Criteria for a representative citizens’ assembly

Given the high profile of some recent (UK) proposals for allotted citizens’ assemblies — including Conservative leadership candidate Rory Stewart’s Brexit assembly and Extinction Rebellion’s global warming assembly — there is an urgent need to initiate an informed conversation on the requisite criteria to ensure that the assembly design is compatible with principles of democratic legitimacy. As we have been debating this topic in depth on this forum for many years, this looks like a good place to start. The underlying assumption of this post is that the legitimising principle is ‘stochation’ — i.e. an assembly that is a portrait in miniature of the population that it seeks to ‘describe’ and that the goal is to increase the fidelity of the representation, subject to cost and other practical constraints. I would suggest the following criteria:

Criteria for the acceptability of the allotment procedure
The criteria for an impartial random-selection algorithm have been covered by Yoram Gat’s recent post. Terry Bouricius has also argued that the need to be seen to be fair might require some sort of public ceremony, as in pre-modern applications of sortition.

Selection pool
Should the same criteria of citizenship be used as for the electoral role, or is there a case to open it to all affected interests? However the Athenians were less inclusive, requiring a higher age for lawmakers than regular citizens along with swearing the heliastic oath.

Voluntary or quasi-mandatory participation?
Athenian legislative juries were drawn from a pool of volunteers, but the 6,000 citizens were a very substantial proportion of the citizen body and there was strong normative pressure for all citizens to serve (those who didn’t were ho idiōtēs).  However the modern take-up of sortition invitations has been in the region of only 4%, so it might well be argued that voluntary participation would generate an atypical sample. If so, should participation be a civic duty (as in jury service) or will incentives and support suffice, bearing in mind that the relevant principle is the representation of the beliefs and preferences of the vast majority of citizens who are not included in the allotment?

Size of the allotted sample
Modern examples of sortition have been for bodies ranging from only twenty to several hundred, and some statisticians have argued that 1,000 is the minimum size for a reasonably accurate representation (Athenian juries ranged from 501 to 5,001). What is the connection between sample size, decision threshold and confidence interval? How big could the sample be before the onset of rational ignorance?

Length of service
On the one hand a citizens’ assembly would need time to gain adequate knowledge on the issue under consideration, whereas on the other there is the danger of participants ‘going native’ and thereby ceasing to adequately reflect the beliefs and preferences of their (virtual) constituents. Should assemblies be convened on an ad hoc basis (as with 4th century Athenian lawmaking), or is there a case for permanent bodies with rolling tenure for members?
Continue reading

Paper: No Stratification Without Representation

One fascinating aspect of sortition is that it treats all groups in the population fairly: If a group constitutes x% of the population, the group’s share in the panel will be x% in expectation (that is, on average over many random panels). Furthermore, it is unlikely that, in a random panel, this percentage will deviate much from x%; this event becomes ever less likely the larger the panel is. Unfortunately, there are practical limits to how large sortition panels can get, which means that a certain variance remains. Had the Irish citizens’ assembly been sampled without consideration for gender, for example, a gender imbalance of at most 45 women against at least 54 men would have happened in about 15% of random panels.¹

One way around this problem is stratified sampling. For example, one could fill half of the seats with random women and half of them with random men. As Yoram wrote in an earlier post on this blog, one can still guarantee that every person is selected with equal probability, and thus, that every group will get its fair share of the panel in expectation. Stratification by gender will obviously ensure accurate representation to the genders. But what happens to the representation of other groups?

My collaborators — Gerdus Benadè and Ariel Procaccia — and I studied this question in a paper that we recently presented at the ACM Conference on Economics and Computation. As Mueller, Tollison, and Willett argued as early as 1972,² stratification can greatly reduce the variance in representation for groups that highly correlate with the feature we stratified on. This is good news since correlation is everywhere; for example, stratifying by gender will help to represent opinion groups related to military intervention, gun control, and healthcare.³ We show that there is no real downside to stratification: Even in the worst case, stratification cannot increase the variance of another group by more than a negligible amount. These results hold up to very fine stratifications, where each seat is filled by a random member of a dedicated stratum. This suggests that we should indeed make extensive use of stratification. In a case study on a real-world dataset, we show that stratification can reduce the variance in an opinion group’s representation by a similar amount as an increase of panel size by multiple seats — even if the stratifier does not know the opinion in question!

The main technical difficulty in the paper is working with indivisibilities. For instance, if we split the 99 seats of the Irish citizens’ assembly proportionally by gender, women should get around 50.66 seats. To ensure that every person is still selected with equal probabilities, we need to randomly “round” the seat assignments, giving women sometimes 50 and sometimes 51 seats. This process is somewhat delicate — rounding introduces new variance, which might lead to some unfortunate group becoming much less accurately represented than without stratification. If one uses the rounding procedure suggested in our paper, this is not the case.

Continue reading

A response to Cody Hipskind, part 3 of 3

Cody Hispkind’s post is here. The previous parts of my response are here and here.

Political activism under a democratic system

A major tenet of democratic ideology is that people are the best representatives of their own interests: when provided sufficient opportunity, each person and each group of people are best able to understand and express their own values and ideas and the actions that should be taken in order to promote these values and ideas. This tenet is in contrast to “republican” ideology which shares with democratic ideology the idea that everyone’s interests should count equally, but asserts that some people (“a natural aristocracy”) are best qualified to determine what those interests are and how they should be pursued, and therefore those people should be in charge.

Elections are a republican, anti-democratic mechanism: they empower an elite to determine public policy for others (whether this elite may be called a “natural aristocracy” is a matter of taste, I guess). That elite should be able to represent itself, the democratic tenet asserts, but is quite unlikely to represent the majority of the people who are very different from it. Sortition, through the process of statistical sampling, creates a body that by representing itself would represent the public at large.

However, the capacity for self-representation is not a spontaneous, automatic capacity. Getting a group of people (or a single person, for that matter) to the state where it is able to represent its own interests effectively is not a trivial matter. From an institutional standpoint, there are clearly some preconditions that need to be met: there need to be enough resources at the disposal of the group so that reliable information can be gathered. There needs to be enough time to discuss matters, determine an agenda, fashion proposals, debate them, amend them, vote on them, evaluate the effect of the adopted policy, reconsider the matter and repeat the procedure over time.
Continue reading

A response to Cody Hipskind, part 1

Thank you Cody for your post. I believe it can serve as a starting point for a fruitful discussion – a discussion that has already started in the comments thread to the post.

I would like to address various points you made separately. Here are the first 3 points, corresponding to the first 6 paragraphs in your post.

1. Class and political conflict

It is an obvious fact of life that different people have different political ideas. If everybody agreed on everything politics would be very easy – any one of “us” would do what “we” all agree upon. Still, there is a tendency, which used to be the dominant line of elite political thought in pre-modern times but is common today especially among liberals, to argue, or at least to imply, that disagreements are over means rather than ends. That is, that “we” all agree what is the common good (or at least would agree what is the common good if some of us overcame our ignorance) and the main difficulty is finding competent people who would be able to achieve that good. This translates to elitism. The elitism used to be explicit with the pre-moderns, the archetype being Plato, it was slightly less explicit with the early moderns, e.g., the American Founders, and it is much more implicit today. Nevertheless, it is often still just below the surface of the democratic rhetoric.

The notion that “we” all agree on goals can be dismissed. The common-goals ideology and the attendant elitism are of course self-serving notions for those in power. They imply that to the extent policy produces undesirable outcomes – say enrichment of those in power and their associates and the impoverishment of much of the rest of the population – then those in power can at most be blamed for incompetence rather than corruption. (Naturally, they would in fact argue that unfortunate as the outcomes are, they are necessary for the achievement of the commonly agreed goals.)
Continue reading

Representation: An ideological and legal fiction

I’ve spent the last few days at an international workshop on ‘Representation in Historical and Transcultural Perspectives’ organised by the Centre for Political Thought at Exeter University. On the final day it was suggested that representation was an ‘ideological and legal fiction’ and none of the participants disagreed. Yves Sintomer gave the final presentation and suggested that his fieldwork comparing Chinese and French systems of representation showed little difference between the two and that the lack of effective representation was an existential crisis for democracy. I had a question for Yves, but we ran out of time, so will ask it here (and draw it to his attention):

One system of representation that would clearly be non-fictional is delegates with legally-binding instructions but this was rejected at the time of the American and French revolutions. Trustee representation (with free mandate) may have worked for a time but didn’t long survive the extension of the franchise and is now rejected by the populist uprising in Europe and America. Virtual representation in Burke’s sense was always fictional due to the dissimilarity between voters and the political class. This would suggest that the only form of non-fictional democratic representation would be when final decision power is vested in a statistically-representative minipublic. Concerns might be raised both about the accuracy of the descriptive representation and the epistemic consequences, but such a system would be non-fictional so long as the microcosm retained ongoing descriptive representation vis-a-vis the target population. This would require large juries, quasi-mandatory participation, short-term service, balanced (exogenous) information and advocacy, and silent deliberation and secret voting, but the representation would not be fictional if it could be demonstrated that multiple samples of the same population generated closely-matching decision outputs. Might such a system be the only way of establishing genuine political representation?

