But the quest itself is not only doomed to failure, but a great way to ensure you have a bad government.
It reminds me of a phrase I came up with in college "The perfect screwdriver is useless if all you have is nails." Perfection requires that you understand the purpose of the device and also have the proper materials.
The problems with a perfect government are the materials. A perfect government wouldn't make mistakes. But government is made of people. People make mistakes.
For this example, I am going to use a sword, not the screwdriver I thought of back in college. Consider designing the perfect melee weapon. You want a nice sharp point, plus a sharp edge. It needs to be made of materials that can attain and maintain these factors. It should be thin and easy to slice or pierce anything, yet strong and able to handle hard use. Lets assume you are an incredible craftsman, capable of doing miracles.
Now try to build it out of wood. Because you are an incredible craftsman, you build a pretty good sword. Oh, it's not as nice as a steel one, but it's pretty sharp and pointy.
Now someone with a crappy club comes along and they break your wooden sword with a single blow. Your sharp, thin wooden blade was a bad idea. You need to consider the materials you have when designing your device. Same thing with Government.
People are error prone. We make mistakes. We have the capacity for evil. We are ignorant. Sometimes we are stupid. You need to design a government that takes all of these things into account.
It is no where near the same kind of government that you would design for holy angelic beings. But such a group wouldn't need government in the first place.
As such, the very idea of perfection becomes an impediment. By it's very nature it encourages you to forget about the fallible, imperfect humans that will have to run the government.
Instead, you need to think about building the Best Possible Government, rather than the Perfect Government. You need to concentrate on the errors themselves and how to deal with them.
Specifically, you need mechanisms to:
- Look for possible human errors.
- Confirm the issues are in fact errors.
- Admit them to the general population (or they will resist step 4).
- Fix the errors - preferably without introducing new errors.