

  • http://www.cognitive-edge.com Dave Snowden

    Actually I said that a complex system has to have a degree of constraint, its not constraint free. In that context the constraints can be determined top down, or bottom up or some combination. I don’t think (as a child of the 60s and 70s) I would risk talking about anarchy as I know how many variants there are.
    I’d also argue that order complexity and chaos should not be represented as a continuum but as a series of phase shifts (think solid-liqud-gas with latent heat).

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    Dave, thanks for the explanation!
    I will think about the phase shifts. I haven’t seen a representation like that, but it sounds interesting.

  • http://profile.typepad.com/ngogerty ngogerty

    Typical complexity, etc. models show the continuum correctly extending on the “order” end of the continuum to an end point labelled stasis or something approximating zero latent energy.
    “order” is a bit of a flawed label in this representation as most would assume that the left end of the continuum order, is a desired state.
    this isn’t typically the case as total order is usually a proxy for absolute stasis (heat death if you will in thermo dynamics). Complexity is an interesting narrow region between zero energy zero information and absolute chaos. “interesting” complexity needs enough entropy etc. to respond and adapt to its environmental shifts. Absolute order isn’t representative of a desired state and thus presents a false percieved optimal choice in the model as presented above. Just my $0.02. but an “optimal” band of complexity presented within region within a continuum ranging from a 0 to infinite entropy state may be more helpful.