Thursday, 28 April 2011

Democracy ... or: My Quaker View on the Referendum

Here it is, the obligatory post on the rapidly-approaching referendum on electoral systems. On May 5th, the UK goes to the polls to decide whether to keep the current, first-past-the-post system, or adopt the really-minor-adjustment Alternative Vote system. Obvious comments include the fact that, given the spectrum of possible systems to choose, AV is a tiny change; the result of this referendum is likely to affect future potential for further reform (views differ on how); short-term effect on individual political parties will drive voting in many cases; and there's so much to say about the actual maths of the systems, and about the lies, misleading statements, and irrelevant arguments made by both campaigns.

I'm not going to do that. You'll know what I support, certainly, and some of the reasons why, but I don't feel the need to re-hash the arguments from democratic principles. I'm voting Yes, and I'll advise anyone who asks to vote Yes, and I think the argument is fine on democratic principles - I'd even vote Yes if I thought it would help the Conservatives and help no-one else, although I'd hate that outcome, because I feel the representation is better with AV. That isn't the point of this post, though. In this post, I'm explaining why I feel that Quaker values and teachings also point to voting Yes. I don't claim to represent Quaker views in general - there's probably no subject anyone could do that for - this is just my understanding, and an explanation of how my understandings support a Yes vote.

For those reading this who are themselves Quakers, or that know Quaker practices well, this might seem odd on first thought; I hope it doesn't, and doubt it will in most cases, but I recognise the possibility. This is because, within our own organisations, Meetings, and so forth, we don't vote. The explanations as to why are varied, but the general gist of it is that we don't think decisions should be made based on what people want - some would say we're seeking God's will, others would say we're looking for the 'best' or 'most right' decision, rather than the most popular, but the general common point, to me, is that popularity doesn't matter in itself. What matters is getting the right outcome. Now, I could write several long-ish posts on Quaker decision making, especially if I include details of what I think about it and why. That's not what we're going for here.

What is worth clarifying, for those who are confused by our non-voting, is that it is still democratic. It's also worth saying that Quakers, in general, vote in elections. I have no evidence, but I'd certainly guess that Quaker turnout is higher, proportionately, than the population in general. Some of you may be stuck on the idea of Quaker decision making being democratic without voting; maybe I'll write more about that another time, but for now let's just say "voting is neither necessary, nor sufficient, for the practice of democracy", accept that you may disagree, and push the point aside for now.

Quakers, in general, including myself, are very keen on democracy in the public realm; our own decision-making practices would be very hard to apply to a group the size of the population of the UK (without an intermediate layer of representation, at least), even if that whole population were familiar and comfortable with our practices. Thus, no-one is about to suggest that such practices are applicable. Voting is the way to do it, and there is no general objection to the traditional practice of representative democracy - as opposed to direct democracy, where everyone votes on every decision. It works, and is workable in large populations. There are problems, of course, and I don't imagine anyone claims any system is perfect.

Much has been said, by very learned figures, about the abstract characteristics of a perfect system for electing such representatives - it should represent views accurately, for instance, and appoint representatives in proportion to the voters supporting their policies (as represented via political parties), it should allow people to be elected without party affiliation, and it should deliver quick, clear results, and so on. No system has anywhere near all of those characteristics, and if there's a relevance in there to Quaker teachings, it's too deep for me to dredge today. So, on to the Quaker principles that are relevant, and how:
  • Equality. Everyone's voice should be heard, and the same fundamental weight attached to it. Now, there's no realistic way to achieve that when electing representatives. Even full list-based proportional representation doesn't quite manage it. However, AV manages it better than FPTP - simply expressing a clear preference for a party who have no hope of winning in your constituency does not render you voiceless in the decision on who will be MP under AV, while it does so under FPTP. Of course the voice is still heard - the number of votes is reported - but it is not given weight. Under AV, every (non-exhausted) vote that isn't for a given candidate acts to prevent that candidate from winning, even if it is for a candidate trailing in a distant fifth place. If the leading candidate is short of majority by 50, and there are over 100 votes for the last-place party, then those votes are preventing that candidate from winning, under AV - if they had not voted, the majority mark would be over 50 votes lower, and that leading candidate would slide past it. Such votes under FPTP do nothing to affect the result.
  • Truth. This is possibly the oldest of Quaker principles - a favourite quote of early Quakers was, if I recall correctly, "let your Aye be Aye and your Nay be Nay". No-one's claiming that Quakers never lie, but the goal is to be open and honest, especially about things that matter. Consider, then, the implicit question on the FPTP general election ballot paper: "Who do you want to be the next MP?" Okay, so it could be read as "Who do you think should be the next MP" - that's up to individual conscience. The distinction isn't important to my point here. The point is that people often have to choose between answering honestly (choosing their honest first choice) and actually affecting the outcome (as per my point on equality above).
Some readers may notice that I've tied these to two of the four traditional Quaker testimonies. For those who don't know what I'm talking about, I'll cover that in more detail another time. The other two are Peace and Simplicity. Peace is difficult to tie in, but Simplicity is an interesting one. AV is certainly less 'simple' than FPTP, but Simplicity is possibly the most subtle of the testimonies, open to interpretation and debate. I would argue that, in pursuing simplicity, one must always have regard for the purposes of things, and lean towards the simpler solutions - no unnecessary complexity. FPTP is sufficiently poor at fulfilling democratic purpose that additional complexity is warranted.

