The Blind Break and the Invisible Hand, Part 3: Consent

Although Peter Stone has asked us not to cite from his draft conference paper, this forum is really just an extension of the debate at the University of London workshop, so I would rather quote it verbatim than run the risk of paraphrasing it and getting it wrong. The following quote is taken from the concluding paragraph (p.16):

If democracy is supposed to be about government by consent of the governed, for example, then sortition looks like an obvious dead end. The arguments for taking elections to represent such consent are quite telling; citizens are thought to consent to be governed by elected officials even if they voted against those officials, or failed to vote at all. But however tenuous the link between election and government by consent, the link between sortition and government by consent is even weaker. There is no sense in which citizens can be said to have done anything to consent to randomly-selected officials; indeed, the whole point of randomization is to remove any opportunity by citizens to influence the selection process.

In Stone’s view sortition is perfectly compatible with the democratic case for equality (which I addressed in part 1 of this post) but can in no sense can be taken as an indication of democratic consent. (Of course citizens may or may not consent to any form of government, including the rule of the one and the rule of the few, but our concern in this post is exclusively with the rule of the many [democracy]). The argument for sortition as a mechanism of democratic consent, described by James Fishkin as “consent by proxy” (Fishkin, 2010) depends on the intermediate notion of sortition as a system of representation. In part 2 of this post I outlined the partial role that sortition might play in this  — partial in that the form of statistical representation that sortition offers is only applicable to democratic judgment, not the initiation of legislative proposals (i.e. disposing rather than proposing). Given that a statistically-representative sample could only decide between the options made available to them, the case for consent would require that proposals should also be the subject of democratic approval and that would require either electoral or direct-democratic initiative. But that is not a function that is amenable to sortition, so is not the concern of this post.

Most advocates of sortition are content with the notion of representation. The reason I’ve taken up the greater challenge of consent is because Manin (1997) claims that sortition was replaced by preference election at the start of the age of representative government on account of the need for consent. Stone’s “there is no sense in which citizens can be said to have done anything to consent to randomly-selected individuals” is a straight paraphrase of Manin’s argument:

However lot is interpreted, whatever its other properties, it cannot possibly be perceived as an expression of consent (Manin, 1997, pp. 84-5, my emphasis).

But above (“tenuous link”) and elsewhere Stone has expressed scepticism over the very notion of consent, arguing that “tacit”, “hypothetical”, “implied” and “assumed” consent render the term of little value. The Lockean argument for consent, satarised below, is predicated on the notion of electoral representation:

Today, a person is deemed to be politically ‘represented’ no matter what, i.e., regardless of his own will and actions or that of his representative. A person is considered represented if he votes, but also if he does not vote. He is considered represented if the candidate he has voted for is elected, but also if another candidate is elected. He is represented, whether the candidate he voted or did not vote for does or does not do what he wished him to do. And he is considered politically represented, whether ‘his’ representative will find majority support among all elected representatives or not (Hoppe, 2001, pp. 283-284).

Consent by the mechanism of preference elections is at best partial, tacit and approximate as it reflects only the consent of the majority and, assuming that people are more likely to consent to policies they approve of (rather than just to the rule of chosen individuals), there is no obvious way for parliamentary representatives accurately to divine the actual views of their constituents (see also the problem of preference aggregation, discussed above). So Manin and Stone’s claim that election is the inevitable, or even the best, way of ensuring that the political system embodies the consent of the people is open to question. Tacit consent might well be expressed as “put up or shut up (or emigrate)”.

An alternative approach to electoral approximation, and one arguably better suited to a mass individualist society, is (sortive) representation by proxy – I may not attend (and consent or dissent) in person but, if the sampling process is accurate, there would be people present who shared some of my preferences and beliefs (in an overlapping manner)* and who could participate on my behalf, and their presence would be directly proportionate to how many people with similar preferences and beliefs there are in the wider population:

A representative microcosm offers a picture of what everyone would think under good conditions. In theory if everyone deliberated, the conclusions would not be much different. So the microcosm offers a proxy for the much more ambitious scenario of what would happen if everyone discussed the issues and weighed competing arguments under similarly favourable conditions (Fishkin, 2009, p. 194, my emphasis).

