This essay makes a few basic assumptions, and builds upon them. If we don't agree on these assumptions, the rest of the essay may not seem logical. Said assumptions are listed here for easy reference:
At some point in the history of the world, man appeared. Man quickly ran out of usable space on our planet, and became increasingly agressive as a result of perceived crowding. Humans started creating governments to minimize the unpredictability resulting from agressive humans in close proximity to one another. Philosophers tried to justify the authority of goverment by inventing societal rights. Philosophers decided that government was indeed created by humans, as the study of History readily confirms. A computer programmer picked up the last sentence and decided to run with it. It appears the programmer with the sentence ran straight into your field of vision.
Like all creatures (and indeed, all objects) in the world, humans have a natural right to exist. A person also has the right to affect himself in any way he chooses, with the obvious caveat that every other person has an analogous right, creating practical limitations to a person's ability to exercise perceived rights. In distinguishing perceived rights from actual rights, the golden rule relies on calculation of the radius of the candidate right from all affected parties: the closest one wins. One may be tempted to apply a more elaborate Mathematical model to the problem in order to solve it more "fairly," potentially resulting in a determination of "percentile ownership" of the candidate right by several parties; such models are fundamentally wrong, however, since shared authority by its very nature is anything but natural. In any case, precise calculation is rarely necessary in practice, as simple logic will generally reveal differences in orders of magnitude between various radii.
Since humans require the services of non-human objects in order to exist on this planet, we're going to have to give people access to these objects in order for the basic right of existance to be meaningful. One possible mechanism for access capitalizes on the observation that a person is at radius zero from himself by definition. Let's invent an attribute called ownership, where being owned by person X simply means that we're defining the radius from person X to that object to be zero. The only other available mechanism (namely, none at all) relies on the lack of greed, which is highly unlikely in a crowded world. Therefore, we're forced to accept the concept of ownership. (This is exactly what Socialism gets wrong, BTW.)
Now that we've agreed that people need to own property due to the fundamental scarcity of resources, the obvious question surfaces: how do we make initial assignments? For purposes of our discussion, let's assume that at some arbitrary point in time, our planet's resources are distributed in some fair way among people. (The mechanics are irrelevant, as long as they lead to an acceptable result: namely, that nobody is left wishing he were in somebody else's shoes.)
Getting the initial condition setup only solves the immediate problem explicitly, relying on people to police themselves in order to prevent the situation from decaying. Again, due to scarcity of resources, we can predict that people are not likely to respect the rights of others, with those who build the most at risk for losing the most if their rights are violated. It should come as no surprise, then, that it's these people in particular who stand to profit the most from an organized governmental infrastructure. However, an organized police force offers potential benefits to all involved, since it provides those who haven't built much with an insurance policy against bandits (defined as people who reject our artificial definition of ownership), which in turn makes building a more attractive option.
The preceeding paragraph sounds great in theory, but there's a pair of minor technical problems, here:
In light of this problem, we need to declare our property rights as some sort of pseudo-natural right, which a government can now justify protecting. What guarantee do we have, though, that the government itself won't become a bandit?
There are two fundamental methods of solving the problem. All real-world attempts are just hybrids of these two:
Since a government is made up of people, and we've already established that we can't trust people not to have greed when resources are finite, it should come as no surprise that there are no guarantees. The best we can do is engineer a government system that is likely to do its job without becoming a bandit itself, and allow people to opt out.
Now, since our government isn't guaranteed to preserve your natural rights, it has no natural right to rule over you. That means you have a natural right to take your share of resources and secede, if the urge strikes you. (It's meaningless to talk about having a natural right to secede only if you're deprived of your rights, since there's no competent authority to judge for you. Therefore, we have to assume your judgement to be correct; we can safely assume that you won't secede if you feel your rights are being protected, since in such a case you have nothing to gain by seceding, but you do make yourself vulnerable to bandits, so you have everything to lose. We can therefore count on you to judge to the best of your ability and to prefer to err on the side of staying than on the side of seceding, and so any government claiming you as a subject has a moral obligation to abide by your verdict. A government that violates your verdict is in direct violation of your natural rights, and is corrupt, by extension.)
