Yesterday, I gave a talk in Cambridge for Invisible College, a week-long residential programme for 18-to-22-year-olds hosted by Works in Progress magazine.1 Some of the instructors this year included Pieter Garicano on European stagnation, Anton Howes on the Industrial Revolution, and Eugenie Reich on detecting scientific fraud.
The students are incredibly sharp, which reminded me of the proviso I gave about my “teaching” experience in my Notes on Taiwan:
Insofar as I can defend my presence there at all, I’m reminded of Steven Pinker’s description of the purpose of nonfiction writing: Write not because you think you are smarter than your audience, but because there is something you have seen that, for whatever reason, they have not yet come across, and you wish to share it with them.
In any case, I decided to talk about proportional representation voting systems.2 Irish elections use arguably the world’s strangest and most mathematically intricate voting system; I’ve been working on an essay tracing its history, about which not much has been written.
The context is that, this year, I’ve been re-reading classics from philosophy (“in the age of AI”, as everyone is supposed to say now), which sent me down a massive rabbit hole about John Stuart Mill and his advocacy for proportional representation voting. Not that anybody cares, but I think you can’t really understand Mill’s political economy and philosophy without understanding nineteenth-century efforts toward electoral reform.
I only talked for about 20 minutes, which was definitely an instance of the truism that giving a short talk is harder than giving a long talk. You can view my slides here.
I included some bonus material about the difficulties of measuring electoral disproportionality when drawing constituency boundaries, and about the impossibility theorems that place hard constraints on the optimal design of voting systems. I’m not really sure why participants need to know this kind of stuff, but, as Talmudic tradition tells us, whoever nerdsnipes one student nerdsnipes the class entire.
As the slides mention, the terms ‘ranked choice’ and ‘proportional representation’ are constantly misused. Ranking multiple candidates does not by itself produce proportionality, and some PR systems only allow you to vote for one candidate.
That's a coincidence, I was just brushing up on my JSM this morning for a piece I'm working on. I'd like to have been at your talk, it sounds interesting.
I have a feeling that systems that combine preferential voting and proportional representation produce a tendency for political parties to occupy the (evolving) middle ground and to take a "broad church" approach to minority views. A very good thing imho. This seems to work reasonably well here in Australia. I also rate compulsory voting for the same reason. I don't think the actual variations are that important provided they are intelligible. We also have an independent respected electoral commission that runs elections and sets boundaries, also great.
It is unfortunate that places that got more primitive single vote forms of democracy basically have no hope of changing them.