94. Elements of a Simulation

Note: This post is the ninety-fourth in a series about government and commercial ethics. Click here for the full listing of the series. The first post in the series has more detail on the book 'Systems of Survival' by Jane Jacobs which inspired this series.

This week, I just wanted to think out loud about what elements would need to be included in a simulation designed to test Jane Jacobs' theory about the 'Systems of Survival'.

The three basic elements I think we need are land, stuff and people.

Based on the territorial nature of the guardian syndrome, it seems as if we need a model that allows for people to interact in a 2 dimensional territory in order to replicate whatever it is about managing land that lends itself to guardian activity.

The interesting variations in the nature of the land that I see are that it could either be finite in size, infinite or it could be finite and also contain various natural boundaries (rivers, mountains, etc.) that would tend to foster distinct political groups forming.

In addition, the land needs to support some sort of resource production, since we will need something for our people to take and trade.

As for the people, they are more complex.

First of all, people will need to have various ethical approaches open to them. Looking at Jacobs' list of precepts, I see the following minimum requirements:

Use force and fraud (take) / Shun force and fraud (trade)
Work hard(er) / Take more leisure
Be obedient / Make your own decision about what is best
Be loyal / Be selfish
Be exclusive / Be open to dealing with strangers
Consume now / Invest to consume more later
Share wealth or dispense largesse / Hoard wealth
Compete / Cooperate

Naturally, putting these options into concrete terms that can be coded into a simulation will be the tricky part.

In addition, for obedience and dispensing of largesse to make sense, we'll need a concept of hierarchy or rank.

For definitions of loyalty and exclusivity to make sense, we'll need a concept of groups or identity.

If taking is an option, we'll need some sort of conflict resolution method, involving individual strength for our people as well as some logic for measuring the increase in combined strength that comes from cooperation. Note that relative strength could also function as a resource which is finite in quantity (although absolute strength would not be).

If trading is an option, we'll need at least two different goods or resources in circulation, that are valued differently by different people. It may be interesting to add another element that is finite in quantity (besides land) to see if it is treated differently.

If investing is an option, we'll need a function that translates investment of time and resources into more/different resources.

People will also need an objective so that we can rank how 'well' they are doing in the simulation. Some possibilities:
Maximize wealth
Maximize consumption of comfort and convenience (resources)
Maximize rank
Maximize status
Maximize honour
Maximize population
Maximize some combination of these.
And of course any of these objectives could be for the person themselves, for the group they belong to or for all people in existence.

Lots of possible combinations - you can see why economists like to simplify and pretend that people only care about their own personal wealth, but clearly this approach will be too limited to either represent reality or to help us develop a simulation that will incorporate the ethical choices listed above.

Finally, we'll need to define how our people's behaviour works over time. Do people just have a fixed set of ethical behaviours and we see who does best. Or do we allow people to modify their behaviour based on the context (trading vs. taking), or to modify their behaviour based on the success of those they encounter, can a leader cause them to change their ethical approach, or do we just introduce random mutation into people's behaviour patterns and see how things evolve.

Obviously, the number of permutations is large, even leaving out all the elements I've no do doubt overlooked here. It seems like it would be best to start with a simple scenario and then gradually elaborate it to take more elements and more complexity into account. But that's a task for another day...

55. One Thing Leads to Another

Note: This post is the fifty-fifth in a series. Click here for the full listing of the series. The first post in the series has more detail on the book 'Systems of Survival' by Jane Jacobs.

----

In a world where everyone is the same and they all just pursue their own self-interest with no regard for what happens to other people, the question of what happens when people with different preference types interact doesn't arise.

But my last post raised the possibility that people might have different 'temperaments' with respect to how they personally are affected by the fate of the people they deal with. As Alan anticipated in the comments on the last post, if there are different sorts of people potentially out there, then a natural question is to try and see what happens if the different types interact with one another and how a collection of different types of people might change over time.

One of the best known ways that a population makeup can change over time is via evolutionary dynamics. People who are more 'successful' with their actions will have more children than those who are less successful, meaning that, over time, more successful strategies will come to dominate.

A common debate in the social sciences is then whether unselfish behaviour can sustain itself over time, given that selfish people might be able to take advantage of the unselfishness of the altruists. It's true that a group of unselfish people will likely outperform a group of selfish people, but then won't the unselfish group fall victim to selfishness from within? The answers are (as usual) it depends, but I won't get into the details any more in this particular post (with a 2 hour episode of wipeout on tonight, time for posting is limited!).

Evolutionary dynamics are not the only way for a population makeup to change over time. Imitation works too. The Czech Republic (for example) isn't a capitalist country because it was outbred by capitalist countries, it's capitalist (arguably, at least) because the population decided to imitate what they felt was a more successful method of doing things. At a personal level, people will imitate what they other people doing around them if they feel those people are successful (see also Bubble, Housing).

A third option is migration. If people are able to move from one society to another, their movements will alter the distribution of preference types within each society. A constant migration of unselfish types to an unselfish society might offset a trend towards successful acts of selfishness within that society, for example.

A fourth mechanism is that the people themselves do not change, but their relative strength of influence does. Maybe an unselfish society contains only one selfish person, but if that person uses their unchecked greed as a means to taking control of the whole society, then the society could change dramatically despite nobody changing their particular nature.

No doubt there are other mechanisms by which the makeup of preferences in a society can change over time - I can't think of any at the moment, but feel free to point them out in the comments.

This is all pretty abstract, but the point is that it would theoretically be possible to model or simulate various ways in which a society of people with different preference types might evolve over time, applying different mechanisms by which behaviours might spread or change or change in influence over time. There are folks out there who have undertaken this sort of work, and I'll cover some of their efforts in future posts.