The consent argument here is, like its Lockean equivalent, predicated on the notion of political representation and I argued the case for statistical representation in part 2 of this post. It is equally dependent on majoritarian assumptions. Referring back to part 2, I would clearly be deemed to consent to the decision in case 1 (the small group in which I participated directly), so in what sense would the consent argument not apply to the cases 2-5? If the case for representation applies in all cases, then it’s hard to understand what would be the additional requirements for consent.**

The confusion on this issue is largely down to the assumption that the relevant unit of analysis is the concrete individual as opposed to the statistical aggregate. Stone demonstrated this confusion at an earlier workshop in this series when he likened consent-by-proxy to him being deemed to consent to the letting of his flat in his absence because the decision was taken by someone “like” him. This demonstrates a double confusion. First of all the business of the legislature is to frame general rules — i.e. the conditions under which apartments may be let, contracts drawn up etc. Under conditions of high levels of homelessness a legislature might rule that privately-owned apartments that have been unoccupied for (say) five years should be sublet by a state agency. The particular decision to let Peter Stone’s flat would be the work of the agency (an executive body), rather than the legislature, which is only concerned with the general rules under which vacant properties may be sublet by the state. And yes, the claim that this did not meet with Stone’s consent (qua citizen that is) would require a refutation of the representation argument, even though wearing his (personal) Peter Stone hat (and Peter has a very nice collection of headwear) he might be a little peeved at the inconvenience. (This distinction between citizen and private individual is as argued by Rousseau and others, the modern incarnation of the private individual perspective being NIMBYism).***

The confusion between persons (“randomly-selected officials” in Stone’s paper) and statistical aggregates is largely the fault of radical proponents of sortition who appear to believe that an assembly elected by lot is exactly the same as an assembly elected by preference votes (except that it would be full of “ordinary” people), so we only have ourselves to blame. Once the principle of statistical representation is correctly grasped, the case for consent-by-proxy requires very little extra work.


* There is no suggestion here of a one-to-one mapping of citizens and proxy “representatives”, as the preferences and beliefs of the target population would be spread over a range of allotted persons. Some allotted persons might share (say) my conservative views on social policy, whereas others might share my liberal views on economic policy. However the law of large numbers would suggest an overall correspondence between the aggregated views of the randomly-selected microcosm and the target population. This presupposes a reasonably large sample, but the upper limit, in practice, would be constrained by the competing principle of rational ignorance (if the sample were too large then allotted members would not take the trouble to attend to the arguments and vote in a well-considered and responsible manner, as their individual votes would be of little consequence.)

** In practice, perceived legitimacy would be equally important, and this would require widespread public education in statistical proportionality along with a proven track record for lot-based decision making in terms of epistemic outcomes — which would have to be, at minimum, no worse than those delivered by expert and electoral decision mechanisms (Mansbridge, 2010; Levinson, 2010).

*** Not In My Back Yard.


James Fishkin (2010), When the People Speak (Oxford University Press).

Sanford Levinson (2010), Democracy and the Extended Republic: Reflections on the Fishkinian Project. The Good Society, 19, 63-67.

Bernard Manin (1997), The Principles of Representative Government (Cambridge University Press).

Jane Mansbridge (2010), Deliberative Polling as the Gold Standard. The Good Society, 19, 55-62.

