Explaining non-participation in deliberative mini-publics

One of the issues dividing commentators on this blog is whether participation in sortition-based assemblies should be mandatory or voluntary — see, for example. Those of us advocating legislative juries based on Athenian nomothetic panels advocate quasi-mandatory participation in order to ensure accurate descriptive representativity. Those, however, who argue for full in-depth participatory deliberation claim that mandatory participation would be ‘disastrous’, as it is hard to see how a a ‘full-charge’ legislature — essentially like existing elected legislatures but with members selected by lot — could function with (in effect) conscripted members.

A paper by Vincent Jacquet in the European Journal of Political Research, examines why it is that the overwhelming majority of randomly-selected persons refuse the invitation to participate in deliberative minipublics. Given that the descriptive representativity of the minipublic (vis-a-vis the target population) is one of the principal rationales for sortition, acceptance rates are extremely poor, ranging from 1% (America Speaks) through 3% (Belgium), 5.7% (Ontario), 6.2% (Netherlands), 7.4% (British Columbia) to 20% in Deliberative Polls. As a result there is a danger that voluntary participation risks the ‘over-representation of better educated and politically active individuals’ as ‘the [overwhelming] majority of the recruited population refuse to participate (pp.2-3).

Stratified sampling can enable some correction but this is generally on the basis of crude population metrics and ‘[voluntary] participation follows the social unequal distribution pattern of political engagement . . . participants are better educated, mostly men and older than the average population . . . participants are more politically interested and have a higher sense of efficacy’ (p.3). Such ‘statistical biases’ might suggest that deliberative minipublics privilege the usual suspects, the added disadvantage being that nobody has chosen ‘descriptive’ representatives and there is no way to kick the rascals out. There is a danger that

Using mini-publics to shape public policies may create new deliberative elites — randomly selected but distinct from the wider population precisely because they have taken the time to deliberate. (p.14)

Jacquet’s paper is the first attempt to ask a sample of refuseniks why they declined the invitation to participate. On the basis of in-depth interviews with thirty-four invitees to three Belgian minipublics convened between 2011 and 2014 (G1000, G100 and CCP) with a raw positive response rate of only 3%, Jacquet developed six explanatory logics for non-participation:

  1. Concentration on the private sphere
  2. Self-disqualification because of lack of competence/expertise
  3. Dislike of group situations, reluctance to speak up
  4. Conflict of schedule
  5. Political alienation
  6. Minipublic’s lack of impact

1, 4 and 6 were the most frequently offered reasons. The first is particularly problematic as it was frequently given by younger people with family commitments and busy self-employed people — groups that tends to be underrepresented in conventional politics. The second reason reflects modern perspectives on the division of labour outlined in Constant’s famous essay and tends to militate against less educated and working-class people, one construction worker explaining that he lacks the skills to deal with the (constitutional) topics discussed in the G1000 (p. 10). University graduates would be less likely to view themselves as incompetent to participate in political deliberation. The same sociodemographic factors might well contribute to explanatory logic 3 (reluctance to speak out). Although many refuseniks ticked the ‘conflict of schedule’ option, Jacquet speculates that this might sometimes be ‘an easy excuse to decline the invitation’ (p.11).

Although scepticism regarding the minipublic’s lack of impact (6) would be a rational reason not to participate, Jacquet was surprised that there was no difference between the private (G1000 and G100) and government (CCP) initiatives. Those of us arguing for legislative minipublics with a binding mandate might well argue that this would overcome the objection. Jacquet notes that none of the participants in the Belgian projects were paid and that DPs secure a higher level of participation due to the honorariums on offer. However it’s not entirely clear from a civic perspective what the difference is between mandatory participation and generous rewards. If persons unsuited (for whatever reason) to deliberate were motivated to attend by generous rewards or legal sanctions the effect would be the same. Participation in trial juries is mandatory and participants are also remunerated, so this might well be the appropriate model in order to ensure that deliberative minipublics accurately ‘describe’ their target population.

15 Responses

  1. Reblogged this on Suefew's Philosophy Blog and commented:
    Agree with this. We cannot create a new elite. And also we would lose out from learning the concerns if those who self exclude.


  2. I’d be surprised if the dropout rate is anything like as high where these things are expected to have an impact. I don’t know for sure that this is the case, but it’s certainly not my impression of Government run citizens juries.

    Here’s one South Australian juror:

    “If you receive an invitation from the Premier it’s a rare thing to pick out of your postbox … It’s a gift.
    You’ve been invited.
    Something about you was chosen.
    Even though it’s randomly selected, you feel ‘Oh wow, I feel so lucky’

    Liked by 2 people

  3. Nicholas,

    Jacquet did speculate that a higher impact factor (6) might have a positive effect on the takeup rate (but it’s possible that this might be offset by perceived lack of competence (2)). He was also surprised by the fact that the acceptance rate was no better for government-sponsored sortition panels.