A Marxist Analysis of Sortition

Welcome to Equality-by-Lot, Cody Hipskind! -Yoram

In this post, I’ll be taking a Marxist approach to the question of sortition. That is to say, I’ll be working from a framework that understands society as being composed of several classes with conflicting material interests, and which understands the state as an instrument by which one class rules over others.

I would also note that though classic Marxists have historically centered location within the system of production as central to the reproduction of class society, for my part I hold with the host of theorists who have shown in the decades since Marx wrote how questions of race, gender, migratory status, etc. are likewise integral to the ways in which the ruling class has reproduced its hegemonic status.

Let me begin by recognizing the benefits which rule by lot has from a Marxist perspective. Within my country of the United States and around the world, it is indisputable that electoral systems produce legislatures which, when taken as a whole, are disproportionately wealthy, white, cis-male, and otherwise more representative of the members of the ruling class than of society as a whole. A lottery system would certainly correct such an imbalance by increasing the likelihood of that members of historically disenfranchised communities would receive an equal voice in the legislature. This would obviously be a desirable outcome.

However, it is my position is that elections provide two social goods which would likely be undercut through a system which chooses officers exclusively through sortition. Specifically:

1. Elections create competition between parties over control of which is to say, over the control of the government.

In this way, elections implicitly recognize the sort of divisions in society I outlined above. If a party does something unpopular–say, decide to lower taxes on the rich, voters can punish that party by throwing them out and choosing a replacement. In practice, this means that the subgroups which compose the working class can play kingmaker between the bourgeoisie parties and, ideally, organize themselves into their own parties in order to take power as a class themselves.

In contrast, under sortition, there’s no way to hold people accountable for their decisions: popular or not, you leave office when your term expires. This has the upswing of producing less competitive and divisive politics. And that’s nice and all, but as a Marxist my goal is not to create a politics without division: its to highlight the material divisions and contradictions under capitalism and to exploit the competition among the bourgeoisie parties in order that the working class can get a foot in the door and eventually seize power.
Continue reading

Cancio: Sortition and the Allotted Chamber as Institutional Improvements to Democracy

Here [PDF] is an automated translation (with a few touch-ups) of Jorge Cancio’s 2010 paper “Invitación a un Debate: El Sorteo y las Cámaras Sorteadas Como Mejoras Institucionales de la Democracia”.

Abstract:

I start off inviting my readers to exercise their imagination and then explaining a proposal of creating new “sortition chambers” on all administrative levels – from a chamber at the same level as the present-day Spanish Congress and Senate down to sortition chambers for each municipality. They essentially would be an addition to present-day institutions and would partake in the powers which are held today by elected representatives and officials, although the proposal envisages that in the short run they could be out-voted by the elective institutions. They would exercise their powers according to deliberative procedures.

After that introduction I offer a short account of present-day theoretical and practical proposals and implementations of sortition-based systems in the political field – highlighting the fact (following Manin) that mainstream discussions on democracy tend to ignore sortition altogether (including those made by the political left), but also making the point that since Dahl and others (inter alia, Burnheim, Goodwin, Barber; and the more recent works now appearing in Imprint Academic) there seems to be an increasing renaissance of this subject. I also point to the practical experiences of the Planungszelle and the voters’ and citizens’ juries.

Thereafter I try to explain why sortition is mainly ignored nowadays – here I follow again Bernard Manin – identifiying the contractualist bias of the XVII and XVIII century revolutions (which sought to establish a rule of consent for being governed) as the main reason for opting for elections instead of sortition as mechanism for the selection of political office-holders. Afterwards, the emergence of the political parties as oligopolical forces in the political market (here I refer to the work of C.B. Macpherson) and their growing distance from society and increasing autonomy vis-à-vis societal needs have led to a “party democracy” or “cartel party” system (Katz and Mair). The facade of autonomous decision in electing public officials covers, therefore, a system where political offers are very limited, allowing me to speak about an “election-fetish”. Contrasting to this evolution I then underline the paradox of identifiyng democracy with elections (and political parties) when we consider that democracy was linked in Athens with sortition whereas elections were considered aristocratic.
Continue reading