52 Responses

  1. PS I should make clear that I am not dealing with consent in the weak form of being deemed to consent to the existence of the prevailing political *system* by either a) willing its establishment via a hypothetical social contract or b) failing to vote with my feet. My concern is the stronger case of being deemed to consent to the actual legislative output of the political system. In indirect democracy this means being deemed to consent to decisions that I did not make myself — in the elective variant either by consenting to the appointment of persons to decide these issues on my behalf (even if I did not vote for them) or accepting a promissory note (election manifesto) which may or may not be subsequently cashed out by the issuing party. All forms of democratic consent, direct and indirect, presuppose the principle of majority rule.

    Consent-by-proxy deals my consent to actual legislative outcomes, as opposed to merely approving the system, approving the appointment of persons or approving a promissory note. Consent-by-proxy is based on the (fully testable) statistical hypothesis that the decision outcome of an allotted body is just a scaled-up version of the small-group direct-democratic decision and that my personal participation or non-participation in the indirect decision would make no difference to the outcome. If I am deemed to consent to the group decision in which I participated directly (irrespective or not of whether the decision goes in my favour), then the same applies to the decision by a statistically-representative body. Note that the phrase “makes no difference” is not analogous to the impotence of individual voters in a large-scale preference election, it is a statement of the statistical fact that the decision will go my way if and only if my own beliefs and preferences prevail in the target population (in which case, ex hypothesi, they will also prevail in the microcosm),* so it makes no difference whether or not I participate in the decision process directly or by proxy. It’s hard to think of a more robust form of democratic consent than this, hence my attack on Manin and Stone’s arguments for the reasons leading to the “triumph of election”.

    *It’s helpful to think of a statistically-representative sample as a holographic photograph — cut the photo in half and all the features are retained, but at a lower lever of resolution.


  2. I wonder if there ever can be (and certainly never HAS been) the “strong” form of consent in any mass society or representative system. After all, in the medical world they speak of INFORMED consent. In other words WOULD the full population (or at least a MAJORITY of it, which is its own retreat from “strong” consent) consent to a given government IF they learned all the information necessary to make such a judgment? All that can EVER exist is either IMPLIED consent, indicated by failure to revolt or emigrate, or consent to the FORM of government, perhaps through a constitutional referendum.

    Consent is essentially a mythology derived from acceptance of the “legitimacy” of a government…which in turn might be derived from such things as longevity, good performance, or revered documents (like a constitution)


  3. If democracy is supposed to be about government by consent of the governed


  4. “If democracy is supposed to be about government by consent of the governed…”

    Democracy is not “government by consent of the governed.” It is government by the governed.


  5. Terry,

    I take it from this that you agree with Peter Stone that systems of government have to be judged purely in terms of outcomes, and that the only value of democracy is that it is a way of generating the cognitive diversity that leads to “good” outcomes. Given that you view the need for consent as something of a noble lie (supported by quasi-religious founding documents) then this explains why you are dismissive of my repeated calls for consistency. This puts you in the same camp as Helene, who had the good grace, when challenged, to acknowledge that she wasn’t a democrat (in the more demanding sense that Arthur references).

    Personally I think we cannot ignore the intrinsic moral argument for democracy. Our guide here must be Rousseau, with his insistence that the preservation of the freedom that is our birthright requires that we should ALL be the author of our own laws. His solution was small primitive republics in which everyone participated directly in the general assembly of the people. All I have done is to suggest a way of applying this principle to large modern states.

    It’s ironic that it requires an avowed conservative like me to argue the moral case for democracy. Am I alone here? (it’s not clear from Arthur’s comment if he is agreeing with or challenging your position). Yoram’s argument is equally instrumental as his case for democracy is utilitarian (the only mechanism capable of ascertaining the interests of the people is a representative sample). He doesn’t appear to mind whether or not the masses consent to taking the medicine prescribed by the allotted sample as it is in their interests so to do (even if it doesn’t taste very nice).


  6. Keith,

    No, I am not merely concerned with “good outcomes” derived from diversity, etc. Indeed a democracy will doubtless make frequent bad decisions (where a king might have made a better one). An increased likelihood of good outcomes is a benefit that goes along with a reduced likelihood of terrible outcomes (likely when an elite few make decisions).

