I’ve had many conversations with people about sortition over the last several years, and I’ve never been happy with the term “sortition.” However, I’ve never found a satisfactory alternative. I suspect other people on this blog have also struggled with this problem.
I know of four terms, in English, to describe the selection process we’re talking about. Here are the problems I’ve encountered with each one:
Sortition – doesn’t have an immediate association with anything that people know about (for example, it’s not clearly related to the verb “to sort,” or to the noun “sort”). Also, it sounds like the word “sordid.”
Random selection – the people I’ve talked to intuitively dislike the idea of “random” anything (except random sampling in statistics) – let alone anything “random” connected to democracy (although, ironically, they do like random selection of jurors).
Selection by lot – sounds archaic, even Biblical.
Lottery (and related terms like Alex Guerrero’s “lottocracy”) – has negative connotations because it’s associated with gambling, and (in the United States) with educational lotteries (a few lucky families win, most lose).
What would be a better alternative? At the moment, I think it would be helpful to talk about “mini-publics.”
The term “mini-public” is free of the negative and archaic connotations of the other terms. People can understand it quickly and easily. And it focuses attention on a body of citizens, rather than on individuals (in my experience, when I’ve told people about sortition, one pitfall is that they’ve immediately started thinking about random selection of individuals).
“Mini-public” could be defined something like this – “a scientific sample of the public – big enough to be representative, but small enough to be manageable.”
Then we could explain how: (1) two kinds of mini-publics already familiar in politics – juries and opinion poll sample groups; (2) several other kinds of mini-publics (Citizens Juries, Deliberative Polls, Citizen Assemblies) have been tried in practice, with good results; and (3) there are a number of promising proposals for new kinds of mini-publics, taking on bigger roles.
I would be very interested in hearing ideas from others on this blog. And if anyone knows who coined the term, I’d like to know that too (I know that Robert Dahl wrote about a “mini-populos” so something like that).
(Note: Thanks to Terry Bouricius for including “scientific sampling” and “mini-public” in a list of possible terms for talking about sortition).
Filed under: Sortition | Tagged: terminology |
Adam Cronkright
Ali
Ahmed R. Teleb
avanderven
avergne
Alex Kovner
azakaras
Brett Hennig
campbellwallace
Simon Threlkeld
Nick Coccoma
codyhipskind
Common Lot Sortitionist
Conall Boyle
constitutionalism
davidschecter
graytmatter
gtridimas
H.J. Hofkirchner
John Burnheim
jgarry20151
johnedwardtaylor
johnhuang865
Jonathan Crock
keithsutherland
Jacob Richter
lukewhito
SimonPek
Nathan El
Nicholas Gruen
Owen Shaffer
Patrick Chalmers
Paul Gölz
paulnollen
peterstone
Pierre-Etienne Vandamme
pronfeld
Emil
rcaze
richwbrown
Roger Knights
roslynfuller
Pierre-Etienne Vandamme
shaharavin
tbouricius
Ted Wachtel
Joshua Laferriere
theodorestathis
Wayne Liebman
Yoram Gat
Oliver Milne
Equality by lot 
I like “representation by statistical sampling”. Due to the use of statistical sampling in opinion polling the term is established in the context of politics and the connotation is positive and relevant.
I also have no problem with “sortition”.
LikeLike
David,
Agree that mini-publics is much better, sortition being more associated with the Stone/Dowlen/Burnheim argument for impartiality, as opposed to statistical representation. I also agree with Yoram’s definition, but would prefer a single (or compound) buzzword. My preference is stochastion, as it encapsulates the meaning of Yoram’s phrase but has the advantage of the same suffix as election. But the problem is you have to explain what it means (at least in the short term), whereas mini-publics is self-explanatory. However there’s something a tad condescending about the term mini-publics. But I think we do need two distinct terms for the two distinct functions of random selection, in order to stop people talking past each other.
LikeLike
I also use ‘statistical sampling.’ It usually goes over well.
It might be nice if the Wikipedia article for sortition emphasized statistical significance more, especially at the beginning of the article. That’s usually where I send people and it would be easy to get the wrong idea at a glance.
LikeLike
Unfortunately statistical sampling is only one half of Yoram’s definition, so you either need a phrase or a neologism. Once we’ve agreed on the term then we need two wikipedia pages (and two blogs).
LikeLike
I like ‘representative sample.’ The principle (and virtue) of representation is pretty deeply entrenched in the American political tradition and who could object to sampling? I think ‘mini-publics’ is too clunky to go viral (so to speak).
LikeLike
In an previous comment I have proposed, for a process that can include alternating random selection with election, but which involves a multistage process of reducing the selection pool until a candidate is selected:
darwinition
eduction
winnowing
winition
fetura — Latin for breeding
I presently lean toward fetura, with eduction second.
LikeLike
Jon,
The trouble is your first preference sounds more like a type of cheese and the second looks like a typo. And any process that ends with a single candidate has nothing to do with representation — so your scheme fits better under the (prophylactic) sortition umbrella.
LikeLike
I could certainly live with representative sampling but its more of a description than a precise concept or (legal) event. Could one speak of a “representative sampling” in the same way that one speaks of a general election? The nice thing about stochastion is that it puts it on a par with election. Given that we are proposing something new, then I think a new concept is in order.
LikeLike
@Keith
“any process that ends with a single candidate has nothing to do with representation”
??
Any process that selects a number of representatives less than the entire population is going to come down to selecting one of each. Nothing in what I write is limited to selecting only one for everything, like the doge of Venice. It is about successive reduction of the size of the candidate pool until one gets to one for each office being filled.
LikeLike
Jon,
It depends on what you mean by “office” — in my proposal (based on Greek terminology) a descriptively-representative legislative jury is established by stochastion, whereas sortition is an impartial process for filling individual magistracies. The former function is positive (the invisible hand of the law of large numbers) whereas the latter function is negative (protecting the political system from factionalism and corruption), hence the need for two distinct terms and two wiki pages. Although the Athenians selected dikastai/nomothetai and archai by the same mechanism (sortition), they insisted that the two categories were distinct. Insofar as I understand your proposal, you are referring to the appointment of archai, so sortition is the correct term. By all means choose a new term for the combination of sortition and election that you suggest, but that’s a separate issue.
LikeLike
@Keith
The traditional definition of “office” is any position of performing official duties, and an “officer” as one who fills that position. It would include representatives of al kinds.
The officers of a court, for example, are:
Judge (president)
Attorney
Witness
Juror
Bailiff
Reporter
All of these, taken together, are the “court”. The judge is not the court. He only speaks for the court.
Generally, mere assistants or clerks are not so considered, as they only advise and don’t make any decisions except as to what to present to those they advise.
LikeLike
I admire all the suggestions so far.
I’ve arrived at the word ‘RATIO’ which could be accompanied by the familiar: simple line drawing of a circle overlapping a square/the golden mean.
Maybe I’m not there yet, but to me RATIO is a word/idea which suggests a universally understood structure akin to snail shells, alongside qualities of strength and fragility, infinite variety yet hopefully, limited growth.
There’s no need necessarily to physically hold a snail shell to appreciate it’s simple beauty. There are many, yet it is hard to picture very many at a time. They’re timeless, poetic and inherently liberating. They can fire the imagination. Different but not unequal.
LikeLike
I am not attracted to neologisms (also Keith, I can’t remember your “stochastion” because it isn’t obvious how it would be pronounced).
I think some variant of “representative sample” is appealing…. the word sample can be used as a noun, adjective and a verb, which is useful as in
1. a representative sample is drawn by standard scientific sampling procedures.
2. (modifying “assembly”) the representative sampled assembly met on Thursday.
3. the list of residents was sampled to form a statistically representative body.
LikeLike
Jon,
You have far greater knowledge of the law than me, but I’ve never heard a juror described as a court official. This would certainly not have been true of the large 4th century Athenian juries — the template for my own proposal — as their prime characteristic (much lampooned by the comedians) was unaccountable power. Political officials are always accountable to the sovereign — not so with jurors (as they are the sovereign, so they are only accountable to themselves).
Barnaby,
The nice thing about RATIO is that it makes an absolute distinction with the guiding principle of sortition (arationality, aka the blind break/lottery principle). The problem is that it only encapsulates one of the words in Yoram’s original definition.
Terry,
The problem with “representative sample” it doesn’t have the unique meaning of Yoram’s phrase, it always needs some contextual qualifier (as in your 3 examples). Yes, you need to repeat stochastion a couple of times out loud to get the hang of it (the last syllable is pronounced like the “tion” in election), as it doesn’t quite roll off the tongue. As it’s something completely new we’re proposing (political representation by statistical sampling) we either need a neologism or else to adopt a classical term. There was no mathematical concept of proportionality in the ancient world, but stochastion derives from the Greek stokhos (“aim”), so that makes it a reasonable candidate, once you’ve learned how to speak and remember it. If we were unfamiliar with the word “election”, it wouldn’t be obvious how it should be pronounced. You just need to get used to it. The affinity with stochastic systems makes it clear what it means, the only problem is the suffix (“tion”). But I think we need that to make the explicit connection with election.
LikeLike
David Schecter asked about the history of the word « mini-public ».I am not historian of the field, therefore « as far as I know ».
*** The word of « minipopulus » comes from Robert Dahl, in After the Revolution ?, 1970, pp 149-53 (to check) and Democracy and its critics,1989 p 340 : « “minipopulus” consisting of perhaps a thousand citizens randomly selected out of the entire demos. Its task would he deliberate, for a year perhaps, on an issue and then to announce its choices.» It is the democracy-through-sortition idea, except that Dahl does not jump: his minipopulus is intended « as a complement », « not as a substitute », of the polyarchic institutions.
*** Fung 2003 coined, as far as I know, the word « minipublic », « following Robert Dahl » (pp 338-339), but as Fung acknowledges his concept is more inclusive, including instances with high content of self-selection >>> Fung (Archon) « Recipes for Public Spheres: Eight Institutional Design Choices and Their Consequences », Journal of political philosophy, 11 (3), pp 338-367, 2003
*** Goodin & Dryze 2006 took the word « mini-public », but in a more restrictive way than Fung « focusing on mini-publics with some claim to representativeness of the public at large », but nevertheless not requiring « statistical representativeness » (p 221) >>> Goodin (Robert E.) and Dryze (John S.) 2006 « Deliberative Impacts: The Macro-Political Uptake of Mini-Publics », Politics & Society, 34 (2), pp 219-244.
*** Taking « mini-public » in the sense of Dahl’s minipopulus is logical if you want to avoid Latin, but we must be aware of the possible ambiguities.
LikeLike
*** « Selection by lot » is bad, because, whatever the Latin etymology, « selection » has an usual meaning of « choice of the best » (in modern times, it seems the first use was for animal breeding), whereas sortition in the democratic thought means a perfect rejection of any idea of « choice of the best ».
*** Yoram Gat likes « representation by statistical sampling ». I used to be wary of any political use of the word « representation » due to the multiple uses of the word and its link to the « representative election » hoax. But I acknowledge that, especially for the young, the « representative sample » concept is becoming usual, and therefore Yoram may be right. But the drawback is the length : it is easier to speak about « democracy through sortition » than about « democracy through representation by statistical sampling ».
*** « Citizen juries » has a good side : to be connected to the old and known system of judicial juries evolved out of the medieval juries of England, and therefore to block the attacks against sortition as belonging to archeology, or utopia, or science-fiction.
*** I agree with David Schecter that « mini-public » is very good as « People can understand it quickly and easily. And it focuses attention on a body of citizens, rather than on individual ».
*** I don’t think we must discard the word « sortition » which has an interesting symetry with election (in the modern sense, restricted to « election by vote »). The word in French doesn’t exist really, but can be introduced (as many other latinizing neologismes were) for choice by lot (in French, « tirage au sort », easily linked with « « sortition).
*** Anyway « representative sortition » or « representation by statistical sampling » are about the process, « citizen juries » or « mini-publics » are about the bodies. I think for both concepts the best is alternating the words, but insisting on « mini-publics » and on « representation by statistical sampling »
*** There is likewise the word, coined by Dahl, of « mini-populus ». But a Latin word is maybe not the best one for popularization.
*** In French, we can use « mini-peuple » without problem. And I think the same in many languages. There is a specific problem in English language, where « people » may be a singular collective name, translation of « peuple » or « pueblo », including in its civic meaning, but is likewise the plural of « person ». In Spanish there is a problem, as « pueblo » may mean village, but I think « mini-pueblo » could work in democratic discussion (I would like to know the feeling of Spanish-language kleroterians).
*** Among people argumenting in France for sortition, some use « mini-peuple », but « mini-public » is more usual. It may be the influence of English language, but I am afraid it may be sometimes the result of some fear of dêmokratia : the sovereign power of the dêmos, Greek world that the French « peuple » translates perfectly.
*** Keith Sutherland’s proposal, « stochastion », as a sub-set of sortition including the idea of representative sample, is interesting (even if somewhat strange as an hybrid Greek-Latin word). But maybe « representative sortition », opposed to the « representative election », would be less abstract, and therefore better – at least at the step of popularization.
LikeLike
Andre,
I intended stochastion as the polar opposite (as opposed to subset) of sortition, in that it’s based on the (positive) principle of stochastic representation as opposed to the (negative) principle of exclusion of reasons. I agree with David that those of us interested in representation by statistical sampling should no longer use the word sortition and I’m personally torn between mini-public and stochastion. I think the former would work well in the short term, but the latter would be better in the long run. The trouble with both representative sortition and stochastion is you still have to explain what they mean, whereas mini-public is self-explanatory.
LikeLike
“Stochastion” is something I wouldn’t attempt to say before hearing someone else say it. The ‘s’ by the ‘-tion’ is throwing me for a loop.
LikeLike
Yes it’s a bit of a tongue-twister. “tion” in stochastion pronounces “chun”; whereas elec-shun. I do think though that election is only easy to pronounce because we’ve heard it so many times and the same would be true of stochastion in the (very) long run!
This is from the glossary in my draft PhD:
Stochastion: (stəˈkæst’shən): a neologism for the formation of a descriptively-representative microcosm of a target population by statistical sampling.
LikeLike
Andre, thank you for the background information on “mini-public,” and for the information about terms in French and Spanish. I recently learned from an Egyptian colleague that when Egyptians hear the Arabic phrase for “random selection,” the Arabic word for “random” is likely to immediately bring to mind the Arabic phrase for “random housing” which refers to the informal settlements around cities like Cairo.
Keith, you wrote that you’re “torn between mini-public and stochastion.” I think we all need at least two terms – one to refer to randomly-selected bodies (at the moment, my favorite in English is “mini-publics”) and another term to refer to the selection process (my current favorites are “statistical sampling” and “scientific sampling”).
LikeLike
Oops – André, I’m sorry I mis-spelled your name in my last post.
LikeLike
David,
Agreed. The problem is that both mini-public and statistical sampling are currently used for bodies with a purely advisory role along with public opinion surveys and deliberative polls. We need something to capture the “representation” element in Yoram’s phrase. This is the “tion” in my neologism (from elec-tion), the “stochas” element referring to the balloting mechanism (randomness). This provides a single word for “the appointment of political representatives by statistical sampling” (election being the appointment of political representatives by voting), the only problem being you have to explain how to pronounce it. If stochastion is the process then I suppose the resultant body would be a stochast (randomly-selected legislative jury).
LikeLike
Yoram,
>I like “representation by statistical sampling”. Due to the use of statistical sampling in opinion polling the term is established in the context of politics and the connotation is positive and relevant.
Given the clarity of this statement, I continue to be puzzled by your refusal to accept the implications for political representation of the aggregate nature of the representation provided by statistical sampling. A single respondent in an opinion poll is just a datapoint, with no autonomous status, and this would also be the case if the statistical sampling process were employed for the selection of political representatives. This is why stochastion can only lead to descriptive representation (standing for); substantive representation (acting for) is the role of empirical individuals, not statistical aggregates. Opinion pollsters don’t attribute any significance to the views of individual respondents, they only become (statistically) significant in aggregate, and the same principle applies to political representatives. Every survey form has an equal value and members of the stochast would be limited to actions that also have an equal value, and it’s hard to understand how this could apply to anything more than voting, without adversely affecting the fidelity of the representation (of the public by the mini-public).
Yoram prides himself in never responding to anything I write, so I invite anyone else to refute this argument on his behalf. The only possible rejoinder that I can imagine is that all the respondents are (more or less) cut from the same cloth, so it matters little which ones get to speak as they all share the same underlying interests. Given the origins of the deliberative democracy movement in Habermas’s Structural Transformation of the Public Sphere I think this is the most likely explanation.
LikeLike
Keith,
The other possible argument is one from fairness. If any selective process necessarily introduces nonrepresentative bias, then redefining political equality to mean having an equal chance of political participation could be defensible regardless of the consistency of the outcomes. It probably shouldn’t be called “democracy,” though.
LikeLike
Naomi,
Yes, but you’re right to insist that it shouldn’t be called democracy, as it has more in common with the “it could be you” slogan adopted by the National Lottery than the Representation of the People Act. Yoram has always insisted that democracy requires representation, so anything that undermines the descriptive accuracy of the mini-public is anti-democratic. So I’m afraid that the argument on the nature of statistical representation needs to be refuted on its own terms.
LikeLike
> Yoram prides himself in never responding to anything I write, so I invite anyone else to refute this argument on his behalf.
Another lie. I usually don’t respond to you simply because I see no use in doing so: I learn nothing from you and you learn nothing from me.
As for your argument, as usual you are regurgitating your old talking points, ignoring any responses. Obviously, government by an allotted chamber and an opinion poll are not the same thing. Peppering this obvious observation with important-sounding jargon does not result in a coherent argument.
My argument for the way sortition results in representative government is here.
LikeLike
Keith, you wrote,
>The problem is that both mini-public and statistical sampling are currently used for bodies with a purely advisory role along with public opinion surveys and deliberative polls.
I don’t see that as a problem. As I continue to talk to people about “mini-publics,” I’m planning to first explain what they are; then talk about the familiar examples (public opinion polls and juries), then talk about newer forms (deliberative polls, citizens juries) and explain that while most of the newer forms have been purely advisory, that there have already been some experiments with mini-publics that are more than that, and many interesting proposals to take the idea much further.
LikeLike
Yoram,
As you rightly say, we’ve been over this many times before, but I suppose it’s worth one final attempt:
>>1. A small group of people, under reasonably favorable conditions, is able to represent its own interests. . . The claim is that when a small group of people, meeting on an a-priori egalitarian basis, has the opportunity to make collective decisions that would promote the interests of the members as they perceive them, then it will tend to do so.
What exactly do you mean by “a priori egalitarian basis”? In the examples that you cite, families are certainly not egalitarian, and status and other inequalities (such as persuasive skills) are very pronounced in any group. At best “a priori egalitarian” can only mean a set of norms that should be aimed at, and deliberative democrats have to go to inordinate lengths to even approximate these norms, using a variety of procedures to monitor and regulate speech acts that would not be acceptable in a political body.
“The interests of the members as they perceive them”. Given that the deliberative body will be dealing with matters that are beyond the normal experience of many/most of the allotted conscripts, there is a potential gulf between esse and percipi. This is unproblematic from a liberal perspective (we are all entitled to enough rope to hang ourselves), but would be ruled out if the decisions of the group were to be binding on everyone else, as there is no obvious reason to believe that the un- or partially-informed preferences of the small group would represent those of everybody else (bearing in mind that you insist that the selection of information and advocacy has to be specified endogenously).
>>This is a situation which most people would be familiar with – group decision making in the family, within a group of friends or with colleagues. “A small group” is taken to be a group in which all-to-all communication is possible. . . a few hundred people seems like the most that would fit the description.”
There is no comparison between the voluntary conversation of a group of friends on matters of immediate interest and several hundred randomly-conscripted strangers, who will be obliged to deliberate on matters that lie beyond their experience and interest. Statisticians would prefer a group with a minimum size of 1,000 — what sort of conversation would they have, and how would you preserve the “a priori egalitarian” conditions, given that the speech acts of each member would have to be equalised in order not to compromise statistical representativity?
PS “a priori egalitarian” conditions are unproblematic, as is the size of the group, if the remit is restricted to listening to the arguments of competing and balanced advocates and then voting. This is the political analogue of the statistical survey example that you cite, not the exchange of reasons/preferences in a small group. Indeed if two respondents to a public opinion survey were found to have collaborated then those data points would be eliminated as no longer being representative of the target population. The free discursive exchange that you are proposing is that of the deliberative democracy tradition, which has little time for the representation of interests, preference aggregation etc.
LikeLike
Thanks David, I thought we were looking for an inclusive term to include all the elements in Yoram’s phrase (representation by statistical sampling), so that you don’t have to go through all the explanatory palaver!
LikeLike
>> A small group of people, under reasonably favorable conditions, is able to represent its own interests
If that were true, most people would represent themselves in court. It is preferable to be represented by someone who knows what they are doing.
>Statisticians would prefer a group with a minimum size of 1,000
This is a bit off topic, but I’m curious. What level of statistical significance would you be willing to accept? Let’s look at the 5% window around 50%. If intrinsic (post deliberation) support for a proposal in the population were 47.5%, what would be acceptable odds of passage due to a statistical fluke? One in fifty? One in a hundred?
LikeLike
Naomi,
>Let’s look at the 5% window around 50%. If intrinsic (post deliberation) support for a proposal in the population were 47.5%, what would be acceptable odds of passage due to a statistical fluke? One in fifty? One in a hundred?
I would be inclined to say that if a group of (say) 300 returned a positive verdict lower than 55% or even 60% that the matter would be referred on to a larger sample (c.1,000). But I’m not a statistician, political scientist, or constitutional expert and I seem to remember that John Garry was arguing for a minimum 1,000 sample size in every instance. But all these numbers are far in excess of the maximum group size for meaningful deliberative exchange (12-24), hence my claim that we have to view statistically-representative bodies as juries whose role is limited to listening to the arguments and then voting in secret.
LikeLike
Sutherland – as usual you are covering old ground.
LikeLike
Naomi,
>> A small group of people, under reasonably favorable conditions, is able to represent its own interests
> If that were true, most people would represent themselves in court. It is preferable to be represented by someone who knows what they are doing.
No. When they choose to do so, relying on expert services is part of the way individuals and groups represent themselves.
To reject the notion that an individual or a group can represent their own interests you would have to claim that it is best for them to have their affairs managed by an external agent even when they refuse to authorize such management.
You would, for example, have to believe that when a doctor decides that it is best for a person to undergo a certain medical treatment then that person should be forced to undergo that treatment even against their wishes.
LikeLike
Yoram,
>Sutherland – as usual you are covering old ground.
Yes, that’s why my appeal was for someone else to refute the argument that statistical representation is . . . statistical.
LikeLike
>You would, for example, have to believe that when a doctor decides that it is best for a person to undergo a certain medical treatment then that person should be forced to undergo that treatment even against their wishes.
Or you could just get a different doctor. Just like how you can elect a different representative. By the way, no one is arguing that elected officials should have the final power to decide policy. We all agree that the decision to “undergo that treatment” should be in the hands of the statistical sample.
LikeLike
You’re both guilty of argument by analogy (families, friends, lawyers, doctors) and conflating substantive (active) and stochastic representation. Statistical processes apply at the statistical (aggregate) level, as that’s what the word statistical means. If it’s true for an opinion poll then its true for any other entity constituted by the same process. The only difference between an opinion poll and a Deliberative (opinion) Poll, is that the latter is better informed. It really is that simple.
LikeLike
> Or you could just get a different doctor. Just like how you can elect a different representative.
So you would have people select a doctor once every few years (from a short list of drawn by the large HMOs) and then be forced to undergo whatever treatment that doctor orders until the next time they are allowed to switch?
> We all agree that the decision to “undergo that treatment” should be in the hands of the statistical sample.
The way you have it, the statistical sample must select a treatment from a short list of possible treatments offered by the different doctors.
LikeLike
> You’re both guilty of argument by analogy
You are guilty of being a pompous fool. Since no two situations are exactly the same, all argument is by analogy.
LikeLike
>So you would have people select a doctor once every few years (from a short list of drawn by the large HMOs) and then be forced to undergo whatever treatment that doctor orders until the next time they are allowed to switch?… The way you have it, the statistical sample must select a treatment from a short list of possible treatments offered by the different doctors.
You have to go to a doctor who actually exists and is licensed to practice medicine. That limits your options. If you go in for a second, third, or nth opinion you are only going to hear a limited number of options. As long as the choices that reach the allotted sample are sufficiently diverse, there is no problem. They are free to pick and choose what they’d like. No one is forcing any particular option. In a lot of ways, having the allotted members make their own proposals is much more limiting. The range of viable options fleshed out by the active subset of members and presented to the full sampling will surely be much smaller.
LikeLike
Yoram
> All argument is by analogy.
Not so, most philosophical argument is attempting to clarify the meaning of words.
LikeLike
Yoram, of course, is trying to have his Habermasian cake and eat it, conveniently ignoring the fact that deliberative democrats disdain grubby things like the representation of interests.
LikeLike
> The range of viable options fleshed out by the active subset of members and presented to the full sampling will surely be much smaller.
Naturally, the allotted are too incompetent and/or too lazy to make sure that a diverse set of options is considered. We should all be thankful that the competent, diligent and benevolent elites are willing to carry the intellectual burden of considering a wide range of ideas and spoon-feeding the allotted with the necessary information to make an informed and considered selection among the choices, including making selections that are detrimental to the interests of those same elites.
Some people’s selflessness and devotion to serving the public interest is truly amazing. We are so lucky to have such people making up a disproportionate part of the ranks of the big and mighty.
LikeLike
Yoram, I don’t know what medicine you are on, but I think you must have missed a dose. I’m used to your insults, but the torrent of sarcasm aimed at Naomi is inexcusable (especially from the forum’s convenor)
> some people’s selflessness and devotion to serving the public interest.
I thought the discussion was on doctors, if so then Adam Smith would argue that they are merely trying to earn their dinners. Given that doctors have spent many years studying medicine, one would expect their suggestions to be more valuable than those of lay persons, irrespective of their devotion to the public interest. The relevant criterion is knowledge, rather than virtue (it’s the job of the allotted assembly to assess the virtues of each proposal)
LikeLike
Naomi’s point regarding statistical significance is crucial. One of the things that amazes me about these debates is the almost total absence of attention to questions of statistical significance. This is despite the fact that one of the core aims is to use the random sample is to infer to the general population from which the sample was drawn, i.e the sample is a proxy for the whole. Inferential statistics is the mechanism for the legitimacy of the entire enterprise and should be seen as pretty important. Market research companies ask at least a 1000 people and don’t let them talk to each other because it would completely screw up the sample; one couldn’t infer from the sample to the whole population and hence the data would be useless for the client. Similarly the ability to infer from 200 people or so is very poor even if they remained independent of each other (i.e. they didn’t talk to each other). The error around any estimate regarding a decision made by 200 would mean a lot more than 50% of the 200 would have to vote ‘Yes’ in order to be able to claim that a majority of the population from which the sample was drawn would vote yes. Hence, a specific super majority is necessary, otherwise we are dealing with completely arbitrary interpretations of what the ‘decision’ is. But once you let the 200 talk to each other, even this large supermajority won’t do any work for you because the undermining of the assumption of the independence of the sampled persons (i.e. their talking) renders any ‘decision’ non generalisable to the wider population. Hence, a pretty big plank of the legitimacy of the entire process has gone.
With respect to terms I like mini-public. Deliberative mini-public is good too. Citizens’ jury links directly to people’s liking of juries and their understanding that juries are randomly chosen. However all juries are made up of citizens and so really it’s ‘political juries’ that might be most accurate. It sounds a bit dodgy though, as political implies bad… Perhaps Policy-Making Juries?
LikeLike
Sorry, forgot to sign the above post on ‘statistical significance’: John Garry
LikeLike
A few months ago I completed a full reread of the archives, from the very beginning all the way up to when I started posting. Just to make sure I didn’t miss anything. And I still don’t understand Yoram’s visceral reaction to anything involving “elites.” It’s particularly odd given that he is willing to trust the powerful unconditionally… so long as they are drawn by lot.
Yoram, you suggest I have contempt for regular people. On the contrary, I suggest you are the one who has made your contempt for the people plain. You seek to deprive 99.9999% of the population of their right to participate in politics and be represented in THEIR government on THEIR own terms on the idea that they are so helpless that almost all of them would allow themselves to be gamed by the “elites” into acting against their own interests. People do not have drop-in equivalents. There is no one else in this world similar enough to me that they can act on my behalf without my consent. I have within my faculties the ability to make a reasonable decision on who to vote for to advocate on my behalf in the legislature even though I haven’t won any lotteries. Winning the politics lottery gives you more time and more incentive to apply one’s self. It doesn’t turn a gullible, manipulatable fool into a wise man. An allotted assembly has DEMOCRATIC legitimacy only because support for proposals is something that can be represented by a statistical sampling in aggregate (with some margin of error, of course). As soon as you go outside of aggregate behavior you lose any and all claim to democratic legitimacy.
Suggesting that the rich and powerful can influence politics enough that they can decide the balance of power by pushing public opinion a few percentage points one way or another is one thing. You can make a serious case for that. However, in the case of a unicameral hybrid, for the elites to control the results they’d need to have almost 100% of the votes locked up. All that is needed is a sufficient degree of diversity of goals/values/ideology/proposals/etc. Elections can give you this with no difficulty.
So then on what grounds do you wish to disenfranchise almost everyone? Hmm? That’s the extreme option. It’s the sort of thing that should only be considered in the absence of any alternative. The sort of thing that could only be achieved if there was a consensus that there was no alternative. Of course that’s NEVER going to happen because there actually are fine alternatives. The public at-large is, and should be, the default… something we only move away from to gain something tangible.
So then. Enlightened us. Please explain plainly what tangible gains are had by locking out all but the chosen few.
LikeLike
John,
My preference would be for a double vote system. If a modest supermajority (say 55%) vote for a proposal it passes into law directly. If a proposal is passed by a smaller margin, it would have to be passed again a few months later, after all of the current randomly drawn members have been replaced (I favor terms of a few months for the allotted with some portion being replaced each week). So two successive samplings would have to be slightly nonrepresentative in the same way. Or one sampling would have to be off by a huge amount. If the odds of a proposal passing due to a statistical fluke are 5%, the odds of it happening twice in a row are only 0.25%. If the odds a bill will pass once are 95%, the odds of it passing twice in a row are 90.25%. I’m not fond of status quo biases, but this strikes me as an acceptable tradeoff.
LikeLike
> You seek to deprive 99.9999% of the population of their right to participate in politics and be represented in THEIR government on THEIR own terms
In what way was it suggested that anyone is deprived of participation in politics?
How is having elites set the agenda a way to have the population “participate in politics and be represented in THEIR government on THEIR own terms”?
It seems that you are sliding toward sloganeering as a substitute for substantive discussion.
LikeLike
John,
I think you are the only political scientist who is active on this forum so we need to take your concerns on statistical significance very seriously. The reason that most of us have plumped for an allotted group of “several hundred” are partly historical (the size of the Athenian council and typical elected assemblies) and partly on account of the rational ignorance threshold (if there are too many participants, then people won’t bother to concentrate as their single vote is of diminished causal efficacy). There are also cost constraints. I believe that Fishkin’s DPs are only around 300 and he does like to claim that the informed judgment of the DP microcosm can be extrapolated to the full population — is he misguided in this respect, and if so why does a competent statistician like Bob Luskin let him get away with it?
Can you put some meat on the bones of Naomi’s suggestion? If you had a silent voting group of (say) 300, how large would the supermajority have to be in order to be a reliable indication of “how everyone would think under good conditions”? To my mind if the vote didn’t achieve the necessary supermajority, then the matter would have to be referred to a new trial with a jury of 1000+.
The trouble with “deliberative mini-public” is that most people equate deliberation with speech acts and there is nothing to say that this is a body with a legislative mandate. I think Yoram’s “representation by statistical sampling” is what we need to try to encapsulate in a single word, and my candidate is stochastion.
>Market research companies . . . don’t let [respondents] talk to each other because it would completely screw up the sample; one couldn’t infer from the sample to the whole population.
Absolutely. This is a knock-down argument for silent deliberation and voting. There is also a long tradition of republican thought that favours this approach.
Yoram [to Naomi]:
>It seems that you are sliding toward sloganeering as a substitute for substantive discussion.
[Pol] pot is up to his usual tricks (calling the kettle black).
LikeLike
Some who are demanding that sorted bodies be statistically representative is that they seem to want each and every decision to be made by such bodies. But as we can see with small juries (12 for trial, 23 fpr grand) that the design objective is not that each jury be representative, but that many juries be representative for an aggregate of many decisions, averaging out over time.
This is discussed in “Jury size matters” at http://constitutionalism.blogspot.com/2007/08/jury-size-matters.html
LikeLike
Jon,
The judgment of the trial jury is epistemic (guilty m’lud) whereas the role of the political jury is to register an opinion/preference, so real-time representativity is required. As with a public opinion poll, where we want to know the opinions/preferences of the whole public on a single issue, a political jury seeks to do the same for political decision-making on an issue-by-issue basis. Diachronic representativity will not do for doxa, although it might possibly suffice for episteme (although a disadvantaged black defendant, convicted by an all-white jury might beg to differ!).
LikeLike
@Keith
First, only a minority of trial verdicts are in criminal cases. Most are in civil cases, where the jury is asked to decide issues like degree of liability and how much if any damages to award. So they do not, for your analysis, differ much from what you call “political” (distributive) juries.
Second, most political decisions that a legislative body is asked to make are not sweeping, earthshaking, or profound, as, say, a constitutional amendment or a budget would be, but decisions about minutiae of legislation containing thousands of clauses, each of which is a separate decision. Moreover, no such decision is permanent, but may be, and likely will be, revised or repealed in subsequent sessions. Many statutes in the U.S. have been amended hundreds of times over 200 years.
And more important decisions commonly require a supermajority to pass, if not as a formal rule then in the ways a deliberative body will tend to form veto groups that, although each might be a minority, can effectively block action.
In politics and law no issue is ever finally settled.
So if the objective is to hold unjust verdicts below some threshold like 5% on the average of a hundred thousand cases a year, one can expect that 1.2 million jurors may achieve that, 12 jurors for each case. Indeed, more likely than if they all met and decided in a single session.
LikeLike
Yoram,
That’s funny. You repeat the same tired trope about “elites” and then accuse me of sloganeering. Just because a person has to have to have some measure of public visibility to win an election does not mean the “elites” have the power to set the agenda. That would be an example of collusion and yes, there are examples of it. Preventing it is collusion of the reasons why I favor PR elections as a general rule and I thought the 7 party debate illustrated quite well the benefits of encouraging a broader playing field. The public sets the agenda through the “elites” in a manner analogous to a plaintiff setting the agenda in a legal fight through their lawyer. Is there agency loss? Absolutely. Loads. Great heaping gobs of it. But focusing entirely on agency loss as though it were as the root of all political evil is simply not justifiable.
Being a member of an allotted body means you have more time and motivation to apply yourself to politics. That’s all. Not being in an allotted body does not make you incapable of doing a decent job. And all we need is for a decent number of people to do a decent job. Didn’t you live in Israel for a while? Can you honestly tell me you could not build an acceptable policy platform by taking apart the platforms of all the parties that managed to clear the threshold in the recent election and copying, piecemeal, just the bits you like? I imagine the great majority of Israelis could. If you can’t you might want to consider the possibility that your views are unlikely to prevail in any democratic institutional structure.
LikeLike
Keith,
I’m neither a political scientist, nor mathematician, but I’m okay with Excel. A few months ago I set up a spreadsheet to solve the binomial probability formula for a few dozen seats past 50% and sum the results. I didn’t share it before because there’s probably an error somewhere (perhaps in the methodology itself) and I didn’t want to make a fool out of myself. Also, it can’t go past 919 seats due to maximum cell number size limitations.
Just put in the size of the assembly and the intrinsic support level in the population (after deliberation) and it’ll give you the odds that randomness will push the results past 50%, 55% and 60%. Or at least that was the idea. I have no idea if these numbers are right or not. They seem reasonable, given the margin of error in opinion polls. Please don’t hate me if they’re wrong. Perhaps Yoram could explain a better way of doing this.
https://docs.google.com/spreadsheets/d/1C4KrkIiW0jiVx99hAGHUqX3E-WqGIjMgOlGcO7ulvwg/edit?usp=sharing
LikeLike
Naomi,
> your views are unlikely to prevail
It is your view (that the parties somehow cover the range of possible policies) that is unpopular. It is held by a small minority of people who have been so thoroughly indoctrinated as to hold on to the electoralist dogma in the face of all the evidence to the contrary. Large majorities all over the world believe that the electoral elite represents its own interests and its own interests only.
But let’s put this issue aside and assume that you are right and indeed somehow the positions of the parties (real positions, not official campaign positions) on the various issues can be used as raw material from which popular policy can be made. Who will carry out the policy? Policy doesn’t follow mechanically from legislation. Unless executives are hired and fired by a representative body and continuously monitored and guided by that body, there is no reason to expect the executives to carry out representative policy, no matter what the legislation is.
> tired trope
Spare me. I know when you are actually engaging with the substance of the matter (as you are doing now) and when you are not bothering to put in the effort and rely instead on slogans.
LikeLike
Jon,
>First, only a minority of trial verdicts are in criminal cases. Most are in civil cases, where the jury is asked to decide issues like degree of liability and how much if any damages to award. So they do not, for your analysis, differ much from what you call “political” (distributive) juries.
Not so, this is still an epistemic judgment as the jurors do not stand to gain or lose personally as a consequence of their decision (as would potentially be the case with a legislative jury).
>decisions about minutiae of legislation containing thousands of clauses, each of which is a separate decision.
I wonder if that is a suitable matter for a randomly-selected jury? — either it should be a delegated housekeeping matter (for the executive) or else the jury should just make the one decision (whether or not to pass the bill). The Athenian council was a collective magistracy, not a sovereign body, yet it still passed routine housekeeping laws and there needs to be a modern (executive) analogue (such as Orders in Council in the UK).
>So if the objective is to hold unjust verdicts below some threshold like 5%
In a legislative court there would be no such thing as an unjust verdict (although decisions that contradicted fundamental human rights would be quashed by non-democratic constitutional checks).
Naomi,
Political theorists like me are happy with vague notions like “statistical representativity”, it’s up to quants like John and yourself to operationalise it — I’m not qualified to comment one way or the other.
Yoram,
>a small minority of people who have been so thoroughly indoctrinated as to hold on to the electoralist dogma
(along with the billions of people who still take the trouble to vote in elections all round the world). But Naomi and myself (along with sundry political theorists and other lickspittles) are beneath contempt for attempting to provide a legitimising narrative for what is clearly an elite conspiracy (according to “large majorities all over the world”). Come the revolution we’ll be the first ones up against the wall.
LikeLike
@Keith
>Not so, this is still an epistemic judgment as the jurors do not stand to gain or lose personally as a consequence of their decision (as would potentially be the case with a legislative jury).
Personal gain or loss is not a valid definition of what kind of judgment it is (unless it is corrupt or non-corrupt). Holding down the risk of unjust verdicts in a trial or unrepresentative judgments in a statute benefits both jurors and the public in the same way, unless the jurors are deciding their salary and benefits. The point of having legislatures being representative is to prevent them from deciding to benefit special interest groups, or burden disfavored groups, more than their proportion of the population. Reducing that public choice problem is the advantage sortition or eduction brings.
>I wonder if that is a suitable matter for a randomly-selected jury? — either it should be a delegated housekeeping matter (for the executive) or else the jury should just make the one decision (whether or not to pass the bill).
And who and how is it decided what is a “housekeeping” matter? Your “housekeeping” decision may be a matter of life and death for someone else. the modern administrative state does something like that, in issuing thousands of regulations every year that have the force of law (contrary, in the U.S., to Art. I Sec. 1 Cl. 1 of the U.S. Constitution that forbids delegation of lawmaking). They are supposed to be a matter of providing details for, or interpretations of, the general statutes, but often go beyond the statutes in important ways. In principle Congress may vote down any of those regulations, but there are too many of them issued too fast for Congress to even read all of them, much less deliberate and vote on all of them. If Congress had to vote up or down on every regulation they would never get to most of them or have time to do anything else. Of course some would argue that regulations should apply only to government actors and not have the force of law for the general public, which is what they are supposed to be, but it is difficult to prevent a regulation that tells a government agent to do something to some member of the public if he does or doesn’t do some action from being a way to tell that member of the public to do or not do that action, and thus regard it as a law governing him.
Much of this discussion is at a level far disconnected from the ways government decisions have to be made to deal with real world problems in the present age. We are not in ancient Athens anymore.
LikeLike
Although it would be difficult to get agreement on the risk of getting an unjust verdict in a jury trial or an unrepresentative statute from a legislature, I would estimate that in the U.S. today at the federal level it is more than 30% in jury trials, mainly because of the ways trial juries are manipulated by judges and prosecutors, and more than 60% in legislation, mainly because of the ways legislators are manipulated by special interest groups. It is the perception of that that has led to so much public anger. The risks are even higher in some other countries, and the public anger in them is often to the point of revolution, unless the public has descended into despair.
Rather than seeking some perfect ideal, we need to look to ways to reduce the risks to a more tolerable level. No human judgment will ever be prefect in all cases. Any reduction in those risks is worth striving for, even if it never reaches what we might consider tolerable.
LikeLike
Yoram,
>It is held by a small minority of people who have been so thoroughly indoctrinated as to hold on to the electoralist dogma in the face of all the evidence to the contrary.
Do you really believe most people in the developed world would say “no” if they were asked, “do you support the principle of representation through election”? It’s entirely possible most people in most countries believe their political system is broken. They aren’t wrong. But there’s a world of difference between rejecting the outcomes of the current system and rejecting the principle of representation through election. There’s a world of difference between believing that the elected officials *in aggregate* fail to represent you and believing that *none* of the elected officials represent you. It’s pretty darn common for individual presidents and prime ministers to have approval ratings of well over 50%. As dysfunctional and unpopular as Congress is, the *individual* members have approval ratings that aren’t too terrible. Why would someone approve of a representative who doesn’t represent them?
>Unless executives are hired and fired by a representative body and continuously monitored and guided by that body, there is no reason to expect the executives to carry out representative policy, no matter what the legislation is.
I agree. The administrators should be appointed and dismissed through ordinary lawmaking procedure. I still think there’s a role for an elected public-facing figure of some sort. But I’m willing to yield and reduce it to a ceremonial position.
I’m not entirely sure how pointing out that people are not drop-in replacements for each other is sloganeering. The people are represented in the sampling only in aggregate and only to some margin of error. That’s it. If there’s a social dynamic between members, as John pointed out, the representivity of the sample will fall apart. If you give power to the individual members to be exercised as an individual prerogative then the sampling loses any claim to democratic legitimacy and you genuinely are disenfranchising everyone who was not drawn into the sampling.
LikeLike
Jon,
>The point of having legislatures being representative is to prevent them from deciding to benefit special interest groups . . .
That’s the blind break (prophylactic) argument. Those of us who argue the case for stochastion claim that it is a way of implementing (by proxy) the considered judgment of everyone on each and every issue, and this requires large juries who deliberate in silence. Unless we separate these two arguments we will continue to talk past each other.
>If Congress had to vote up or down on every regulation they would never get to most of them or have time to do anything else.
Yes, that’s why we do need to make the (imperfect) distinction between laws and regulations. One solution might be to aggregate the regulations into multi-clause bills and to leave it to the advocates (and media) to draw the attention of the jury to clauses that they might find objectionable, rubber stamping all the rest.
>Much of this discussion is at a level far disconnected from the ways government decisions have to be made to deal with real world problems in the present age.
Yes, that’s why I limit my own role to pointing out philosophical distinctions, leaving it to others (like yourself) to flesh them out for real-world application. But this means we need to agree on what we are talking about — you have argued the case for sortition as a prophylactic, so we’re not going to agree on the implementation if we have a basic disagreement on whether random selection is for sortition (the blind break) or stochastion (representation by statistical sampling).
Naomi,
>I’m not entirely sure how pointing out that people are not drop-in replacements for each other is sloganeering. The people are represented in the sampling only in aggregate and only to some margin of error. That’s it. If there’s a social dynamic between members, as John pointed out, the representivity of the sample will fall apart. If you give power to the individual members to be exercised as an individual prerogative then the sampling loses any claim to democratic legitimacy and you genuinely are disenfranchising everyone who was not drawn into the sampling.
Hear, hear. Strangely enough this argument appears blindingly obvious to most people (who do not have a pre-theoretical commitment to nineteenth-century political anthropology), both on the level of the use of words and direct experience of real-life group dynamics, in which “a priori conditions of equality” are only a Habermasian dream.
PS I can’t make my mind up if the correct word is representivity or representativity. WordPress seems to privilege the latter.
LikeLike
>>I wonder if that is a suitable matter for a randomly-selected jury? — either it should be a delegated housekeeping matter (for the executive) or else the jury should just make the one decision (whether or not to pass the bill).
>And who and how is it decided what is a “housekeeping” matter?
I’m inclined to agree with Jon here. However, delegated lawmaking power doesn’t bother me. I trust the various regulatory agencies far more than Congress. As long as the legislature can intervene quickly enough and congruency between the lawmaking process and the executive dismissal process is maintained I see no problem with the free delegation of lawmaking powers.
In any case, many substantive details of a major new law will end up being contested. It’s probably better to contest them beforehand. I’d much rather give the advocates the ability to bounce ideas off the sampling and gradually, iteratively, build up larger bills rather than passing them in one go.
LikeLike
This discussion has moved way off topic, and I didn’t want to contribute to that… but I think I should respond to a couple of Keith’s questions. I am not authorized to “represent Yoram ;) but let me try to defend a fully authorized randomly-selected mini-public.Remember that my OWN scheme (multi-body sortition) does not do this…as I want to have separate (randomly selected) bodies tackle each separate function, because merging functions harms the product.
I think the key is that EXISTING elected chambers mix together the agenda-setting, policy development, fact-finding, deliberating, and decision-making functions. The argument for a full-charge sortition mini-public is to simply swap out the method of selecting the members of the representative body so as to avoid the consistent bias of elected chambers. The goals is not to have all outcomes EXACTLY match the decisions that would be made if all citizens could be engaged in deliberation perfectly (as Keith feels is essential), but rather, as Constitutionalism (Jon?) wrote above: ” that the design objective is not that each jury be representative, but that many juries be representative for an aggregate of many decisions, averaging out over time.” With this goal in mind, a full-charge sortition mini-public would do a better job than any elected body at accurately representing the general population. (Note that MY scheme far exceeds this low standard).
In other words all of Keith’s objections to speech acts by allotted representatives are JUST as applicable for ELECTED representatives. Choosing between a handful of candidates who have been vetted by powerful elites (in the nomination process) assures that many good policy options will never see the light of day. They won’t be mentioned in the campaign nor discussed by the media, even though they might well surface in a statistically representative allotted agenda setting group.
The act of deliberating (and lobbying, etc.) distorts the ELECTED representatives from how they presented themselves in the election campaign. Some representatives will be more skilful persuaders, etc. (all of the points Keith makes). This distorts the outcomes of an elected chamber every bit as much as it would an allotted one. No two ELECTED chambers would be any more likely to make identical decisions as any two allotted chambers. The standard Keith wants to apply to a sortition mini-public has NEVER been applied to elected chambers. But, some of Keith’s points have merit, and that is why I want each element of the legislative task (agenda-setting, policy development, fact-finding, rule-making, debate, decision-making) to be carried out by a separate body well-suited to the task (and, yes, the final decision made by a jury that only listens and votes by secret ballot). This includes allowing absolutely ANYBODY who wishes to participate in the proposal development process (I call them Interest Panels), but only a final policy jury decides yes or no.
LikeLike
Naomi,
It occurs to me that if the reason you think it is useful to have parties present legislation proposals is to increase the range of ideas considered, then you would not insist that the choice be made from those proposals alone. The range of ideas could be further increased by having allotted delegates amend those proposals as they see fit and/or add their own proposals.
LikeLike
Terry,
I have no wish to defend our existing (all electoral) arrangements — that’s why I’m an advocate for stochastion as a crucial element in a mixed system of government. But I think you do need to address the issue of ‘perceived’ legitimacy (or empirical legitimacy, as Jane Mansbridge puts it). I agree that the choice of candidates/policies is determined by party elites, but their choices are constrained by the need to put forward candidates/policies that are likely to receive public support and voters do exercise their free choice. In an all-sortition system the vast majority of citizens would be left disenfranchised, without any choices whatsoever. The argument (by John Burnheim and yourself) that perceived legitimacy will be a product of epistemically benign outcomes is, unfortunately, untestable (without first instituting the kleristocratic putsch).
Regarding Jon’s claim regarding the diachronic representativity of small juries, he has already outed himself as a Blind Breaker — sortition being merely the prophylactic aspect in the hybrid that he refers to as ediction. But the concern of this thread is political representation by statistical sampling, so his remarks are off-topic.
As for your role as Yoram’s representative on earth, his own defence of full-power sortition does not just claim that its no worse than our existing electoral arrangements. His argument is an ideal one, based on a thought experiment extrapolated from the notion of small groups (families, friends etc) representing their interests under conditions of “ex-post equality”. The problem is that when the problems with his thought experiment are pointed out he refuses to engage in debate, accusing his intellectual opponents of re-visiting old ground, dogmatism etc.
LikeLike
Sorry, that should have read “a priori egalitarian”
LikeLike
Yoram,
>The range of ideas could be further increased by having allotted delegates amend those proposals as they see fit and/or add their own proposals.
While this would increase the range of proposals that go to a vote it would also damage the sampling’s ability to vote on behalf of the people.
I have absolutely no problem with having some portion of the advocates drawn by lot as long as they are a separate group of people from the ones who cast the votes. We’ve previously discussed the possibility of having a “none of the above” option on the ballot in legislative elections. If 30% of the voters vote “none of the above” 30% of the seats go to people drawn by lot from the general population. Something like that for the advocates would be fine with me. In fact, I really like that option. Or simply drawing a hundred advocates by lot in parallel with the elected advocates would be okay too. As long as there are enough elected advocate seats to represent a high degree of policy diversity through the ballot box and the fidelity of the voting sample is preserved I’m fine with opening up the floor.
Terry,
The difference is that, in an elected legislature, the majority of those interested gave their written consent to the winners. They then retain the power to punish the winners if they misbehave. Neither is true in a body drawn by lot.
LikeLike
Naomi,
>As long as there are enough elected advocate seats to represent a high degree of policy diversity through the ballot box and the fidelity of the voting sample is preserved I’m fine with opening up the floor.
Then why not open it up to everybody, via some sort of direct-democratic competitive process (crowd-sourcing, online petitions, information markets etc). The allotment of policy advocacy rights has no representational value, so why would we want a process dictated purely by chance? Out of the 100 or so allotted, only a tiny number would be likely to be able and willing, so it strikes me as entirely happenstance. If randomisation is deemed important (for prophylactic reasons) then rather than narrowing down the direct-democratic proposers by votation, you could pick a few out of a hat.
LikeLike
I must admit, I’m attracted to the “none of the above” concept. If you don’t like your options, you can just vote in favor of tossing for it. And there may be prophylactic benefits too. Sure. If nothing else it does no harm. I’m trying to find something that will meet with the approval of most of us. Yoram has previously said he has no strong feelings on unicameralism vs bicameralismism. Splitting functions between different groups of allotted members should presumably fall into the same vein. And presumably the allotted voters will be able to stand on their own two feet and pick and choose the options that make sense to them, so having elected advocates in parallel shouldn’t be too bad.
I’m not convinced the initiative is very useful in a unicameral system. We don’t have to discourage fragmentation and we don’t need to get the ratios between the various movements, ideologies, values, etc. exactly right. So it shouldn’t be very difficult to capture everything that could be represented with an initiative in a well-designed electoral system.
LikeLike
The danger is mixing up again the various potentials of the lottery process (as a political theorist, my job is to clarify the distinctions). If you mix them up then you run the risk of all kinds of non-sequiturs, like the (erroneous) argument that replacing election with sortition will put an end to factionalism and corruption. If someone opts for none of the above then she gets nothing — the goal of constitution designers should be to make sure that there will always be something of value on offer.
LikeLike
The option does not offer the voter representation, but it does offer her a blind break. If someone values a break over representation, and *chooses* that, as far as I’m concerned, that’s their business, not mine. We must be careful with the voting members, as they will vote on behalf of everyone. But having the portion of the public who would prefer a break over representation representated as a proportionally sized break in the advocacy seems entirely reasonable.
LikeLike
Naomi,
Yes, that does indeed seem like a much more reasonable arrangement – although of course still containing a significant oligarchical element.
By the way, regarding the “none of the above” procedure which you suggest, shouldn’t those who don’t vote at all be counted with the “none of the above”? Why should they be disenfranchised?
There is also reason to consider having those who voted for anyone other than “none of the above” disqualified from the sortition pool. Why should those people get two shots at representation?
LikeLike
Yoram,
>shouldn’t those who don’t vote at all be counted with the “none of the above”? Why should they be disenfranchised?
Because that was their choice.
>There is also reason to consider having those who voted for anyone other than “none of the above” disqualified from the sortition pool. Why should those people get two shots at representation?
As Naomi made clear, this is just a case of the blind break. It’s a variant on Russian Roulette, not political representation.
LikeLike
Naomi,
You wrote:
>The difference is that, in an elected legislature, the majority of those interested gave their written consent to the winners. They then retain the power to punish the winners if they misbehave. Neither is true in a body drawn by lot.
Firstly, citizens can consent (by referendum) to be governed by sortition with mini-publics just as validly as by electing specific people.
Secondly, The notion that elections can hold representatives accountable is only a myth. I can give numerous reasons for this, but will just quote Prof. Alex Guerrero explaining one of the key reasons. After discussing fixing problems like money in politics, news ,media, gerrymandering, voter registration hurdles, etc. he notes:
“Even if these problems were addressed, they would succeed only in making elections fair. But meaningful accountability requires not just open and fair elections; it also requires that we are capable of engaging in informed monitoring and evaluation of the decisions of our representatives. And we are not capable of this. Not because we are stupid, but because we are ignorant: ignorant about what our representatives are doing, ignorant about the details of complex political issues, and ignorant about whether what our representative is doing is good for us or for the world.
“Our ignorance means that representatives can talk a good game, and maybe even try to do a few things that benefit the majority of us, but the basic information asymmetries at the heart of the representative system ensure that, for many issues — defence manufacturing and spending, policy that affects the insurance and pharmaceutical industries, agribusiness policy and regulation, energy policy, regulation of financial services and products — what we get is what the relevant business industries want. In the presence of widespread citizen ignorance and the absence of meaningful accountability, powerful interests will effectively capture representatives, ensuring that the only viable candidates — the only people who can get and stay in political power — are those who will act in ways that are congenial to the interests of the powerful.”
You can read Guerrero’s full essay here:
http://aeon.co/magazine/society/forget-elections-lets-pick-reps-by-lottery/
LikeLike
“Sample” – short form
“Citizen Sample” – long form
“Small sample” – when a clear majority emerges
“Large sample” – when a small sample is not reliable enough
“Polical Sampling” – the political decision process based on the efficient selection of a representative sub-set of all the citizens.
LikeLike
The trouble is none of these captures all the elements of Yoram’s phrase (representation by statistical sampling). I think we either need to keep the original four-word phrase or else resort to neologism. My preference is for the latter on account of the parallel with election (everybody knows that election stands for representation by preference balloting). Over time a single word (like stochastion) would gain the same kind of recognition. The trouble with sortition is it has no inherent meaning and it also refers to the selection of persons by an arational process and that’s the opposite of stochastion, which is an entirely rational process (being based on the stochastically-determinate ratio of the properties of the sample with the whole population).
LikeLike
Terry,
>Firstly, citizens can consent (by referendum) to be governed by sortition with mini-publics just as validly as by electing specific people.
Yes, but (even in the event of that happening) the issue is not the original social contract, it’s the ongoing consent to policies and the persons that implement them. I agree with Fishkin that the consent-by-proxy of an allotted microcosm is more meaningful than the tacit/approximate/hypothetical consent instituted by election, but that places very serious constraints on the actions of the allotted microcosm (in order not to undermine its representativity). Absent these constraints (as in Yoram’s proposal) then electoral consent is better than nothing.
LikeLike
Terry quoted Alex Guerrero about why elections don’t provide sufficient accountability,
““Even if these problems were addressed, they would succeed only in making elections fair. But meaningful accountability requires not just open and fair elections; it also requires that we are capable of engaging in informed monitoring and evaluation of the decisions of our representatives. And we are not capable of this. Not because we are stupid, but because we are ignorant: ignorant about what our representatives are doing, ignorant about the details of complex political issues, and ignorant about whether what our representative is doing is good for us or for the world.”
Terry and I wrote, “Having one opportunity every 2-6 years to “throw the bums out,” when legislators may vote on hundreds of bills each year, is an accountability mechanism that cannot possibly work.” (the paper can be found here – http://stwj.systemswiki.org/?p=1407)
LikeLike
Oops – I left out a sentence. I meant to say that I wanted to another reason why elections don’t provide meaningful accountability. Aside from ignorance, it’s the infrequency of elections, compared to the frequency of the decisions that they’re supposed to provide feedback on.
LikeLike
A
LikeLike
Mini-Public is accessible. I like it a lot, so I intend to use it for a local Magna Carta anniversary event on June 14th 2015. Hope you’ll all come – I need your support!
Perhaps the most interesting and fun way to see how people feel exploring new methods of suffrage in the 21st century is to invite them to try it out, amongst a bunch of friends & strangers alike over tea & crumpets.
“LiberTeas” was initiated by Parliamentary Outreach UK to celebrate Magna Carta & it can be held by any individuals or groups nationally (but you’re welcome to help me).
My version of the event [in SW Herts] will be ‘free to enter’ and will take place in an AONB by our local chalk river.
LikeLike
Terry,
We’ve been over this, and yes, I’ve read the article. Elections offer accountability with considerable agency losses. We can go another twelve rounds over how great those completely unquantifiable agency losses are, but claiming that they are well and truly 100% is not defensible. We all agree that elections are too clumsy a tool to decide the balance of power. That said there are a few elected officials (even in the US) I actually kinda like. Not many, but a few. I reject the idea I’d be better represented by an individual drawn out of the phonebook sight unseen over someone an individual or organization I choose myself. At least if the field is broad enough, that is. If you feel we’d do better with allotted advocates, then by all means, vote none-of-the-above. That’s your choice. I’ll be represented on my terms and you can be represented on your terms. Is not letting everyone make that choice for themselves the fairest option possible?
A none-of-the-above option is much more elegant than having binary and periodic referenda which would have many of the same basic problems as referendum democracy and invite a destabilizing number of constitutional revisions. If Yoram is right, then bootstrapping a pure sortition system would be quick and easy. If Keith and I are right, and elections are basically indispensable, then at worst we’d have done no harm. Experience can decide what debate on this blog never will.
LikeLike
Naomi,
You wrote,
>Elections offer accountability with considerable agency losses. . . . but claiming that they are well and truly 100% is not defensible.
I don’t think Terry is claiming 100% agency loss – just enough to make sortition preferable to election.
>I reject the idea I’d be better represented by an individual drawn out of the phonebook sight unseen over someone an individual or organization I choose myself.
All the sortition proposals I know of are about being represented by groups, not individuals. I’d suggest that you would be better represented by a statistically representative group, whose members didn’t owe anything to donors and weren’t seeking re-election, than you would be by a group that was chosen by elections. This would be especially true with single member districts and first past the post elections – in the U.S., huge numbers of people are “represented” by people that they did not vote for. But I think it would still hold true with multi-member districts and proportional representation.
You wrote,
>Experience can decide what debate on this blog never will.
Absolutely! I would love to see people’s ideas about how to set up experiments that would test these ideas on a small scale – the next steps to Citizens Juries, Deliberative Polls, Planning Cells, Consensus Conferences, and Oregon Citizen Initiative Review.
LikeLike
>I don’t think Terry is claiming 100% agency loss
Of course not. But as it is unquantifiable, the best we can do is discuss it in the vaguest terms. And as we have no modern experience with powerful allotted bodies we can’t even do that much for sortition. I’m sorry if I’m overly terse but the nature of the discussions on this blog can be trying at times.
>All the sortition proposals I know of are about being represented by groups, not individuals. I’d suggest that you would be better represented by a statistically representative group, whose members didn’t owe anything to donors and weren’t seeking re-election, than you would be by a group that was chosen by elections
All the proposals where individual members take an active role involve both representation by group and representation by individuals. I believe this is the point Keith has been trying to hammer home for the last five years.
In a unicameral set-up, where elected members propose, and allotted members vote, the elected members constitute representation by individuals and the allotted members constitute representation by group. There is no balance-of-power decided by election and so you aren’t represented by the group of elected officials in aggregate. The question is whether you can find someone out of the large number of viable candidates that can represent you.
LikeLike
Naomi,
>All the proposals where individual members take an active role involve both representation by group and representation by individuals. I believe this is the point Keith has been trying to hammer home for the last five years.
Exactly. That’s why I’m uneasy about the none-of-the-above option as a way of selecting political advocates. I think we need to be very clear regarding the different roles of sortition and stochastion — selecting individual advocates by sortition may well return persons who “didn’t owe anything to donors and weren’t seeking re-election”, but in an individual advocacy role they would have no (stochastic) representative mandate and would be extremely vulnerable to ex-post corruption. Stochastion will return a representative sample but this rules out active (substantive) representation — for that we need election or direct-democratic initiative. The advantage of election, is that it includes a measure of accountability, albeit in a less than optimal manner but entirely absent in a stochastic or direct-democratic context.
PS — I have no problem in principle with introducing voting members by stochastion on a none-of-the-above basis, but I’m not sure how it would work with two different categories of members sitting alongside each other, one who could advocate and vote and the other who could only vote.
LikeLike
None of the above and other unrepresented voters
Naomi wrote:
> Is not letting everyone make that choice for themselves the fairest option possible?
While I don’t think mixing elected and randomly selected members into a joint assembly is a good idea, I want to challenge the logic that only those who select “none-of-the-above” should be represented by allotted members. Setting aside the logical argument that all of the eligible voters who don’t turn out at all should automatically be represented by allotted members, let’s focus only on those who DO show up at the polls. Since only those who succeed in electing a candidate of their choice are represented by elected members shouldn’t both
A) those who mark the box labeled “none of the above”
but ALSO
B) those who voted for a candidate who does not win a seat be instead represented by allotted seats?
In other words. … in a situation where no voters select “none of the above” and 51% of voters happen to vote for winning candidates, why should they get 100% of the seats. Aren’t the 49% who failed to elect any members entitled to SOME sort of representation? Now, in a system of proportional representation (like STV) the number of voters who fail to elect a candidate will be relatively small, but in American or British style winner-take-all single seat elections (with a plurality threshold) the winners may have gotten votes from well less than 50% of those who cast their ballots for a candidate.
LikeLike
Keith,
Any corruption on the part of the allotted advocates would have to be approved by the voting sample which would, presumably, do a better job of opposing corruption (on account of rational ignorance issues) than the population at large. I do think a purely elected advocacy would be better. However, any real world political system is going to be filled with compromises that are nasty from a theoretical standpoint. If this is a compromise that can move us past the endless bickering, then so be it.
Terry,
Sure. I’m fine with limiting the lottery pool to those who choose none-of-the-above. Some people would prefer an equal vote. Others would prefer an equal chance. Allowing everyone to have what they would prefer seems like an acceptable compromise. That said, it would be impossible to do with a fully secret ballot. I don’t think that the quality of representation in the allotted subset would be good enough to make a difference. It’s a minor matter to me, as proposal diversity would be assured through the elected advocates. Ensuring that the proposal introduction process is not monopolized by a narrow majority of the population is what really matters. But if that’s a sticking point, so be it.
LikeLike
Terry,
Terry,
>Aren’t the 49% who failed to elect any members entitled to SOME sort of representation?
Yes, that’s a fair point, so long as the selection principle was stochastion. Selection of individual (full-mandate) parliamentarians by sortition is not a representative principle. This looks quite close to Naomi’s plan for a bicameral system, with elected members proposing and debating and allotted members deciding. The nice thing is there would be no automatic majority — the largest party would need to persuade the allotted members (who would vote in secret) of the merits of their proposals. There would also have to be a decent number of allotted members, in order to reduce the risk of their votes being bought. It would also mean that people would be less inclined to vote tactically during elections, as voting for the losing party would still achieve some form of representation. In the aftermath of a close election the casting votes in legislative debates would be in the hands of the allotted members, rather than at the whim of small parties in coalition negotiations (or the manufacturers of hanging chad machines).
What wouldn’t work would be Filimon Peonidis’s proposal for a part-elected, part-allotted chamber, with both categories having the same rights.
LikeLike
“To include all in the republic, Rousseau had to make all silent”.
Nadia Urbinati, Democracy Disfigured (Harvard UP, 2014, p. 41).
LikeLike
Naomi,
> If Yoram is right, then bootstrapping a pure sortition system would be quick and easy.
I don’t know what makes you say that. This is very far from my position. Maintaining democracy is always a matter of vigilance against those who try to grab positions of disproportionate power, so it is never quick and easy.
In fact, an important reason that allotted bodies must be sovereign (rather than constrained by pre-defined procedures) is because they must have the flexibility to keep redesigning the system and tweaking the rules to address flaws in the system and to counter oligarchical challenges.
LikeLike
Yoram,
The point is that, if experience shows its desirability, there’s a path to bootstrapping a pure sortition system. I, and I imagine most people, will fight it tooth and nail.
Terry,
Also, as far as I’m concerned, if the number of votes that fail to elect a candidate exceeds 1-2% then the electoral system designer did a crap job. It’s not hard to do a lot better than either FPTP or STV.
LikeLike
15 years after spending another 15 studying and teaching political theory I finally see this subject of “sortition” seemingly taken seriously by academics – at least from what I could find on the web. Coming back to the original proposal of the topic, I don’t know how uninviting or misleading “sortition” may sound in English, all the other options sound less precise and worse in taste. There is a latin rooted word, “aleatory”, that might be of some use: what about something like “aleatory choice” or “aleatory election”, to keep a term dear to the democratic glossary?
LikeLike
How about “aleation”, to encompass both sortition and merit screening (which is not necessarily by voting)? Word seems to be available.
LikeLike
I think we need to come back to David’s original post, which focused on the need to distinguish the aleatory from the representative function of sortition. The thing about a mini-public (David’s original preference) is that the microcosm is held to be a representation of the whole, not so with most people working on random selection, who are only interested in insulating the selection process from reasons (such as merit screening). So let’s keep sortition for the aleatory function and look for a new word to encapsulate representation by statistical sampling (my suggestion being stochastion). Jon should certainly coin a new word for his own combination of sortition and merit screening, as I’m not aware of anyone else advocating the Venetian balloting system in the modern world.
LikeLike
Hi tdguedes,
> 15 years after spending another 15 studying and teaching political theory
Did you teach sortition? Where? What were your experiences with that?
LikeLike
It is funny that almost every one-word option or neologism offered seems problematic to me …”sortition” sounds like sedition, “stocastion” looks like a typo and is a mystery about how to pronounce, “aleation” sounds like alienation, “Sample” is too limited in meaning, “eduction” also looks like a typo, etc.
Have we thoroughly examined the option of “policy jury” or “legislative jury” (“citizens jury” is trademarked and all juries are made of citizens, so is redundant). “Jury” has some positive connotations, and incorporates the concept of random selection. However it also implies very small size. Maybe “jury assembly?”
Then of course, there is the option of popularizing the ancient Greek word for legislative jurors…”nomothetai” … though I don’t find a singular noun for the body of nomothetai, only the plural equivalent of legislators. Maybe apply Nomothete to the whole body rather than a single juror??? Nomothetai has the advantage of pedigree (unlike a neologism) and thus historic legitimacy. But just like “sortition” it requires constant explanation, of course.
p.s. I like Jon’s filter and random choice approach for certain uses…such as selecting executives, or professional staff, etc, but think it is not appropriate for policy/legislative bodies, as it abandons representativeness in a vain search for some notion of “bestness” (may be legislative competence, expertise, trustworthiness, civic virtue, etc.).
LikeLike
Don’t forget the Latin word “fetura” for “breeding, which could be used as a noun, although one might want to modify it into “fetury” to match “jury”, then use “feturation” for the process, corresponding to “election”. I also like “nomothete” but it is a tad strange-sounding.
Part of the reason I keep suggesting some kind of merit screening for legislators (as well as executives and judges) is my two-year experience working with members of Congress on Capitol Hill. That experience impressed me with the greater need for legislators to be competent than to be “representative”, as that term is being used in this forum. Contrary to popular perception, I found members of Congress to be all-too representative of their constituents, who, if a random sample of the voters replaced them, those would probably make similar decisions and respond to pressures from the same special interests. It is all very well to say that legislators don’t need to be experts or especially virtuous, but only consult experts, in making their decisions, but members of Congress do consult what they think are experts, and too many of those witnesses don’t know much, and what little they think they know is wrong. The problem is that members and their staffers do not have enough of their own competence to tell the real experts from the pretenders.
I vividly recall one such experience in particular. I was working with some staffers to arrange a Senate hearing (on the overpopulation problem), and the members wanted witnesses who were prominent, with impressive credentials. My suggestions for witnesses were ignored. On the day of the hearing, I was standing on the side of the room. I would scribble a question for a member, and a staffer would take it to the member, or I would scribble an answer for a witness, and another staffer would take it to the witness. Sometimes the member or the witness would get it wrong (perhaps my handwriting was bad) and it was all I could do to keep from laughing, but no one else in the room noticed or was laughing. So the members had put on their show, to appease some of their constituents, and moved on.
But apparently I was the only one in the room with “domain knowledge” of the subject matter, and except for a few scribbled notes, I had little real influence on the proceedings. The staffers who collected my notes knew, but they had almost no influence either.
Major social or economic decisions are bad enough, but can you imagine such incompetence applied to defense policy or a decision whether to go to war?
So pardon me if I don’t put as much weight on a decisionmaking body being “representative”. I want them to be wise.
Contrary to the notion of the “wisdom of crowds”, I find (representative) crowds to more often be herds, that seek the safety of the middle of the herd, and who are too prone to follow charismatic but incompetent leaders off the edge of a cliff.
LikeLike
Terry,
If you talk with classicists, the problem with nomothetai is it actually just means lawmakers and legislative decision making by randomly-selected juries only occurred for a short period during the 4th century — before that time nomothetai were voted in by the council and/or assembly. Many historians of the period (Hansen and Blackwell being the prime exceptions) are not very aware of this process, and we have all been rather guilty of bigging it up as it was adopted largely for reasons of administrative convenience, rather than representation. And most decisions were still made in the Assembly and Council. So not only do we need to explain it to the general public, we also need to justifiy it to the experts.
As for stochastion (you misspelt it in your post), the only problem I can see is the initial difficulty in pronouncing it, which passes quickly once you’ve said it a couple of times. The same would be true of election if we weren’t used to hearing it all the time. And the word does suggest exactly what we are looking for — representation by statistical sampling (election meaning representation by preference balloting).
Jon,
You don’t need to convince (most of) us of the benefits of impartial selection on merit, but that’s not the point of David’s original question. By all means find a neologism for your synthesis of sortition and preference balloting but you’re not going to convince the likes of Yoram, Terry, Naomi, David and myself that descriptive representation and the wisdom of crowds are not worthwhile goals. Dowlen, Stone and Boyle, however, would agree with you and that’s why we need to separate out the sortitionists from those advocating representation by statistical sampling. Otherwise we’ll just carry on talking past each other.
LikeLike
Jon,
Let’s play with a simple example. In my home state (Georgia) most people live in or around the capitol city (Atlanta). Now, Atlanta’s mass transit system is very mediocre. It’s not as bad as other mass transit systems in the region, but it is certainly not adequate given the city’s population and the extent to which the city has sprawled out. People who live in the city largely want to extend the system. People outside are generally opposed. So who should win? What gives that decision its legitimacy in the eyes of not only the winners but also the losers?
LikeLike
Classic zero-sum situation (in the short term), perhaps solvable, if the players are willing to compromise to keep the game going, as with the “divide and choose” solution (one cuts the pie and the other one chooses his piece). Some version of this is now used in most legislatures, whose ongoing dispute is over when to make the cut and who comprises the “player” on each side.
But what makes a decision “wise”, as distinct from “representative”, is the time-span for applying the utility function. We tend to call a decision “wise” if posterity, perhaps a thousand years hence, approves the decision.
We can also describe a constitution of government as “well-designed” if it accords distant posterity a greater weight than the present generation, what economists call the “discount rate”. A discount rate of less than about 18% per generation will give more weight to the benefits for the third generation and beyond than for the present and next generations. (Integral of the utility function over time — the area under the curve.)
So to answer your question, to the extent we can anticipate it, what will Georgians (if Georgia still exists) of the year 3015 think is the best decision the lawmakers of 2015 could have made.
A good constitution benefits the long term interests of its citizens, not the short or mid-term.
LikeLike
Jon,
Your personal definition of wisdom perfectly demonstrates the futility of filtering for “wisdom.” The further in the future one projects, the less knowable are the unintended consequences…or even the desirability of the INTENDED consequences.That future generation may HATE the world we crafted and assured they would inherit (in our wisdom.) I agree that ONLY looking to the short term interests is a bad idea, but looking too far ahead is only make believe. So YOU might prefer to filter legislators for such distant future wisdom, but most people would reject that as being wise. (I am not saying YOU are advocating anything horrible like this, but killing off many present day people, through mass murder or eugenics programs is seen as evil, rather than wise, even if it paved the way for happy people centuries in the future.)
LikeLike
We kill off many present day people whenever we engage in war, with the expectation that posterity (or at least our posterity) will be better off if our side wins. Of course we cannot know every long-term consequence of every decision, but we can assure ourselves that we can identify some decisions that will turn out to be bad, such as those that restrict the resources and opportunities remaining to posterity, most of which decisions focus only on short-term payoffs, and avoid those decisions. A decision not to let tyranny prevail seems a good bet, because tyranny tends to persist for millennia, doing much injury to future prospects.
Is Mutual Assured Destruction (which is still U.S. doctrine) a wise policy? Was the future made better by dropping the bombs on Hiroshima and Nagasaki? Was the invasion of Iraq a wise move? When Chou En-Li was asked what he thought of the wisdom of the French Revolution, he is reported to have replied, “It is too early to say.”
To get back to Naomi’s question, will distant posterity think it was was a good decision to extend mass transit systems, or perhaps that we should do something completely different? I’m not providing an answer here, just suggesting what kinds of questions lawmakers should ask and sincerely try to answer.
LikeLike
Jon, why don’t you start a new thread on this (very important) topic? I agree with you that long-term epistemic outcomes should be paramount — and this is a very difficult problem to address, but this has nothing to do with David’s original question — what word should we use to refer to representation by statistical sampling?
LikeLike
About Keith’s proposal : « stochastion »
*** Maybe « stochation » would be simpler ; and it seems easier to spell (for me, at least) ; a blend of the ancient Greek « stokhos », which meant, in logical vocabulary, « conjecture », as opposite to perfect knowledge, and gave « stochastic » in probability science ; and of French (and next English) « election », which had the general meaning of choice, including « election by lot », before taking in 19th century the restrictive meaning of « election by vote » ; or, as Keith proposes, « tion » could be for « representation ». The letter « s » in « stochastion » is not necessary, it corresponds to the Greek adjectival form « stokhastikos » derived from the noun « stokhos»
*** A good thing in this word, its etymology : the Greek « stokhos », which meant, in logical vocabulary, « conjecture », is therefore linked to the idea of (imperfect) knowledge, and, in the probability field, to the knowledge with a degree of precision, whereas the other words are etymologically linked to the idea of disorder, chance, unpredictability : « aleatory » comes from Latin « alea », the game of dice, « random » comes from Old French « randon », impetuous running (cognate to English « run »).
*** I don’t agree with the idea of restricting the word « stochation » to mini-publics with effective powers, as opposite to advisory ones. The basic idea is the same. Even if Dahl says, without precision, that his « minipopulus » system is intended to function as « a complement », « not as a substitute » to the polyarchic institutions (« Democracy and its critics » 1989 p 340), it is clear we have only a step more to go to dêmokratia-through-sortition, or stochation. And reading Dahl’s text, he was very near of such an idea. I am convinced it was the basic reason of Habermas rejection of Dahl’s proposal, which he deemed « abstract » and « utopian ». I think it would be a bad idea to put a gulf between Dahl proposal and the straight dêmokratia-through-stochation idea.
LikeLike
Andre,
Stochation would be ideal — and much much easier to pronounce. I agree that advisory vs binding powers is a separate issue — especially given my own preference for a mixed system of governance.
P.S. Can you give me a reference for Habermas’s rejection of Dahl’s mini-populus? I’m currently struggling through the Structural Transformation of the Public Sphere.
LikeLike
In terms of my original question, here are my thoughts after following the discussion:
1. I think we need an “audience-friendly” term for the process of random selection, and another one for a randomly selected body.
2. I like the term “mini-public” for a randomly selected sample of the people as a whole. People seem to grasp it quickly and easily. In terms of Terry’s suggestions, “nomothetai” is going to be hard for many English speakers to pronounce or understand. I like “policy jury” or “legislative jury” for a more specific meaning – a body that only decides on an issue (as a criminal or civil jury does) – that would fit with people’s existing understanding of “jury,” but not with bodies that might set agendas, develop proposals, review and revise proposals, and so on.
3. In terms of clarity, I think it’s probably most useful to keep the term “mini-public” broad like this, and then explain sub-categories – for example, some “mini-publics” are advisory while others have decision making power, some are one-off while others are ongoing.
4. I like the term “statistical sampling” for the selection process. People understand it much more quickly than “sortition” (or probably “stochastion”), and it doesn’t have the negative connotations of “random selection,” nor the ancient/Biblical feeling of “selection by lot,” nor the negative associations of “lottery.”
5. I suspect that we will communicate most effectively if we describe our subject matter in terms of the bodies first, and the selection process second. I’ve been experimenting with this in conversations with people who have never heard of sortition, and so far I’m getting better results than I did when I started with the selection process.
6. This may not address Keith’s distinction between two different rationales for sortition (descriptive representation vs. avoidance of corruption and such, aka “the blind break versus the invisible hand). So maybe, in addition to terms for the bodies and the selection process, we need “audience friendly” terms for the main rationales for sortition.
LikeLike
David,
Election is political representation via “a formal and organized choice by vote of a person”. It’s derived from the Latin eligere‘ (to pick out). I don’t know exactly at what time in history its meaning became audience friendly.
Stochation (pronounced like vacation) would be defined as “political representation by statistical sampling”, so there’s a symmetry with election. It derives from the Greek stokhos (conjecture) and therefore suggests probabilistic knowledge. I know it isn’t currently user friendly, but if we want to make a radical proposal for political change then it needs its own name. OK we need words like mini-public and statistical sampling to help explain it, but we still need a name, and sortition and aleation refer to something else. Just as with election, people will learn in time what it means.
LikeLike
Well Divined
Clairvoyant: (after a reading) That’ll be forty
Archbishop: I only have a fifty
Clairvoyant: Peace be with you [takes note in clasped hands while bowing head ]
Archbishop: What about my change?
Clairvoyant: Change must come from within
LikeLike
David,
You suggest that mini-public would be an audience-friendly term for “bodies that might set agendas, develop proposals, review and revise proposals, and so on.” I can’t think offhand of a historical example where these roles have been carried out by the public (they are generally the province of politicians or bureaucrats), so why do you think this idea would be “audience friendly”?
LikeLike
Keith Sutherland asks for a reference about the rejection by Habermas of Dahl’s « minipopulus » as « abstract and somewhat utopian ». I have only the French translation therefore I hope it can be found easily in the English one.
Habermas, “Between Facts and Norms”, chapter VII, part III, end of the fifth paragraph (p 343 of the French translation)
LikeLike
About Keith Sutherland observations about Classicists and the role of lot in the Athens of Demosthenes.
*** Many Hellenists are more interested in the Athens of Pericles than the Athens of Demosthenes. They have their reasons, including the idea of the « decadence of Athens » we discussed about ; but we as kleroterians are interested specifically by the Athens of Demosthenes, because it is the one detailed historical example we know of mass use of alloted citizen juries.
*** Many Hellenists interested in the political history of 4th century Athens are not specially interested in the legislative juries. Here there is some good reason. Athens was an ancient society, not a modern one ; a largely static society. The social data did not evolve quickly as in our modern world and a conservative-minded democrat as Demosthenes could consider without absurdity that the legislative work must be kept as small as possible (let’s trust our ancestors’ wisdom !). In our modern dynamic societies, legislation and regulation must adapt to a changing world ; this branch of politics is very important, it was not the case in Demosthenes time. We have no quantitative data on legislative work by Athenian alloted legislators, but it may have been small. Thus, the low historical interest about nomothêtai is understandable, but we as kleroterians have specific reasons of interest, as we are thinking about the modern world. Due to the Athenian experiment, legislation through alloted bodies does not appear as belonging to utopian dream or sci-fi fantasy.
*** The Assembly was much more important than the legislative juries, because legislative work was probably small and often did not carry heavy consequences, whereas the Assembly decided about peace and war, i.e. about the freedom and the life of the City and its inhabitants, specially when the Macedonian power was threatening the Greek Cities. A very rough analogy, the Assembly had the role of USA President, the alloted juries the role of Supreme Court and Congress ; during the Second World War, the President role was paramount, but we must not say it is the one power in the US constitutional system.
*** Even if the legislative work was less important than in modern times, it is impossible to consider that the political use of lot was a minor thing in the Athens of Demosthenes. We should have to forget the paramount role of judicial juries. They could crush any decree as illegal, any law as inconstitutional, and they could sentence politicians to big fines or worse for many reasons, including having put forward proposals of decree or laws deemed illegal, inconstitutional or seriously detrimental to the City.
*** I maintain that in the Athens of Demosthenes the judicial juries (dikastêria) had a very important role in the political history of the City ; and I doubt Keith Sutherland could find an Hellenist who would say that it is not true. Therefore it is an historical truth that Athens was partly a democracy-through-alloted juries.
*** Coming back to the legislative juries, Keith says the system was adopted « largely for reasons of administrative convenience ». It is always possible to speculate about the reasons of the Athenian statesmen, and of the citizens, as we have no data about this institutional mutation and the related deliberations. But we must consider that this reform was perfectly in line with the other 4th century institutional changes : the judiciary power basically taken from the Assembly and given to the juries ; the big political trials by juries taking the former role of the ostracism by the Assembly (the ostracism, although always in the law, fell out of use) ; the clear distinction of laws and decrees, with the paramount role of juries in assessing the legality of decrees and the constitutionality of laws. In every case, a big part of the sovereign power was given to alloted juries.
LikeLike
Andre,
I agree with all your comments — my only point was that to use the word nomothetai for our modern proposal is of little value, as it isn’t user-friendly and would lead to some puzzlement among classical historians — for all the reasons that you provide. It’s one thing using it as a historical precedent (I do that all the time), but it isn’t much good as a name for our programme — your suggestion of stochation is much better.
LikeLike
Keith, you wrote,
>You suggest that mini-public would be an audience-friendly term for “bodies that might set agendas, develop proposals, review and revise proposals, and so on.” I can’t think offhand of a historical example where these roles have been carried out by the public (they are generally the province of politicians or bureaucrats), so why do you think this idea would be “audience friendly”?
I’m not saying that giving these functions to mini-publics would necessarily be “audience friendly.” I’m only saying that I think the term “mini-public” – for “a statistical sample of the public” – would be much more “audience friendly” than terms like “sortition” or “random selection” or “stochastion.” I expect a lot of people to have big reservations about mini-publics doing the things you wrote above – but then, there’s a lot of discontent about the way elected politicians do these things.
I should clarify that I’m talking about the challenge of how to introduce the idea of mini-publics/sortition to people who are not familiar with it.
Btw, it seems to me that the nomothetai, the BC Citizens Assembly, and the Irish Constitutional Convention all developed and revised proposals.
LikeLike
David,
The proposals for the nomothetai resulted from the votes of all citizens in the assembly and were not developed or revised by the nomothetic jury (mini-public), who only voted them up or down. I agree that the public would be familiar with this aspect of jury voting, but not with mini-publics that set agendas, develop proposals, review and revise proposals. I don’t have Dahl’s book to hand at the moment but I think his proposal was closer to the jury model (perhaps that’s another reason that Habermas disliked it).
PS We’ve dropped the word stochastion in favour of Andre’s suggestion of stochation (pronounced like vacation) as it’s more user-friendly.
LikeLike
Keith,
Dahl’s description of his “minipopulus” (page 340) was a major source of inspiration for my multi-body sortition model. He imagined each minipublic existing for perhaps one year. One might focus only on setting the agenda of issues, while others would focus on a single issue each, deliberate for a year and announce their choice. He suggests they would bring in experts, commission research, hold hearings, discuss and debate.
LikeLike
*** Keith Sutherland writes : « The proposals for the nomothetai resulted from the votes of all citizens in the assembly ». Right, but that does not mean that there was a dual legislative system, as in a parlementary bicameral system, where for instance a law must be approved first by Representatives and afterward by Senators. The Assembly had not to decide whether they approved the proposal, but « whether or not the proposal had sufficient merit to be brought before the Nomothetai » says Douglas M. MacDowell (The Law in Classical Athens, 1978, chap III). I think MacDowell here is right, this is a different idea, and the difference is highlighted by a fact : the Assembly appointed five persons to defend the old law. Do you imagine the Representatives deciding to substitute the old law by a new, and appointing before the Senators the defenders of the old law they had just rejected ? It would be very strange. When Demosthenes wants the (judicial) jury to crush the « Law of Leptines », he put forward a counter-proposal ; the jurors must trust the Assembly not to exclude a proposal which must be considered as serious, given its supposed implicit approval by the jury.
*** Right, this process gave nethertheless to the Assembly a de facto veto power against any legal change. In an ancient society where, as I said, the idea of changing the law was always somewhat dubious, that could be accepted. It cannot be so in a modern dynamic society.
*** Some may think the Assembly did not easily devolve power upon a citizen jury. I am afraid they think too much about political bodies with politicians as permanent members, and each body extending his power as widely as possible. But the ekklêsiastês (the member of the Assembly) was the same citizen as the dikastês (the member of a jury); he did not give power to « others », but to people like him ; to him some day maybe (excluding the subject of age limit – which does not appear to have been a hot political issue). And the ekklêsiastês had a very good practical reason to leave the job to the dikastês : it reduced his own time-consuming political work. In a modern dêmokratia, for the same reason, we can trust the members of a mini-populus to devolve easily power to more specialized kinds of mini-populus. If there is a risk, it is rather the risk of such devolutions producing lack of coherence, which will need specific institutional design.
LikeLike
*** Keith Sutherland writes : The proposals ( …) were not developed or revised by the nomothetic jury (mini-public), who only voted them up or down ». Right : it was the one possible way with a jury of two hundred jurors. It could be different with modern technology.
*** We must underline the face-to-face deliberation on the Agora could occur during all the legislative process. Any proposal, from the beginning, had to be written and exhibited on the monument of the Eponymous Heroes, in the Agora. And the length of the legislative process allowed for much discussion.
LikeLike
Keith,
While I am still wary of using a neologism, with Andre’s shortening, “stochation” is less objectionable. How would one construct the other parts of speech in parallel to “election,” in light of the existing word stochastic?…
The election will be held tomorrow
The community elected five hundred people
She is an elected member
this is an elected body
the electoral commission is tallying the votes
The stochation will be held tomorrow?
The community stochated five hundred people?
She is a stochated member?
this is a stochastic body?
the stochastic commission is tallying the votes?
LikeLike
Andre
Happy again to agree with the thrust of your argument. I was not suggesting that the Athenian system was bicameral, merely that:
1) Any citizen could propose a new law.
2) All citizens decided whether to refer the law for deliberative scrutiny.
3) The final decision was made by an allotted microcosm.
These are the three principles that I’m seeking to emulate in my proposal for a modern analogue of 4th century practice. I agree that the Athenians were keen to empower citizen juries (this is the main reason Aristotle claimed that 4th century practice still constituted a “radical” democracy).
>In an ancient society where, as I said, the idea of changing the law was always somewhat dubious, that could be accepted. It cannot be so in a modern dynamic society.
As a conservative I don’t accept that, and I would argue that words like “progressive” indicate a Whiggish teleology that should have no place in post-Darwinian discourse. Modern societies are certainly dynamic, but changing the laws every five minutes is just change, not “progress”.
>If there is a risk, it is rather the risk of such devolutions producing lack of coherence, which will need specific institutional design.
I Agree. That’s why there will never be a fully sortition-based democracy, other institutions will be required for coherence and stability and this will involve a combination of election and appointment on merit.
>We must underline the face-to-face deliberation on the Agora could occur during all the legislative process. Any proposal, from the beginning, had to be written and exhibited on the monument of the Eponymous Heroes, in the Agora. And the length of the legislative process allowed for much discussion.
Yes indeed, that’s where the Habermasian notion of the free exchange of opinion in the public domain comes in, along with Condorcet’s argument for significant time intervals between the various stages of the legislative process. It took a long time to change the laws in the 4th century (there were some seven stages involved), and conservatives like me would advocate something similar for our modern demokratia.
LikeLike
Terry,
That’s an interesting challenge. The two examples that don’t work are 2) and 3) and I think that is to be expected. In the case of 2) “stochated” is a transitive verb and not really appropriate for an impersonal process; in 3) I’ve long argued that there is no such thing as an individual “stochated” member as descriptive representativity only applies at the aggregate level. So it’s meaningful to refer to a stochastic body, but not to a stochated member.
This also serves to illustrate the need to clearly distinguish between active and descriptive representation. The former does things (in the transitive sense) whereas the latter just is. Statistical representation is not just the selection of political representatives by a different balloting method, it’s a different kind of representation from the substantive acts undertaken by elected officials. This is why ordinary-language philosophy has a practical role to play beyond the ivory towers — clarifying the use of words does have important entailments in the real world. Words are not just labels for objects and events, they can actually tell us what we need to do.
LikeLike
Hello, this is my first post here although I am very interested in the discussions.
What if there is a popular election between “Mr. Election” and “Mr. Stochation”? I guess “Mr. Election” would always win – he looks good, he sounds good. I don’t like pretty much the way the word “stochastic” sounds, I would prefer not to use it in a casual conversation. I guess only few people out there have any idea about it’s meaning. So can we just invent a brand new word with little or no meaning but a nice and beautiful word that could win popular elections? A word that one would be happy to say? Then we could just add meaning to the word. A humble example for a new word – egalation, derived from egalitarian and election. It could be an abbreviation either.But with some creative we could find a much better word.
LikeLike
If you want an appealing term, try “discorruption”, to emphasize the value of sortition (or aleation, eduction, or feturation) for reducing the undue influence of special interests.
Or we could use the word “jury” but apply it to lawmaking. We’ve been using “legislative jury” but we could use “legisjury”, or “legjury”.
Then for executive officials that are selected using some aleatory process, we might use “execujury” or “exejury”.
LikeLike
Delyandimitrov,
>What if there is a popular election between “Mr. Election” and “Mr. Stochation”? I guess “Mr. Election” would always win – he looks good, he sounds good.
Surely the only difference is that we’ve become accustomed to hearing the word election? I don’t see why that shouldn’t also apply to stochation — it has the same number of syllables and has structural and etymological parallels with election.
Jon,
>If you want an appealing term, try “discorruption”, to emphasize the value of sortition (or aleation, eduction, or feturation) for reducing the undue influence of special interests.
While this is clearly true ex ante, there is no reason at all to believe that a full-mandate legislative assembly, selected by lot, would not be wide open to the undue influence of special interests. Allotted members would only be incorruptible if their role was limited to voting in secret (as in the Athenian legislative courts). All your suggestions conflate Yoram’s phrase (representation by statistical sampling) with the prophylactic role of sortition.
LikeLike
Keith,
It might be a matter of taste. If most of the people like stochation than it is certainly a good option. We might need some sort of a survey to find out.
P.S. My first name is Delyan :)
LikeLike
Terry, Andre,
Dahl devotes just three paragraphs of his 1989 book to the mini-populous, so he’s clearly doing little more than flying a kite. Does he develop the idea any more in his later work and does he respond to Habermas’s criticisms?
In addition to being representative he insists that the min-populous should be “able to take advantage of the best available knowledge to decide what policies were most likely to achieve the ends it sought” (p.340). He doesn’t explain exactly how this might work in practice, but on the previous two pages, where he discusses the epistemic gulf between the demos and the policy elites, he acknowledges that the success of Polyarchy III depends on “the diversity of views among policy specialists and the relative weakness of their common interests as a ‘class'” (p.339). Given the aforementioned epistemic gulf, this would suggest that information provision is an exogenous function and well-balanced provision is dependent on intrinsic diversity. I’m not sure how you (and Yoram) would reshape Dahl’s proposal to fit your view that the the interests of the political class are somewhat more homogeneous than Dahl suggests.
Bear in mind also that Dahl’s proposal is “not for a substitute for legislative bodies but a complement” (p. 340) — ie the minipopulous would have an advisory role. If he had wanted to replace elected institutions then I imagine he would have been somewhat more rigorous in his proposals and might have given a little more consideration to the problem of representation.
LikeLike
@Keith
>”While this is clearly true ex ante, there is no reason at all to believe that a full-mandate legislative assembly, selected by lot, would not be wide open to the undue influence of special interests. Allotted members would only be incorruptible if their role was limited to voting in secret (as in the Athenian legislative courts).”
Judicial juries do not vote in secret among themselves, but with respect to the outside world. The main ways jury tampering are avoided is by having them serve for short periods of time, and controlling contact with outsiders. Anyone who serves for an extended period can be compromised.
And legislators are more unduly influenced before they take office than after. The special interests invest to place their already convinced minions in key positions. Corruption while in office usually occurs from prospects of future gain rather than immediate gain, although re-election is itself such a gain. Members might have to accept a periodic audit of their finances and be taxed 100% of any gains in excess of, say, 10% per year, for the ten succeeding years after leaving office. That would take care of most of it.
Corruption is not just initiated from special interests. In the U.S. we have developed the practice of pairs of legislators introducing bills designed to scare the constituents of the other, so that each can demand donations from his constituents, citing the other guy’s scary bill as a threat which his constituents need to re-elect him to defend them against. U.S. Rep. Robert Matsui once openly did that, warning his donors not to reduce their donations, which was reported in the Sacramento Bee about 1998. It is less bribery than extortion. My own member of Congress, that I ran against in 1974, was reported to regularly ask for money when asked for constituent services, money that went into his pocket.
LikeLike
Keith,
Yes Robert Dahl was free-wheeling about sortition in a possible future democracy, and for his purpose had no need to carefully consider the optimal structure. I don’t think we need to debate exactly what he imagined (he wasn’t playing a prophet), and we’ll never know since he died last year. But here’s my take…While he imagined them being supplemental to elected polyarchic legislatures, I don’t think he proposed that they be merely advisory…
By using the words “complementary” and “supplemental” he seems to mean that they would be in addition to and perhaps that that the range or number of issues these numerous allotted assemblies tackled might be less than a typical legislature (perhaps the issues most people care about), so a traditional legislature would still exist as well.
But he clearly imagined them fulfilling all of the traditional deliberative legislative functions (rather than just voting yes or no).
LikeLike
Terry,
I agree Dahl wasn’t proposing a purely voting body and can understand why you took his proposal as the template for your own multi-body model. Regarding the complemental issue his exact words on p. 340 are “I see the institution of the minipopulus in Polyarchy III not as a substitute for legislative bodies but as a complement”. If it isn’t a “legislative body” then its role can only be exemplary or advisory. If he had intended a legislative role for the minipopulus then I think he would have taken considerably more care in examining the implications of a purely statistical mandate.
Jon,
I agree that allotted legislative jurors should only serve for a short period of time (ideally on a case by case basis). If the office period was longer and members had the right to sponsor bills then special interests would soon find ways of getting round your proposals to monitor their income, especially in a globalised economy. Payments could be channelled via third parties, offshore tax havens etc. Re-election is a double edged sword in that it is also an incentive for politicians to behave themselves as opposed to just feathering their own nest.
LikeLike
Keith,
The more I think about it, the more convinced I am that short fixed terms are needed. I would like to see six month terms with continuous rotation. Some countries mandate a six month long maternity leave so it’s probably not an unreasonable length of time. The average member at any given point (in both the first and second votes if a second vote is needed) would have served for three months. That’s three months of listening to debates and following politics full-time. That should be enough time for the novelty of the experience to wear off and for the members to both develop a modest understand of the issues of the day and hopefully maintain a Nash equilibrium. It’s also hopefully enough time to at least partially null out undue media influence and for rhetorical excesses to lose some of their edge.
The details of proposals can then be developed by many advocates with the sample choosing from the presented options rather than having to approve an all or nothing proposal crafted by the first advocate out the door with an acceptable plan. The allotted members can make a decision on overall principles and then work their way through the details. And if, in going through the details, they realize the overall principles are less reasonable than they thought, they can go back and correct them. If a new sample had to be drawn each time they’d end up rehashing the same territory over and over. If we leave the details up to the executive and the overall principles are flawed then there’s a problem. I believe it is important for the allotted members to have as full an understanding of both the details of what a proposal would entail and the broader context as is practical. I don’t think a few days spent on a single issue is enough for that.
LikeLike
Might want to incorporate such methods as brainstorming (for solutions) and questorming (for questions) http://pynthan.org/documents/questorming/ into the procedures for juries of various kinds.
LikeLike
Naomi,
I’m agnostic as to the length of the term, so long as the activity of the allotted jury is restricted to voting in secret. There’s nothing magic about the number two, so I don’t have any problems with either-or-or-or . . ., but would need some convincing that allotted conscripts would be well equipped to work their way through all the details. The devil (in the form of sinister interests) could easily find his way back in through the details, especially given your proposed feedback from details to overall principles. It would also unduly privilege the few random jurors with the necessary skills and knowledge and this would adversely affect the representativity of the sample (ditto regarding Jon’s proposal for brainstorming, if it were limited it to the random brains in the sample).
>If we leave the details up to the executive and the overall principles are flawed then there’s a problem.
Yes indeed, the law of unintended consequences would suggest that the default position should be to leave well alone (of course I would say that, given my oft-stated conservative prejudices). Given that the long-term interests of advocates/parties will be to improve the confidence of the public in the soundness of their proposals, this should encourage them to get both the overall principles and the details right in the first place. I’m persuaded by Hansen’s criticism (2013, p. 106) of Aquinas for only describing the carrots and ignoring the sticks, when he re-presented Athenian democracy for the modern world (the Athenians were unforgiving of persons who gave poor advice to the demos). Hopefully this might lead to a serious reduction in the number of poorly-considered new bills — it will be the bad advisors who will take the flack in the long run as jurors are entirely unaccountable.
LikeLike
I was thinking more in terms of pure yes/no votes. The sample will have a working memory and can go back and forth and work things out gradually. If the sample’s opinion on fundamental principles shifts as it goes over the details, then that shift will be reflected in future votes concerning principles.
In a unicameral system responsibility falls almost entirely on the sample. Surely you’ll be able to find *somebody* with proposal powers willing to introduce some ill conceived bill. Even in the US where we have a relatively narrow range of views represented in Congress there’s a litany of ridiculous things introduced for various reasons each session. And we’re aiming for greater diversity here. It’ll be up to the sample to sort out the mess. Which is why I think it is so important that the members serve long enough to get a lay of the land, politically. In a bicameral system it’s different, the only things to reach a vote in the allotted chamber are things that the average elected politician supports and surely, most of them will be reasonable and responsive to conventional electoral pressures. But if we have a hundred people with proposal powers there will be a few oddballs no matter the selection method. Furthermore, assigning blame and translating that blame into electoral consequence is a complex matter. There’s enough agency loss in the process to let a few oddballs thrive. I don’t see how we can rely on the electoral stick without limiting the elected officials to acting only in aggregate.
LikeLike
Naomi,
Who’s suggesting “a few hundred people with proposal powers”? In my model proposal rights are limited to a) the manifesto commitments of the victorious political party/parties, b) online petitions that have gained 100,000 signatures followed by a successful public votation and c) government ministers (for housekeeping bills). I’m sure these imperfect filters will still let through a litany of ridiculous things, so it’s the task of the allotted group to throw them out. I’m agnostic as to how long the group needs to serve in order to gain the skills required to reject the turkeys; ditto with the level of detail that they should be required to scrutinise.
Although my original proposal was unicameral, I’m increasingly persuaded by your argument for elected and allotted houses working in parallel. This is (broadly speaking) the classical republican paradigm (which was inverted by the American revolution). In the classical (Harringtonian) model the “aristocratic” senate does the proposing and deliberating and the “democratic” house of commons makes the final (up/down) decision. This assumes that election is the aristocratic selection principle and stochation the democratic one.
LikeLike
Some of these schemes seem detached from legislative reality. Anyone undertaking reform in this field needs to consider how any scheme would work with extreme examples of actual legislation that is being debated in actual countries. For example:
1. How, how much, and when to tax, borrow, spend, and regulate fiscal and monetary policy, investment, and banking.
2. Whether and how to authorize going to war in various theaters against various enemies.
3. (US) How to structure and whether to approve the Trans-Pacific Partnership (trade policy).
4. Immigration and border protection. When, how, and how many of whom from which countries are to be admitted as refugees, tourists, workers, or naturalized.
5. How and whether to structure and fund a national health care system (NHS in UK, Obamacare, Medicare, Medicaid, VA in U.S.)
6. Public education policy. Teach what to whom and how, and how to measure results.
7. Energy, mining, extraction, disposal, and environmental policy.
8. R&D and IT policy.
9. Criminal legislation. What to penalize, when, where, and how.
10. Court system. What courts to establish and maintain, and where and how much to fund them, and whom to appoint as judges.
11. Election structures and procedures.
12. Agriculture policy. What to support and what not, and how.
I could go on, but that is enough to work on for the moment.
LikeLike
Jon,
I agree that what we refer to as governance covers all the big issues that you have listed. Unfortunately our current arrangements don’t work very well in most of the areas that you reference — for example both the (UK) Labour and Conservative parties have concluded that they need to bind their own hands on fiscal policy as this is too important to be left to purely electoral forces. And the recent record on whether or not to go to war is not an edifying spectacle. For this reason I am proposing a mixed system of governance, not a democracy, and this has the following implications:
1. Fiscal policy: Given that it should be laws that rule, rather than men, the default position should be balanced budgets and this can be constitutionally mandated. All legislative proposals have fiscal consequences and the chancellor/treasury secretary has the obligation to outline these consequences — if the legislature wishes to increase spending in one area then it has to be reduced in another (or taxes increased). The chancellor would be legally bound to deliver a balanced budget and would have to resign if this were not achieved (over a specified number of years), triggering a fiscal cliff scenario, automatically shutting down whole areas of government spending.
2. The decision whether to go to war: it strikes me that a microcosm of those who (potentially) have to fight should be the ones who make such a decision, having listened to the competing arguments of the politicians with expertise in foreign affairs. This strikes me as an improvement on leaving the decision to statesmen eager to carve a place for themselves in the annals of history.
Regarding issues 3-12, are you suggest that we would start off with a blank slate? Every state has existing policies in all these areas and the role of the legislature is purely to amend them (ideally as little as possible). So they would all be amenable to the process that I outlined in my last comment.
LikeLike
If we were to go to a constitutionally very different way of selecting legislators then yes, all existing legislation and policy would have to be reconsidered anew, as much of it already is.
Most existing ;legislators don’t really understand all of the legislation they pass, nor have read all of it. This is partly because they have to spend so much of their personal time raising money to get re-elected. But it is also because much of it is beyond their comprehension.
Remember the infamous words of Nancy Pelosi: “We have to pass the bill to find out what’s in it.” Such reckless legislation demands that we find another way to select legislators, with the ability and willingness to do the job thoroughly.
LikeLike
Jon,
>If we were to go to a constitutionally very different way of selecting legislators then yes, all existing legislation and policy would have to be reconsidered anew.
Why? Stochation is just a different mechanism for instantiating political representation and has no inherently revolutionary properties. All existing laws would remain in place. I get very worried by Year Zero proposals to rip everything up and start from scratch.
LikeLike
Keith,
Fair enough. I was picturing proposals coming from a set of elected officials rather a party’s internal mechanisms. Party discipline would probably be extraordinarily low, so that might make sense.
Would you restrict the ability to introduce money bills to the chancellor? I have my doubts on your mechanism to reduce deficit spending. In a conventional parliament MPs face serious consequences for triggering the resignation of the cabinet. However, the allotted members face no consequences for precipitating the resignation of the chancellor. The extent to which the mechanism alters their behavior will depend on how much the want the current chancellor to remain in office. If it takes a few years for the automatic resignation to be triggered, I doubt their behavior will be appreciably altered. You’re probably better off just imposing a constitutional ban on deficit spending, if that is your goal.
Jon,
I agree it takes a measure of knowledge to make reasonable decisions on those matters and it may be the case that the members of an ad hoc body could not do better than a coin toss, on average. But after a few months of experience I suspect they will be able to do well enough in voting on gradual changes to the status quo. At least their odds will be better. I’m reluctant to have them remain in office for more than a few months, though.
LikeLike
Naomi,
Anyone would have the right to introduce a money bill (subject to the democratic constraints mentioned earlier); the chancellor/treasury secretary would be legally obliged to balance the books but how that is to be achieved would be a democratic, not administrative, decision. The role of the delegated government is merely to cost every project and point out its implications. As always I’m vague regarding institutional mechanisms but I imagine most citizens would want to avoid the fiscal cliff consequences of the resignation of the chancellor, so I think this would be a genuine constraint.
I appreciate that zero-deficit fiscal policy is moving into what might be seen as partisan politics but given the unaccountable nature of jury-based governance, I can’t see any alternative than laws with an embedded constitutional status. As well as fiscal laws, there would also be a need for civil- and minority rights constitutional safeguards. The majority threshold to alter such laws would need to be very high.
LikeLike
It is largely election that drives members of a legislature to lump together many often unrelated bills into large complicated ones, because each such omnibus bill is the result of “logrolling” by members and their staffs (and in a aleatory legislature the staffs would be much more important, unless the process for appointing them was also aleatory). The result would be many much smaller bills, and all the tradeoffs would have to renegotiated anew, if there even were many tradeoffs.
A few months in office would not be nearly enough. It takes longer than that just to learn parliamentary procedure, staff an office, and become acquainted with colleagues.
This is why I complained that much of this discussion is detached from reality. It needs to be done in the light of experience with actual national legislative bodies, which face very different problems than state or local ones. I spent two years on Capitol Hill, during which I often seemed the only one able to read fast enough or comprehend well enough to understand much of the legislation. The intellectual limitations of most members was not just funny, it was dangerous.
LikeLike
Jon,
Why would there be log-rolling in a stochation-based system and where would the pork come from? It strikes me that there would just be individual bills and the task of reconciling them is that of the delegated government, not the legislature. And why would allotted jurors need staff? Obviously bills need to be drafted, but this is a job for the civil service — by all means allocate draftsmen to each bill by lot, in order to ensure impartiality.
LikeLike
Constitutionalism,
If Nancy Pelosi really said that about passing bills, is there a not a robust process for pre & post-legislative scrutiny? Sounds familiar. At least there’s a slim chance flawed bills can be amended or even repealed at a later date.
This form of due process may not be possible in the event the current “free-trade partnerships” (TPP & TTIP etc ) being negotiated in secret are passed.
Just as UK Gov’s 2014/15 Infrastructure Act, or Bill H.R 6 seem expedient, grass-roots initiatives & NGO’s may need to be much bolder, more collaborative & expedite in demonstrating a fair, accountable, clean & efficient infrastructure across the board.
We share common goals, therefore it should be much easier to agree common sense rules spared of temptation and risk.
LikeLike
Logrolling is not just about pork. Resources are always a constraint, so negotiation and compromise will always be necessary. Do we spend the next tax dollar on defense, health care, or R&D? Executives cannot reconcile legislative differences without becoming a legislature, and if the civil service drafts bills then the real decisions would be made by entrenched officials accountable to no one.
We already have the problem in countries with civil services that the civil servants are often more devoted to the out-of-power legislators that appointed them than to the current members. You can then have a shadow government that still runs the country even though they are not the ones who are supposed to.
And it can take forty years or more to learn enough to competently draft legislation, if even then. Governmental decisions are attempts to intervene in complicated legal, social, and economic systems that no one fully understands, and unintended consequences dominate. See http://constitution.org/ps/cbss.pdf
LikeLike
Watch Pelosi say it at https://www.youtube.com/watch?v=hV-05TLiiLU
LikeLike
Imagine the Pelosi logic applied to a declaration of war. “We have to pass the declaration to find out who we are at war with.” But that is just what the vague Authorization for the Use of Military Force (AUMF) resolutions do, that authorize action against “terrorists” and leave the sole discretion for the designation of “terrorists” to the president.
LikeLike
Jon,
Yes, that’s all true, and what you describe is a mixed system of government, not a democracy. But in my proposal legislators (out-of-power or otherwise) do not appoint civil servants, this is an independent estate. Radical (jury-based) democracy needs to counterbalanced by other (non-democratic) institutions, whose interests lie in the ongoing stability of the political system.
The embedded fiscal constraints would mean that every bill proposing an increase in expenditure on xxx would have to contain a proposal to reduce spending elsewhere (or increase taxes) and this would be the fully-costed proposal that voters selected to refer for allotted scrutiny. That way voters would no longer be able to opt for motherhood and apple pie and then leave the restaurant without picking up the bill (thereby leaving it for their grandchildren to pay).
LikeLike
How would you enforce “embedded fiscal constraints”? No constraint will be implemented if there is not an effective enforcement mechanism, and who will the enforcer be? A constitution or other law is not “a machine that would go of itself” to borrow a title from a book. http://www.amazon.com/Machine-that-Would-Itself-Constitution/dp/141280583X/ref=sr_1_1?s=books&ie=UTF8&qid=1431616859&sr=1-1&keywords=a+machine+that+would+go+of+itself
LikeLike
Jon,
I’m just a political theorist so that’s not really my problem! What I had in mind, though, was something similar to the law that requires the Governor of the Bank of England to write to the Chancellor every month inflation deviates from the prescribed bands. The only difference in the case of failing to meet the annual fiscal constraints would be that the Chancellor’s employment contract would be terminated rather than her just having to put pen to paper. The enforcer would be the highest court in the realm.
LikeLike
Naomi and Jon demonstrate the tension between uncorrupted representativeness (requiring short terms of service) and experience/competence (gained from longer duration of service). Both are needed, but cannot be contained by the same group of people. Thus the notion of short duration legislative juries (mini-publics)… that ideally would each tackle a single topic, to prevent log-rolling deal-making and the corrupting psychology of sustained power, on the one hand, and separate proposal generating or at least revision and perfecting bodies that serve for longer terms (but have no power to adopt), combined with extremely long-serving professional staff specializing in drafting and research.
LikeLike
Jon,
>A few months in office would not be nearly enough. It takes longer than that just to learn parliamentary procedure, staff an office, and become acquainted with colleagues.
As you know, Keith and I are talking about a sample that does absolutely nothing besides voting on proposals. A few months may not be enough, but a few years (much less decades) is problematic in the own right. As Terry said, there needs to be two different sets of people.
>Executives cannot reconcile legislative differences without becoming a legislature, and if the civil service drafts bills then the real decisions would be made by entrenched officials accountable to no one. … And it can take forty years or more to learn enough to competently draft legislation, if even then.
Hear, hear. This is one of the reasons I find myself drawn to Keith’s arguments in favor of unicameralism. In it there’s a natural dispute resolution mechanism. Different parties, executive officers, and perhaps others as well, can duke it out on a somewhat level playing field. They don’t need to be representative in aggregate.
Terry,
The problem is that policies don’t exist in a vacuum. The people who have the final say should know what’s going on.
LikeLike
To expand on my previous post (sorry for the typos, I was in a hurry and using a new phone) I have no problem with the executive wielding great delegated legislative powers. I would be perfectly fine with most legislation coming from various agencies. The problem is in relying on the executive to perform core legislative functions.
Given that the executive’s neck is on the line it is difficult to imagine them feeling limited to a purely objective empirical analysis of the facts. There would be pressure to take an active role in resolving conflicts rather than simply pointing out the existence of those conflicts. This pushes them into the world of value judgements, rather than empirical judgements. Politicization might be a safe bet in that case. Ultimately there must be an institution that makes both value judgements and empirical judgements to some extent. Value calls are necessarily political in nature, and so this institution must be tainted by politics. At present this is the executive. If we wish to quarantine politics within the legislature we must make sure the legislature is capable of handle value judgements and resolve conflicts between values and empirical realities on its own.
One option is to have an elected house as part of a bicameral system. The elected officials would have considerable experience and manage things much as they do now, but are checked by an allotted house. We can say with an exceptional degree of confidence that the results would be an improvement on the status quo, which is not something that can be said for very many of the systems proposed on this blog. Another possibility would be a unicameral system, which would place the coordination burden square on the shoulders of the allotted members. To handle it I imagine they would need to stick around long enough to get their sea legs. It’s riskier, but there are more advantages, I believe. But I could be wrong.
LikeLike
Naomi,
>One option is to have an elected house as part of a bicameral system. The elected officials would have considerable experience and manage things much as they do now, but are checked by an allotted house.
Yes, I’m increasingly persuaded that this is the way to go. But would both houses meet separately (as in the classical republican paradigm) or would it be a case of a single assembly with two different kinds of actors — the elected ones who would make the proposals and the allotted ones who determine (or veto) the outcome? The problem with the former proposal is that it’s hard to understand how the allotted members could make a sensible decision without hearing the arguments; the problem with the latter being that the whole process would be politicised. I would go for the second option, which is in effect elected rather than appointed advocates, the important point being that only allotted members have voting rights.
The other factor should be that ministers (who have to implement the policies) should also have speech rights in the unicameral assembly, otherwise there is a significant risk of impractical dogma-driven policymaking (values overrunning the empirical). This is assuming that you are not calling for a fusion of the executive and policy making functions, as in a parliamentary system.
LikeLike
I believe your second type of bicameralism is what I was thinking of when I wrote “unicameral.” At the moment, my ideal system would consist of a single assembly with two types of members, 100-200 elected, 300-500 drawn by lot, with powers segregated between them. Executive officers would hold the same range of powers in the assembly as the elected members. The elected members would serve for several years and the allotted members several months. If a bill clears a supermajority threshold it goes into force immediately, otherwise a second vote must be held after the allotted sample has completely turned over.
In any case, I don’t necessarily see a problem with having separate debates in two separate houses. Supporters and opponents in the elected house could simply stage courtroom-style debates in the allotted house. However, the dynamic between the parties, between the two houses, between the different executive departments and agencies, and between the executives and both houses would all be different and complex and dependent on things we are not in a position to predict at this point. For example, the old electoral seesaw between parties in power and out of power may or may not be preserved. In Switzerland the existence of an easy to use initiative/referendum drives an eternal grand coalition. But Switzerland is merely one case study. In California there’s no evidence of consensus politics. But then again, as a subnational entity, California is a much poorer example. Who the heck knows? The dynamic in the single assembly model should be much more flat. All the actors (save for the allotted members, of course) can make proposals. Their conflicts — be they between different parties, different executives, or between executives and parties — are all settled by debate in-chamber followed by a vote.
I don’t necessarily see how a one assembly model would result in excessive politicization.
LikeLike
Ideal. I’m very encouraged that two people (you and me) who started off from very different perspectives have listened to each other’s arguments and arrived at a common conclusion. It’s (almost) enough to make one believe in deliberative democracy!
LikeLike
Keith and Naomi,
I think it would be a useful contribution if the two of you wrote up your new conclusions, and your rationale, and published it, in this blog or elsewhere.
LikeLike
Yes, good idea. If Naomi is interested she could write the institutional part and I’ll put it in the context of the classical republican theory of mixed government.
LikeLike
The main problem I have with partisan elected officials making the proposals for the allotted segment to vote on is the tendency for people to take sides and defer to the judgment of the elected “leaders.”
As an example, researchers presented subjects with a proposed policy dealing with “the poor” and fabricated quotes from leading Democrats and Republicans either supporting or opposing the proposed policy. Participants who identified with one party or the other generally agreed with whatever position the faked statements of their party leaders. The preferences of participants flipped based on which version of the fake statements they were given.
A goal of having mini-publics is to give members the incentive to do the hard work of thinking for themselves rather than taking the mental shortcut of getting in line with their favored party.
LikeLike
Terry,
I share your anxiety here, but remember that in the proposal that Naomi and I are making (modelled roughly on the Deliberative Polling procedure) allotted decision makers are presented with the full arguments/information, rather than just a couple of soundbites. And I think US political experience is exceptional in that European countries do not have the same kind of bipartisan polarities. And the broad range of proposal sources (and the decoupling of the legislative and executive functions) would likely lead to a wide multiplicity of political parties that would correspond more closely with the beliefs and preferences of individual voters.
LikeLike
*** Keith Sutherland said (May 11, 2015)« as a conservative »
« Modern societies are certainly dynamic, but changing the laws every five minutes is just change, not “progress”. »
*** First I must specify that I understand « law » as « nomos » in the Second Athenian Democracy, i.e. any general rule issued by the political power. For instance we can have a law regulating the supposed dangerous psychotropic substances ; and another law which gives a list of these supposed dangerous psychotropic substances. In France nowadays this kind of list is issued as « administrative regulation » (« arrêté ministériel »), but it is for me a law. Note that theorically all alcoholic beverages might be prohibited, by only adding « ethanol »on the list …
*** Such a list in an ancient society could change very slowly. Not in a modern society, where technology produces new substances ; and where world communication may introduce hithertho unknown substances – as the ayahuasca coming from Amazonia and forbidden in France since 2005 (by an addition to the list ; with a fuss about religious freedom).
*** Maybe a conservative will stick to the same basic principles he chose when he was young; but in a modern society, he cannot stick to the law as it existed when he was young.
*** Many things have changed since I was young. Internet could not be imagined, for instance. Some people in France said : « we have very good laws about press ; we have only to use them for the Web ». But it is not possible, the things are too different. Some others, in such cases, say « the Courts will adapt the law ». It means actually « the Courts will create the new law ». Even in a dêmokratia where the last word would be given by citizen juries, including in the judicial branch, I would not like the idea. That destroys the « rule of law » and the legal safety, as, to know whether you are lawful, you have to wait and see if you are declared guilty by the judges!
*** I conclude that Demosthenes, as a conservative, could consider that legislative activity had to be exceptional ; but that is not possible in a modern society.
LikeLike
Yes, good point
LikeLike
The comments are seeming need to disperse to allow for effective debate of particular points. I wanted to comment on the original idea of the most appropriate terminology. From a marketing perspective to advocate for the idea I have no doubt lottocracy is the most catchy and self-explanatory term. Many people like lotteries too. Are explanations can be in other terms such as the creation of mini publics through sortition or random sampling. I think the use of the term mini public does create an important emphasis on the non-individualistic character of the assembly.
LikeLike
Satkhalsa,
The trouble with lottocracy is there is no reference to political representation and (in the UK) it has a sense of randomness in the pejorative sense (as in “postcode lottery”). Agree that minipublic does (rightly) put the emphasis on the assembly, not the individual members (lottery winners). My own preference is for the creation of legislative juries by stochation, but this would all take some time to gain public recognition. But, as we’re suggesting something new, then a brand new term seems quite appropriate.
LikeLike
Keith, David,
Sorry for the late reply. It’s been crazy. 60hr work weeks crazy.
There are still a few aspects which make me uncomfortable. I’ve long believed the relationship between the executive and the legislative process to be the single most difficult aspect in constitutional design and it probably needs further consideration before we can declare victory.
Terry,
Some portion of the allotted members will surely vote one way or the other due to a sense of loyalty to the elected members, however, as long as the number of allotted members who are open to persuasion is large enough there shouldn’t be a problem. I imagine the portion open to persuasion will be fairly high as the members will have little choice but to sit and listen to arguments for days on end.
LikeLike
Naomi,
The relationship between the executive and the legislative process is indeed difficult to work out. So maybe you want to work on that some more – and when you and Keith feel more confident about your conclusion, write it up and publish it.
LikeLike
The problem of loyalty to elected members would certainly not arise in the UK, where MPs (wrongly, in my view) are viewed in a very poor light. Party membership is at an all-time low and voters change their allegiance with each election.
LikeLike
Keith,
Is there actual research showing what portion of the electorate CHANGES their party preference over time? In the U.S. at least, I would estimate that 90% of voters stay with the same party year after year, and the swings in election outcomes come from two factors, 1. that small percentage of swing voters who DO change preference, but mainly from 2. differential turnout rates (what portion of supporters of each party bother to go and vote in a given election.) My guess is the same holds true for most of the UK, except when there is a rare realignment (like in Scotland).
LikeLike
I’m not a political scientist, so I’m not really in a position to say, but the common understanding of UK politics since 1979 is very different. Although there are still a large number of “safe” seats, both main parties compete very hard for the median voter. If it were not for the single-member plurality system, most voters would be uncommitted, so I would agree with Naomi that the likelihood of allotted members in a mixed parliament voting with “their” elected representatives is small — most would make their own minds up after listening to the arguments. The rational ignorance argument would also support this conclusion, as there would be no obvious disincentive to pay attention, as each vote will have significant causal power.
LikeLike
I suspect the number of people who vote for the same party year after year in both countries is fairly high. However, I also suspect it is largely due to the need to avoid shooting one’s self in the foot from a policy standpoint. The big parties have adopted similar electoral strategies but I don’t believe they have become entirely redundant from an ideological standpoint in either the US or the UK. This is very distinct from loyalty, though I absolutely believe that is a factor as well. I’ve heard of people (on this side of the pond) always voting for the same party because it’s the party their parents always voted for. It’s just what they do. And certainly, in the US for sure, there’s an undercurrent of loyalty for popular political figures. The question is whether this is enough to be a problem. It would surprise me if this were the case. I know when I was selected for jury duty the magnitude of the responsibility shocked and terrified me to my core. It’s hard to see the average allotted member engaging in much cognitive free-riding given that they will know exactly what is at stake.
LikeLike
Naomi,
>I’ve heard of people (on this side of the pond) always voting for the same party because it’s the party their parents always voted for. It’s just what they do.
That was certainly the case in the UK until 1979; now it is considerably less prevalent.
>I know when I was selected for jury duty the magnitude of the responsibility shocked and terrified me to my core. It’s hard to see the average allotted member engaging in much cognitive free-riding given that they will know exactly what is at stake.
Ditto, and I was deeply impressed with how seriously my fellow jurors took their responsibility, even though they did not stand to gain or lose personally as a result of their decision. It was just a civic responsibility that was taken very seriously by all of us. And it struck me that the cognitive resources of the group would have been insufficient to arrive at the right verdict in the absence of the balanced adversarial exchange (the guy just looked guilty, and desperately needed the skills of his defence lawyer, whereas the natural empathy of most of the jurors for the victims [it was a commercial fraud case] unduly affected their judgment, IMO)
LikeLike
I’m a bit late to this discussion. I’m starting to think that “sortition” is a perfectly fine word with a good Wikipedia article. Yanis Varoufakis used the word with Russell Brand recently. I think we should also get the word “jury” into the conversation where possible since people generally view juries favourably. It is also worth reminding people that we already have sortition via nullification of the law. Juries are capable of putting laws on trial. But in practice this right is suppressed.
LikeLike
Hi Mike,
> Yanis Varoufakis used the word with Russell Brand recently
That sounds interesting. Do you have a link?
LikeLike