    I’m sure that the reaction you quote is fairly typical for those who accept the invitation, but this is a tiny minority and there is no good reason to extrapolate this response to refuseniks. As far as I know this is the first qualitative survey on why citizens fail to take up their allotment so we need to take the reasons cited very seriously and assume that the positive responses that you mention are atypical.

    What puzzles me is the hostility expressed by some sortinistas to mandatory participation — legislative jury service would be a very responsible civic duty so it would be entirely reasonable for participation to be just as obligatory as in any other form of jury service. Membership of the nation state is compulsory for all its citizens and we cannot choose which laws we wish to obey, so the same principle should apply to the microcosm charged with making the laws that we have to abide by. Lawmaking by atypical enthusiasts is a form of aristocracy (especially considering which social categories are privileged), not representative democracy.

    Liked by 1 person

  4. > If persons unsuited (for whatever reason) to deliberate were motivated to attend by generous rewards or legal sanctions the effect would be the same.

    What an absurd statement. “Attendance” is very far from participation. One can be forced to attend but cannot be forced to participate. Even worse, someone who is forced to participate is very likely to see their forced participation as reason to take personal advantage of their position of power.

    The kind of false analysis offered in this post is an example for why not every “experiment” with sortition should be hailed as a step forward toward democratization of government using the sortition mechanism. An improperly designed or improperly implemented allotment setup must be clearly rejected by democratically-minded sortition advocates or it may very well lead to a loss of credibility and creation of negative popular sentiment toward the idea of sortition.


  5. Yoram,

    Why is bribery acceptable but not civic obligation? The latter may well be preferable as it can be validated by republican norms, as opposed to pure self interest. As for participation, according to Pitkin all descriptive representatives can do is vote and that can, and should, be mandatory. We have yet to hear what you mean by the word “representation”, as invited by Terry’s observation (that you have failed to respond to) that your vision of sortition is a form of aleatory oligarchy.


  6. So by “bribery” you mean “payment for services provided”, and by “civic obligation” you mean “forcing unwilling people to make crucial decision for society”. You are a joke, Sutherland.


  7. Yoram,

    As you well know, my model is based on trial-jury practice, which involves both mandatory service and payment for services provided. It’s essential to ensure that the vast majority of those called to serve actually turn up — in order to ensure accurate statistical representation, rather than the jury being a subset of the small minorities described in this paper. You have yet to tell us (fourth request, and counting) what you mean by representation and how your perspective is compatible with the absence of accurate statistical representativity.


  8. Perhaps I am being repetitive, but how obligatory service should be depends on the nature of the task and duration of the mini-public. For a short term service (no more than a few weeks) in a mini-public that will hear pro and con testimony on a proposed bill and then give a final vote, service should be quasi-mandatory (deferments for cause). For mini-publics that will spend more than a year debating bills, proposing amendments, etc. it is impractical to expect all called citizens to attend or participate. In this case, a more volunteer body (not just those who proactively offer their service, but those who agree to serve once called) makes sense… so long as the final vote is by the more representative short duration body.

    Liked by 1 person

  9. Terry, I think we both disagree with Yoram that long-serving quasi-voluntary assemblies would be representative and that’s why I’m asking him to clarify what he means by the word “representation”.


  10. *** Actually Jacquet’s study is not very interesting relating to democracy-through-minipublics or even of an hybrid model. The six reasons could be guessed by anybody, and were considered in our previous discussions – except maybe factor 3 « dislike of group situations, reluctance to speak up » as far it may be distinguished from factor 2 « self-disqualification ».
    *** Jacquet says his study is only qualitative. But actually it is likewise quantitative (fuzzy quantitative) when he distinguishes if the presence of a factor is « often » or « not often ». This quantitative side of the study faces two obstacles. First, the one encountered by all studies founded on sel-assessment : the bias of the answers. It is easier to put as main reason of non-participation the child care, or a rational political discurse, than a pettier reason. Second, the non-representativity of the interviewed persons, as the study covers persons who refused to participate to the minipublic – but accepted to participate to Jacquet’s study ; Jacquet used some kind of stratification – « ensuring a certain balance across the different types of profile » (p 7), but we know that stratification cannot generate real representativity.
    *** Anyway the weight of the factors may depend strongly of the circumstances of the considered minipublics (e.g.: no indemnity) and of their lack of political impact – factor 6, but playing likewise on factor 5 « political alienation », which includes (p 9) « participation seen as an elite-driven manipulation ».
    *** All the considered minipublics had around 3 % participation ; even the more badly designed system of democracy-through-minipublics of the volunteer kind would probably be far of that ; which confirms the article is not very useful from an (ortho-)democratic point of view.
    *** I think that in this blog we have discussed all the relevant points. Sure, some factual points have to be ascertained by experiments ; but that will need to be in a democracy-through-minipublics, or at least in an hybrid system including minipublics with strong and clear political impact.
    *** And the strength of the factors of reluctance to participation may evolve with time.

