Reading "Sciences of the Artificial" by Herbert Simon a few weeks ago, I was struck by his distinction between normative and descriptive computer models. Basically, normative models explain how things should be, as when we say, "Governments provide the greatest good for the greatest number," in contrast to a descriptive model which would say something like, "Most governments throughout history have been corrupt and have carried out the oppression of the masses."
His point about computer models is that the normative, simplified, idealized sort are easier to make because it's often difficult to find mathematics to describe the very weird and complex things that can happen in reality. But, of course, descriptive models represent reality better and thus - if the logic is sufficiently deeply structured - they are better able to predict the future.
In my mind, any bureaucracy or corporation has the same abstract structure as a computer program, and thus they can also be considered to be models of human behavior. The major difference is that what is written in a corporate charter, the mission statement of an NGO, or the constitution of a country is signed onto and agreed to by a bunch of people, and these people will call the police and put somebody in jail if they don't behave in accordance with the model. We don't tend to punish natural phenomena when they don't abide by Newton's or Einstein's laws (except by ignoring them), and if the stock market doesn't obey the predictions of some wall street quant's models, then it's the model that is considered defective.
This distinction that I see humans as a whole making rests on the understanding that humans are different and separate from nature. If we move towards breaking down that distinction, it brings up the possibility of making our government charters more descriptive, rather than some idealized story about the way things should be. The group who wrote the US Constitution took a large step in this direction by realizing that every individual will tend to seek more power, and then pitting them against each other with checks and balances.
If we look at the US Government today, the largest aspects that aren't accounted for by the Constitution are the lobbyists and the costs of the election process. Without getting too much into how these processes are inefficient, immoral, or whatever, I'd like to merely propose the academic question: what would a more descriptive Constitution look like?
The essential function of lobbyists is to bring information to Congress and to help it form legislation that will help certain groups do their jobs. If the lobbyist is representing a mining company, he or she will probably skew things in ways most individuals find terrifying: increasing the amount of Chromium 6 intake considered to be "healthy," lowering corporate taxes, etc. But if the lobbyist is working on behalf of GreenPeace, most individuals will benefit from a peaceful sense that nature will still be there when next time we get a vacation in 2026 to go stare at it for a couple weeks. When they're working for large, wealthy individuals or organizations, lobbyists also give Congress lots of money.
Essentially, working this into the Constitution would essentially involve a major expansion of the Congressional Research Service. The public policy research arm of the US Congress, its budget last year was about $90 million, compared with approximately $2.8 billion spent on lobbying. Even without adding in normative features like ensuring that somebody is lobbying for public interests, and even if we include currently hidden features like expensive lunches, junkets, and bags of money that indubitably get exchanged, putting this whole process into law would make it obvious to everyone how to come to Washington and participate in the policy debate.
Similarly, codifying the electoral process might carve into stone some pretty annoying and destructive practices, but once they are in stone we can see them clearly and see better whether a candidates act evil because of their personality or because they are simply being warped by the process.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment