More on self-organisation

... and on what they tell us about non-self-organised units.

Older, related posts:

I had recently the (not very entertaining) opportunity to look more closely at rather temporary self-organising. The previous posts were about self-organising into an actual organisation; this time I'm thinking of very short-term self-organising  i.e., working together for a few minutes or hours in pursuit of a common goal.

Problem #1 to overcome:
Is there really a common goal? Often times there isn't. Some dudes are just too stupid for that.

Problem #2 to overcome:
Does a standard template for behaviour exist?
Multiplayer games often have this in PvE (player versus environment) settings, for example. Self-organisation along a consensus template appears to be really easy and common.

Problem #3 to overcome:
It really takes some confidence to pull on one string. It's amazing how much one person with a reputation can turn a disorganised group into a team following his or her tactical directions. The success rate is fine if the group pulls on one string, which makes it even more astonishing that they usually don't.

The confidence and template things have interesting parallels in the top-down organisation of a military.

Trust / confidence is a major component of good leadership and of effective co-operation between leaders of one level. One of the advantages of the German army in WW2 was that its officers were very often trusting each other because they were able to anticipate the other's behaviour. This also touches on the template thing. German military writings of the pre-'45 period often had tactical examples, but many of the drawings had a "Kein Schema!" on it, indicating that it was not a template to be used at all times. Still, there was so much of a tactical consensus after the summer of '40 that many actions were much smoothed because the involved leaders already knew the tactic's basic features and the thinking behind the tactic in advance.

The best ways for short and temporary self-organisation appears to be an earned reputation or a rather simple and standardised challenge. The military can make use of the reputation in temporary small units, but it often has to replace it with the artificial authority of a superior rank. Simple and standardised challenges are rare and rather relevant for support troops than combat troops.

It's intriguing to think about how a dedicated reputation system in a military organisation would work out. Imagine you complete your basic NCO course and get the choice to which unit you want to go. You get to know some basic information about different company commanders; which courses they completed, their scores, their reprimands (not necessarily a bad thing!), their track record in simulated free play exercises including the debriefings...and said company commanders would get the same data on you.
The result would probably be a lot of gaming the system and a information overload-induced limitation to the own formation, but it would also result in some crack units and some crap units. That would be an invitation for creative destruction; the 10% worst units (judged by their free-play exercise results) could be destroyed ("disbanded") with all personnel fired ("retired"). Next period; new fresh company commanders create the desired qty of companies again. This sounds a bit like soccer leagues to me.

S Ortmann

P.S.: I found this in my draft article list. It may have been published otherwise and be accidentally reverted to draft when I fixed a typo or such.  There were no comments archived yet, so I'm not sure but I think this is an old text, likely 2011. It's not great, but doesn't need to be lost in the drafts archive either.


  1. A very good sum-up of the problem, which I tend to consider from a political point of view.

    I would add one aspect, that is present in the three "problems" presented, and which is the point of transcendance.

    Fighting is a difficult thing, first as such, and then because of the individual risk involved. The key point is what motivates people to do it.

    A "goal" is too operational, it is limited to the immediate future. The reason to fight is more complex. Even when there is an obvious situation (an enemy is approaching your position / village) several options present themselves to the individuals, before they decide to unite their efforts. That would be surrender, flight or fight (and then, which fight : purely defensive, harassment before the attack of the village, after the attack, retarding manoeuvers etc.)

    What will be done depends on what the individuals view as the best outcome for all concerned. For instance, some Volkssturm units fought bitterly against the Soviets as they thought it would enable their own families to escape to the Western side.

    So the main aspect of self-organisation would be a common view of the situation : what is the end situation we strive for, and what is the right thing to do. The latter aspects encompasses moral questions (do we take prisoners ? What about innocent bystanders, or women and children of the enemy ?) which are strongly determined by society, and by technical aspects pertaining to « the most appropriate methods to fight ».

    The aspects stemming from society is what I call « transcendance » in that it encompasses everything the society does (how it lives, fights, dies, enjoys itself, views the optimal state etc.). I think this is what the « weaponization of religion » (muslim fundamentalists, evangelists etc.) is about.

    More thoughts on this could be gained by reading about studies around such concepts as « prisoner's dilemma », cooperation vs. Concurrence etc. ,like this article


  2. Wallenstein gave his best companies the right to elect their leaders. I think this might be a good idea, set a goal for performance ambitions and give out an organizational reward. To achieve that goal the people must all pull together and working that out allows to establish the suitable structure to keep it that way.
    That is a major problem for the modern structure, but you could take a Prussian approach and make all officers look the same as a kind of brotherhood. They do have the knowledge to lead, but can they convince the troops of being led by them and do they then together achieve the goals? Anyone remember Hannibal?