Cristina Lafont, Professor of Philosophy at Northwestern University, presents her anti-sortition position as being based on participationist ideology. In a debate with Brett Hennig and Samuel Bagg, which took place in August 2022, Lafont initially makes the standard participationist arguments:
[T]he very idea of having something like a lottocracy, where we change the political system, my main concern is it is not democratic. It is a way of empowering the few, the very few the tiny, tiny few randomly selected people to do the thinking and the deciding for the rest of the citizenry. Whereas the citizens really are just supposed to blindly defer to whatever decisions they make. They have no formal tools of holding them accountable or of collectively shaping which political agenda we are going to have. They just can only blindly refer to whatever those very few people decide, and to me, that is really not democratic. Blind deference is quintessentially a non-democratic relationship of political inequality where you have just decision makers who are not accountable, they can decide anyway they like as they see fit, and then you have people who just follow and obey and have no other way of shaping their decisions. That’s my main concern.
I think lottocracy is not only non-democratic, but actually the reason why democracy was important in the first place was because if we want to really have a better society, you need the citizenry to be behind it. You are going to have to convince them anyway. The shortcut of having a very few deciding and being informed, deliberating while the rest have no idea, no clue being manipulated, being exposed to fake news. The idea that the shortcut will help us get better results, actually, it won’t work because that’s not enough.
All of this is standard participationist dogma: It’s not democracy if not everybody is making every decision, there is no reason to trust the system unless one is informed about every decision and takes part in every decision. It is actually more accurate to call this sloganeering than a dogma, because what “being informed” or “participation in decision making” is never defined, so it is mere rhetoric. Any meaningful definition of being informed or participation would exclude the possibility of implementing the participationist ideal, so it is essential to leave the matter as vague slogans.
In reality, of course, people trust very many systems which do not meet the absurd conditions set up by the participationists. The essential requirement for trust in a system is that it delivers positive outcomes. This is the reason the citizens of Singapore trust their system, and that the Americans used to trust their system in the decades following WWII.
Later, however, Lafont moves beyond the participationist dogma and simply makes the assertion that sortition cannot be used because people are easily bought.
The idea that it’s harder to capture those few representatives in the citizens assemblies than politicians was not very plausible. I think that precisely because they have zero accountability and because they won’t ever necessarily have such power again – it’s a one-off chance in your life – they are very extremely likely to be captured by the revolving door, by promises about what jobs … not bribes. It’s not bribes. It’s not violence. I don’t think the powers that be have to captured that way. No, it’s the usual way. You are simply invited by Murdoch, by the Koch brothers and you just see the possibilities of having a job in the future that will be extraordinary. I think the possibility of capturing a randomly selected person to change their lives by having opportunities is so much higher than with politicians.
Setting aside the obvious point that the system can be designed so that “job offers” from interested parties would be illegal, this argument reveals a lot about the participationist (or at least about this participationist’s) mindset. It turns out it is a mindset that is based on the idea that people are immoral self-serving creatures and thus cannot be trusted. The typical person, Lafont confidently asserts, can be instantly corrupted by “an invitation” from some right-wing evil-doer. Lifelong values and habits of thought are instantly forgotten at the prospect of “a job”. In fact, Lafont thinks that politicians are relatively immune to this type of corruption, since they, for some unspecified reason, prefer the political career to “a job”. We are lucky we have politicians to keep us normal people away from position of power.
The bottom line is that Lafont, like all participationists, is an advocate for the oligarchical status quo. Their absurd slogans regarding what would make a democracy leave no way forward for a meaningful reform of the current system, so we are stuck by default, with that current system. Well, this current system certainly has its advantages if you are a well-placed academic.
Filed under: Academia, Elections, Participation, Sortition |
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
Jochen Krattenmacher
johnedwardtaylor
johnhuang865
Jonathan Crock
keithsutherland
Jacob Richter
kkoreilly
lukewhito
SimonPek
Nathan El
Nicholas Gruen
Owen Shaffer
Patrick Chalmers
Paul Gölz
peterstone
Pierre-Etienne Vandamme
pronfeld
Emil
rcaze
richwbrown
Roger Knights
roslynfuller
Pierre-Etienne Vandamme
shaharavin
tbouricius
Ted Wachtel
Joshua Laferriere
theodorestathis
towardsautonomyblog
Wayne Liebman
Yoram Gat
Oliver Milne
Equality by lot 
I think Cristina would be surprised to be labelled a “participationist” (in the sense of Pateman?). This is her concern:
“my main concern is that [lottocracy] is not democratic. It is a way of empowering the few, the very few the tiny, tiny few randomly selected people.”
For the decisions of the few randomly selected persons to accurately mirror the informed preferences of the many is a non-trivial problem, that cannot be resolved by Gattian syllogism.
LikeLike
“my main concern is that [lottocracy] is not democratic. It is a way of empowering the few, the very few the tiny, tiny few randomly selected people.”
Is she not concerned that elections are not democratic, but are a way of empowering the few, the very few the tiny, tiny few elected people. If not, why not? It seems her concern is universal to any representative democracy.
Oh, Keith, can you explain “Gattian?” I’ve seen you reference it before but a search turns up nothing for me.
LikeLiked by 1 person
Lance,
Keith is referring to Yoram Gat, the editor of this site, who has a unique definition of democracy. While he advocates sortition for a legislative body, he also is not a “proceduralist” in that this body can function however they choose, with the key factor in deciding if a system is “democratic” being whether the population is satisfied with the policy decisions. (Yoram will correct me if I made a mistake).
LikeLike
>elections are not democratic, but are a way of empowering the few, the very few the tiny, tiny few elected people.
The problem is not the principle (voters choosing who they wish to represent them), it’s the curtailment of the choices available — hence our insistence that superminority (along with citizens’ juries) is the key to democracy.
>Gattian sylogism
What I was referring to was Yoram’s method for demonstrating the democratic legitimacy of random selection, based on a logical syllogism involving a small group of friends deciding which restaurant to attend.
LikeLike
Terry:> The key factor in deciding if a system is “democratic” being whether the population is satisfied with the policy decisions.
Terry and I have pointed out to Yoram that this would also mean that theocracies, dictatorships etc. could be democracies, so long as “the people” were happy with the outcomes.
LikeLike
Thank you for your interest in my thoughts about democracy and sortition. I am not being ironic – I’d be happy to discuss those thoughts on a open-minded basis.
For the “syllogistic” argument for the effectiveness of sortition as a democratic mechanism, see: https://equalitybylot.com/2013/09/29/a-theory-of-sortition-part-2-of-2.
For the idea that democracy is most authoritatively measured by the consent of the governed, see: https://equalitybylot.com/2020/10/26/consent-of-the-governed.
LikeLike
OK, thanks Terry and Keith, I didn’t connect Yoram’s last name to Gattian.
Keith, the point I was making was that Lafont’s objection to sortition was applicable to any method of empowering a few to represent the many. She then goes on to complain that they have no means of holding their representatives accountable, a frequent objection that I have not been impressed with, given how little accountability elections provide, but that is still a separate point. All representative democracies empower a tiny few. Sortition could empower more at less expense than elections, but it would still be a small fraction. That is by definition for any representative democracy. In short, she is making a silly point for rhetorical purposes.
LikeLike
Lance,
It depends what you mean by “representation”. Hanna Pitkin, who’s book is generally taken as the starting point for the modern debate, draws a categorical distinction between a) descriptive representation and b) the active representation of interests. The descriptive representation mandate is constrained by the Law of Large numbers, so there is no such creature as a (singular) “descriptive representative” and it is hard to understand how policy proposals could be generated in a democratically legitimate manner. In the latter case the numerical constitution of the representers matters less, so long as there are sufficient to broadly match the beliefs and preferences of the target population.
The mistake is to think that the two forms of representers are equivalent, but empowered by different systems of balloting (lot or preference election).
Accountability is the flip side of the cybernetic feedback loop between the representer and the represented. Under superminority, a representer who made epistemically bad proposals and/or failed to reflect the beliefs and preferences of the plurality of the electorate, would be unlikely to get a second chance. Under “pure” sortition, there would be no accountability at all.
LikeLike
Keith,
It depends what you mean by “representation”.
What I mean by representation is that a small minority, or in Lafont’s words, “the few, the very few the tiny, tiny few” represent the population and make decisions. Thus her point is trivial, true by definition of representation, no matter how the representatives are selected. She emphasizes the small number with the repetitive “tiny few”remark as a rhetorical flourish. It is intellectually dishonest as the number of representatives is proportionally the same with elected officials as sortition chosen officials.
The accountability argument is a separate issue, but I don’t believe elected officials are accountable. In the US senate, senators stay in for a lifetime passing unpopular policies.
Would you be in favor of selecting members of mini publics based on a representative sample of the range of opinions, rather than a demographic descriptive sample?
LikeLike
>Would you be in favor of selecting members of mini publics based on a representative sample of the range of opinions, rather than a demographic descriptive sample?
This is what John Dryzek calls discursive (as opposed to deliberative) democracy and, in principal, it’s the ideal way of selecting policy proposers and advocates. The trouble is who gets to decide which discourses merit representation and the weight that should be afforded to each one? And what if the randomly selected rep that supports your preferred discourse has piss-poor rhetorical skills, which renders her far less able to persuade her peers to take up the good cause? This is why Alex and I think an elected policy-making assembly with a low superminority threshold is, in practice, the best solution.
Our model presupposes the argumentative theory of reasoning which valorises confirmation bias and other cognitive tools for the proposing/advocacy role. But when it comes to evaluating the different discourses the citizens’ jury needs to be a quasi-mandatory random sample of short-term duration, and several hundred (or even 1,000) strong. All such a large group can do is to weigh the arguments and determine the outcome.
The trouble with sortition-based deliberative democracy (which conflates proposing and deciding) is that it’s based on the (quaint) Burkean ideal of the deliberative pursuit of the ‘laws of God and nature’. Even Habermas admits that this is the exception rather than the rule.
LikeLike
The trouble is who gets to decide which discourses merit representation and the weight that should be afforded to each one?
Perhaps this could be done with generative AI. H. Landemore has a paper on using AI to collate arguments and institute a direct democracy. I don’t understand enough about the workings of AI to form an opinion.
This is why Alex and I think an elected policy-making assembly with a low superminority threshold is, in practice, the best solution.
It certainly would be an improvement. The strength of your proposal – that it would be more palatable for adoption because it leaves the political parties intact, is also what bothers me – it leaves the political parties intact. I thought the best way for me to look at it is that you developed a vaccine that doesn’t destroy the virus but protects the population from its harmful effects.
Hopefully!
Our model presupposes the argumentative theory of reasoning…
I’m a big fan of Mercier, et al, having read their paper before I discovered sortition and evaluated sortition through that light. I was glad to see Terry referenced it. I’d like to read the paper by Mercier and Landemore but cannot find a site where a non academic can access it.
LikeLike
The trouble with sortition-based deliberative democracy (which conflates proposing and deciding) is that it’s based on the (quaint) Burkean ideal of the deliberative pursuit of the ‘laws of God and nature’. Even Habermas admits that this is the exception rather than the rule.
One simple formation of Habermas’ requirement is that for a norm, law or decision to be valid, those affected by it should have a voice of consequence in the decision. Taking that as a starting point, it’s an easy extrapolation to choose the system that maximizes the informed voice of those affected.
LikeLike
Lance,
The trouble with political representation via AI is that it is essential that any constitutional innovations should have the confidence of the citizenry. Happy with your virus analogy, as we have always argued that election is a way of quarantining elite power. My uni library rights have expired, but I’ll have a look and see if I downloaded the Mercier/Landemore paper previously. “All affected interests” is little more than a normative ideal, given the transnational nature of modern life.
LikeLike
Lance,
Here is how to bypass the paywall and read the Mercier and Landemore article. Go to the article here http://www.jstor.org/stable/23260333
and you will see an option JSTOR offers to set up an account for independent researchers for free that lets you read up to 100 articles per month.
LikeLike
The trouble with political representation via AI is that it is essential that any constitutional innovations should have the confidence of the citizenry.
Yes, this would not have immediate trust. Perhaps down the road.
LikeLike
Terry,
Thanks for the link!
LikeLike
Speaking of AI, here is a seminar being offered on using it to enhance deliberation. I could see it as a tool. Also, it could be used as a deliberative poll just to check how well CA assemblies match deliberative public opinion, but the public would not have the degree of information on which to base their deliberation.
https://democracywithoutelections.org/seminars/
LikeLike
The problem with using AI to select which “discourses” to represent is the ability of elites to manipulate the algorithm. A earlier incarnation of Google’s Bard (Bart?) behaved like Homer (Simpson) designing his ideal car, as it hallucinated images of female popes and black Founding Fathers, because it had been told to privilege “diversity”. Chat-GPT can show you puppies playing in the snow but refuses all requests to include violence in an image. It’s one thing convincing a sceptical public that a random selection algorithm is unbiased, no chance of trusting Big Tech to select which discourses it wants to represent. How could you trust tech companies that have a previous record of censoring dissenting voices on Covid vaccination, climate change and gender realism?
And even if you could get the selection process right, you have the problem with the wildly different presentational skills of the (randomly selected) spokespersons for each discourse. Election with a low superminority threshold would be much better.
LikeLike
Helene Landemore on using AI to augment and aid mass deliberation:
https://www.law.nyu.edu/sites/default/files/Helen%20Landemore%20Can%20AI%20bring%20deliberative%20democracy%20to%20the%20masses.pdf
LikeLike
As AI becomes more ubiquitous, I think the trust will inevitably increase. Perhaps in England the trust in politicians is higher than Big Tech, but here in the US it is so low one would have to scour the prisons to assembled a less-trusted collection of scoundrels. Personally, even now, I’d trust an agenda or agendas composed by Chat GPT over anything I’d expect from the likes of Chuck Schumer, Mitch McConnell, Joe Manchin, or Marjorie Taylor-Greene. I wouldn’t be surprised if the general populace shared my opinion.
An open-source platform could alleviate some fear. Just saw today that election transparency was made available to the skeptics in Idaho and they appear to be satisfied.
LikeLike
I apologize for comparing prisoners unfavorably with politicians. I do realize some prisoners are innocent.
LikeLike
Bear in mind Helene’s principal interests are epistemic. Democratic legitimacy is very much a secondary concern.
LikeLike
LikeLike
> transparency
“AI” is, by definition, the antithesis of transparency. AI is defined as “a machine doing something complicated in an explicable way”.
LikeLike
“AI” is, by definition, the antithesis of transparency. AI is defined as “a machine doing something complicated in an explicable way”.
So you have extensive experience designing AI systems?
LikeLike
we should also remember that there is no obvious connection between deliberation and democracy. One could even argue that the compound term is an oxymoron.
LikeLike
Just noticed my typo above. It should of course be “in an *in*explicable way”.
LikeLike
AI systems can be relatively simple to explain. AI/machine learning` ingests in data. The data must be labeled, by humans, as “correct” or “incorrect”. The data must be categorized by humans. The AI then creates a model based on what the humans have said is correct vs incorrect, good vs bad.
For problems with no clear “right” answer, such as political problems, what is correct then? What is good? An AI will just regurgitate what its master told the AI was good or bad.
So if we ever built some AI to “do politics for us”, the AI will just be doing the bidding of its master. He who controls the AI, and told the AI what was good and evil, therefore controls the state.
AI is then a red herring. In the future, should AI be used in government? Who cares? What matter more is the man behind the curtain. I want the man behind the curtain controlling the AI to be a democratic deliberative body formed by lottery. And that’s the place AI should only ever be used. In an advisory capacity, servient to the true sovereign of a state.
LikeLike
The Cortica site has information 0n how they are using AI to augment and enhance deeper listening and communication. It’s worth a visit for those wishing to gain further understanding on this topic:
https://cortico.ai/platform/
LikeLike
An interesting vignette on the power of group deliberation by historian and political philosopher Martin Jay:
The homely example I would offer from my own experience concerns my service on jury duty, when we deadlocked six to six at the first ballot. Without spelling out the details, I was in favor of a guilty verdict, but by the end of the third day of our sober and careful discussion, I and the other five others on my side were convinced by the arguments of one of the other jurors that we couldn’t convict beyond a reasonable doubt. What made the experience particularly moving was that the persuasive reasoning came from a postal worker, who had succeeded in convincing a professor at a prestigious university and five other professionals through the sheer power of his arguments. I suppose no one had come in with a strong investment in a prior outcome, but still, what made the experience so powerful an instance of communicative rationality working to reach a meaningful consensus became clear after we announced our verdict. We had exit interviews with the attorneys for both sides, and the prosecutor admitted that he had a weak case and thus tacitly confirmed our judgment. Jury deliberations are, of course, imperfect models of public discourse in general—our interests weren’t involved and we had strict rules of admissible evidence—but they draw on the same aspirational imperative as deliberative democracy—giving justifications and listening to the justifications of others before decisions are made.
LikeLike
Lance,
I can report a similar anecdote, the only difference being that I was the juror who managed to persuade my peers not to convict. Needless to say there was a re-trial and the unfortunate defendants were sent down without further ado.
To my mind this shows that if we want to borrow the jury model for citizens’ assemblies we need to take steps to ensure that the decision output is invariant between different samples of the same population. Otherwise, how will we know which decision is the “right” one? This strikes me as more (democratically) important than the exchange of reasons per se.
Although I still believe that the first verdict was epistemically correct (it was a complex commercial case of fraudulent intent), I think the second verdict was democratically correct, in that most jurors would not have had the particular business experiences that led me to believe the defendants were innocent.
LikeLike
Socrates was convicted by majority vote, and a different sample would have likely done the same. So the verdict was democratically correct, although us moderns might have reservations about its epistemic value. We need to decide whether we are interested in the democratic potential of sortition or “communicative rationality and meaningful consensus”. The latter discourse style is more characteristic of oligarchic councils (see Bagehot’s discussion of efficient power).
LikeLike
To my mind this shows that if we want to borrow the jury model for citizens’ assemblies we need to take steps to ensure that the decision output is invariant between different samples of the same population. Otherwise, how will we know which decision is the “right” one? This strikes me as more (democratically) important than the exchange of reasons per se.
I am not so sure. Your quotes indicate that “right” is a contextually dependent notion, why should it be invariant?In the example you gave, you were able to provide novel reasons to the group unlikely to be repeated consistently across many different distinct group collections. Isn’t it better to have even just one group that has a unique insight and solves an issue because of that rather than none?
One of the uses Cortico makes of AI, if I understand it properly, is to identify different arguments/proposals and categorize them, then putting the most popular ones on the agenda for further consideration. If that method was used in physics, Einstein’s relativity theory would not have made it on the docket.
LikeLike
Lance,
Contextual dependency from an epistemic perspective would mean Socrates being crowned with a wreath of leaves, rather than drinking the hemlock. But if we are interested in sortition as a way of enhancing democratic decision making then the sample must be an accurate portrait in miniature of the target population, hence the need for invariance. Helene’s concerns are epistemic, and democratic legitimacy takes a secondary place.
Regarding AI, I find the arguments of Yoram and johnhuang865 persuasive.
LikeLike
We need to decide whether we are interested in the democratic potential of sortition or “communicative rationality and meaningful consensus”.
I don’t see the two in conflict. Sortition is democratic, in that everyone has an equal chance of being seated to the deliberative body, and they use communicative rationality to arrive at decisions.
Martin Jay’s story was insightful in that it runs counter to Bagehot’s belief in the need for elite decision-makers for efficiency. The more general counter to his thesis is that his elite does not reflect the perspectives of the populace and so will necessarily be inefficient.
LikeLike
Lance:> Sortition is democratic, in that everyone has an equal chance of being seated to the deliberative body.
Democracy means that the people (in their collective capacity) have power (isonomia), which is a very different matter. What you are referring to is isegoria, the other side of the democratic coin. And in large modern societies equal right is not sufficient to ensure equal speech, as representation is necessary for both democratic norms.
My reference to Bagehot was regarding the maximum size of a deliberative group, not how it is constituted.
LikeLike
Contextual dependency from an epistemic perspective would mean Socrates being crowned with a wreath of leaves, rather than drinking the hemlock.
You are assuming that the current perspective is the epistemically correct one. Socrates himself willingly obeyed the judgement, believing in the procedural justification for it, even though feeling they were in epistemic error. In any case, I’m not understanding your point, are there advocates of sortition who claim that the decisions of a deliberative body chosen via sortition could never be wrong?
But if we are interested in sortition as a way of enhancing democratic decision making then the sample must be an accurate portrait in miniature of the target population, hence the need for invariance.
I still fail to understand this need. If we take sample groups of physicists at any point in history, we would not get invariance of opinion among all the groups. Why is this so important to you?
Regarding AI, I find the arguments of Yoram and johnhuang865 persuasive.
If we are deciding by vote, I abstain. I believe AI is still a nascent field, more investigation is warranted before committing to any adamant position. I did not find John’s description reflective of what either Landemore or Cortico were advocating for AI use, and so not relevant to my concerns.
LikeLike
Lance:> are there advocates of sortition who claim that the decisions of a deliberative body chosen via sortition could never be wrong?
That depends on whether their perspective is epistemic or procedural.
>If we take sample groups of physicists at any point in history, we would not get invariance of opinion among all the groups.
Two large random samples of physicists would probably come to the same conclusion. Bear in mind also that political societies, unlike professional groups, are what Oakeshott referred to as compulsory associations, and the current paradigm is that the people, in their collective capacity, should have power. If the ballot mechanism is lot, rather than preference election, then there is no way of knowing the will of the demos, if different samples return different verdicts. If there is significance variance, then persons, rather than the people, have power.
>I did not find John’s description reflective of what either Landemore or Cortico were advocating for AI use
I think your argument fails epistemically as well as procedurally, as you acknowledge that Cortico’s AI would not have put Einsteinian relativity on the docket. Superminority (assuming a low threshold) will always have space for innovative proposals, unlike AI (assuming it hasn’t passed the singularity threshold).
LikeLike
“significant” not significance
LikeLike
The other difference between physics and politics (apart from interests) is that in science, the paradigm (in the Kuhnian sense) is hegemonic. Few scientists would dispute relativity, evolution etc. This is less true of politics, where there is genuine disagreement over what is (epistemically) right.
LikeLike
The other difference between physics and politics (apart from interests) is that in science, the paradigm (in the Kuhnian sense) is hegemonic. Few scientists would dispute relativity, evolution etc.
That is true, but my assertion was not that physicists disagreed about everything, agreeing on paradigms is true by definition. My claim was that at any point in history, there are physical theories that they do not agree on, thus random sample groups of physicists would mot arrive at invariant conclusions. I guess I was unclear.
To further clarify, random groups of physicists would not come to the same conclusions on the interpretation of quantum mechanics. Politics is not unique in producing disagreement. Looking for invariance in human affairs is looking for a unicorn. I see no reason to demand such an impossible standard for sortition chosen deliberative assemblies to fulfill.
LikeLike
I think your argument fails epistemically as well as procedurally, as you acknowledge that Cortico’s AI would not have put Einsteinian relativity on the docket. Superminority (assuming a low threshold) will always have space for innovative proposals, unlike AI (assuming it hasn’t passed the singularity threshold).
‘You could dial in the super low threshold for either. The question is would or when that would be done. Elected officials would be much less likely to advocate for paradigm shattering proposals, especially if that were to incur disfavor by their party.
An AI collecting various proposals and arguments from many groups that rotated their members after each deliberation could track the acceleration of an idea through the collection and tag ideas that made rapid headway. Political parties quash ideas that vary far from their doctrinaire positions. I know your superminority threshold ameliorates this, but I don’t think it would be to the same degree as the AI enhanced lottery chosen bodies. Still, both would be vastly superior to the current system.
So if we ever built some AI to “do politics for us”, …
It’s not a matter of if, it’s already being used. Although AI is not “doing politics for us” it is being used experimentally by MIT, Cortico and others, and by current political parties sometimes much more nefariously.
I want the man behind the curtain controlling the AI to be a democratic deliberative body formed by lottery. And that’s the place AI should only ever be used. In an advisory capacity, servient to the true sovereign of a state.
On this John and I agree, which is exactly what Landemore is advocating for and the above mentioned are constructing.
LikeLike
Lance,
There will be a marked difference in the deliberative style of expert and lay groups, even if both are selected by sortition. A randomly-selected group of quantum physicists would include supporters of different theoretical models. They are unlikely to be persuaded to change from a Copenhagen to a many-worlds interpretation, so the deliberative style would be similar to a group of politicians, or other professionals. If the samples were large enough, each group would contain a similar number of advocates for the different interpretations of quantum theory. Anyone who has attended an academic conference will know that the chance of overall agreement is absolute zero.
By contrast, a lay group, who knew little about the topic under deliberation, would be start off in listening mode and their decisions would be strongly influenced by information cascades and other social-psychological factors. High-status speakers and the rhetorically gifted would be able to influence the group and change minds (nobody wants to be the odd one out). Whilst this may or may not have epistemic merit, it will make the outcome of the debate random in the pejorative sense, so it will not be possible to determine which sample is the representative one.
Regarding the choice between discourses selected by Superminority and AI, my principal concern is democratic legitimacy. In the former case the discourses are selected by voters, rather than Zuckerberg, Musk, Altman, Gates or other Oz-like characters. As for the likelihood of a group of random citizens being able to exercise sovereignty over an AI, I agree with Yoram (who is a professional data scientist):
“AI” is, by definition, the antithesis of transparency. AI is defined as “a machine doing something complicated in an [in]explicable way”.
Remember, also, that the Singularity has moved from science fiction to an imminent possibility, so this would leave the machine in charge of itself. http://books.imprint.co.uk/book/?gcoi=71157100174580
LikeLike
Keith,
It is true that the different groups of physicists “deliberating” about quantum physics would arrive at invariance via consistent disagreement. “Deliberation” is in quotes because the deliberations already took place in academic journals, colloquiums, and “in the wild.” Little would be gained by its continuance. The various positions are akin to ideological stances. Given enough iterations, mini-publics could also arrive at decisions reflective of steadfast ideological views held throughout the demos. It would be more than with the physicist sample groups, because those iterations have already been done, but it’s not, in theory, unachievable. The “marked difference in style” is not anything endemic to either group, but merely reflective of the extant that each has deliberated among others with conflicting views.
If we go back in time, say to 1905, and set up different deliberative groups to determine the best explanation for the propagation of light given the Michelson-Morley experiments, we’d find more variance. True, there would still be a high degree of consistency, given a large enough sample size, between the supporters of a fixed luminous aether, and a dragged one, but if Einstein was chosen by lot (doubtful he be elected) at least some of the groups that he was in would dismiss with the aether altogether. Enough iterations and eventually relativity theory would win out. Science, simply put, is deliberation with experimental observation. It may be invariant at any given point in time, but variant throughout time, variance increasing during times of paradigm shift.
Einstein, in the above example, would be the equivalent of a high-status, or rhetorically-gifted speaker, except his status and giftedness would be the logic of his arguments. Such gifts are also present in the public at large, such as in the example of the postal worker in the jury. They are not all sock puppets at the mercy of rhetorical savants and high-status individuals. Moreover, scientists are not immune to bias, witness the suppression of Bohmian mechanics, or the publication of the sexual behavior of bonobos, as well as numerous instances of corporate-sponsored “science.” Apparently, members of CAs in China have resisted acquiescence to the wishes of high-status individuals. More research would be useful to understand the extant of these issues . Of course, elected individuals are placed into their positions of power on the basis of such traits, so it is not a unique fault to deliberative representatives chosen by lot, if it is a significant factor at all.
LikeLike
Lance,
Given your interest in the Mercier-Landemore theory of argumentative reasoning, I’m surprised that you are promoting the classical notion that the purpose of deliberation is to arrive at the truth. Whilst that might apply to physics (although Kuhnians would beg to differ), what is the “right” answer to a political problem (epistemically speaking)?
>Given enough iterations, mini-publics could also arrive at decisions reflective of steadfast ideological views held throughout the demos.
The trouble is that the temporal need for multiple iterations would mean that the minipublic was no longer a portrait in miniature of the target population, as it would have gone native. And the resulting equilibrium would be both path dependent and artificially hegemonic, on account of the social pressure to conform with the majority perspective. Bear in mind that the (aggregate) wisdom of crowds presupposes independent agents and this is destroyed by their active participation in the exchange of reasons.
>Apparently, members of CAs in China have resisted acquiescence to the wishes of high-status individuals.
In the Zeghou DP, the infrastructure options were presented by the local CP and the participants chose ones other than those preferred by the party. This is entirely in line with the argumentative theory of reasoning, as the lay participants were judges, not advocates.
LikeLike
It is important to keep the distinction between active give-and-take deliberation (useful for crafting proposals), and internal “deliberation” within in each individual member’s mind on a large jury (useful for judging proposals). I sense that the three of us (Land, Keith, and I) agree on this point. There is no way separate random samples of citizens, nor re-runs of history to end up with different mixes of ideology in current political parties (under the super-minority proposal), would generate the same proposals. With an infinite number of variations in proposals, they would never repeat. We only suspect they would all generate somewhat reasonable proposals that had some chance of being ratified by a random jury. The HOPE is that ANY random jury, presented with a yes / no decision would make the same yes or no. REPLICATION is desirable for a jury , and nonsensical for proposal generation.
LikeLike
Given your interest in the Mercier-Landemore theory of argumentative reasoning, I’m surprised that you are promoting the classical notion that the purpose of deliberation is to arrive at the truth.
I’m not, where did I say anything about truth? All I brought up were procedures for reaching agreement, or realizing clearly that there was no agreement.
Whilst that might apply to physics (although Kuhnians would beg to differ), what is the “right” answer to a political problem (epistemically speaking)?
It seems that you promoted truth when you expressed the opinion that the Athenians decision to execute Socrates was epistemically incorrect. My interest is not in finding truth, but finding solutions to problems that most people can support. That will vary as time and circumstances change. Also, how decisions are arrived at affects the support they receive, this also varying by time and place.
>Given enough iterations, mini-publics could also arrive at decisions reflective of steadfast ideological views held throughout the demos.
The trouble is that the temporal need for multiple iterations would mean that the minipublic was no longer a portrait in miniature of the target population, as it would have gone native.
I don’t know what that means. Are you saying there would need to be millions of citizens involved and hundreds of thousands of iterations? I don’t think it would take that many before a clear front runner would emerge, but perhaps you have something else in mind.
And the resulting equilibrium would be both path dependent and artificially hegemonic,
Again, not sure what you are saying.
on account of the social pressure to conform with the majority perspective. Bear in mind that the (aggregate) wisdom of crowds presupposes independent agents and this is destroyed by their active participation in the exchange of reasons.
I think conformity of opinion during deliberation is greatly exaggerated. I rarely experience it in groups I’ve been in, there is only one case that really sticks out. It certainly is not a problem on this forum! If you have such a dire view of deliberation, why do you have it as your final step in deciding between the various proposals? It seems that all the good work done by the superminority requirement on proposals would be undone as all the lemmings followed each other down the cliff.
I’m still pondering your statement that the wisdom of the crowd is destroyed by the exchange of reasons. Do you believe that is true for individuals as well?
>Apparently, members of CAs in China have resisted acquiescence to the wishes of high-status individuals.
In the Zeghou DP, the infrastructure options were presented by the local CP and the participants chose ones other than those preferred by the party. This is entirely in line with the argumentative theory of reasoning, as the lay participants were judges, not advocates.
Yes, they were capable of independent thought. This in spite of having to be brave enough to disagree with party officials. Seems like far more pressure than what is applied in the various staged conformity experiments.
LikeLike
Regarding the choice between discourses selected by Superminority and AI, my principal concern is democratic legitimacy. In the former case the discourses are selected by voters, rather than Zuckerberg, Musk, Altman, Gates or other Oz-like characters.
As for the likelihood of a group of random citizens being able to exercise sovereignty over an AI, I agree with Yoram (who is a professional data scientist):
I would argue that one thing we do know is that extremely wealthy individuals from Oz have an inordinate voice in electing representatives, I don’t see that as democratically legitimate at all.
That the Oz men would be selecting the discourses has moved from pure speculation in your mind to fact, when the only fact at hand is the one I mentioned.
Rather than continuing the discussion on general terms and conjuring up vague fears, I think it would be helpful to look at individual instances of how AI use is being suggested. The first would be to transcribe the spoken word into written text, a laborious job for humans that can be done rapidly by AI. Do you see an opportunity for the Four Horseman of Oz to ride in and corrupt the process at this step? If not, which step in Cortico’s AI assisted process are you suspicious of and why? Please be specific.
I am relishing in the irony of Yoram being designated an elite!
“AI” is, by definition, the antithesis of transparency. AI is defined as “a machine doing something complicated in an [in]explicable way”.
A truly Gattian definition, they are always unique! This is more an amusing quip than a definition. Nowhere, have I ever seen inexplicability defined ad a necessary attribute of an AI system. Even Yampolskiy, no Pollyanna when it comes to AI, would not agree to that. Some systems are inexplicable, especially LLMs but it is important still not to conflate that with not being transparent. If AI could not be designed to be transparent or explicable how could the Oz men program it to accomplish any of their dastardly deeds? If they can design it for a specific task, then that design could be required to be made transparent if it was used to augment political decision-making.
LikeLike
I realized I may have been mistaken when I ascribed deliberation to the final decision-making step in your allotted body. I reread the paper and there is no description I saw of how that body makes it’s decision.
LikeLike
Happy to agree with Terry once again, except:
>There is no way separate random samples of citizens, nor re-runs of history to end up with different mixes of ideology in current political parties (under the super-minority proposal), would generate the same proposals.
Repeatability isn’t a prerequisite of Superminority, only the citizens jury. The democratic justification of electoral superminority is that voters get to choose (from a wide slate of parties), and that the diversity is not annihilated by the 50+1 proposal threshold.
LikeLike
Lance:> My interest is not in finding truth, but finding solutions to problems that most people can support.
Alex and I would fully concur with this pragmatic approach.
>Are you saying there would need to be millions of citizens involved and hundreds of thousands of iterations?
No. Our model is the 4th century nomotheatai, where numbers were typically 501 and the deliberation never exceeded one day. The 50%+1 decision threshold applies to the final decision, so no iterations are required and there is no sense that unanimity (or even agreement) is necessary.
>If you have such a dire view of deliberation, why do you have it as your final step in deciding between the various proposals? It seems that all the good work done by the superminority requirement on proposals would be undone as all the lemmings followed each other down the cliff.
Deliberation in the final step is silent — the jury just weighs the competing arguments and determines the outcome. None of the jurors knows how any of the other lemmings vote. This is where Fishkin departs from the Habermasian deliberative ideal.
>I’m still pondering your statement that the wisdom of the crowd is destroyed by the exchange of reasons. Do you believe that is true for individuals as well?
Yes, relying on the evidence cited by Surowiecki in his book.
[Zeghou DP participants] were capable of independent thought. This in spite of having to be brave enough to disagree with party officials.
All they needed to do was to vote (in secret) to say whether they preferred clean water or a motorway. And the Bloomfield Track CA revealed that most opinion change was at the information, rather than deliberation, stage.
LikeLike
Lance:> I would argue that one thing we do know is that extremely wealthy individuals from Oz have an inordinate voice in electing representatives, I don’t see that as democratically legitimate at all.
Agreed. But Superminority is designed to undermine this, and the evidence from proportional representation suggests that the greater the number of parties, the less the influence of the greenback (until this is undermined by the 50+1 decision threshold). Yoram is in a better position to comment on the multitude of tiny parties in Israel, but I can’t imagine they are all backed by the big shekel.
Regarding AI, my concern is perceived legitimacy. Whilst proposing rights under electoral systems are not perfect, at least citizens get to choose which scoundrel best represents their beliefs and preferences (Lafont and Urbinati would describe this as representative isegoria). Bearing in mind that sortition will disenfranchise the isonomia right of the overwhelming majority of citizens, I can’t imagine anyone choosing to arrogate their proposing rights to a machine as well.
LikeLike
The Bloomfield Track CA was n Australia. I was referring to this deliberative poll:
https://deliberation.stanford.edu/news/deliberative-pollingr-zeguo-township-china
LikeLike
Yes I’m familiar with both. I only mentioned Bloomfield Track because it’s the only case I’m aware of that tested whether it was the information or deliberation stage that led to most of the changes in beliefs and preferences. Bob Goodin cites it in his book on the topic.
LikeLike
I mentioned Zedou as a deliberative CA that resisted high -status pressures. They deliberated on budget matters.
Goodin makes much of the single case in Australia where the participants didn’t feel that deliberation influenced their decision. I’m not comfortable drawing conclusions from a sample size of one, but all one could draw from it is that deliberation did not harm or help.
LikeLike
Regarding AI, my concern is perceived legitimacy.
AI recently entered the “panic cycle” due to Chat GPT-4. As people become more comfortable with it and there understanding grows, so will their trust, unless AGI is reached and annihilates us all.
Whilst proposing rights under electoral systems are not perfect, at least citizens get to choose which scoundrel best represents their beliefs and preferences (Lafont and Urbinati would describe this as representative isegoria).
I have never gotten that choice. None of my preferred scoundrels ever have a ghost of a chance of winning. Given that, and the astronomical chances of casting the deciding vote between scoundrels I don’t even like, voting is a complete waste of time for me. Also, I do not believe it is representative, as many lower income people do not vote. I know a young mother who works two jobs to try to support her family. She isn’t willing to stand in line to vote for Twiddle Dee Dee over Twiddle Dee Dum. I don’t think this is an uncommon story. She would have real power if selected through sortition to a paid assembly. Now she has none, and no possibility of it. Isokratia is zilch, if you prefer Greek.
Bearing in mind that sortition will disenfranchise the isonomia right of the overwhelming majority of citizens,
“The rule of the people has the fairest name of all, equality (isonomia), and does none of the things that a monarch does. The lot determines offices, power is held accountable, and deliberation is conducted in public.” – Herodotus
I can’t imagine anyone choosing to arrogate their proposing rights to a machine as well.
Once again, AI doesn’t decide anything in the proposed uses. Humans are making the decisions, AI is only used to help manage and understand the numerous deliberations. This is a strawman argument.
LikeLike
Lance: could you email me at keith@imprint.co.uk — there’s something I’d like to discuss offline.
>I’m not comfortable drawing conclusions from a sample size of one.
Agree.
>None of my preferred scoundrels ever have a ghost of a chance of winning.
Agree. That’s why we need superminority (representative isegoria) alongside large, randomly selected juries (representative isonomia) to establish modern demokratia. All sortitionists agree on the latter, but disregard the fact that discursive representation is needed to achieve isegoria in large, multicultural poleis.
LikeLike
That’s why we need superminority (representative isegoria) alongside large, randomly selected juries (representative isonomia) to establish modern demokratia.
I do think the superminority idea is an important addition to dialogue around democracy. At first it would most likely be instituted, as you say, leaving political parties intact. They no doubt would vehemently oppose it. I see a real possibility to it leading to a “withering away” of political parties.
All sortitionists agree on the latter, but disregard the fact that discursive representation is needed to achieve isegoria in large, multicultural poleis.
I had to look up discursive representation, which led my to Dryzek. Foundations and Frontiers is bound to be a near future read. Thanks, I am learning from my dialogue with you, which may not be readily apparent given my combative style.
LikeLike
Lance,
My own guess is that political parties would be attracted to superminority, as it would enable them to better reflect the beliefs and preferences of their supporters, rather than compromising them to get over the 50%+1 threshold (either alone, or in coalition). I also can’t imagine demographic governance without political parties (pace the title of my first book), as I agree with Lamont and Urbinati that they are the modern equivalent of the Athenian rhetor. Ho boulomenos may have been the classical principle, but the modern version has to be representative — anything else would just be blooming, buzzing confusion.
I think we have all learned from this exchange, which I would characterise as robust, rather than combative. (I could make a remark about the provenance of the withering away meme, but probably better not).
LikeLike
Hello again, Keith,
My own guess is that political parties would be attracted to superminority, as it would enable them to better reflect the beliefs and preferences of their supporters…
That depends on who they view as their supporters, to wit, the provenance of the withering meme.
I also can’t imagine demographic governance without political parties (pace the title of my first book), as I agree with Lamont and Urbinati that they are the modern equivalent of the Athenian rhetor.
On this side of the pond I see the chief aim of political parties being to sustain their power, which they do by exacerbating division. The US is bordering on being a dysfunctional state if not already worthy of that title.
Here is an interesting study on how the populace views the responsiveness of the government to their needs. So-called democracies don’t compare well with autocracies:
https://www.allianceofdemocracies.org/wp-content/uploads/2024/05/DPI-2024.pdf?utm_source=substack&utm_medium=email
As far as external and internal deliberation, the simplest description of our differences is that I’m more concerned that individual bias and the propaganda instilled by power elites will result in poor decisions than I am worried that deliberative bodies will succumb to group-think or yield to people of high status. Group-think is already present via propaganda, external deliberation is a tool to ameliorate it. I’m concerned that relying purely on internal deliberation at the decision-making stage would be no better than conducting a poll of the population to make the decision. If possible to imagine, the manufacturers of consent might find an even higher gear to influence randomly-selected non-deliberative decision-making bodies.
Ho boulomenos may have been the classical principle, but the modern version has to be representative — anything else would just be blooming, buzzing confusion.
On this we agree, the only question is how to best overcome the limitations of representation.
On AI, to express our differences similarly, your main concern seems to be AI taking over the decision-making process, or perhaps it is only the perception by the public of such. My main concern with AI is its use in the wild to distort public understanding on issues and candidates. It is already distorting the ability to discern fact from reality. See this article on what is taking place in India:
https://www.bbc.com/news/world-asia-india-68918330
Again, it is not a matter of whether to use AI in the political process, it’s already being used. The question is who will use it, and how it will be used. I don’t want to battle a well-equipped propaganda army with Stone Age tools. The AI arms race is already underway, the genie is out of the bottle, how do we best direct it?
LikeLike
Lance,
Alex and myself are acutely aware that current electoral systems encourage power seeking as an end in itself. I think you underestimate the radical transformation that political parties would undergo in order to survive in a superminority/citizen-jury ecosystem. In our book project (work in progress) we provide the necessary supportive argumentation.
>That depends on who they view as their supporters
The people who vote for them.
>I’m concerned that relying purely on internal deliberation at the decision-making stage would be no better than conducting a poll of the population to make the decision.
I think you underestimate the intelligence of your fellow citizens. The argumentative theory of reasoning suggests that the “evaluation” module is a powerful one and that we are all equipped with it.
>the manufacturers of consent might find an even higher gear to influence randomly-selected non-deliberative decision-making bodies.
I’m sure they would try very hard, but that’s why we need a diversity of elites to beat each other up. We need agonism, not consent.
>your main concern seems to be AI taking over the decision-making process, or perhaps it is only the perception by the public of such.
Both. But remember that the context of our conversation is using an AI to select the discourses that need representation, and the evidence that I cited earlier suggests that the tech companies have an extremely bad record in this respect. I agree with your other concerns, but I think the diverse advocates returned by the superminority electoral process would be best equipped to tackle them.
LikeLike
Alex and myself are acutely aware that current electoral systems encourage power seeking as an end in itself. I think you underestimate the radical transformation that political parties would undergo in order to survive in a superminority/citizen-jury ecosystem. In our book project (work in progress) we provide the necessary supportive argumentation.
Looking forward to it.
>That depends on who they view as their supporters
The people who vote for them.
I think the people who donate to them, especially the large donors are the ones that call the shots. The voters only occasionally can create a large enough unified opposition to overcome the donors clout. Currently, Biden is threatened with losing his bid due to his handling of the situation in Gaza. Perhaps he has no way out, perhaps the pro Israel voters would toss him if he softened his stance. That is not clear to me, but long observation of politics in the US has convinced me that the voters have little say in determining policy. The Gilens and Paige study only confirmed an obvious truth.
>I’m concerned that relying purely on internal deliberation at the decision-making stage would be no better than conducting a poll of the population to make the decision.
I think you underestimate the intelligence of your fellow citizens. The argumentative theory of reasoning suggests that the “evaluation” module is a powerful one and that we are all equipped with it.
That is not my understanding of the ATR. The theory is that reasoning works well in argument, in external deliberation, but is not so effective in solo evaluation due to human biases. Even internal deliberation, where imaginary foes are created, is less effective than external. Since AI is on the table I asked for a summary in case I had grossly misinterpreted the theory:
…the argumentative theory of reasoning, which posits that the primary function of human reasoning is to argue and persuade in social contexts rather than to enhance individual cognitive processes.
Key points of the theory include:
Social Nature of Reasoning: Reasoning is more effective in group settings where individuals can present arguments and counterarguments. This social aspect helps to refine ideas and promote better decision-making.
Bias Towards Confirmation: Humans are naturally inclined to seek out and interpret evidence that confirms their preexisting beliefs. This bias is less of a flaw and more of a feature that enhances argumentative effectiveness, allowing individuals to construct stronger arguments to persuade others.
Epistemic Benefits of Deliberation: Group deliberation and discussion can lead to improved collective decision-making, as the exchange of diverse viewpoints and critical engagement can correct individual biases and errors.
Role of Institutions: Effective institutions are crucial in facilitating productive deliberation. Institutions that promote open dialogue, critical scrutiny, and equal participation can harness the benefits of argumentative reasoning for better governance and decision-making.
Critique of Individualistic Models: The theory criticizes models that view reasoning as an individual cognitive process aimed at truth-seeking. Instead, it emphasizes the role of social interaction in shaping and enhancing human rationality.
The argumentative theory of reason highlights the importance of dialogue, debate, and collective reasoning in human cognition and decision-making processes.
It still seems to me that most of the effort to select decision-making members by lot would be for naught if they are only going to make decisions based on their individual perspectives. Taking a non-deliberative poll of the population, letting them choose between the various agendas, would be more expedient, given the likely similarity of outcome.
>the manufacturers of consent might find an even higher gear to influence randomly-selected non-deliberative decision-making bodies.
I’m sure they would try very hard, but that’s why we need a diversity of elites to beat each other up. We need agonism, not consent.
Aye, there’s the rub. Elections are not the best system for generating diversity, fortunately the superminority requirements ameliorates some of its faults. The allotted body is diverse, but since they do not deliberate together, in what sense are they “beating each other up?” I’d be concerned that they would only select the proposal that best matched their respective biases. Five proposals are likely to receive the distribution of votes that they would in the general population, if the sample size of the assembly was sufficiently large to represent that population. Again, just conduct a poll of what you want is popular opinion to make the decision. It is deliberation that can change minds, as in the two stories about the jury decision.
>your main concern seems to be AI taking over the decision-making process, or perhaps it is only the perception by the public of such.
Both. But remember that the context of our conversation is using an AI to select the discourses that need representation
The suggestions put forward by Landemore, on the usage of AI that MIT and Cortico are developing, do not enable AI to select the agenda. It is only an aid to transcribe, highlight salient points (again selected by humans, hopefully allotted), and establish percentages of beliefs in various proposals. AI doing the actual selection is a different topic than what I am interested in exploring. That might be something worth considering in the future after using it extensively as an augmentation for human deliberation and decision-making, or it may not.
LikeLike
Lance,
I think Gilens and Page overstate their case: see the long debate we had on this forum: https://equalitybylot.com/2014/04/17/commentary-on-gilens-and-page-average-citizens-have-no-political-influence/ But remember Alex and I are not seeking to defend electoral democracy in its current incarnation.
I need to re-read the Mercier/Landemore paper before responding to you points on ATR.
>The allotted body is diverse, but since they do not deliberate together, in what sense are they “beating each other up?”
It’s the advocacy elites, overflowing with confirmation bias, that beat each other up. The role of the jurors is to judge who is the most persuasive. They deliberate in the latinate rather than germanic sense (weighing arguments, rather than deliberative stimme).
>AI is only an aid to transcribe highlight salient points . . . and establish percentages of beliefs in various proposals.
OK, but quis custodiet ipsos custodes? The AI is a black box, and lay people do not have the expertise to expose any algorithmic biases. Remember that Helene’s primary concern is deliberative quality, not democratic legitimacy.
And we have yet to hear how the discursive diversity of the population can be represented, absent AI or elections under superminority.
LikeLike
Lance,
My understanding of the entailments of the ATR for sortition-based democracy are somewhat different. Forgive me for a longer response than usual:
The ATR hypothesizes that reasoning serves two survival-related functions: a) convincing people and b) evaluating the arguments of others – ‘thereby allowing communication to proceed even when trust is limited’ (Landemore, 2013, p. 126). The theory – developed as an evolutionarily-plausible alternative to the classical (Cartesian) model of reasoning as a way of updating and correcting one’s own beliefs – is based on the distinction between performing speech acts and evaluating the performative utterances of others:
“According to this theory, individual reasoning works best when used to [a] produce and [b] evaluate arguments during a public deliberation. It predicts that when diverse opinions are discussed, group reasoning will outperform individual reasoning.” (Mercier and Landemore, , p. 243)
According to Mercier and Landemore the argumentative theory of reasoning privileges intra-group exchanges as opposed to individual ‘deliberation within’. However they appear to agree with (Goodin, 2000) that ‘internal deliberation can sometimes be very similar to public deliberation’ (Mercier and Landemore, 2012, p. 251). What matters is that ‘exposing people to disagreement and debates increases their ability to entertain different opinions . . . either by witnessing a debate or by being part of one’ (ibid., p. 252). The important factor is not participation in speech acts so much as ‘the presence or expression of dissenting opinions in deliberative settings’ (ibid., p. 254, my emphasis). (Mercier & Sperber, 2017) also point out that as a species we are much better at evaluating reasons than producing them – I may not be able to see the beam in my own eye, but I can find the mote in yours. Appeals to authority underlying this claim cited in The Enigma of Reason includes William Blackstone’s arguments on the value of juries, backed up by field work demonstrating that they do a good (epistemic) job. Laboratory studies also indicate that we are better able to find the flaws in our own reasons when we believe those reasons to have been produced by someone else, thereby supporting the case for the division of labour between persuaders and evaluators.
Bernard Manin, in his brief review of experiments in Deliberative Polling, argues that the success of the DP in avoiding group-think and polarization is on account of the ‘presence of diverse and conflicting views amongst deliberators’. He conjectures that the introduction of the graphe para nomon into Athenian democracy was because ‘the adversarial proceedings were required during the second hearing on grounds of their superior epistemic merits’, (Manin, 2005, pp. 8, 17).
Given that the argumentative theory of reasoning suggests two distinct mechanisms this would also suggest two categories of persons in a deliberative context – proposers and disposers. The former category
“should only be concerned with . . . [arguments] that increase the plausibility of a conclusion one is trying to defend, or those that decrease the plausibility of a conclusion one is trying to rebut. Representations that have the opposite effect are of no direct value if one wants to convince one’s interlocutor. Reasoning should therefore be directed towards these valuable representations, and, as a result, it should display a strong confirmation bias” (Mercier and Landemore, 2012, p. 241, emphasis in original)
The confirmation bias, however, ‘mostly affects the production, and not the evaluation of arguments’ and can be ‘checked, compensated by the confirmation bias of individuals who defend another opinion’, hence the necessity for ‘the presence or expression of dissenting opinions in deliberative settings’ (ibid., pp. 251, 253, 254). The requisite cognitive tool for those whose task is to evaluate communicated information (i.e. disposers) is ‘epistemic vigilence’, the polar opposite of the confirmation bias (Mercier, 2013; Sperber et al., 2010).
Mixing the two cognitive styles in the same persons, as Landemore and most deliberative democrats advocate, is sub-optimal as while ‘reasoning is mostly truth oriented when it come to evaluating arguments, it is still biased by the function of producing arguments’ (ibid., p. 127, my emphasis). Whilst it is possible for the same agents to oscillate between the two different cognitive styles . . .
“After reviewing and weighing for ourselves the reasons for and against a given action, we come to a conclusion. We then take a position. However, when we speak in public in the course of deliberation, we share only the part of information that supports our position.” (Manin, 2005, p. 16)
. . . nevertheless, cognitive dissonance theory would suggest that it is more difficult to modify one’s evaluation of an argument once having expressed an opinion. E.M. Forster’s dictum ‘how do I know what I think until I [hear] what I say’ could be taken as contributing to the confirmation bias – it’s much harder to change your mind once you know what you think. Speech acts intended to persuade (whether oneself or others) are inherently partisan and Landemore makes a good case for the confirmation bias as a valuable cognitive tool for proposers and other rhetorical persuaders (but not for disposers).
LikeLike
Keith,
It will take me some time to get around to reading through the debate you posted and addressing many of your points as I am unusually busy just now. I will only respond to one at the moment.
OK, but quis custodiet ipsos custodes? The AI is a black box, and lay people do not have the expertise to expose any algorithmic biases.
The algorithmic workings of the AI may be opaque to most of us, but the output is transparent. One doesn’t even have to know basic arithmetic to see if a discussion was transcribed properly. Transcribing something like 100,000 one hour deliberations would be an enormous task for us, but an AI can quickly accomplish it. Humans reading through the transcripts could highlight salient points, then do a search to count how often such appeared. The search function is something we all use probably everyday now without thinking we have given up tremendous freedom, or that there are nefarious actors manipulating it. If someone has such a fear, they could check manually to test the AI’s accuracy by sampling.
I think it is important to talk specifics on this issue to gain any understanding of the risks and rewards.
LikeLike
Thanks for the clarification Lance. I can see how AI might be an aid for transcribing conversations and other administrative tasks. Unfortunately this doesn’t have any relevance to democratic representation, hence my reluctance to consider AI for the agenda-setting role in a legislative body. I think it’s important to realise that deliberative “democrats” are, on the whole, indifferent to issues of democratic legitimacy — they simply want to improve the quality of deliberation. As this would apply to a group constituted by a wide variety of selection methods, I don’t see that it has any relevance to sortition per se.
LikeLike
I fully agree with Keith’s interpretation of the argumentative theory of reasoning, AND the implications for dividing tasks. Once a person has spoken (taken a stand) on an issue, that act psychologically hobbles their ability to absorb and evaluate contradictory information and change their mind. (How often have two people started arguing about something they both have strong opinions on (even worse, perhaps pinned their tribe and personal identity to), ended the argument with one saying “you were right and I was wrong.” It can happen, but is not typical. (More likely go away thinking the other guy was a real jerk, who just couldn’t see reason)… the mote and beam analogy. Here is a link to chapter section from my book that makes essentially the same points Keith does, but perhaps in slightly more layman’s language (though I cite many of the same sources).
https://open.substack.com/pub/democracycreative/p/the-argumentative-theory-of-reasoning?r=cvh3h&utm_campaign=post&utm_medium=web
LikeLike
That’s very nicely put Terry. I particularly liked:
“I had an eye-opening experience while serving on the Commerce Committee in the Vermont House of Representatives. Rather than following the usual procedure of taking testimony from one witness at a time, on one occasion we had a panel of opposing lobbyists, essentially conduct an organized argument in front of the committee for a bill on which none of us had prior stances. The lobbyists would call each other out, since they knew when an opponent was intentionally cherry-picking data, or skirting an important issue, which none of us less-informed committee members could detect.”
It’s interesting that democrats (Terry, Keith) and deliberationists (Helene, Lance) draw such different entailments from the same scientific theory. IMO this is because the cognitive bias of deliberationists makes them assume a Habermasian position, come what may. No doubt they would retort that I’m equally biased by the electoralist paradigm. But it strikes me that Terry’s own experience acknowledges the value of “opposing lobbyists” and that the way to fully democratise the decision process is to ensure a wide range of discourses via the superminority principle. It takes professional expertise to know if someone is cherry picking or skirting, but once the opposing lobbyist has pointed this out, the jury is in a position to judge the outcome. We accept that this is the role of the trial jury, but deliberationists are loath to apply the same principle to (say) a legislative jury, as they are so beholden to the Burkean ideal of ‘the pursuit of the laws of God and nature’ (Pitkin, 1967, p.169), even though the ATR rules that out as a Platonic pipe dream.
LikeLike
[…] published some anti-lottocracy paragraphs from Christina Lafont recently. It spurred me to read her book and compose a rebuttal, which I’m in the midst of. In […]
LikeLike
Terry, you said:
Once a person has spoken (taken a stand) on an issue, that act psychologically hobbles their ability to absorb and evaluate contradictory information and change their mind. (How often have two people started arguing about something they both have strong opinions on (even worse, perhaps pinned their tribe and personal identity to), ended the argument with one saying “you were right and I was wrong.” It can happen, but is not typical.
I have ,and I have seen, people change their minds too many times to count. My experience differs from yours. It seems to me that you are claiming that deliberation is a waste of time. If so, why include it at any step? It seems that a variety of proposals could easily be gathered from the public at large, media, political writers, then put before the public as referendums and let them decide. Why mess with elections or lotteries, since deliberation is so useless by your account? It would be simpler and less expensive to just count opinions, since they won’t change very often through deliberating.
LikeLike
Keith,
I think it’s important to realise that deliberative “democrats” are, on the whole, indifferent to issues of democratic legitimacy — they simply want to improve the quality of deliberation.
Remember that Helene’s primary concern is deliberative quality, not democratic legitimacy.
Could you point out a passage or passages where Helene takes such a stand? I have never seen that, but no doubt I have not read her as thoroughly as you. I did read Can AI Bring Deliberative Democracy to the Masses? several times, but one needn’t read past the first paragraph, where she states that both deliberation and mass participation are equally important for democratic legitimacy. Perhaps her views have changed.
LikeLike
Lance:> Could you point out a passage or passages where Helene takes such a stand?
This was during a long email exchange while she was advising me on my PhD thesis, and wondered why I was so hung up that the minipublic should accurately (and consistently) reflect the informed preferences of the masses, who would be disenfranchised by the aleatory coup. Read her first book (Democratic Reason) and you’ll see that her concerns are purely epistemic. She distances herself from Estlund’s book Democratic Authority, which does treat with democratic legitimacy (both formal and perceived). Epistemic and deliberative democrats have embraced sortition eagerly, on account of its [questionable] representative claims, as it means they don’t have to bother with such matters.
>I have seen people change their minds too many times to count.
People will change their minds on issues which do not affect them directly, but it’s much easier to do so by (internally) weighing the competing arguments of others. That’s why Bernard Manin, who’s 1987 paper led to the deliberative turn in democratic theory, is now advocating debate, rather than deliberation. And the distinction between advocates and judges is exactly what you would expect from the ATR.
LikeLike
>deliberation and mass participation are equally important for democratic legitimacy.
Bear in mind that few people on this blog are arguing for mass participation. A long time ago I invited Carole Pateman to join the ed board of a periodical that was to be entitled the Journal of Deliberative and Participatory Democracy, but she pointed out that the two have nothing in common, so I dropped it.
LikeLike
Read her first book (Democratic Reason) and you’ll see that her concerns are purely epistemic.
From the Amazon intro to Democratic Reason:
“Democracy as a form of government is therefore valuable not only because it is legitimate and just, but also because it is smart.”
These are the only kind of statements I have seen Helene make, not choosing one over the other but claiming that either both are equally important aspects for democratic decision-making, as in the article on AI mentioned above, or the epistemic component is an additional reason for choosing more inclusive deliberative decision-making. I still would be curious to see a passage where she states a preference for deliberation over democracy.
Ultimately it doesn’t matter what Helene believes, and it is no doubt best for us not to speculate. If you are still in contact with her, and could ask her directly, I would be interested in her response to your query. I will limit myself to my own views and state that they are both that deliberation and inclusiveness are necessary components of any true democracy. I also believe the US fails at both.
Bear in mind that few people on this blog are arguing for mass participation.
Yes, Helene is exploring the possibility of both in Can AI Bring Deliberative Democracy to the Masses?
Note it is put forward as a question. I won’t speak for people on the blog, but it does seem desirable to include as many people as wish to be included and is practically possible. I realize there is problems with the former, the issue of unrepresentative participation. AI, is being offered as a possibility by Helene to expand the practical possibility of the latter.
A long time ago I invited Carole Pateman to join the ed board of a periodical that was to be entitled the Journal of Deliberative and Participatory Democracy, but she pointed out that the two have nothing in common, so I dropped it.
I have long thought that the lack of democracy in the workplace is a major component in the dire state of political participation and understanding. I was unfamiliar with Ms. Pateman, so thank you for the introduction. I cannot agree with her assessment on the unrelatedness of deliberation and participatory democracy, however. The situation in the workplace, and in schools and religious institutions, I might add, are the perfect examples of institutions that that promote Kant’s nonage of the population, and makes them unfit to govern themselves. I have little interest in a “democracy” that simply garners non reflective opinion as its basis for making decisions. That is not democracy at all, in my view, but a somewhat sophisticated form of plutocracy, as the plutocrats have majority ownership of the information dissemination organs. I am unclear what her definition of participation is. It seems she seeks more than merely expressing one’s opinion through a vote.
People will change their minds on issues which do not affect them directly, but it’s much easier to do so by (internally) weighing the competing arguments of others.
That probably is the case, although I have witnessed many times where people come around to a group consensus through group deliberation. You and Terry appear to hold contradictory views, as you both are now arguing that external deliberation is ineffective at changing minds where earlier you both were arguing that it produces conformity of thought because the participants change their minds under the influence of high status or eloquent deliberators. Which is it?
LikeLike
Lance,
No, you misunderstand my view. I apparently have not made myself clear enough. I have an entire chapter of my book about the importance of active give-and-take deliberation within a diverse and representative mini-public to seek out possible win-win consensus policies, and if not able, to craft reasonable policies that the population as a whole would endorse if they could all somehow have the time and motivation and resources to become well-informed (like the mini-public). Within that mini-public some members will soon take stands and become less able to change their stands…. but the group as a whole may be open minded and able to judge the arguments of others. However, there is also a danger of groupthink and information cascades that takes the entire mini-public down a bad path. That is why, as a check and balance, a SEPARATE jury gets to learn, hear all the pro and con arguments and judge the final draft bill.
You can find my book chapter (chapter 11) on all of these points about deliberation here in the book archive:
https://democracycreative.substack.com/p/the-trouble-with-elections?r=cvh3h&utm_campaign=post&utm_medium=web
I should note that chapter 11 also implicitly explains why I don’t favor electoral parties setting the agenda and drafting bills for a jury to consider, as in the “super-minority” proposal. Parties have relatively like-minded, self-selected people who are not representative of the population, are often narcissistic, and who rarely constructively engage with or listen to people they disagree with (and demonize them if they do). So the policy proposals parties produce will almost never seek win-win options, nor be optimal. The population will be forced to choose the lesser of evils, both in the election AND then again if selected for a jury (less evil policy this time).
It is possible that if the super-minority system were in place and party politicians had no real power (only the power to propose), that over time many of the current power-hungry narcissists who currently fill the ranks would eventually abandon party politics, so the nature of parties could change. It is hard to anticipate the effect.
LikeLike
Lance,
My overriding concern is for the representation of the vast majority of citizens who would be disenfranchised by the aleatory coup, rather than just “as many people as wish to be included”. That may have worked well at the time of the origins of the public sphere in eighteenth-century coffee house culture, but democracy in large multicultural states is, of necessity, representational and the only question is what combination of institutions can best achieve this.