    If I am understanding Arthur’s point…I agree. Mere “consent of the governed” is NOT a definition of democracy. The governed could consent to an all-powerful king or dictator, but that doesn’t make it a democracy. Democracy refers to some sort of “self” government by the people themselves, not elite rulers.

    Since it is impossible and undesirable for all the people to participate in all decisions (what a waste of effort and time), the best democracy can achieve is taking turns with statistical representation with equal opportunity for all, as was crudely practiced in the reformed 4th century democracy of Athens through sortition.

    I DO agree with you, however, that this comes closer to reflecting consent of the governed than mass elections ever could…that just isn’t my top concern, since it is at best a mythology.


  7. Currently, we in the West are living under constitutional oligarchies. We choose or legislators and magistrates by means of election. If we were to replace election with sortition we would still be living under constitutional oligarchy. Citizen participation will have diminished. The quality of government will have improved, dramatically.


  8. Terry,

    I’m glad that we agree on the need for both instrumental and intrinsic justifications of democracy. Self-rule, as you rightly say, is the intrinsic justification, not consent. I couched part 3 in the language of consent on account of Manin and Stone’s claim that sortition cannot possibly indicate consent (whereas election does, albeit in a tacit and approximate form). Would it be fair to say that consent is a necessary, but not sufficient condition of self-rule? I can’t see how it is logically possible to rule oneself and yet not consent to that rule (although it is possible to consent to the rule of others). I would like to think that the three parts of this post have indicated the possibility of self-rule AND consent to legislative outcomes via sortition. I do have another paper that couches the argument in explicitly Rousseauvian terms, concentrating on the issue of self-rule. But I’m glad we agree that purely functional explanation is not good enough.


    >If we were to replace election with sortition we would still be living under constitutional oligarchy.

    That would be true for rule an allotted body with full proposing and disposing powers (indeed I’ve referred to rule by such a body as klerotocracy — dictatorship by persons selected at random). But why do you claim oligarchical status for an allotted body exercising a purely statistical mandate? I’ve tried to demonstrate in these three posts that this is only true if one thinks of an allotted body in terms of the persons that compose it, rather than its statistical properties. If all those persons do is listen, deliberate within and then vote then they are not oligarchsl they are simply proxies for how everyone would vote if subject to similar conditions.

    >The quality of government will have improved, dramatically.

    I’m sorry that, like Peter Stone, you are only able to justify such a body in functional terms rather than the moral right of all citizens to rule themselves.


  9. “But why do you claim oligarchical status for an allotted body exercising a purely statistical mandate?”

    The number of people who govern determines the form of government. When one person governs, the government is monarchy. When a few people govern, it is oligarchy. When all the citizenry governs, it is democracy. In the United States 535 people speak for over 300 million people, one voice per 600 thousand, clearly an oligarchy.

    >The quality of government will have improved, dramatically.
    “I’m sorry that, like Peter Stone, you are only able to justify such a body in functional terms rather than the moral right of all citizens to rule themselves.”

    Keith, I don’t understand this sentence. It seems like a non sequitur. I don’t understand the context. I am speaking of the quality and integrity of the government. Elections lead to lying, greed and corruption. Where there are elections, hidden forces govern. If we eliminate elections we eliminate much of the concomitant evil.

    It is my understanding that the whole purpose of governmental reform is better government, i.e. government that serves the common good.


  10. Arthur,

    >The number of people who govern determines the form of government.