What about going beyond the testimonies? Is there anything else Quaker teachings can tell us about the pros and cons of different electoral systems? Well, there are several widespread Quaker concerns, often with links to the testimonies, that may be relevant:
  • Social justice
  • Environmental concern and awareness
  • Conflict resolution
I can't say that AV has immediate relevance to any or all of these. Arguments can be made based on the current electoral landscape, certainly, that social justice is served by a change to the voting system ;many view the Conservatives as the major party with least regard to social justice, and they would be the biggest losers under most models of voting under AV. This largely depends on opinion and models, however, and is not (as far as I have seen) an inherent benefit of the Alternative Vote system.

There is a link to the concept of conflict resolution, however. Where a majority of parties don't agree on a preferred outcome, sometimes you can find an outcome which is at least acceptable to all; in an election, that is unlikely, but you will usually find that there's a compromise that most will agree on. Consider 10 friends going out for a meal. 3 would like curry, 2 each would like Italian, Chinese and Thai, and one wants to go to a kebab shop. It may be that all 7 who did not prefer Curry would really rather not have curry, and would end up ordering steak and chips even though they are in a curry house. Now, it might be that all of them would agree on going to somewhere like Wetherspoons, where there's a wide range of styles of food; AV doesn't give that sort of option. However, suppose the kebab fan would choose Thai as a second preference. You now have 3 each for curry and Thai, and 2 each for Italian and Chinese. Under the proposed implementation of AV, you'd pick one of Italian and Chinese randomly, and eliminate it. The two who like Chine both like Thai as a second option; this would leave 5 for Thai, 3 for curry, and 2 for Italian. A clear lead for Thai, but it's still not a majority - you need 6 out of 10 for that, as a majority must be more than 50%. So, the Italian fans are definitely out of luck. Now, if both prefer curry over Thai, we have a dead heat, and the system would select randomly. It's important to note that, with the numbers involved in public elections, ties are very unlikely in that instance. However, suppose one has no preference between Thai and curry; whether the remaining Italian fan prefers Thai or curry, Thai now has a majority of those who express a preference between the remaining options, so Thai wins. If the remaining vote went to Thai, it's 6 for Thai, 3 for curry, out of 9 votes; if they went to curry, it's 6-4 out of 10, and Thai still has a majority. That's a better resolution to the conflict than forcing all to have curry just because the plurality chose it.

In an ideal situation, to resolve a conflict, you'd try to find something most acceptable to all - this has been represented in democratic theory using something called the Condorcet Criterion. This states that the winner must be a candidate who would beat all others in a one-on-one vote. AV doesn't do that, but there aren't many systems that do, and all are more complex. However, AV will not elect a Condorcet loser, a candidate who would lose in a one-to-one vote with each and every other candidate, while it's not even an outlandish possibility under FPTP. That's because of vote splitting, which AV reduces (or, ideally, eliminates); to give a concrete example, consider a vote between Liberal Democrats, Labour, and Conservatives. At least before the current coalition, both Liberal Democrats and Labour are seen as left-wing parties, supporting and promoting at least some left-wing ideals; Conservative are right-wing. The left-wing vote would be split between Labour and the Liberal democrats, leaving the Conservatives only needing to get more votes than one of them. Under AV, one might expect most Liberal Democrat voters to choose Labour as their second preference, and vice-versa, leading to whichever of the two had the most first preferences as a likely winner, in the situation where Conservatives get a plurality, but not a majority, of first-preferences. In such a circumstance, the Conservative candidate would most likely be a Condorcet loser, but would win happily under FPTP, but not under AV. The Condorcet loser is obviously a bad outcome under most ideals of conflict resolution, and thus AV has advantages over FPTP under those ideals.

There's likely lots more that could be said, but I hope I've given a good outline as to how Quaker ideals and teachings point to supporting AV over FPTP. These arguments will mean the most to Quakers, of course, but hopefully others can see the value of the Quaker principles I've discussed, and can see their value as arguments in the referendum debate even when considered on entirely secular grounds.

    No comments:

    Post a Comment