A government that makes the correct decisions is optimal. A government that tries to make the correct decisions is asymptotically optimal. A government that prefers to make incorrect decisions is suboptimal by design. Corruption means being suboptimal by design. To illustrate the three terms with a simple example, suppose somebody wants to steal something from you. An optimal government puts 100% into preventing the theft. An asymptotically optimal government tries to put 100% into preventing the theft. A corrupt government doesn't necessarily try to prevent the theft at all, and may even aid it.
Our second assumption (at the top of the page) established that a person has a natural right to be his own dictator. This form of government is incorruptible, and is therefore the ideal. Our goal, then, should be to put as many decisions as possible within the jurisdiction of the individual's own Dictatorship. Any government that doesn't set this as a goal is not asymptotically optimal. Put another way, an asymptotically optimal government tries to be as small as possible.
When two (or more) people are about the same radius from a resource, contention can occur. Contention is defined as disagreement about the shortest radius. Without an external government, contention is resolved using war. (War is defined as disregard of the shortest radius rule.) In the interests of preventing war, we create an external government. The external government can't be a dictatorship because conflicts of interest make it virtually impossible to find the perfect dictator, so we create a Democracy as a last resort. This government has a moral obligation to stay as small as possible, knowing that in the best case scenario it's only asymptotically optimal (and in the worst case, much worse). Therefore, a non-corrupt Democracy will make every effort possible to regulate as little as it can get away with.
Pretend our world has five people. Person A invents a telephone, and cooperates with person B to run a line between their properties. Person E now wants to join the network, but the line will have to run through the property of either person C or person D, neither of whom wants anything to do with this newfangled network. What do we do? The correct decision is "nothing," which is exactly what an asymptotically optimal government will attempt to do. In practice, though, this tofu problem is used by people A, B, and E to outvote people C and D; the result is a corrupt expansion of the government to create a concept known as "eminent domain." (It's a tofu problem because it's not a real problem at all. People A, B, and E have no right to run lines through the properties of people C or D. If that means people A, B, and E can't have a conference call, that's the harsh (but entirely fair) reality. (Remember, if geographic proximity to his friends were of any concern to person E, he would've wished to have been in the shoes of person C or D back when resources were initially assigned.) The expansion is corrupt because the clear intent is to make an incorrect decision: theft of the natural rights of people C and D under "eminent domain," or the mistaken assumption that those rights never really belonged to them in the first place, and therefore are okay to steal from them now for convenience.) At this point, the correct thing for people C and D to do is to secede from the corrupt government and form their own government. The correct reaction for government ABE would be to respect the natural right of people C and D not to be governed, and in particular not to be governed by the corrupt government of people A, B, and E. In practice, of course, corrupt government ABE will become the aggressor by starting a war with the newly formed government CD, after mislabelling the secession of people C and D with a newly coined (and entirely artificial) term: rebellion. From here, there are five possibilities:
Unfortunately, only the first possibility above ends right, and even if CD wins one war of survival against an aggressor ABE, there's no guarantee that after ABE regroups it won't become a bandit again. Perhaps more unfortunately, though, notice the deep rift between the dry facts and the media language commonly used to "describe" each case: a dear friend of mine (a very wise man whose presence I miss every day of my life) who survived Hitler's rather severe dislike for live Jews some 70 years ago expertly calls the phenomenon "machbeset milim," which literally translates into "word laundry." It's the science of taking the correct words to describe a harsh reality, running them through the washing machine, and then taking them out looking spanking clean. This is normally regarded as an inherent strength of "rich" languages. In fact, the "richness" of a language can easily be defined as the ease of doing "word laundry" with it. Clearly, if your goal is to understand facts, your best bet is to take a page from the Mathematician's rulebook: keep your language as poor (and precise) as possible, and avoid washing machines like the plague. (Richer isn't always better, eh?)