    It looks like I’ve failed to achieve the purpose of these three posts, which was to demonstrate that “persons” is the wrong level of analysis for sortition in large modern states. The appropriate level is statistical aggregations of preferences and beliefs, “persons” merely being the carriers of these preferences and beliefs, selected by the invisible hand of the law of large numbers to approximate to the preferences and beliefs of the target population. Talk of persons made sense in the ancient world, where all citizens had the opportunity to take turns to rule and be ruled in turn. It also makes sense in terms of your own book which, if I remember correctly, involves so many and varied sortitions as to attempt to replicate the conditions of the ancient world. Talk of persons also makes sense for the selection of rulers by preference election. But it’s simply the wrong level of analysis for representative government via sortition — in my proposal the tiny number of citizens selected by lot are not governors they are just carriers of aggregate preferences and beliefs, in order to ensure that decision-making is the product of the informed preferences and beliefs of all of us. This imposes tight constraints on what these citizens can and cannot do, in order to maintain their status as accurate carriers of preferences and beliefs. As soon as they start to behave as persons (indulging in speech acts) they forfeit their status as representative carriers. I guess we share different perspectives on this as you are a psychologist, whereas I am, by training, a sociologist. Sociologists, as you know, have no interest in persons.

    >It is my understanding that the whole purpose of governmental reform is better government, i.e. government that serves the common good.

    That is one purpose of government — the belief that good government is the whole purpose of government puts you in the same utilitarian camp as Yoram Gat (and Plato). Yoram has consistently made this claim and his argument for allotment is that good government is government that is in the interests of the masses and sortition is the only way of discovering the interests of the masses. This may well be the case, but it would be possible in theory for another body (an aristocracy, an epistocracy, a knowledge engine or even a random dictator) to coincidentally come up with the same practical policies as an allotted assembly. Would the latter bodies be democratic, if their output was functionally identical? The answer has to be no, because they would not instantiate the other (intrinsic) purpose of democratic government — self rule. This is a moral undertaking (the preservation of the freedom that is our birthright) which requires *everybody* to decide. To my mind the only way of enabling this moral purpose in large modern states is by de-personating the decision-making body, so that it makes no difference which concrete individuals are included or excluded. (The other method is direct democracy but this would be ruled out on epistemic grounds [“better government” in your parlance.)


  11. In other words sortition cannot be applied as a band-aid to existing democratic systems.


  12. The blind breakdancers do in in fact claim that sortition can be used to protect the integrity of existing political systems; my criticism of the University of London workshop was that this has no obvious connection with democracy and their proposals amount to little more than applying a band-aid.


  13. > breakdancers

    Your name calling habit is not funny, just unpleasant.


  14. > — the belief that good government is the whole purpose of government puts you in the same utilitarian camp as Yoram Gat (and Plato)

    As the human species approaches extinction, we need the best government we can get as soon as we can get it.

    Good government is definitely not the only purpose of government. It is the best we can hope for with an oligarchic form of government. In a true democracy, where citizens legislate on their own behalf, each individual is raised to his highest level of development and so is society as a whole.


  15. Arthur,

    Agree regarding the need for true democracy, the relevant question being how to achieve it. As you know I am sceptical regarding your plan for such an inclusive sortition, that everyone would participate without paying the price of indifference and rational ignorance.


  16. Keith,

    In a true democracy, everyone legislates on their own behalf. There would be no sortition. There would be thousands of local councils debating national issues, open to all citizens. Sortition would apply to magistrates, the people who run the parks and the subways and the water supply.


  17. Arthur,

    We clearly differ on the need for representation in large political communities. In my Rousseau paper I argue that a randomly-selected legislature with a brief limited to inner deliberation would not be in breach of J-J’s requirement that everyone should legislate on her own behalf, but this requires refocusing from the level of persons to properties that only apply at the collective level. The trouble with your model is that it neither delivers equality nor maintains the freedom that is our natural right. Running the parks and subways is not an act of the sovereign legislature and those who decide macroeconomic policy at the national level are a lot more equal than those who decide how much to charge for street parking in their own neighbourhood.


  18. > The trouble with your model is that it neither delivers equality nor maintains the freedom that is our natural right.

    When everyone legislates on their own behalf. That is political equality.

    > freedom that is our natural right.

    I am not sure what the word “freedom” means nor can I see how deeming it a “natural right” makes it any more attainable or desirable.


  19. Arthur,

    >When everyone legislates on their own behalf. That is political equality.

    Of course it is! But it’s not possible in large states (other than via ill-informed and capricious referenda), hence the need for representation. Sortition is a mechanism to achieve statistical representation and this is the main focus of this blog. If you don’t believe that representation is necessary then you’re addressing the wrong audience.

    Freedom (in a political context) simply means choosing our own rules (subject to the proviso that the choice that predominates will be that of the majority). Rousseau phrased this rather differently — voting with the minority was an indication that you were mistaken as to your real willing. But that’s led to all sorts of unsavoury consequences, so probably best to rely on bog-standard majoritarianism.


  20. Arthur >When everyone legislates on their own behalf. That is political equality.

    Keith > Of course it is! But it’s not possible in large states (other than via ill-informed and capricious referenda), hence the need for representation.

    If we have local councils, thousands of them, with the citizenry debating and voting on national issues in these local councils, then we have democracy in a large state.


  21. Arthur,

    We’ve been over this ground before. Assuming that people have a minimal degree of rationality, they will soon figure out that their own little vote counts for next to nothing amongst the members of the thousands of councils, so won’t bother to turn up and do the necessary work to inform themselves. You can’t trump rational ignorance by dividing the electorate into thousands of councils. Rousseau was sceptical that his general assembly model would work for tiny and thinly-populated states like Corsica, so why on earth do you think it would work in large modern states?

    The default position of this forum is the necessity for representation via sortition; you really ought to take your plea for direct democracy somewhere else.


  22. Keith,

    We have been over this ground before and my response this time is what it was previously: It is participation, not voting, that creates citizens. And further, people vote in elections where there are 150 million voters. They do turn up. And why wouldn’t they continue to turn up if their deliberation actually contributed to the final outcome?


  23. Arthur,

    In addition to the unsolvable problem of rational ignorance when millions of people are involved, that Keith raises, what about the sheer amount of “man-hours” you propose society divert to participation? Why should we ALL participate (thorough thousands of sub-councils) in the decision about whether the Transportation Safety Board should get a 2.78% increase in its budget next year? And if it is only those who CHOOSE to participate, then the bias of self-interests will overwhelm “average citizens” who could care less about 99.9% of all governmental decisions that must be made, especially when their vote will make no difference any way (due to sheer numbers). Sortition is the only way to combine political equality and over-coming rational ignorance.


  24. Arthur,

    All of us on this blog believe in Equality by Lot, not Equality by dividing something up into microscopic pieces that are of no value to each recipient. What you have in mind is equivalent to the parable of the five loaves and fishes but, unfortunately, we no longer live in the age of biblical miracles.

    Get real.


  25. PS It would be good if we could have a conversation on the issue of consent — the subject of this post. It’s plausible that citizen-subjects might freely consent to the rule of a wise dictator, if that led to consequences that were in their “obvious” interests* (peace, security, impartiality, justice, prosperity, freedom of religious practice, etc). This is the Hobbesian notion of consent — trading in one’s political freedom to end the warre of all against all. (Arthur and his neo-Machiavellian colleagues would simply deny that this is possible, owing to the fact that the grandi are motivated by the will to oppress. This is history as psychopathology.)

    But under what conditions might free and equal citizens consent to the rule of other persons (elected or selected by lot)? Would the above consequential benefits suffice, or is “republican” consent something more demanding? I would argue that it is, in that it is predicated on self-rule, rather than submission to the will of another person or persons. The problem is how to establish self-rule in large political communities and this is where I think the notion of consistent outcome comes into force (irrespective of whether the outcome is “good” epistemically speaking). One of the problems with our existing political institutions is inconsistent outcomes over time (cycling between the conservative and liberal poles every 5-10 years), so consistency would appear to be important for both intrinsic and consequential reasons. Personally speaking I would consent to the rule of an allotted assembly iff a) it delivered consistent outcomes and b) it made no difference whether or not I was one of the sample selected. But I would not consent to the rule of samples of randomly-selected persons who delivered different outcomes, based on who just happened to end up in the sample. This might well be described as aleatocracy (rule by random chance).