    Liked by 1 person

  11. By way of clarification, in The Concept of Representation (1967) Hanna Pitkin identifies the two principal forms of political representation. Pitkin’s distinction has become the standard tool in representation studies and political theory in general, mirrored for example in the work of Philip Pettit. The two forms are as follows:

    1. “Acting for”. According to this model, candidates for election make representative claims, which are evaluated by voters who then select the claims that best match their preferences. The fidelity with which successful candidates implement their claims are then re-evaluated at subsequent elections. Although politicians frequently forget their election promises when in office, a new breed of politicians (Donald Trump being an obvious example) is seeking to implement them more faithfully. There is no reason in principle why a candidate needs to share the same socio-economic profile as her supporters, which is why it is possible for the representative claim of a billionaire property developer to be well received by poor and disadvantaged citizens. Sortition cannot implement this form of representation as no choices are made and there are no mechanisms to ensure retrospective accountability.

    2. “Standing for”, also known as descriptive/statistical representation or stochation. Representatives are chosen at random, on the understanding that a body several hundred strong would be a portrait in miniature of the target population. Pitkin is sceptical that such a body could act for the target population in any respect other than voting, as individual speech acts are not subject to the law of large numbers.

    Jacquet’s paper has demonstrated that in existing sortition schemes only a small minority of those selected by lot take up the invitation and, as a consequence, the resulting group fails to “stand for” the target population. This being the case it’s hard to understand the nature of the representative claim for non-mandatory sortition as it does not correspond to any of the standard uses of the term in the literature. If Yoram claims that such a group is representative then he would need to explain exactly what he means by the word.


  12. Andre

    >the strength of the factors of reluctance to participation may evolve with time.

    Then you have a chicken and egg problem as the only democratic mandate of the randomly-selected body is its accurate statistical representativity. I imagine this would require that the vast majority of those selected by lot participated, so it’s much easier to stipulate quasi-mandatory participation.


  13. Keith,
    In your criteria for a mini-public you focus primarily on the issue of statistically accurate representation (which you argue is defeated by active deliberation among members). However, there is other features of sortition. One is the enhanced problem-solving ability of diverse groups over homogeneous ones. Secondly, the ability to seek win-win solutions when a group is free from antagonistic partisan electoral imperatives. And thirdly, the anti-corruption potential. (I am using the term “anti-corruption” in a the broadest sense, rather than the narrow bribe-taking sense). The problem solving and win-win seeking are lost if a mini-public cannot deliberate.

    To expand on the anti-corruption potential…Under all existing political systems power ends up being concentrated in the hands of a few who seek to extend their own power. Sometimes as individuals, and sometimes as a party, class, or whatever. Sortition, uniquely, provides the possibility of eliminating such concentration of power through both random selection and regular rotation. Yoram’s notion seems to be that by assuring that existing elites will not continue to monopolize decision-making in society, that nearly any other “random” group would do better for society as a whole, than an entrenched elite. The accuracy of the representation is less critical to him than the freedom from elite domination. I share this view, and see the statistically accurate mini-public as a wonderful combination to have with the anti-corruption, problem-solving and win-win seeking features (which I argue need to be obtained from separate mini-publics with distinct functions, duration, and modes of selection.

    Liked by 1 person

  14. Terry,

    I agree that sortition has other potential benefits — both epistemic and prophylactic — but my concern is with democratic representation. The reason I keep banging on about this is that Yoram also claims that statistical representation is his desideratum, but this clearly cannot apply to a group in which only a tiny percentage of those chosen by lot accept the invitation. It may or may not be the case that a combination of payment and genuine power could drastically reduce the number of refuseniks but the track record of existing sortition experiments is very poor in this respect. I’m puzzled as to why Yoram has not responded to your earlier claim that his sortition model is a form of oligarchy, leaving you to defend a viewpoint that you clearly regard as problematic.

    >nearly any other “random” group would do better for society as a whole, than an entrenched elite.

    That observation may well be true for a random (rather than “random”) group, with a tightly-constrained mandate, as in our legislative juries, but the assumption that self-selecting persons will “do better for society” than elite members has dubious epistemic credentials and is democratically illegitimate. It would also appear to rely on an archaic Mosca/Pareto elite theory that fails to accurately reflect modern polyarchies. A large number of American voters clearly believe that their interests are represented by a rich property developer. Perhaps you think this is a case of false consciousness — what would your reaction have been if your old comrade Bernie Sanders had won the Democratic primary and the subsequent presidential election?


  15. […] seek another term than “representation” to describe their project, as the term is being used in a way that is unfamiliar to theorists working in the field of political […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: