I came up with almost exactly this idea several years ago -- I called it augmented representative democracy. I figured that software programmers would immediat

I came up with almost exactly this idea several years ago -- I called it augmented representative democracy. I figured that software programmers would immediately latch onto the concept, because they are familiar with the concept of inheritance. You elect a legislator to vote on your behalf, and in many situations that legislator's vote is the same as your own, yet you retain the right to cast your own vote on any given issue. But there are several objections that I could not overcome: 1) In our free, open, representative-style government, each legislator's voting record is a matter of public record, but each individual citizen's voting record is confidential. If we were to switch to augmented representative democracy, the question arises: Do we make each citizen's voting record public or private? Both options come with potential problems. For instance, if all votes are private, we lose out on transparency, and anonymity tends to embolden people to make some pretty nasty choices. (Exhibit A: Any online forum.) But if all votes are public, it could invite retribution that the average citizen is not equipped to handle. 2) It's hard enough to monitor roughly 535 federal legislators, to make sure they're not taking bribes or kickbacks in exchange for their votes on particular pieces of legislation. Could you imagine if you had to instead monitor all 207.6 million eligible voters? True, it also becomes more difficult to influence a significant number of them through nefarious means. (Suppose you need to bribe five senators to tip the scale in your favor on a particular piece of legislation. So you offer them each $1 million. Now suppose every eligible voter got to weigh in. You'd need to bribe more than 10 million of them, assuming they all voted. And a $1 bribe isn't nearly as attractive as a $1 million bribe.) But on bills where the vote is really, really close ... there is really no viable way to keep everybody honest. 3) A legislator's workload is (or should be) a full-time job. It takes a lot of time to read through bills and understand them. It takes even more time to fully consider its broader implications and its potential unintended consequences. As part of that process, you'll likely have to engage in discourse with fellow legislators; evaluate expert testimony; listen to the concerns of constituents, trade groups, lobbyists, and other organizations; and weigh the potential for the law to be challenged as unconstitutional. All of that takes time, and it's unrealistic to expect every eligible voter to assume that responsibility for the purpose of voting on policy directly. 4) In practice, the cases in which this augmented representative democracy would result in a vote different from the legislator's would be relatively few -- and yet there would be a whole lot of extra effort required to support the system. Basically, you would need an issue where the legislator's vote is different than what the people who elected him would expect (and the legislator, if he wants to be re-elected, is only going to do that sparingly), and you would need a substantial turnout of people willing to overrule him. Considering how few people vote in general elections, that's a tall order. 5) When you break it down, augmented representative democracy is really direct democracy, and not true representative democracy. And one objection to direct democracy -- take it for what it's worth -- is that there's a danger of mob-mentality policy decisions. If you look at some historic decisions, at least here in the U.S., the legislature was a bit ahead of the curve, compared with the population at large. So ... I guess the question is ... could direct democracy have derailed or delayed something like the civil rights legislation of the mid-20th century?