    * But are there “obvious” ways of judging what is in the interests of “the masses” without (as Yoram claims) assembling a minipublic? I imagine most Syrians would answer in the affirmative and would, with the benefit of hindsight, find the Hobbesian answer an improvement on their current malaise. I don’t agree with Peter Stone that we should give up on the notion of consent, but we need to be a bit careful in how we specify “republican” consent.


  26. Terry,

    There are two critical issues that concern everyone, the budget, war and peace. All citizens should be involved in debating and voting on these two critical issues. Most other issues are derivative of these two critical issues. And bear in mind that the frame of reference is national government, not local government.


  27. Keith,

    You actually made me chuckle. Thank you. I think it was the “Get Real.”

    Your style of debate is to repeat your position and ignore what your opponent has to say.

    I point out that in the US 150 million people — your “microscopic pieces” vote in elections. Why can’t these same people debate and vote on more substantial issues?


  28. Keith,

    This time I actually laughed: “Arthur and his neo-Machiavellian colleagues.”

    You put in a plea for “a conversation on the issue of consent.” Such a conversation will not be about democracy. In a democracy citizens do not consent to be governed. They govern.


  29. Arthur wrote:
    “I point out that in the US 150 million people — your “microscopic pieces” vote in elections. Why can’t these same people debate and vote on more substantial issues?”

    Yes, mass elections and mass participation in meetings of millions of people (sub-divided into smaller groups) are equally ineffective and undemocratic. Since virtually every participant has essentially zero impact on the outcome, few will bother to become well informed. Lots of people choose to participate in elections for different reasons… Some feel it is a civic duty, an expression of communal solidarity, some treat it like cheering for their favorite sports team, and some are deluded into thinking their vote matters.

    Both systems allow those with control of media and money to determine the outcome (within a narrow range, if not precisely). Both provide the illusion of popular sovereignty, though using different methods.


  30. Arthur,

    >I point out that in the US 150 million people — your “microscopic pieces” vote in elections. Why can’t these same people debate and vote on more substantial issues?

    People can (or, more accurately, could) debate and vote on whatever they choose. In order to understand why people won’t behave in the way you prescribe you really need to read the literature on rational ignorance. Then you can decide for yourself whether or not there is a case for statistical representation by sortition. If you decide there is a case then your contributions to this blog will be very welcome; if you decide that you still want to argue the case for direct democracy then there’s not much point making that case here. The only reason this tiny community exists is because we’ve given up relying on direct democracy as a form of democratic judgment (although some of us still reserve a role for citizens’ initiatives), and I can’t imagine that you’ll have much luck trying to win us back over to the dark side of the force.

    >Arthur and his neo-Machiavellian colleagues

    This is on account of the reliance throughout your book on explaining the behaviour of elite actors in terms of psychopathological concepts. This puts you in a similar camp to John McCormick (I thoroughly recommend his book Machiavellian Democracy).


  31. Terry,

    You are mixing apples and elephants. Mass elections governed by sound bites and media propaganda is not the equivalent of a working democracy.

    People become a mass in large numbers when they are treated like objects, have no sense of self and are serving no useful purpose. In a true democracy none of this pertains. By participating, discussing, having a say in critical matters, on a national scale, they become individual adults for the first time.


  32. Keith,

    I will be happy to forsake participation on this blog, but on one condition: that you and other contributors stop misappropriating the word democracy when you mean oligarchy, i.e. representative government. When you use the word democracy you create a conversation on that subject. If you don’t want to talk about democracy then you shouldn’t employ the term.


