Clinical research offers support of a pet theory of mine
I have long had this pet theory that society would be better off if our legislative bodies were generally composed of common citizens chosen by lot. In such a system, the executive branch would consist of people who rise through the ranks of managing successively higher levels of government, just as modern executives in the private sector generally attain their position. But the legislative branch would be a large body--probably at least 1000 people so statistics almost guarantee you'll get a random sample representative of the whole population--that would function much like a board of directors that wields the ultimate authority to dictate what the executive branch should be trying to accomplish.
Simple observation leads me to believe our system of elected representatives pretty much guarantees that we will be governed by liars and hypocrites whose dominant attribute is the ability to say whatever they believe people want to hear and sound really sincere at it. Government by lot would no doubt produce a number of completely incompetent people (and perhaps there would be the option for popular election or some other method to cull 10% or so from those randomly selected to weed out the truly crazy), but with a large enough sample, the competent people would almost certainly far outnumber the incompetent and allow the incompetent to be steered along. Besides, majority decisions would not require everybody be on board, and a majority requirement would prevent small groups of lunatics from accomplishing anything--unlike our current system, it would appear.
And so, with extreme satisfaction, I read this article in The Economist discussing recently published psychological experiments about the effects of power on people's moral compasses. The study "primed" people for different states of mind--either a state of mind of one who has earned and deserves power, a state of mind of somebody who does not have power and does not deserve it, and a state of mind of somebody who received power fully aware that they did not deserve it. This is accomplished by having people randomly formed into groups and assigned to write about times in their life when they experienced a situation that made them feel the way the researchers were intending them to feel--a method that has been used time and time again in psychological studies and shown to have significant validity.
After being put into the appropriate frame of mind, the subjects were asked questions or put in situations designed to tease out the subjects' feelings about ethical behavior as applied to themselves and others. It turns out those who were in a frame of mind reflecting earned power held other people to a significantly higher standard of ethical behavior than they held themselves. Those who felt they had little power generally held themselves to roughly equal ethical standards as they applied to others. And the group that was thinking of a time they'd been given power that they did not deserve held themselves to a significantly higher ethical standard than they applied to others.
It is not hard to see the parallels with our system of government. People who campaign and win an election no doubt tend to feel that they have earned their position of power. As such, psychologists would conclude, regardless of other personal characteristics, that they are likely to feel entitled to cheat the system and play by a different set of rules than apply to "everybody else." On the other hand, people who are selected by lot for a position of power would not merely be likely to view themselves as equal to others, they would in fact be likely to hold themselves to a higher moral standard than "everybody else."
Combining the selection method with reasonably high salaries (i.e. keep the current salary/benefit structure, or even raise it a little bit) would increase the "undeserving" feeling for those chosen and would likely increase the effect. Plus high salaries (and the promise of a future pension for service completed proficiently) would help insulate the legislators from the effects of being bought by corrupt parties.
In the comments with the article, there is an interesting comment discussing George Washington. Apparently Washington's personal diary and other comments recorded at the time indicate Washington felt completely undeserving of his appointment to lead the colonial army. His career prior to that point had been unremarkable and his appointment was largely for political reasons. As such, the research would indicate he would be more likely hold himself to a higher ethical standard than those around him, and be uninterested in seizing further power through any means available. Indeed, that seems to be exactly what happened.
But while Washington's intentions, and those of many other founding fathers, may have been good, the system they established has turned out to be one that does not propagate leadership by those humble enough to use it responsibly. It would appear that the best way to achieve a system of government "of, by, and for the people" is not through direct election, but by lottery.