>You and Terry appear to hold contradictory views, as you both are now arguing that external deliberation is ineffective at changing minds where earlier you both were arguing that it produces conformity of thought because the participants change their minds under the influence of high status or eloquent deliberators.
My understanding of the ATR is that committed advocates will not change their minds even if hell freezes over (nearly 20 years of debating on this blog has confirmed this), but that juries will, especially if they refrain from speech acts. The Superminority Principle (SP) will diversify the policy options, but parties will continue to choose eloquent persuaders as that is why evolution has developed this cognitive module. If there is a choice between naturalistic and ideal explanatory models, I would opt for the former. The cybernetic link between proposers and disposers suggests that the natural selection metaphor applies to both. Win-win does not require consensus — that’s a hangover from the (discredited) classical theory of reason, that can be traced back to Plato.
Terry:> So the policy proposals parties produce will almost never seek win-win options, nor be optimal.
SP would suggest that, so long as the proposal threshold is low enough, the win-win compromise will be selected for, purely on the strength of the LLN. As to whether SP will diminish the number of power-hungry narcissists, we’ll just have to wait and see. But it’s important to remember that Alex is proposing SP for the executive branch as well.
LikeLike
Keith… I don’t know all your short hand (so other readers won’t either) ATR is the “argumentative theory of reasoning” of Sperber et al.
SP = ?
LLN = ?
LikeLike
Glossary!
ATR: Argumentative Theory of Reasoning
SP: Superminority Principle
LLN: Law of Large Numbers
LikeLike
Terry,
> Within that mini-public some members will soon take stands and become less able to change their stands…. but the group as a whole may be open minded and able to judge the arguments of others. However, there is also a danger of groupthink and information cascades that takes the entire mini-public down a bad path. That is why, as a check and balance, a SEPARATE jury gets to learn, hear all the pro and con arguments and judge the final draft bill.
The word “some” is left out of this earlier statement of yours, thus the confusion:
Once a person has spoken (taken a stand) on an issue, that act psychologically hobbles their ability to absorb and evaluate contradictory information and change their mind.
I agree that that sometimes happens, but it is not an inevitable outcome of group deliberation, which I will take up later in my reply to Keith.
As to the danger of groupthink, I believe that is way overblown. The experiments you referred to in your book have little in common with a facilitated group-deliberative process which greatly reduces such dangers. As you pointed out in your book, there has not been evidence of such occurrences in deliberative polls and CAs.
The groupthink I believe is a danger to thoughtful decisions is that caused by the corporate control of the public sphere. Relying strictly on internal deliberation to make the determinate decision has a far more dangerous possibility of introducing such groupthink into the process than relying on group deliberation. If you feel you must eliminate group deliberation, I’d feel more comfortable eliminating it, or reducing it to the modicum necessary, at the proposal stage, rather than the policy choice stage. There is little that needs to be decided at that stage anyway, if using Keith’s model. The lack of power the proposers have, will, as you say, possibly reduce the number of narcissistic office seekers. I think it will also be the reason it will be fiercely opposed by those who currently hold office. Only time will tell.
LikeLike
Lance:> The groupthink I believe is a danger to thoughtful decisions is that caused by the corporate control of the public sphere.
If you take a look at the group psychology literature I think you’ll find that it’s more widespread, and generated primarily by internal factors (nobody wants to be the odd one out). Take a look at Surowiecki’s book for some interesting examples. As for the need for moderators (whether human or AI), if the group has the right to make binding decisions, then you can’t escape quis custodiet? (presupposing democratic norms).
LikeLike
Keith,
None of the systemic flaws Suroweiki points to that result in group conformity are present in a well run CA or mini -public.
If you know of an actual case where a CA has made a decision due to such systemic flaws please point it out.
LikeLike
Keith,
My overriding concern is for the representation of the vast majority of citizens who would be disenfranchised by the aleatory coup…
Ha! While I appreciate the clever and creative language, it is hyperbolic. I am unaware of any sortition advocate advocating for the forceful adoption of sortition. Perhaps there are some unwilling to openly admit such. I have only heard scenarios where the public majority wishes it and adopts it. True, a minority would then feel disenfranchised. They would then be in the same boat as I and many others are in the current system. I was introduced to sortition by reading about a poll where 65% of the respondents believed randomly selected people would perform better than those currently in congress. You can’t please all the people all of the time, let’s just try to please most.
…“as many people as wish to be included”. That may have worked well at the time of the origins of the public sphere in eighteenth-century coffee house culture, but democracy in large multicultural states is, of necessity, representational and the only question is what combination of institutions can best achieve this.
Not sure it is of necessity representational. The advocates for AI enhanced deliberation are looking at the possibility of expanding participation by orders of magnitude. We still may be a long way off from including everyone that wants to participate, but I see no reason to make absolute, eternal statements.
>You and Terry appear to hold contradictory views, as you both are now arguing that external deliberation is ineffective at changing minds where earlier you both were arguing that it produces conformity of thought because the participants change their minds under the influence of high status or eloquent deliberators.
My understanding of the ATR is that committed advocates will not change their minds even if hell freezes over (nearly 20 years of debating on this blog has confirmed this),
Come now, this blog is a far cry from a deliberative body. There is no moderation or facilitation. The site owner has admitted to not understanding what deliberation is, which does not prevent him from violating Bob Dylan’s dictum to not criticize what you can’t understand. Your understanding of ATR does not square with the experimental evidence gathered by Fishkin and his colleagues at Stanford. Confirmation bias doesn’t imply that people will never change their mind. I have never seen Mercier, Sperber, or Landemore make such a claim. Then again, you also claim that everyone will change their mind in a group deliberation and conform to information cascades, high status and eloquent persuaders, so I am still unable to understand your position. Is it that some will not, and others will conform? That may be so, but why would allowing internal deliberation only produce better results? Less, conformity, possibly,but probably more people sticking to their adamant positions. Confirmation bias is less easily overcome via internal deliberation than through group deliberation.
I believe the examples conjured up by people to conclude that minds aren’t changed through deliberation are mainly one-on-one dialogues, or occasionally parties of three. Even adding a third party facilitates agreement. This blog is primarily one-on-one conversations with little collective agreed upon goals. I recall my introduction to Fishkin’s work came about through reading an article in the New Yorker, Why Facts Don’t Change Our Minds. It was the most depressing article of 2017. Fishkin’s work, however, was the one sliver of hope offered. When people were assembled to create policy, they were able to lessen the gap between them. Perhaps this is what you would call conformity, but it could also be labeled depolarization.
The Superminority Principle (SP) will diversify the policy options, but parties will continue to choose eloquent persuaders as that is why evolution has developed this cognitive module.
Seems like those eloquent persuaders could bring about group conformity even if the second step was limited to independent internal deliberators.
If there is a choice between naturalistic and ideal explanatory models, I would opt for the former.
I’m unclear what you are defining as naturalistic vs ideal in regards to our topic. Clubbing each other over the head is one naturalistic way to reach agreement. I prefer a more ideal method.
LikeLike
Lance:> If you know of an actual case where a CA has made a decision due to such systemic flaws please point it out.
How would you even know (absent a control group) the extent to which the deliberations of a minipublic have been distorted by information cascades etc. The nearest we have is the multiple Texas utility DPs, which came to very different decisions on the same topic. My guess is that the variance was on account of systemic flaws.
LikeLike
Keith and Terry,
More on ATR, confirmation bias, and groupthink,
Abstract from Reasoning Is for Arguing: Understanding the Successes and Failures of Deliberation by Mercier and Landemore:
Theoreticians of deliberative democracy have sometimes found it hard to relate to the seemingly contradictory experimental results produced by psychologists and political scientists. We suggest that this problem may be alleviated by inserting a layer of psychological theory between the empirical results and the normative political theory. In particular, we expose the argumentative theory of reasoning that makes the observed pattern of findings more coherent. According to this theory, individual reasoning mechanisms work best when used to produce and evaluate arguments during a public deliberation. It predicts that when diverse opinions are discussed, group reasoning will outperform individual reasoning. It also predicts that individuals have a strong confirmation bias. When people reason either alone or with like-minded peers, this confirmation bias leads them to reinforce their initial attitudes, explaining individual and group polarization. We suggest that the failures of reasoning are most likely to be remedied at the collective than at the individual level.
(emphasis mine)
Keith said (again bold emphasis mine):
According to Mercier and Landemore the argumentative theory of reasoning privileges intra-group exchanges as opposed to individual ‘deliberation within’. However they appear to agree with (Goodin, 2000) that ‘internal deliberation can sometimes be very similar to public deliberation’ (Mercier and Landemore, 2012, p. 251). What matters is that ‘exposing people to disagreement and debates increases their ability to entertain different opinions . . . either by witnessing a debate or by being part of one’ (ibid., p. 252). The important factor is not participation in speech acts so much as ‘the presence or expression of dissenting opinions in deliberative settings’ (ibid., p. 254, my emphasis). (Mercier & Sperber, 2017)
If it is sometimes very similar to public deliberation, then the implication is that most of the time it is not, and when it is not it is subject to confirmation bias, and yes, groupthink due to mass media indoctrination. I am not convinced that confirmation bias can be overcome merely by exposing the internal deliberators to written material, or even recorded debates, due to the findings described in the New Yorker article mentioned above. In a nutshell, when people are exposed to facts that contradict their beliefs, their beliefs grow stronger. Perhaps, if Fishkin-like experiments were run with internal deliberators commissioned to come up with policy solutions we could settle this. You may know of such experiments, if so, please pass them on.
I asked chat GPT some questions about internal vs external deliberation and the ATR. Forgive me for relying on an AI bot, but I am unable to copy and paste directly from the literature, and I am a slow typer.
Summarize the distinctions and effectiveness of internal vs external deliberation according to the argumentative theory of reasoning
The argumentative theory of reasoning posits that human reasoning evolved primarily for argumentation, aiming to persuade others and evaluate their arguments. This theory contrasts two main types of reasoning: internal deliberation and external deliberation.
Internal Deliberation
• Definition: This involves an individual reflecting and reasoning within their own mind.
• Characteristics: Often prone to confirmation bias, where individuals seek out and favor information that supports their pre-existing beliefs.
• Effectiveness: Generally less effective for objective reasoning because it may reinforce existing biases and lead to overconfidence in one’s own ideas.
External Deliberation
• Definition: This occurs when individuals engage in reasoning through dialogue and debate with others.
• Characteristics: Facilitates the presentation and critique of arguments, promoting critical thinking and exposure to diverse perspectives.
• Effectiveness: Typically more effective for reaching well-reasoned conclusions as it involves challenge and scrutiny from others, which can reduce biases and errors.
In summary, according to the argumentative theory of reasoning, external deliberation is more effective than internal deliberation due to its capacity to mitigate biases and enhance critical evaluation through social interaction.
Does external reasoning lead to groupthink?
External reasoning, when done correctly, generally serves as a countermeasure to groupthink rather than a cause. However, under certain conditions, external reasoning can still contribute to groupthink. Here’s a nuanced view:
External Reasoning and Groupthink
1. Promotion of Diverse Views: Ideally, external reasoning involves the exchange of diverse perspectives and critical evaluation, which helps prevent groupthink. This process encourages individuals to question assumptions and consider alternative viewpoints, leading to more robust decision-making.
2. Conditions Contributing to Groupthink:
• Homogeneous Groups: If a group lacks diversity in perspectives, even external reasoning can lead to groupthink. Members might echo each other’s thoughts, reinforcing a dominant viewpoint without substantial critique.
• Dominant Leadership: In groups where a leader’s opinion is overly influential, members might conform to the leader’s views during external deliberation. This can suppress dissenting opinions and critical analysis, fostering groupthink.
• Social Pressures: In environments where dissent is discouraged or penalized, external reasoning may not effectively counter groupthink. Members may self-censor or align with the majority to avoid conflict, limiting the benefits of external deliberation.
Effective External Reasoning Practices
1. Encouraging Open Dialogue: To mitigate groupthink, it is crucial to create an environment that values and encourages open dialogue. This includes actively seeking out and considering dissenting opinions.
2. Structured Techniques: Methods such as the Devil’s Advocate approach or the use of structured debate formats can help ensure that external reasoning remains critical and diverse. These techniques require participants to argue against prevailing opinions or explore multiple sides of an issue.
3. Inclusive Leadership: Leaders should foster a culture where questioning and critical thinking are encouraged. This includes being open to feedback and not penalizing dissent.
4. Diverse Group Composition: Ensuring a diverse group composition in terms of backgrounds, expertise, and perspectives can enhance the effectiveness of external reasoning and reduce the likelihood of groupthink.
In summary, while external reasoning is generally an effective tool to combat groupthink, its success depends on the context and the way it is practiced. Homogeneity, dominant leadership, and social pressures can undermine the benefits of external reasoning, potentially leading to groupthink. Therefore, fostering an environment of open, inclusive, and structured deliberation is key to leveraging external reasoning effectively
Does internal reasoning lead to group think?
Internal reasoning does not directly lead to groupthink, as groupthink is a phenomenon that occurs in group settings. However, internal reasoning can contribute to groupthink under certain conditions. Here’s how:
Internal Reasoning and Groupthink
1. **Confirmation Bias**: Internal reasoning often involves confirmation bias, where individuals favor information that aligns with their pre-existing beliefs. When individuals within a group predominantly engage in this type of biased internal reasoning, they may come to group discussions already aligned in their thinking, reinforcing a homogeneous perspective.
2. **Reinforcement of Beliefs**: If individuals privately reinforce their beliefs without external challenge, they might become more entrenched in their views. When these entrenched views are shared within a group, it can lead to a consensus without sufficient critical evaluation.
Groupthink Dynamics
1. **Lack of Diverse Perspectives**: Groupthink thrives in environments where dissenting opinions are suppressed or absent. When group members have similar internal reasoning patterns, the group is less likely to encounter diverse viewpoints, increasing the risk of groupthink.
2. **Social Pressure**: In group settings, social pressure to conform can discourage individuals from voicing dissenting opinions. Even if someone internally questions the group’s direction, the fear of social isolation or conflict may prevent them from speaking up, further contributing to groupthink.
Mitigating Groupthink
1. **Encouraging External Deliberation**: Promoting open dialogue and debate within a group helps to counteract the effects of biased internal reasoning. By encouraging external deliberation, groups can ensure that multiple perspectives are considered, reducing the likelihood of groupthink.
2. **Fostering a Culture of Dissent**: Creating an environment where questioning and critical evaluation are valued can help mitigate groupthink. Encouraging members to challenge prevailing views and consider alternative solutions promotes healthier decision-making.
In summary, while internal reasoning alone does not cause groupthink, it can contribute to the conditions that facilitate groupthink when coupled with social pressures and a lack of diverse perspectives in group settings. External deliberation and a culture of open dialogue are crucial in mitigating these risks.
I will add that most of the “failures” of external deliberation resulting in groupthink illustrated by the various psychological experiments are when the conditions suggested for overcoming it are not fulfilled. The experiments are valuable in identifying processes and conditions to ameliorate or overcome groupthink, not as reasons for throwing the baby out with the bath water.
LikeLike
Lance:> I am unaware of any sortition advocate advocating for the forceful adoption of sortition.
There are many on this blog who seek to fully replace elections with sortition, thereby disenfranchising most citizens. I never suggested the use of force, but coups take many different forms (including the long march through the institutions).
Not sure [democracy] is of necessity representational.
Then it would have to be plebiscitary. If it’s limited to the minority who choose to participate, then it’s a variant of oligarchy.
>Is it that some will not, and others will conform?
Activists will stick to their guns and jurors will be subject to group dynamics. The ATR is a dualistic theory, but deliberationists concatenate both categories together.
>minds aren’t changed through deliberation
But jurors’ minds are changed — the challenge is ensuring that the changes match the considered preferences of their peers. “What everyone would think under good conditions”, as Fishkin puts it.
>Seems like those eloquent persuaders could bring about group conformity even if the second step was limited to independent internal deliberators.
Of course, that’s why we need to ensure the range of persuaders is a reasonable match to the ideological diversity of the citizenry. Each party wants to win the debate, and will choose the most eloquent persuaders, but it’s the jury that decides the outcome.
LikeLike
That’s a long one Lance, and time is a finite resource.
>According to [ATR], individual reasoning mechanisms work best when used to [a] produce and [b] evaluate arguments during a public deliberation. It predicts that when diverse opinions are discussed, group reasoning will outperform individual reasoning.
But we all agree on that. The issue is whether the two cognitive modules are so opposed that, paraphrasing Madison, combining them in the same body of men leads to corruption. Although the wisdom and crowds and collective wisdom sound synonymous, Helene (rightly) concludes that they are entirely different. The former is aggregative and, using Rousseauian terminology, reveals the will of all (a simple majority will suffice), whereas the sort of deliberative group that she champions is seeking the general will (where unanimity is the ideal). It’s worth noting that Rousseau was no democrat and was resolutely opposed to deliberation in the general assembly (he reserved it for the government). Alex and I are argue that we should accept the Rousseauian distinction but adopt a pluralist model for the government, reserving the decision rights to the sovereign assembly.
Can I recommend Andrew Dobson’s book Listening for Democracy (alongside Alex’s essay Turn Down the Volume on Citizens’ Assemblies).
>if Fishkin-like experiments were run with internal deliberators commissioned to come up with policy solutions we could settle this.
Not really, because we deny that it’s the job of randomly-selected bodies to come up with anything. Their role, as in the trial jury, is to adjudicate between the proposals on offer, because:
>According to [ATR], individual reasoning mechanisms work best when used to [a] produce and [b] evaluate arguments during a public deliberation.
Note that the reasoning mechanisms are still individual, it’s only the setting that’s public.
LikeLike
Lance:> I’m unclear what you are defining as naturalistic vs ideal in regards to our topic. Clubbing each other over the head is one naturalistic way to reach agreement. I prefer a more ideal method.
Deliberative Democracy is a normative project designed to reveal the laws of god and nature via the cultivation of the ideal speech situation. The ATR is a description of the cognitive toolbox that evolution has equipped us with to optimise our chances of survival. It involves the adaptation of war-war into jaw-jaw, but the same dynamic is involved (however distasteful Habermasians might find that to be).
LikeLike
One key point where Keith and I disagree is his idea that he denies “that it’s the job of randomly-selected bodies to come up with anything.”
I think it is a terrible lost opportunity to limit sortition to the yes/no jury phase. Having groups of like minded, successful partisans consulting experts they agree with on designing a policy proposal for a jury to consider forfeits one of the greatest benefits of sortition… diverse perspectives and life experience engaging in give-and-take deliberation will sometimes find win-win consensus ideas that no party would think of, or favor. This “collective intelligence” is crucial. While partisans, special interest groups, odd-ball policy wonks, and ordinary people can all submit ideas to an allotted review panel, it is that body’s cobbling together of diverse ideas and discovering synergies, etc. by the diverse allotted body that is a crucial step. Lance seems to think that is the good stopping point, let this body also make the final decision. There are several reasons why this is problematic. I mentioned information cascades and groupthink, but one that I haven’t mentioned in this thread is size. There is a practical limit for all-to-all give-and-take deliberation… perhaps 150? But that is far too small to have reliable statistical representativeness of the population. We need the jury to be many times larger. But this is perfect, since they are not going to express anything to each other… just learn about the topic, listen to the best arguments the supporters and opponents of the final proposal can muster, and vote by secret ballot, to use the “wisdom of crowds.”
LikeLike
Terry:> I think it is a terrible lost opportunity to limit sortition to the yes/no jury phase.
Yes, that’s true from an epistemic perspective (see Helene’s first book). But my concerns are purely democratic — Dahl’s insistence that the demos should have total control of the policy agenda. I don’t think that’s possible without limiting proposing rights to elected political parties (with a low superminority threshold). Whereas your proposal is (loosely) based on ho boulomenos, Alex and I argue that isegoria in large modern states has to be as representative as the final decision right (isonomia). I’m puzzled as to why you accept the latter argument, but not the former.
PS Helene’s examples for collective intelligence (and selective genealogy) are dubious, but no time to go into that now.
LikeLike
Note that Terry’s proposing group would fall under the category of epistocracy (collective intelligence”), the principal difference from other such bodies (whether elected or appointed on merit) being that the volunteers select themselves.
LikeLike
The idea that anyone who wishes and contribute ideas, may have epistemic benefits, but is one of the most fundamentally democratic principles, known in ancient Greece as isegoria. This, and the associated principle of ho boulomenos (anyone who wishes) feeds INTO the allotted review panel. These rights (sometimes equated with freedom of speech) go deeper than that individual right. The COMMUNITY benefits when all are allowed to add information and perspectives. A fact known by one individual may be critical for making a good democratic decision. Thus these are BOTH individual rights and rights of the demos.
Now as for the amount of speaking, we simply need a fair democratic process for elevating useful information and weeding out trolls. Before AI, my idea was to simply have all submissions being read by a small random sample, with any submissions that nearly all thought were a waste of time not moving up the chain, in a series of steps until the best (diverse and including minority disagreeing) comments made it all the way to the randomly selected Review Panel. With AI (if it can be transparently double-checked for impartiality and accuracy) this could be sped up by orders of magnitude by generating summaries, and uniting similar proposals, etc. The idea is that anyone in the society can contribute, speak and try to educate and persuade. Every submission will be read and weighed, but nobody needs to listen to all those submissions. After the review panel crafts a final proposal a very large statistically
This is in sharp contrast to the superminority scheme where only an oligarchy of partisans can speak or submit ideas. The fact that these oligarchs have roped in enough people to cast a ballot for them merely repeats the utter failings of current electoral schemes that suffer from rational ignorance. If a voter strongly believes in policies A and B, and is unalterably opposed to policies C and D, but no party agrees with that mix, how are they to vote? (IF THEY EVEN HAVE ANY IDEA HOW THE PARTIES LINE UP ON EACH OF THESE). Perhaps the voter is anti-abortion, pro-union, favoring urgent climate action, and staunchly anti-immigrant, but no party agrees with that list. Forcing voters into pigeon-hole issue combinations is anti-democratic. Issues should be tackled à La Carte, not as oligarchic party leaders choose.
LikeLike
Terry:> The idea that anyone who wishes and contribute ideas, may have epistemic benefits, but is one of the most fundamentally democratic principles, known in ancient Greece as isegoria.
Sure, but in practice few did, and the majority of speakers were rhetores. We both accept, given the difference in scale of ancient and modern poleis, that the final decision (isonomia) should be taken by a representative sample, yet you refuse to accept that isegoria requires representation, such is your confirmation bias against the electoral principle. (A paper from a classicist arguing that election was fundamental to ancient Greek demokratia is currently under review for the first issue of JoS.)
>Forcing voters into pigeon-hole issue combinations is anti-democratic. Issues should be tackled à La Carte, not as oligarchic party leaders choose.
Agree. A la carte is fundamental to the superminority principle. Oligarchy comes in various shapes and forms, and democratic norms presuppose that we get to choose between our oligarchs, rather relying on those who put their hands up (or who design the AI algorithm). But I’ll ask Alex (who also knows a thing or two about data science) to respond to your last paragraph.
LikeLike
That’s a long one Lance, and time is a finite resource.
I’ve been busy and have not been able to keep up with this thread. Just a few thoughts in catch up:
Yes, Chat GPT 4.0 is loquacious, once again, my apologies. In my defence, I would offer that learning Greek and deciphering academic language is quite demanding on my limited resources, both temporally and intellectually. It has been educational, however!
…but coups take many different forms (including the long march through the institutions).
That is usually called social change. Plain English for the word “coup” is reserved for illegal, forceful replacement of the government. Even the disenfranchisement of the majority that you speak of would be unlikely. It would be hard to adopt sortition without majority approval, and probably a large majority at that (due to opposition by entrenched power). As mentioned, the poll I saw claimed that 65% of the US population felt that citizens chosen at random would be preferable to the current congress. Perhaps it is that 65% that is being disenfranchised by the current representational system that is oligarchic.
I am hesitant to disagree with you and Terry about your interpretation of ATR, but I cannot find any corroboration of that interpretation anywhere. Perhaps you could provide a link, page number, or quote from Mercier, Sperger,mor Landemore that makes the claim that internal deliberation is as or more effective than external. Everything I have come across claims the opposite. Here is a YouTube interview by Robert Wright with Mercier on why that is so:
The explanation occurs at 31:55.
Looks like the link is not going to load, but it is the only YouTube with those two so won’t be hard to find.
LikeLike
Keith, you said:
Deliberative Democracy is a normative project designed to reveal the laws of god and nature via the cultivation of the ideal speech situation. The ATR is a description of the cognitive toolbox that evolution has equipped us with to optimise our chances of survival. It involves the adaptation of war-war into jaw-jaw, but the same dynamic is involved (however distasteful Habermasians might find that to be).
I don’t know what deliberative democrats believe as a whole, but Habermas is well aware of the turn from metaphysics from Nietzsche through Heidegger to the Postmodernists. I’ve never seen him describe ethical discourse as you are. In fact, although he believes in continuing the Enlightenment project abandoned by his Frankfort school colleagues, and does follows in Kant’s footsteps, he does not justify deliberation by an appeal to universal reason in the Kantian sense, of an individual capacity endowed by the creator or Nature that allows us to perceive normative truths. Quite the contrary, he believes normative truths are negotiated through the deliberative process. His only moral truths are his principles of ethical discourse, and those are not metaphysical endowments but logical outcomes of what is implied when one agrees to deliberate, or take communicative actions. May I suggest a book for you? Moral Consciousness and Communicative Action, is an excellent introduction to his ideas on this matter. Habermas would be quite at home with ATR, as am I, as both stress the importance of public deliberation for arriving at normative agreement.
LikeLike
Hi Terry,
When it rains it pours…
Lance seems to think that is the good stopping point, let this body also make the final decision. There are several reasons why this is problematic. I mentioned information cascades and groupthink, but one that I haven’t mentioned in this thread is size.
Wait now, I am not opposed to having people deliberate internally. I don’t wish to stop them from doing that at all, and don’t believe that is even possible. People will always internally reflect within a public deliberation. It is you and Keith who want to stop public deliberation based on your fears of information cascades and groupthink, of which there is little evidence for that in regards to deliberation on normative matters. If the chosen body was trying to guess the weight of a steer or predict the price of pork bellies a year from now, I’d agree with your preventing them to deliberate publicly. I think it is a serious epistemic deprivation to censure public deliberation at the decision-making stage when attempting to arrive at normative agreements, for the same reason you believe it is unwise to do so at the proposal stage, but even more so. If you want to have separate bodies propose and depose, so be it, but don’t prevent the decision-makers from talking to each other.
There is a practical limit for all-to-all give-and-take deliberation… perhaps 150? But that is far too small to have reliable statistical representativeness of the population. We need the jury to be many times larger. But this is perfect, since they are not going to express anything to each other… just learn about the topic, listen to the best arguments the supporters and opponents of the final proposal can muster, and vote by secret ballot, to use the “wisdom of crowds.”
What do you think is the minimum size for a reliable statistical representation? Why not just institute a Swiss style referendum if you want to access the wisdom of the crowd? If you don’t believe they will be educated enough on the issue, what makes you think private deliberators chosen by lot, whether paid or unpaid, will “hit the books?” I guess you can test them. Then, how do you correct for the confirmation bias that Mercier describes in the interview I linked to above? Even if, as you say, only 150 is the limit for deliberation (and AI enhanced deliberation might increase that limit by orders of magnitude), it would not be any worse than the non-deliberative body that you and Keith are proposing.
LikeLike
Thanks for the link Lance, will listen to it and come back to you. I think the main issue is that Terry and I (unlike most deliberative democrats) are arguing for allotted groups with statutory decision powers — i.e. that can come to a decision that is legally binding. This means that, presupposing democratic norms, the sample has to be genuinely representative of the target population, and statisticians would generally opt for a minimum of 1,000. But that takes us well over the rational ignorance threshold, hence the preference for smaller sample sizes.
I’m sorry that you think Terry and I are opposed to deliberation. All we are saying is that the deliberative mode of the final decision body should be weighing options, rather than exchanging reasons. Just for clarification, here’s the OED on the two different meanings of the word deliberation:
LikeLike
Hi Keith,
This means that, presupposing democratic norms, the sample has to be genuinely representative of the target population…
While I agree that a sample size large enough to be a statistically valid representation would be desirable, I’m not convinced it would have to be to be legitimate. Wouldn’t legitimacy have to come from its acceptance by the populace, whether it’s genuinely representative or not?
and statisticians would generally opt for a minimum of 1,000. But that takes us well over the rational ignorance threshold, hence the preference for smaller sample sizes.
This is where AI-augmented deliberation comes in. I believe 1000 deliberators would be well within its abilities, but I might be mistaken.
I’m sorry that you think Terry and I are opposed to deliberation. All we are saying is that the deliberative mode of the final decision body should be weighing options, rather than exchanging reasons.
I don’t believe the two of you are against deliberation, but only against it (in your second sense, what I have been calling external or public deliberation) at the decision-making stage. If you feel it is important to get rid of public deliberation (PD), I’d opt for eliminating it at the first stage. PD is not essential to obtaining a number of proposals. I don’t know how policy writing works in the UK, but my understanding is that few bills are written by elected officials in the US. Staff, and frequently various think tanks and lobby groups, write proposals. Asking for proposals on any given issue should no doubt produce a flood of candidates. The most popular could be determined via referendum, or electronic polling, or through elected officials voting if you wish to keep them employed, and winnowed to the superminority requirement. Eliminating PD, at the stage of striving for normative agreement, seems like a serious, and unnecessary frugality.
LikeLike
Lance:> Wouldn’t legitimacy have to come from its acceptance by the populace, whether it’s genuinely representative or not?
That’s the Yoram Gat definition of democracy, and it would mean that theocracies and autocracies could be more democratic than representative political systems.
>I don’t believe the two of you are against deliberation, but only against it (in your second sense, what I have been calling external or public deliberation) at the decision-making stage.
That’s the Rousseauian perspective. Terry and I are adopting the trial analogy where it’s the advocates who deliberate (in the external sense) and the jury that decides (by deliberating within, followed by a vote).
LikeLike
Hi Keith,
Lance:> Wouldn’t legitimacy have to come from its acceptance by the populace, whether it’s genuinely representative or not?
That’s the Yoram Gat definition of democracy, and it would mean that theocracies and autocracies could be more democratic than representative political systems.
You did say,”presupposing democratic norms” so it is understandable that you would think I was speaking of democratic legitimacy, but I was talking about legitimacy, per se. I’m not inclined to follow the neoliberal crusaders in their efforts to impose liberal democracy on whatever country they deem needs improvement. I’m unwilling to say all the monarchs of Europe, emperors of Asia were all illegitimate because undemocratic. Obviously, I don’t believe that theocracies or autocracies are democratic, but they can still be legitimate if they have the support of their people. This interesting study by the Alliance of Democracies indicates that Western-style electoral democracy is not faring well among its people, even in some key democratic norms, compared to oligarchies around the world:
https://www.allianceofdemocracies.org/wp-content/uploads/2024/05/DPI-2024.pdf?utm_source=substack&utm_medium=email
Although the government of China might be less democratic than the US, I don’t believe it is less legitimate. Interestingly, democracy is more important to Chinese than US citizens, they view their government as more responsive to the needs of the majority than the citizens of Western democracies, and more Chinese believe their country is democratic than most Western democracy citizens.
>I don’t believe the two of you are against deliberation, but only against it (in your second sense, what I have been calling external or public deliberation) at the decision-making stage.
That’s the Rousseauian perspective. Terry and I are adopting the trial analogy where it’s the advocates who deliberate (in the external sense) and the jury that decides (by deliberating within, followed by a vote).
I don’t know how juries work in the UK, but in the US, the deliberation is as a group, and they continue to deliberate externally until they reach a consensus, or in certain civil cases a majority decision on damages.
