I like your thoughts, and am particularly charmed by the random selection parliament. But I don't agree that no skills are required, and as such I am also no fan of the judge system, which we don't have here. Recognizing that I might be culturally biased in that, and in thinking about democracy in general, let me first give my own perspectives on all of the systems described above.
To talk about these systems, I will also take the position that politics is about the allotment of resources, and democracy is the ideal of each person having an equal say in that politics.
1. Direct DemocracyThis system, ideally the settling of each and every political decision through a direct vote, has some difficulties which can be overcome. But also some flaws which probably can't. The difficulty of course is access: It relies heavy on all people having access to the relevant sources of information to come to a decision, and then access to whatever medium you use to cast the votes. If you use the internet for example, in an e-democracy, then everybody must have equal access to the internet. Well, that can be done, and it would be very positive.
But there are 2 flaws I see that make it problematic: First, there's influence, as Fiona mentioned. How can you ensure people become the well-informed experts on everything, when they can be bombarded with biased information, by whatever broadcasters have the most wealth? Information distribution is not equal; and the influence of (biased) information matters when it is not your own life that is obviously affected by it. I add that constraint, because I think the effect of propaganda is a little less when people are directly affected. But then it is still possible to misrepresent or re-frame the issue, if you have a massive ton of exposure compared to the opposition. Which is why the elections in America are about whoever has the most money: You can't buy votes, but you can buy influence.
A second problem that I've seen in the referendums we've had here is the actual formation of the question. Who gets to do that? Referendums are rare here, but we had two on the European Constitution a few years ago. And they were voted against twice. The government clearly wanted it, and in the end, they adopted it anyway, "with a different title", and without referendum. The first referendum was problematic because people widely complained that they didn't understand what they were supposed to vote on: The issue was too difficult and complicated. It was a whole set of laws and consequences which were phrased badly, and too extensively, and they were also too abstract for most folk. The second time around, the government had spent several hundreds of millions of euros to distribute the information in an easier manner; in understandable language and bullet points. And still the people rejected it. For me personally, the proposal had some fundamentally good provisions, but also some fundamentally bad ones, which led me to vote against. In this case, the proposal for the European Constitution was popularly rejected by a small majority, but it made clear to me that you need to think very carefully about the question itself, and how compositions of issues could be used to smuggle in unpopular measures. On the other hand, sometimes a compromised proposal might be needed to get a majority to support it.
But then, I suppose these are problems that all vote-based democratic processes suffer from.
2. Proportional RepresentationThis is the main system we have in the Netherlands. We currently have 10 parties holding seats in the second chamber. Which means that there are 10 party programs that have a voice, and that have a chance to form a majority government with each other. But that is difficult: these party programs are often not very compatible. We saw that after the last elections, when the National-Socialists got 15% of the vote, which made them a serious player in the negotiations, while many of Wilder's top priorities are fundamentally incompatible with most of the other parties. And so we now find our government in a unique position: It is formed by a minority of Christian Democrats and Neo-Liberals, with the "toleration support" of the National Socialists, which means they have agreed to some concessions between the 3, but they supply no ministers to the government.
A big problem with this system of coalitions is, indeed as Fiona mentioned, that it is hard to recognize whether or not your elected party delivered on their promises. And similarly, when the policy of a coalition has created problems, it's proven difficult to hold the correct parties responsible: The Right over the last decade has won major support by discrediting the Left as causing the "failures of the Purple Cabinets". Which is funny, because they were IN that cabinet, and now they are winning major support by blaming it. And by funny, I mean frustrating.
Perhaps it's not a bad idea to test the coalition manifest through a referendum or election (with multiple possible coalitions to choose from?). But that doesn't take away the problems, I think.
