Proportional representation : yes please, maybe

Recently I decided to design a new political system based on a truth-finding system that's proven to actually work : science. Here I'll give a the brief summary for those who don't want to read the full thing, and some afterthoughts.


How science works

The essence of the scientific method is that it emphasises examination of the evidence and the arguments presented, while downplaying the importance of who to trust or how many people already believe a counter-argument. It uses extended cognition to carefully examine as many possible relevant data points as possible in a multitude of different ways, which also helps keep things relatively calm and unemotional. Individual scientists are more-or-less free to say whatever they like in most venues, but anonymous peer review means that's not true when it comes to publications. Anonymity helps encourage skeptical inquiry, promoting competition without undue hostility, while editorial oversight ensures that both author and reviewer play by the rules. Dogmatic groupthink is avoided by the need to publish interesting, unexpected results, while peer review acts to restrict people from publishing whatever the hell nonsense they dreamed up while smoking something illegal.

It's a clever balancing act. It doesn't always work on individual cases, but the system does very well indeed on a large scale. The system both selects people who actually have some degree of intelligence in the first place, and, more importantly, provides them with tools to reach better conclusions than they might as mere individuals.

A key aspect of research is that it's always changing - mistakes occur and are tolerated at almost all levels, including factual errors, misinterpretations, and methodological shortcomings. All are subject to continuous improvement, though a surprising number of people seem to think "mistake" is the same thing as "being an incompetent twit". Hint : it isn't. The only kinds of mistakes that are very much not tolerated are those of deliberate deception. In this way, the system has a peculiar kind of pseudo-stability, always being close to the most rational conclusion currently possible (I call this the efficient consensus hypothesis). It rarely if ever reaches "the" truth, if there is such a thing; it just does the best it can possibly do. Conclusions are thoroughly tested by the community at large, making it extremely difficult to fool. Though inherently unstable, it would seem to be remarkably robust.


Lessons for politicians

My suggestion to improve politics was to take this system and transplant it wholesale into the political arena. Have party allegiance represent a broad set of common values, but almost entirely castrate the capacity for party machinery to dictate what politicians say publically - let them be free to express their individual views, otherwise no-one will trust them (the scientific consensus emerges because the system not only tolerates but actively encourages diverse views - it's not much use if you pre-select everyone who already agrees with you). Have their proposed laws be subject to a process analogous to peer review, fostering cross-party co-operation and anonymous scrutiny. Allow MPs of all parties to propose legislation and totally eliminate the government's ability to veto it - everything proposed must be voted on, though sometimes only by selected commissions (chosen by an independent, external council) and not the whole House. Fund the whole political system publically, forbidding donations of any kind. Impose much harsher sanctions for those found to be guilty of deceipt, but encourage trials of proposed laws whenever possible : that is, tolerate mistakes but not willful deception.

Now I have, in this little exercise, almost completely ignored the role of money. This is a deliberate choice, as "how to allocate money" and "what to do with the money available" are not necessarily the same type of problem. The latter can and should be done by specialists, whereas the former needs generalists to judge which department needs what. I do not propose any way to deal with that. Nor do I offer any solution to the problem of using pointless ministerial positions as loyalty bribes to hold MPs hostage to party leaders; though I suggest that either the number of minsters be cut drastically, or every single MP be given some official unique position. And perhaps most importantly, I haven't tackled the thorny issue of local government, nor did I say much about the electoral system.

My point was that we already have an established procedure for getting to the most rational viewpoint possible. This works well not only in compiling encyclopedias, but even in the messy world of front-line research. The system makes a strength of diversity of views, rather than trying to stamp them out, and everyone gets a fair say and genuine influence, rather than the winner-take-all approach of contemporary politics. It's not especially necessary that political reform follows the precise structure I describe, only that the system make use of the general principles of consensus through diversity of genuinely representative voices.


Afterthoughts

I've always been in favour of emphasising the "democratic" aspect of representative democracy. If I vote for a party based on their policies, I expect those policies to be implemented pretty much as stated. Having them compromised by negotiation with others would seem to defeat the purpose of elections. Of course, we do need the "representative" aspect as well, because their are plenty of fine details for which I'm more than happy to concede that expert examination is required, not to mention a myriad of points that few people are at all interested in, but the main policies ought to be the voter's choice.

