troubles come in legions

asking silly questions, getting policy answers

One of the many little stories and anecdotes that’s stuck in my mind for decades relates to some long-lost magazine article I read about the French Foreign Legion. I might have misremembered it (I certainly can’t find the source), it might have been misreported in the first place and the Legion might have changed its policy, so don’t treat this as a fact of any kind, but I think it’s descriptive of something which does happen quite a lot.

The anecdote related to the way in which the Legion tried to deal with the fact that the language of command was French, they wanted to avoid the troops forming cliques based on language, but lots of the Legionnaires, being foreign, didn’t speak French. So what they did was to significantly encourage the use of the French language rather than native tongues, and to give everyone a booklet with a bunch of common phrases that might be used in everyday military life, and when off duty.

What the author claimed to find, when talking to soldiers who had been in the Legion for an extended period of time, is that they had begun to lose fluency and facility in their original native language, but that this wasn’t because they had become particularly good French speakers. Instead, he got the strong sense that their mental and conceptual universe had shrunk, to the scope of the things and situations covered in the booklet.

There are a lot of reasons to be worried about the over-use of polling in politics today; I tend to agree with those who say that politicians ought to be leading public opinion rather than just trying to reflect it. I also worry that the entire science of polling is in a slow-motion crisis with respect to response rates, and that for the moment, people aren’t taking the likely degradation in quality and replicability anything like seriously enough. But I think it’s potentially even more dangerous that something similar to the affliction of these apocryphal Legionnaires might be affecting public life.

In order to carry out opinion polling, you need to formulate a standardised question which can be asked in reasonable time. Which means that you’re putting a potentially quite complicated issue through a really tiny information filter. Even if you’re being a lot more sophisticated and conducting focus group research, you can only discuss anything at the level of detail that it’s reasonable to expect a bunch of ordinary citizens to engage with over the course of a couple of hours.

This sort of variety reduction is obviously bad for obvious reasons, but it’s also bad for the non-obvious reason that the kind of information which gets filtered out is likely to be the operational and practical detail; the “plumbing”. You wouldn’t expect to get ideas about plumbing and implementation from a polling exercise or focus group, of course, that’s what technocrats and the civil service are for. But when the political leadership is so strongly inclined to focus on these particular sources of information, there’s a constant danger that their mental universe will shrink.

I think this might explain the tendency we’ve seen over time, that we seem to be getting political leaders who regard policy as a black box, with a few simple levers to dial up and down independently. While regarding the technostructure that has to make things work as a “Blob” which gets in the way of the levers. It’s no good blaming the general public for saying that they want good things but don’t want to pay for them – that’s just what the words “good things” and “pay” mean. The big problem here is that the people who deal with the kinds of questions that you can’t reasonably ask the general public have a greatly diminished voice, and only get brought in at the end of the process to say “yes”, “no” or “its gunna cost yer guvnor”.

Which makes me think we might want to expand the scope of the kind of questions for which Citizens’ Assemblies are used. As well as issues that are universally recognised to be complex and multi-layered questions affecting all sorts of areas of society, it might be a good idea to bring them in from time to time on utterly mundane decisions of government. You’d have the same approach of providing a budget for technical advice and explaining the practical issues, and the same output of a considered consensus report for the ministerial decision makers to engage with.

It's not so much that a Citizen’s Assembly approach would give us better decisions on the specific issues they considered; I think that if we did it a few times a year we might get better decision making in general. A lot of the purpose of all sorts of research is as a “spiritual exercise”, rather in the way that Jesuits are encouraged to spend hours at a time considering episodes in the life of Jesus. Getting our decision makers into the habit of understanding that plumbing exists and that it affects the popularity and outcomes of their policies would be an important change.

(PROGRAMMING NOTE: The comments spammmer has now been reported and nuked, thanks very much to everyone who told me about them. I would not have known otherwise, because it turns out that the Substack implementation of blocking is a great protection for impersonators - I couldn’t see their comments until I logged out, at which point I couldn’t report them, because I was logged out! Just in case anyone thought differently - when I comment on the website, it shows a little “Author” tag, and I am very very unlikely to be promoting investment advice. Thanks again everyone).