My overriding concern is for the representation of the vast majority of citizens…
This may be the crux of our difference. My primary, or at least one of my major concerns, is to lessen the extreme polarization in the US. About a third of the population doesn’t believe the last presidential election was legitimate. I’m not convinced that more of the same will make that any better. I know you have your Brexit issue across the pond, but things seem much more unified there compared to here. Perhaps I am mistaken. Another concern is the widespread political apathy in the population, perhaps spurned by the polarized animosity of those that are political. A system where officials, remote from the general public, are elected every few years as representatives engenders apathy and disinterest in politics, whereas a system where anyone could be chosen as a representative would empower the average citizen and I believe reduce apathy. Through many generations it could actually transform the population into more mature actors of self-governance. Elected representatives place the decision-making responsibility onto an “other” and relieve the burden of being interested, informed or learning how to reach agreement on difficult issues.
“In the end, we will not remember the words of our enemies, but the silence of our friends.” – MLK
LikeLike
Lance > ” It is you and Keith who want to stop public deliberation based on your fears of information cascades and groupthink”
I have not been clear enough. My notion is that a diverse allotted body would take input from any who wished to offer it on the particular topic at hand, engage in active give and take deliberation, likely seek win-win possibilities, but eventually agree on a draft law. Keith favors elected political parties developing proposals instead, but I am all about diverse active deliberation. This body (I call a Review Panel) would have a substantial workload, so probably certain sorts of people would decline to serve (or decline to do the work) and it would not be very large (to allow good deliberation). This means that body would not be accurately representative of the population. (There is also a risk of groupthink, etc., but those are additional concerns, that compound my main concern).
That is a major reason a large final policy jury makes the final decision after hearing the best arguments for and against the draft law (and as Fishkin advocates, perhaps developing their own clarification questions in small groups). This body would be quasi-mandatory (like jury duty), but serve for a very brief time. Their decision would fairly reflect what the society as a whole would decide if everybody could somehow have the time, motivation and resources to become adequately informed. (I hope this earns “legitimacy.”)
As for “legitimacy” I am struggling with that section of my book right now. Ultimately you are correct… legitimacy is a matter of public feelings about the government. Philosophers invent criteria… but… I find it interesting that so may fairytales, and Disney properties have a “rightful prince” that we accept as “legitimate” because of being born to the former ruler (Lion King, et. al.), and the audience willingly accepts this legitimacy. Elections only have legitimacy because we were all born into societies that told us they are legitimate. I guess that is all the enlightenment philosophers meant by “consent of the governed,” not rising in revolt or emigrating. Transferring legitimacy to a government system based on a different basis of legitimacy is a tricky problem.
LikeLike
Thanks for the clarification Lance — Yoram always speaks of legitimacy in democratic terms, so I assumed you were taking the same view. I’m not holding a (normative) candle for democracy — my preference is for politeia, but with the (mini)demos always having the final word. That’s certainly not the case with autocratic and theocratic regimes. (I think Yoram, Terry and yourself possibly share a similar political provenance, whereas I’ve always been more of a liberal constitutionalist.)
>I don’t know how juries work in the UK, but in the US, the deliberation is as a group, and they continue to deliberate externally until they reach a consensus, or in certain civil cases a majority decision on damages.
Sure, but the job of trial juries is to arrive at the truth (guilt or innocent), whereas the remit of legislative juries is to articulate informed preferences.
LikeLike
Lance,
> Obviously, I don’t believe that theocracies or autocracies are democratic, but they can still be legitimate if they have the support of their people
So, as you have things, support of the people is the criterion for the legitimacy of a government, but not for democracy. If it is not the support of the people that makes a government democratic, what does?
LikeLike
The demos must have power, rather than just nodding its approval. According to the Gat-Hilt theorem, electoralism would be democratic so long as the survey data from the Pew Research Center indicated a high approval rating.
LikeLike
> The demos must have power
Unfortunately, this literally-derived definition is useless unless there is an agreed criterion for what “having power” means. Whether the Athenians would believe it or not, in modern times some insist that holding “fair and free elections” means that “the demos has power”.
I cannot find a better criterion for “having power” than being satisfied with how things are run or being satisfied with how things turn out (which of course are not exactly the same thing, but in reality, people are satisfied with how things are run if and only if they are satisfied with the outcomes).
Outcomes are the ultimate arbiter of where power lies. When person A finds that they are constantly satisfied with how things are, even when the outcomes run counter to what can be thought of as person B’s interests and wishes, then it would be hard to convince person A (or anyone else) that it is really person B that “has power”, even if person B is the one that is seemingly making decisions.
LikeLike
According to the Gat-Hilt theorem, electoralism would be democratic so long as the survey data from the Pew Research Center indicated a high approval rating.
I don’t have time to give an extensive answer but this did elicit a hard laugh from me so thank you. I’m afraid I cannot share in any prestigious awards resulting from the application of the above theorem as it is incorrectly named, for reasons stated in my previous post.
LikeLike
Yoram:> When person A finds that they are constantly satisfied with how things are, even when the outcomes run counter to what can be thought of as person B’s interests and wishes, then it would be hard to convince person A (or anyone else) that it is really person B that “has power”, even if person B is the one that is seemingly making decisions.
That’s a good paraphrase of Harrington’s (1656) theorem:
“Two [girls] have a cake yet undivided which was given between them that each of them therefore may have that which is due. ‘Divide,’ says one to the other, ‘and I will choose, or let me divide, and you shall choose.’ If this be but once agreed upon, it is enough, for the divider dividing unequally loses in regard that the other takes the better half. Wherefore, she divides equally and so both have right.”
Harrington, like Alex and myself, was an advocate for the mixed constitution, in which the divider is elected and the chooser balloted by lot. Although the first principle is aristocratic and the second democratic, the latter always has the whip hand. The crucial difference between this and the Gat-Hilt (2024) theorem is that the chooser has political power, rather than being fobbed off with bread and circuses
Qu’ils mangent de la brioche
LikeLike
>some insist that holding “fair and free elections” means that “the demos has power”.
That’s true in the tautological sense. The problem is not the electoral principle per se, it’s the restricted choices available, and the absence of a properly-functioning feedback loop between the makers of representative claims and the choosers.
LikeLike
> The problem is not the electoral principle per se, it’s the restricted choices available, and the absence of a properly-functioning feedback loop between the makers of representative claims and the choosers.
Now you are expanding the vague notion of “having power”, in terms of other terms (“restricted choices”, “properly-functioning feedback loop”) that are at least as vague.
Until you describe “having power” in a way that skeptical observers can agree on who has power in particular cases, this is just running in circles.
LikeLike
Yoram:> vague
OK, so let’s operationalise it. We both agree that policy choices should reflect the (informed) beliefs and preferences of the citizen body on each policy domain. Say if there were 10, 20, 50 or 100 (pick any number) proposals on a particular policy issue. In our model this would be by superminority, but other options are available (Terry’s agenda councils, prize competitions, Chat-GPT etc). Then a large randomly-selected jury listens to the advocates’ arguments and choses the one they prefer, by simple majority. (In practice it would be better to have 3-6 proposals). Under superminority, feedback occurs both ways — the parties have to anticipate the preferences of the juries and unsuccessful parties will lose the right to make future proposals. The demos has power over both the agenda and the decision, as Dahl required.
LikeLike
As far as I can tell you are barking up the wrong tree. Other than by assertion, how do we know that some proposed arrangement (yours, Terry’s or anyone else’s) “reflects the informed beliefs and preferences of the citizen body”?
More importantly: say that you manage to convince me and lots of other people that some arrangement is “democratic”, and finally it is implemented. Then, it turns out that people are unhappy with the outcomes that this arrangement produces and want it replaced with something else. What then? Do we still accept it as democratic?
LikeLike
Yoram,
I’m doubtful about Terry’s claim (as it overprivileges engaged citizens) but the superminority argument is based on the fact that the proposals are chosen by the same population using two complementary representation principles. In the first instance everyone gets to choose, in the second it’s a representative sample. That’s about as good as it gets.
>[If] it turns out that people are unhappy with the outcomes that this arrangement produces and want it replaced with something else. What then? Do we still accept it as democratic?
Yes. Citizens are entirely at liberty to change their minds, at which point the process is repeated. They are also entitled to shoot themselves in the foot. Democracy is just a way of deciding who has power — any epistemic benefits are contingent. If citizens choose to replace the demokratia with alternative arrangements (dictatorship, epistocracy, oligarchy, AI or whatever) then it is no longer a democracy (by definition).
LikeLike
So is it your claim that your system is a democracy whether or not the people are happy with it?
LikeLike
Yes, but if people are unhappy with any particular policy outcome this will trigger another round of lawmaking.
If the concern is more existential then the demokratia is unlikely to last long. If the demos has kratos then the people are at liberty to dissolve the demokratia.
Our proposal is only normative in the sense of privileging the equality of all citizens (especially the silent majority), but this requires two variants of the representative principle, so might be better described as politeia.
LikeLike
> Yes.
So, since you assert that your system is “democratic” it must be so, whether people find it serves them or not. I am glad we got this cleared up.
LikeLike
Yes that’s right. The principal role of political theory is to clarify the meaning of words, and etymology is a valuable aid. This is the case even if you adopt a normative perspective — see for example Hansen’s chapter on Democracy as an Ideology. The focus is entirely on equality, rather than whether there is anything intrinsically good about democracy per se.
LikeLike
The point is that you see yourself as an authority on “clarifying” to the people that under a certain situation they have power, even if their own subjective experience is that they do not. This not only seems like an obviously anti-democratic position but is also completely unworkable in practice since your arguments (as are those of all other political theorists) are very far from having the power to create a consensus around a clarified concept of democracy. Thus no “clarification” is taking place and the concept of “democracy” remains useless in science.
But there is another point, that I think is related. The notion that you can write an argument that would definitely show that a certain system is “democratic” based on some sketch of its structure is akin to the notion that you can write an argument that would definitely show that the air conditioning system is well-functioning based on a sketch of its structure. The ultimate criterion of whether the air conditioning system is well-functioning is whether the room is at a temperature that is comfortable to its occupants. No analysis of the structure of the system can ever override this criterion. It is because there is a consensus around this criterion that the notion of a well-functioning air conditioning system is useful.
LikeLike
Yes, that’s right. It’s a structural argument — rather than a claim that democracy is a well-functioning political system. Political theorists are concerned with clarifying the meaning of concepts, and any normative claim is limited to its egalitarian nature (we can be equally poor, equally miserable or whatever).
LikeLike
Well, the type of argument you are offering fails on both fronts. It obviously fails to define “democracy”, since the normative power of democracy is a major component of this concept. But it also fails to provide any scientifically useful concept whatsoever since it is far too vague and arbitrary to generate a productive discussion.
(In your defense it may be said that these failure are not your own. You are merely extending a tradition that exhibits the same failures as your arguments do.)
LikeLike
> the normative power of democracy is a major component of this concept.
Quentin Skinner published an important paper several decades ago arguing that democracy had (wrongly) become a “hurrah” word, whereas it is a purely analytic concept. And there’s a lot more to political science than reproducing public opinion polls.
LikeLike
Political science should be a lot more than the endless rehashing of clichés.
LikeLike
We’ve been here before, but I’ll try a different summary… “Democracy” is an ideal of community self-governance based on equality, and the population as a whole having power, rather than an elite and persistent class only having real power. There may be many particular designs that are intended to achieve this democratic ideal (I presented mine in my paper on multi-body sortition). Any design may have defects that turn out to make it fail to achieve democracy, but the word democracy is all about a SYSTEM that is based on, and intended to achieve those ideals. Many of us suspect that a well-functioning democracy would generate good policies that most people would like. But using the metric of whether the people like the outcomes may be useful for deciding if the government is “good” but not if it is democratic. When the people have power they may (for many different reasons) happen to make some bad decisions. But one benefit of democracy is that when the outcomes are bad, since the people have power, they will change those policies.
Yoram’s idea of relying on how satisfied the people are with government policies is an important metric, but unrelated to whether a system is democratic. A society might have a benevolent oligarchy who hold all power, and where the population is essentially powerless, and with the people being completely unequal with the elite. But because the oligarchs happen to be benevolent, they may deliver wonderful policies that the people love. That does not make the system democratic at all.
If my dog loves me and all of my policies about food and walks and playing, but has no actual power in the relationship, that does not make us equals. Liking does not mean power or equality.
LikeLike
Terry,
I agree 100% (as would most of us cliche rehashing political theorists). It’s important that participants in forums like this have a clear understanding of the concepts that they are relying on for their arguments. The principal concept still in need of clarification is representation. “Representation of interests” and “a representative sample” are very different concepts, and there is no necessary connection between the two.
LikeLike
Terry,
> But using the metric of whether the people like the outcomes may be useful for deciding if the government is “good” but not if it is democratic.
Three inter-related related points, then:
1. But then what is the metric to measure the democratic-ness of a system? Other than asserting that “this system is democratic”, what do we have? We have been stuck in this democratic-by-assertion situation for two centuries, and none of the arguments you are providing in favor of your own systems has the promise to change this. Unless a metric for democracy is developed, the endless debate on what system is democratic is destined to remain as fruitless as it has been.
2. The democratic-vs-good distinction is a-historical since the entire interest in “democracy” is the idea that it is a system that serves the people. If a democratic system is not good, why should we be interested in it? And if a different system is good, isn’t that system more interesting than democracy?
3. In fact, the democratic-vs-good distinction is not only a-historical, it is a contradiction in terms because the notion that a system that empowers people could in fact be not good is antithetical to the democratic ideology (and is part of the classical elitist anti-democratic ideology). For such a situation to occur we would have to accept that people are unable to judge what is good for themselves so that unbeknownst to themselves, they are better off being ruled by a system that they reject. Thus, to make this distinction, one must first reject the notion of democracy.
LikeLike
Democracy is a goal that real governments can only approach… move towards or away from. There are two fundamental principles: equality and popular power. If a political class of people have more influence or power than others in making public policy decisions (and this includes deciding the agenda of what sorts of decisions need making), the system is not democratic. An undemocratic oligarchy may be based on birth, wealth, celebrity, technocracy, etc. All of which can be locked-in through elections (elections can even enhance the power derived from privileged birth). For democracy, real power needs to be shared by the general population, rather than concentrated in a political class.
But simply due to practical concerns like time efficiency and information overloads, it is impossible to have all people involved in all decisions, so the only practical way to make decisions is through delegation to subsets of people. Elections tend to concentrate power unequally in the hands of a certain sort of people. Only lotteries with rotation can fairly and equally create democratic decision-making subsets.
It is my belief that democracy will over time tend towards the best decisions in the opinion of the population, and indeed, that is why I advocate for democracy. But since we have never gotten very close to a real democracy, it is impossible to be certain about the output. It is certainly possible that most people are mistaken about the best means to achieve their desired goals, and that even if the people have real equality and real power, that the people will make decisions that have so many unintended consequences that they dislike the results of their own handiwork. However… and this is key … at least the people do not have a hidden agenda (such as every oligarchy I can imagine) that seeks to thwart those optimal outcomes they would prefer. So, democracy is our best bet to have good outcomes.
LikeLike
Yoram:> Unless a metric for democracy is developed, the endless debate on what system is democratic is destined to remain as fruitless as it has been.
You are right to point out that Terry and I (and political theorists in general) rely on structural and etymological definitions. As for the metric, given that we all agree that demokratia in large modern poleis can only be achieved via sampling, then I stick by my repeatability metric. If different samples return different outcomes then the system is not democratic (the relevant norm being the equality of all citizens — especially the vast majority not included in the sample). Whether or not it leads to eudaimonia is an entirely separate issue, and nothing to do with democracy per se. Terry is right to argue that any hope for optimal outcomes is little more than a best guess.
I’m afraid if you are going to stick to your outcomes metric you need to acknowledge that your concern is eudaimonia, rather than demokratia (remember that Marx viewed himself as an Aristotelian — and the latter was no democrat).
