Thursday, October 27, 2005

Being late...

....and very sorry for it! What a delightful and fun reading! I very much enjoyed Clark’s clear, illustrative writing. The examples he gives to explain constructs which otherwise seem hard to grasp are great.

I couldn’t decide on a heading for my post, so I will give my choices as subheadings.

Social Psychology meets Andy Clark
Co-op living – emergence gone wild?

After reading chapter 4 and the explanation of direct and indirect emergence, I could not help but to think of the co-op where I currently live at as a case of emergence. At my co-op there are 120 people but there is no central figure who coordinates the operations. Democratic and student owned is what my housemates like to call it, meaning nobody is in charge. Tasks and responsibilities are distributed. So it seems to me that the co-op is an emergent system. However, the system does not work very well, precisely because there is nobody in charge and the local interactions which are one characteristic of emergent systems take strange forms. A good example is the kitchen. There are cleaning crews for the kitchen every day except on Saturday nights and Sunday mornings. While our kitchen is usually decently clean, by noon on Sunday, there are piles of dirty plates, heaps of trash in all corners and the surfaces look like gremlins fought a war with our fridges. That is despite the fact that everybody is supposed to clean their own mess. So, what happens between Saturday and Sunday? This is my explanation: On Saturday afternoon, an innocent individual places a cleaned plate beside the sink because no rack is immediately available. A fellow co-oper sees the plate and puts his plate on top. The next fellow co-oper runs by in a hurry, sees the hint of a pile and sets his uncleaned plate on top. Three hours later, a mess has emerged. It is astonishing to see the similarities in behavior between the co-opers and the tiny ants building arches!

Where does social psychology come into the picture? Well, there is a phenomenon in social psychology which is called bystander-effect. In German, the word denotes a diffusion of responsibility. In very simplistic terms, bystander-effect means: The more persons there are, the less likely a single person feels responsible because responsibility becomes distributed. The emergent property is that nobody takes action. This is precisely what happens at the co-op! At one time on a typical Saturday afternoon there are certainly two persons in the kitchen at the same time and one sees the other stack their uncleaned plate on top of the pile or becomes the witness of a food processor explosion splattering the walls with pieces of what was supposed to be a banana milk shake. However, the culprit and the bystander do not communicate - a local interaction, comparable to one based on an inhibitory link in a neural network, does not take place. So, my question is: The co-op seems to be an emergent system, but it does not work. How can we make such a system work without installing video cameras and inflicting punishments?


Blogger gfp said...

I have been thinking about this. First, I think it's important to note that the emergent approach is more of a tool of description than of change formation. That said, to change emergent cognition, wouldn't you need to look at the forces (environmental and social) on the persons and the individual conscious and determine what forces are influencing what? Then you could address certain forces to increase the likelihood of the results (a clean kitchen). That's my stab at it....

12:45 PM  
Blogger Sean McCarthy said...

This is a really interesting post, as it takes the theoretical idea and places it in a practical context. Emergent situations in the non-human world - say a colony of ants - do have the kind of triggers that create a functioning environment. Chemical markers, such as pheremones, catalyze discreet work processes so that the larger colony works effectively. In your co-op environment no such triggers exist, which calls for management of some sort. I don't think that an emergent and a managed system are necessarily incompatible: in some ways, management can create a working emergent system - the setting of rosters, for example. Simply put, the emergence needs to be managed.

I am not too sure that I agree with Greg on his point that the emergent approach is a tool of description rather than change, however. Emergence is all about change - but from the bottom up. I think it can be engineered as much as any management system - it just requires a different approach to how the system works - for example, how can bottom up and top-down approaches be synthesized?

1:38 PM  
Blogger jmj said...

God I hate communal kitchens.

It sounds like the problem is that the kitchen users depend on the nightly cleaning crews to keep things tidy. Instead of behavior changing on the weekends, the exact same behavior, Resident X leaving a dirty plate in the sink, becomes compounded because the cleaning crew is not around to keep its effects (a dirty kitchen) in check. Perhaps if you eliminated the cleaning crews, the piles of dirty dishes would encourage everyone to clean up after themselves, even on the weekends.

2:50 PM  

Post a Comment

<< Home