Ideas, hacks, representation by sampling and political theory

Posted on January 12, 2022 by Nicholas Gruen

https://twitter.com/ockhamsbeard/status/1481137490920882178

In response to an exchange of tweets I wrote what seems like a long post on Twitter — lasting 7 tweets —which is a short post here. In any event it tries to crystalise something I think is important in the way I see things — and in how I see them differently to those who give more weight to political theory than I do.

Seems to be working well in New Zealand. But while such topics occupy the minds of the political ‘thinkers’, that’s because academia in particular is so given to ‘big debates’ with ideal types with long histories in the literature.

I think much more progress is possible by paying less attention to theory and the endless set-piece debates between this and that (say FPTP v PR) and more attention to specific hacks which look like they could make a major contribution

https://www.themandarin.com.au/103093-what-is-a-policy-hack/

What is a ‘policy hack’?

In the language I developed in that article, juries are both an ‘idea’ and a ‘hack’ — which is to say they intimate a whole repertoire of possible institutions based on a different idea of what makes someone ‘representative’. (here representation by sampling not election)

And they are the ‘hack’ because they provide a concrete action that can be taken. I think there’s a lot to be said for bringing citizens’ juries into our understanding of checks and balances. Not only are they a different way to do democracy.

They’re time honoured institution. So, in seeking the populace’s support, we wouldn’t be asking them to back some professor’s theory but rather the chain of legitimacy back to Magna Carta and beyond and into people’s trust of their neighbours (and distrust of politicians).

I’d LIKE to think that greater PR here would improve things, but I just don’t know. New Zealand has done some good things since greater PR, but nothing DIFFICULT that I can think of. And the alternative is Italy which doesn’t appeal.

I think we can point representation by sampling at specific problems our system has and, in so doing give ourselves a very good chance of making them a lot better.

Filed under: Uncategorized |

8 Responses

keithsutherland, on January 12, 2022 at 1:04 pm said:

In the language I developed in that article, juries are both an ‘idea’ and a ‘hack’ — which is to say they intimate a whole repertoire of possible institutions based on a different idea of what makes someone ‘representative’. (here representation by sampling not election). And they are the ‘hack’ because they provide a concrete action that can be taken.

I don’t think anyone would deny that. But if the representative claim is based on sampling, then any hack would need to comply with basic principles of statistics (most modern CAs fail this test). And the role of juries has always been to make decisions (guilty or innocent etc), the word “adjudicate” meaning “make a formal judgement on a disputed matter” or “act as a judge in a competition”. Prior to recent (ab)usage by deliberative democrats and sortitionists, I know of no example of juries originating policies. Hack away as much as you like, but please lets make sure we don’t conflate apples and pears.

LikeLiked by 1 person
Conall Boyle, on January 12, 2022 at 1:54 pm said:

Nor do I know of any CJ being given any other powers than ‘advisory’.

Although the Brexit Referendum was technically advisory, P.M. Cameron promised beforehand that he would abide by the result. Has any CJ been given such a promise?

LikeLiked by 1 person
keithsutherland, on January 12, 2022 at 1:57 pm said:

I don’t think so Conall. And, if the representative status of the CJ or CA is questionable it’s just as well!

LikeLiked by 1 person
Nicholas Gruen, on January 13, 2022 at 1:15 am said:

We’ve discussed your view on statistics before Keith. It isn’t a big deal for me, because, as with the jury in a legal case, I don’t want to give much power to narrow majorities on citizens’ assemblies. I do want to
1) have a standing citizens’ assembly as some way of surfacing the considered opinion of the people — as opposed to their ‘raw’ opinion as represented by opinion polls.
2) give that assembly some power to appoint other assemblies as the one in East Belgium has.
3) crank up the power of that citizens’ assembly to act as a check and balances on the system we have where they reach a reasonable consensus — represented by — say — a 60% majority as outlined here.

LikeLiked by 1 person
keithsutherland, on January 13, 2022 at 7:22 am said:

Nick,

It wouldn’t take much to convert a simple majority (51%) in a deeply divided polis into a “reasonable consensus” (which you claim would be 60%). Which way it would go would depend on a number of factors, including the acceptability of voluntary participation, the choice of experts and advocates, the (elite) cultural zeitgeist etc. This is particularly true when the organisers have a particular goal in mind (e.g. undermining Brexit). So, if the legitimising principle is mirror representation, I’m afraid there’s no escaping basic statistical principles (and good practice).

LikeLiked by 1 person
Liam Jones, on January 13, 2022 at 9:13 pm said:

I partly agree.

Political philosophy has a bad habit of getting caught up in abstract debates over false ideals and vanishing up its own rear. Practical changes in the short term are immediately beneficial in a way perfect forms can never be.

On the other hand, ideal types can help clarify our values and goals even if they never come to pass. They help us see where we’re going.

I think we all agree that current citizen assemblies are far from a perfect form of sortition. Is that a show-stopper?

Democracy in the 18th century was far from perfect even by the standards of electoral democracy. But it evolved by a series of hacks and battles into something better (if not exactly good).

On the other hand, manipulated referenda — an imperfect form of direct democracy — serve as a cloak of legitimacy for dictators.

Which path will citizen assemblies go down? I don’t know. To the degree that they can overcome the failures of modern politics, they might be a route towards better sortition. But for some elements of the design — especially if they’re manipulable — it would behoove us to draw a line in the sand.

LikeLiked by 2 people
Nicholas Gruen, on January 14, 2022 at 2:53 am said:

Thanks Keith

Do you think the Australian constitution would be better off or worse off with the proposal I suggested for it in this piece?

The first power I’d give them is a subtle, indeed sneaky one. The secret ballot was known in the nineteenth century as the “Australian ballot” after South Australia and Victoria instituted it in 1856 to prevent citizens being intimidated into voting against their will. So it’s fitting that we take the Australian ballot further and use it to loose the bonds of careerist intimidation within our Parliament. I’d give any super-majority of the people’s chamber the power to compel a secret ballot of the other chamber(s) on the matter in question. In fact, a secret ballot isn’t unknown to Parliament today. The Speaker of the House and President of the Senate are elected by secret ballot, and that fact has changed outcomes. 4 If this didn’t resolve the deadlock, I’d use a mechanism like S57 of our constitution uses for resolving a deadlock between the House and the Senate after an election. If, after three months, the deadlock between the houses remained, the matter would be resolved in a joint sitting between them.

LikeLiked by 1 person
keithsutherland, on January 14, 2022 at 6:32 am said:

Nick,

A chamber of 201 is certainly an improvement, especially as you view its function is the prophylactic, rather than representative. Peter Stone has a paper forthcoming in Constellations that makes a similar argument for the protection of democracy. But my principal anxiety is anything that undermines sortition as a representative tool, so I would have concerns if your chamber was (effectively) constituted by voluntary sortition with long-standing participation.

LikeLiked by 1 person