And yet....

The scientific system is neither representative nor democratic, but inasmuch as it's either, it's infinitely more emphatic on the "representative" aspect. As far as we decide on what's correct, no-one makes a mandatory decision imposed on the others - the consensus view just emerges after prolonged scrutiny. The consensus is not all-encompassing either, but only those parts that the majority can agree on; there are plenty of problems for which there are about as many solutions as there are researchers tackling them. So the system of getting to the most sensible decision would seem inevitably at odds with the need to give people a choice, meaning that in such a system, voters would get less say in policy than they do currently.

This leads me to the much more popular idea of proportional representation. Here's the current UK Parliament :

I've made the Liberal Democrats pink just so they're easier to distinguish from the Scottish National Party.

And here's what it would look like (maybe) with proportional representation :

Quite different. The Tories remain the largest party, but would not have a majority. The SNP would shrink drastically and the Lib Dems expand considerably. In most UK elections, it would seem that a hung Parliament would result and we'd get coalition governments every time. This is good news for anyone wanting a more consensus-based approach, but bad news for anyone wanting more direct democratic choice in decision-making.

I'm not yet ready to commit to saying that the system should be a simple proportionality, but I am more than happy to say, "the current one ain't right". It ought to at least be roughly proportional, if not exactly, but it isn't even that. Overall, 52% voted for pro-Remain parties, yet the Tories have a thumping majority. In Scotland, the SNP won 45% of the votes and 81% of the seats, whereas Labour had 19% of the votes but barely 2% of the seats. This is getting silly. Is it really a democratic system when national issues are decided by local elections, and the results are so starkly at odds with how people actually voted ? Arguably not, meaning that the whole advantage of giving people a more direct say in decisions is negated. If a party doesn't need a majority of votes to win a majority of seats, then most people aren't getting any meaningful say at all. What's the point of allowing a plurality of voters such extreme control over the majority ? In such a situation, proportional representation starts to look a lot more attractive.

I should add that until very recently there was a sort-of temporal balance to the whole thing. There was a fair-ish chance that any damage done by one party could be undone by the next, keeping the system from lurching too far in any direction : even if a party lost, it could at least claim to be mitigating the worst excesses of the others by dragging them in their own direction in order to persuade voters. The success of the centrist incarnations of Labour under Blair and the Conservatives under Cameron are both testament to this.

But a new and altogether grimmer political era has dawned, one with distinctly authoritarian tendencies. The Opposition have been out of power for a decade and are likely to be out for another five years or more. So that aspect of the system too appears to be broken. The argument for PR, or at least a system that's more proportional than the current, is strengthened further.

Yet there are major caveats. First, there is absolutely no guarantee that people would vote the same way under a PR system, so the actual resulting Parliament could be completely different. Secondly, while hung Parliaments force the necessity of compromise, more is needed to actually facilitate how this happens in a sensible way : you can't just bang a group of radically polarised politicians together and lock them until they agree, unless you just want to scrape their mangled remains off the walls. Thirdly, PR doesn't prohibit any party from winning a majority, though it might make it harder. More is needed to ensure that democracy is prevented from falling into a tyranny by majority.

(As an interesting side note : the SNP did incredibly well under FPTP, whereas the Brexit party's share went up and down like a yo-yo, finally collapsing into nothing. The Lib Dems have seen similar trials and tribulations : left, right, or centre, how you campaign under different systems really does matter.)

So PR, or something similar, might be necessary but not sufficient. Politicians need to learn to work together, to both compete and collaborate without falling into petty tribal disputes or pointless unanimity of opinions. This is not asking the impossible - far from it. We already have such a system, and it's used routinely. It's true that it's a lot less interesting to watch than the theatrical displays of politicians bragging about their metaphorical penis size that passes for modern political debate, but I, for one, am sick of living in interesting times. Yes, ideally, I would rather have more direct democracy - but this manifestly doesn't work. So if it's a choice between a representative, cooperative, consensus-based, functional system that limits voter choice but gives everyone some degree of influence, or a dysfunctional, unrepresentative, winner-takes all approach which gives voters plenty of direct control, I have to say that overall I'd prefer the former. Policies that actually make sense but a bit less personal control ? Sounds good to me, thanks.