LikeLike
Terry,
> Democracy is …
This, unfortunately, is exactly the vague and clichéd rhetoric that has kept democratic theory from “clarifying” the concept of “democracy”.
Whether deliberately or through mere habit, you have not answered the crucial question: how do we know that under a given system “the people has power”? Until we have a clear answer for this question, proposals for “democratic” institutional designs (or, indeed, actually existing systems) cannot be evaluated in an agreed-upon way, and their value remains a matter of personal taste (or social convention).
To be specific: You believe the standard electoralist system is anti-democratic. (And I agree.) But why? What are the indications that the system is not democratic? Did you just one day decide to examine the institutional design and happened to came to the conclusion that it is anti-democratic? Or is your view and about the electoralist system in fact ultimately due to the discontent people feel toward it?
LikeLike
Yoram:> how do we know that under a given system “the people has power”?
As you know, I favour a system that involves two forms of democratic representation (choice and sampling). If the representation is based on sampling, I stick by my repeatability metric. If different samples return different outcomes then the system would not be democratic (as those not included in the sample are not represented).
>You believe the standard electoralist system is anti-democratic. (And I agree.) But why?
Because a) in FTTP systems the choices are too limited; b) in proportional systems choices are destroyed by the need for 50%+1; c) distinct issues are conflated; and d) the advocacy and judgment rights are conflated. So my objections are to the details, not the electoral (choice) principle (and have nothing to do with outcomes).
LikeLike
Keith,
I agree that the final yes or no decisions on presented laws should be replicable between different samples. But that is a necessary but certainly not sufficient requirement for a democracy. Control of the agenda, and control of creating the proposals is also key. You favor an electoral system to select an ongoing elite of partisans for generating proposals, while I favor sortition. Your system is more akin to the non-democracy Sparta, in which an elite (aristocrats) decided what policies to present to the Spartan populace for a vote. While partisan elections affect the peripheral edges of that partisan elite, they do not put ordinary people into any position of power for this step. Mine is more like the Athenian model in which anyone might make a proposal, and the randomly selected Council of 500 made revisions to language and established the agenda for the population to consider (originally in the ecclesia and later by the randomly selected nomothetai).
Yoram,
Nobody argues that the population as a whole, or statistically representative samples, or that ordinary people play any role at all in 99.9% of public policy decisions in electoral systems (only in rare referendums can it be argued they had any role… and that is distorted by elite manipulation and spending, etc.) With nearly all political power held by the elites, the only “equality” the electoral system offers is a de minims, symbolic vote that virtually never matters. The only reason this is called “democracy” is because the word has been redefined by the elites to mean a system with one-person one-vote to minimally shape the edges of the elite who are far above the population in power. In other words everyone agrees (if pressed) that political equality is essentially symbolic only, because a political elite holds all power. This reality is well known by the population. public opinion polls regularly show that a vast majority of people in most electoral systems (especially in my own United States) believe that politicians do not serve the ordinary people, but instead serve themselves or a powerful elite. Yes, I expect many of these people draw the conclusion because they are dissatisfied with government policies. But is is not that output that makes a system democratic or not. It is whether the ordinary people or an elite hold power, and whether there is a political class that is completely unequal to the broad population.
LikeLike
I have not been clear enough. My notion is that a diverse allotted body would take input from any who wished to offer it on the particular topic at hand, engage in active give and take deliberation, likely seek win-win possibilities, but eventually agree on a draft law.
This body (I call a Review Panel)…that body would not be accurately representative of the population.
That is a major reason a large final policy jury makes the final decision after hearing the best arguments for and against the draft law (and as Fishkin advocates, perhaps developing their own clarification questions in small groups).
Thanks, Terry, for the explanation. I believe I understood your stance on what I was addressing, but it is good to see some of the details. Your first body, the Review Panel, would engage in public, or external deliberation (let’s call it PD, ED has unfortunate connotations). The final policy jury would not, though you are considering allowing them to break into small groups to formulate questions to ask, I assume, the experts.
Might I suggest you allow PD in the small groups as well as institute something that they do at MIT and Cortica? Each small group begins by each person stating a core value and relating a personal story to illustrate it. Core values are not something that are often laid on the table, and even if they should be they are often axiomatic and not open to justification. See Paul Katsafanas, The Philosophy of Devotion, or an excellent treatment of this. One of the members of the Cortica presentation I attended was a German fellow who had been involved with CAs for over ten years. He was very happy about the story telling and stated that the “scientists” just don’t get it, that they focus on argumentation, but felt that it is the group involvement and sharing of personal information that makes CAs effective.
After much criticism Habermas relented and admitted storytelling as a legitimate contribution to reach normative agreement, and not merely rational argument.
LikeLike
Terry:> Your system is more akin to the non-democracy Sparta, in which an elite (aristocrats) decided what policies to present to the Spartan populace for a vote.
You underestimate the feedback loop between proposers and disposers in our hybrid model. Assuming a modest superminority threshold, the winning coalition will be the one that successfully anticipates public preferences. As Ledru-Rollin put it “There go the people. I must follow them, for I am their leader.” Given that you accept the need for a representative principle (sampling) for disposers in large poleis, why don’t you allow the same for proposers? I don’t see how it’s possible to follow Dahl’s requirement that the demos have exclusive control of agenda-setting without elections (albeit in a modified form). Modern isegoria is better understood as a principle (that requires a representative mechanism) than an individual right (as in small ancient poleis).
>[In classical Athens] anyone might make a proposal, the randomly selected Council of 500 made revisions to language . . .
Thanks for that clarification. Some people on this forum have imagined the Boule to be a collective agenda-setting body, although it is hard to understand how a group of 500 strangers could deliberate together. And, according to Finley, speech acts were almost always the province of (elite) rhetors.
Lance:> it is the group involvement and sharing of personal information that makes CAs effective.
Sure, but what has that got to do with demokratia, as the overwhelming majority of citizens will not get to participate in storytime? In Britain “Jackanory” is now generally used in the pejorative sense:
“Jackanory, jackanory” said by someone in the sing-song tones of the theme tune indicates that they think that someone else is making up or “stretching” a story, i.e. lying (Wikipedia)
Remember that the ATR (and the hermeneutics of suspicion) evolved as a mechanism to overcome the lack of trust between strangers.
LikeLike
Of course some people might argue that the ATR is a Just So story (oops!)
LikeLike
Keith,
Lance:> it is the group involvement and sharing of personal information that makes CAs effective.
Sure, but what has that got to do with demokratia, as the overwhelming majority of citizens will not get to participate in storytime?
The overwhelming majority don’t need to participate. Legitimacy is obtained if the majority approve of the process.
In Britain “Jackanory” is now generally used in the pejorative sense:
“Jackanory, jackanory” said by someone in the sing-song tones of the theme tune indicates that they think that someone else is making up or “stretching” a story, i.e. lying (Wikipedia)
You have lamented that agreement is seldom reached through argumentation and gave this forum as an example. I’d point out that there is little incentive here to reach agreement, plus much of the intractability has to do with axiomatic beliefs, or what Katsafanas terms sacred values. Those will not be easily changed though argumentation since, by definition, they are adopted as a bulwark against argumentation. See his book, or this interview:
Remember that the ATR (and the hermeneutics of suspicion) evolved as a mechanism to overcome the lack of trust between strangers.
I have seen it posited that that was the initial impetus for religion. Reason, according to ATR evolved for argumentation, not to overcome suspicion, but to advance the carriers of genetic information survival, and thus their replication chances. If reason evolved to decrease suspicion I’d say it did not evolve well.
LikeLike
Lance:>The overwhelming majority don’t need to participate. Legitimacy is obtained if the majority approve of the process.
And how would you know? — especially as approval is likely to be issue-dependent. Your criterion sounds a lot like the speculative history/mythology served up by early-modern social contract theorists.
>You have lamented that agreement is seldom reached through argumentation and gave this forum as an example.
I merely observed (no reference to the desolation of Judah) that the tiny minority gripped by confirmation bias would never change their view. That’s why Terry and I argue for simple majority decision making by dispassionate citizens’ juries, without the need for agreement. (Your reference to axiomatic beliefs [Katsafanas’s sacred values; Collingwood’s absolute preconditions] is a good way of describing confirmation bias.)
>Reason, according to ATR evolved for argumentation, not to overcome suspicion
The ‘argumentative theory of reasoning’ (ATR) hypothesizes that reasoning serves two survival-related functions: a) convincing people and b) evaluating the arguments of others – ‘thereby allowing communication to proceed even when trust is limited’ (Landemore, 2013, p. 126).
As for the origins of religion, there are many hypotheses — my favourite candidate is Rene Girard’s mimetic theory.
LikeLike
Hello again Keith,
Lance:>The overwhelming majority don’t need to participate. Legitimacy is obtained if the majority approve of the process.
And how would you know? — especially as approval is likely to be issue-dependent. Your criterion sounds a lot like the speculative history/mythology served up by early-modern social contract theorists.
I guess I don’t understand your question. How would I know if the majority approved? The same way we know whether they approve of any policy, candidate or referendum. I don’t see a problem here, or if there is one it would equally apply yo your proposed system, or anyone’s for that matter. You think it is not possible to know what the majority approve of in society?
>You have lamented that agreement is seldom reached through argumentation and gave this forum as an example.
I merely observed (no reference to the desolation of Judah) that the tiny minority gripped by confirmation bias would never change their view. That’s why Terry and I argue for simple majority decision making by dispassionate citizens’ juries, without the need for agreement. (Your reference to axiomatic beliefs [Katsafanas’s sacred values; Collingwood’s absolute preconditions] is a good way of describing confirmation bias.)
No, confirmation bias is something else. Sacred values are axiomatic, not defended, inviolable. The operations of confirmation bias could be employed in their defense, but they needn’t be. Where are you and Terry going to find “dispassionate citizens’ juries?” You will need a lamp far brighter than Diogenes’.
>Reason, according to ATR evolved for argumentation, not to overcome suspicion
The ‘argumentative theory of reasoning’ (ATR) hypothesizes that reasoning serves two survival-related functions: a) convincing people and b) evaluating the arguments of others – ‘thereby allowing communication to proceed even when trust is limited’ (Landemore, 2013, p. 126).
Yes, trust is limited, not established, suspicion still remains, it is not overcome by argumentation. Argumentation is a poor method for establishing trust, if even possible via argumentation alone. Trust is established through interacting with others in the lifeworld. The sharing of stories I mentioned, and religious affiliations are short cuts, stand-ins, for such real life experiences.
LikeLiked by 1 person
Lance:> How would I know if the majority approved?
Your criterion is approval of “the process”, and that would suggest some sort of constitutive moment (hence my reference to social contract mythology). This is different from the consent provided by electoral procedures, where voters select parties/policies. Citizens demonstrate their consent to the process by participating in elections but how would they show their approval of your process? If by plebiscite, how would that consent be withdrawn if they no longer approved and how would you know if the (lack of) approval was to the process or issue-specific factors?
>Where are you and Terry going to find “dispassionate” citizens’ juries?
Any impartiality is a purely aggregative function, and because the selection procedure does not privilege those with an interest in the issue.
>Argumentation is a poor method for establishing trust
According to liberal democratic theory, politics is essentially agonistic and does not presuppose trust. That’s why the ATR is so applicable.
LikeLike
I suspect any fundamental shift towards a sortitional democracy would need some sort of authorizing referendum (such as a constitutional ratification). This is because the electeds would be VERY unlikely to simply give up their position and power on their own. Likewise, a coup seizing power would be EXTREMELY unlikely to think sortition should replace them. Whether an authorizing referendum would be repeated every ten years, or whatever is an interesting question… Electoral democracies never do repeat authorizations once they are in power… they only have constitutional referendums that involve relative tinkering around the edges with manipulations of the process, and never whether to use elections. A sortition democracy might use a periodic massive allotted constitutional convention to reaffirm or eliminate sortition… but that again (like with system based on elections) could be seen as a snake eating its own tail … using the system to ratify that system. So probably periodic re-ratification referendums would be appropriate…. A referendum could theoretically be acceptable since the question is almost outside the realm of “being well-informed”… its simply about how people feel…do you trust and want to keep this system. Of course the danger is that authoritarian strong men and plutocrats would likely push to restore elections, which allow them to rule.
LikeLike
Terry:> A referendum could theoretically be acceptable since the question is almost outside the realm of “being well-informed”… its simply about how people feel
It would be interesting to see how the question would be posed, given that the vast majority of citizens would be asked to arrogate all political power to a small group of volunteers, on the basis of faith in the power of communicative rationality and storytelling. By contrast, the system proposed by Alex and myself only involves a change in the proposal threshold for elected policy advocates and an extension of the (well-established) jury principle.
LikeLike
Keith,
Lance:> How would I know if the majority approved?
Your criterion is approval of “the process”, and that would suggest some sort of constitutive moment (hence my reference to social contract mythology).
Has nothing to do with social contract theory. Why do you think that?
This is different from the consent provided by electoral procedures, where voters select parties/policies. Citizens demonstrate their consent to the process by participating in elections but how would they show their approval of your process? If by plebiscite, how would that consent be withdrawn if they no longer approved and how would you know if the (lack of) approval was to the process or issue-specific factors?
By asking them to approve the process.
>Where are you and Terry going to find “dispassionate” citizens’ juries?
Any impartiality is a purely aggregative function, and because the selection procedure does not privilege those with an interest in the issue.
That’s not impartiality, that’s a collection of people with different partialities. They would tend to stick to their opinions despite contradictory evidence since no PD is allowed, and confirmation bias is given free rein. You will get a slightly more informed poll of public opinion, but that is all.
I would not consider such non-deliberative bodies true citizen juries. Would you and Terry, if charged with a crime, prefer your method where each juror only listens to the arguments of the opposing counselors, or one where they also publicly deliberate?
>Argumentation is a poor method for establishing trust
According to liberal democratic theory, politics is essentially agonistic and does not presuppose trust. That’s why the ATR is so applicable.
Yes, because it works not by overcoming distrust, but in spite of it. Personal interaction is required to establish trust. This is where exchanging stories and shared values, or establishing an understanding of diverse values comes in ala Scheuerman, and Habermas acknowledged.
LikeLiked by 1 person
Lance:> Has nothing to do with social contract theory. . . By asking them to approve the process.
Hence the social contract parallel, the only difference being its mythical status. My point is that it involves a constitutive moment, rather than the ongoing consent that is the rationale behind the electoral principle.
>That’s not impartiality, that’s a collection of people with different partialities.
Correct, hence my “aggregate” qualifier.
>Would you and Terry, if charged with a crime, prefer your method where each juror only listens to the arguments of the opposing counselors, or one where they also publicly deliberate?
There is no difference in principle between the two options. The only reason for the jury’s (in camera) deliberation is the need for unanimity in trial juries, which doesn’t apply to democratic politics.
>Personal interaction is required to establish trust.
Sure, but the issue is whether the disenfranchised masses would trust the deliberating body. My hunch is that they would more likely conclude that they have gone native, and this scepticism could only be overcome by demonstrating that different deliberative groups came to the same decision. Why would anyone object if it can be demonstrated that their presence/absence makes no difference to the outcome? This would rule out wild-card procedures like storytelling, along with other speech acts that were subject to information cascades etc.
Bear in mind that the Habermasian deliberative model was designed for the (voluntary, plural) public sphere, not political decision making. A procedure intended to democratising deliberation (an intra-group normative project) has purloined sortition (a method for the statistical representation of large populations), and this has had fundamentally anti-democratic consequences. (Bear in mind that both Habermas and Dryzek are contemptuous of the Fishkinian project.)
LikeLike
Lance:> Has nothing to do with social contract theory. . . By asking them to approve the process.
Hence the social contract parallel, the only difference being its mythical status. My point is that it involves a constitutive moment, rather than the ongoing consent that is the rationale behind the electoral principle.
Huh? There has never been any consent to the the electoral principle, at least not in the US. Did you watch our presidential debate? A senile war criminals vs a pathological lying convicted felon. Electoral principle? Maybe you meant unprincipled elections.
LikeLike