What I like about the proportional system is that I can vote for relatively nuanced parties across the entire spectrum, which all have a fairly realistic chance to get into the government. Which as far as I can tell is not going to happen in:
3. First Past The Post It seems to me that these systems tend to gravitate to having only 2 major parties that have any realistic shot at the government, meaning that a vote for a third party is literally thrown away: A vote for an independent lefty party in the US or UK means you didn't vote for the slightly more lefty party out of the 2 major ones (i.e. Democrats and Labour resp.), which would have had a chance of getting into government. Thus, dissenting parties are discouraged: The less parties at your end of the spectrum, the more will vote for the remaining one, whatever it is. And when there's only 2 effective parties left, then it makes sense to move as close together as possible, because that way you get the maximum number of votes. Everyone to the outside of that position has no realistic choice but to choose the one closest to them. A bit like when you are trying to guess an integer with 2 people, and whoever is closest gets the prize. The first one has to make a really good guess. But for the second one, it makes no sense to guess at all: he just needs to guess whether the opponent's guess is too high or too low and guess the integer right next to it. Which means in politics that the manifestos are necessarily similar, with some exaggerated minor differences. Sticking to too many points too far from the other one manifest means that all the people in the middle between those two positions might go to the other party, which means you don't get to rule at all. The parties in a First Past The Post system are necessarily compromised to get (just) over 50% of the vote, through self-censorship.
What happened in the last elections in the UK is an exceptional situation. Suddenly, there's a third "guesser" in the game, and it became a reduced PR democracy. Reduced, because the new kid was in a sad position: It was quite immediately bullied into accepting an abysmal coalition agreement, which, as I understand, tossed overboard all of the main election promises. That will probably be used against them in smashing their credibility at the next elections. Had they not caved, they may have been held responsible for sabotaging government formation, which was a pretty heavy consideration on their part, from the reports I heard here.
On the other hand, now that it has been shown that a third party ís realistic in the UK, perhaps a new dynamic will evolve, where the 3 parties re-position themselves on the spectrum, with a little less self-censoring. Unlikely, because it seems Cameron is quite successfully riding the Lib Dem horse, in an accelerated fashion, towards Neo-Liberal Tory policies, with the added option of blaming the horse for bad results.
By the way, if the UK situation is an example of a FPTP system moving towards PR, there is a current in our PR system going to blame Leftists for voting for multiple different small factions, so that the Right could more easily form a government last time. But our PR tradition is stronger than that, and the current was not that strong. We'll have to see how that (American?) influence develops next elections...
4. Mixed SystemsThe Swiss have a system of a general election and many weekend referendums. I didn't know. And I am not versed in how it works. I also heard of the Nordic inclination towards referendums. When are things decided through referendum and what is left to the gov? Clearly, our government doesn't take referendums that seriously, as I discussed earlier. So I am quite curious to learn how those are integrated.
We do have the ability for petitions to force parliament to discuss an issue. I am not so convinced about the force of that... Fiona calls it "continued engagement", but these things are single issue, plus signing a petition is not the same as engagement in any meaningful way. In fact, I've never heard of a successful petitioning action that made any difference. Though I've heard of plenty petitioning lists being offered to the parliament, they are often for issues that the government already knows about; that has already been protested for/against; and that they have decided not to care about. I don't see how forcing a (re-)debate is going to realistically contribute to a cause the government is against. If anything, the whole construction seems to serve more like a pacifier, quite literally: You don't agree with us? Well, get 20,000 signatures. You got the signatures? Okay, we'll discuss the matter again in a month or so. After the debate: Nope, we still don't agree with you. You don't agree with us? <repeat>.
5. Radical systems5a. Random Selection Participation
As said before: I like the idea. The problem of influence remains, but it would have to be continuous and across the entire population, so that they come into the parliament with the right wrong ideas. Also: what about buying members of parliament? Would that have consequences for the temporary parliament?
Convenient link to corruption thread..5b. Federated random selection participation
The same as above, except broken up into syndicates/federations, each dealing with smaller portions of the political task. Same problems hence exist, but, firstly, there's a lot more organizations to influence by malicious forces, and the participants would be more directly involved with the body in which they represent. Now the test is: if politics is the allotment of communal means, then how do those means reach other factions? That seems quite unlikely, if they are like tribes. There may be multiple levels of these factions, each employing voluntary association, and where necessary random selection participation, including one at a national level, if the nation matters; even one or more European ones above that; perhaps even global. Let me ask: is there a reason the politics should be restricted to national levels and necessarily be across all aspects and all professions of society?