  33. Arthur,

    Democracy was a term devised in antiquity to describe rule by the many. In small poleis this was unproblematic, as all citizens met up all together to make decisions directly and took it in turns to govern. Democracy then disappeared for 2,000 years until the birth of representative government. The new challenge was how to enable rule by the many in largepoleis in which it is no longer possible for everyone to meet up together and/or take turns. The fact that rule by the many was not the stated intentions of the original founders of representative governments is incidental to this point — a similar case can be made for the development of Athenian democracy.

    So, no I will not refrain from the goal of seeking means to establish rule by the many (democracy). Ironically this necessitates “treating people like [preference-bearing] objects, without a sense of self”, for reasons outlined in these three posts.

    There are many forums devoted to ways of establishing direct democracy, so if that is your goal you should choose one of them to develop your arguments. But there’s no point engaging in debate with a bunch of people who are committed to statistical representation as a means of establishing democracy. And many of us (I’m thinking in particular of Terry and Andre in this respect) have a deep knowledge and respect for Athenian political practice.


  34. Arthur,

    > Mass elections governed by sound bites and media propaganda is not the equivalent of a working democracy.

    Our point is that mass politics are necessarily governed by sound bites and media propaganda.

    But if not, what in your opinion is the cause of this sad situation? Why don’t people discuss policy and select candidates and vote for them in a way that is closer to your ideal?


  35. >Why don’t people discuss policy and select candidates and vote for them in a way that is closer to your ideal?

    It is naive to expect that when people are treated like idiots, systematically denied access to the truth, that they will behave like reasonable, intelligent individuals. There is an excellent book by Benjamin Barber that I strongly recommend. The title is STRONG DEMOCRACY: PARTICIPATORY DEMOCRACY FOR A NEW AGE. If you read this book you will get a comprehensive answer to the question you raise.


  36. > It is naive to expect that when people are treated like idiots, systematically denied access to the truth, that they will behave like reasonable, intelligent individuals.

    Treated by whom? People in power? How will this state of affairs change? Elected government and big media will always have the incentive to dissemble and manipulate. If this is enough to make people abandon the high ideals that you pin your hopes on, then the system you advocate is unstable and therefore unsustainable.


  37. > Elected government and big media will always have the incentive to dissemble and manipulate.

    This is true. We all agree, which is why I am proposing a form of government where there are no elections, a form of government in which citizens govern themselves. It is called democracy.


  38. Arthur,

    Barber’s book was important in launching discussion (I read it about a decade ago, and skimmed it recently). But he is a bit vague and doesn’t go far enough. One can either have full participation, OR meaningful deliberation, but not both all the time (mass participation destroys deliberation). Sortition allows for a neat combination of these two democratic ideals through sequential participation through short-term service on allotted bodies.


  39. Arthur,

    Also, The Athenian democracy (4th century) elevated sortition in the form of legislatoive juries (nomothetai) above the mass participation assembly (eklesia). The assembly voted to strip itself of all law-making authority, and vested the sole right of adopting laws in the randomly selected law-making juries. Sortition was the CORE of Athenian democracy.

    You can read a bit more about this in my paper, “Democracy Through Multi-Body Sortition: Athenian Lessons for the Modern Day” published in the Journal of Public Deliberation. Here is the link:


  40. Terry,

    It is my view that democracy was in decline in the 4th century for reasons which Kitto outlines. When I think of democracy I think of Athens in the 5th century. The eklesia was at the center of political life and Greek civilization was at its height.


  41. > I am proposing a form of government where there are no elections

    So no centralized government at all? How are decisions which affect many people (say, building a bridge, or constructing a computer factory, or setting environmental regulations) made?


  42. Arthur,

    The relevant issue is why when democracy was restored at the start of the fourth century the Athenians made the decisions to diminish the role of assembly, increase the role of sortition and institute the direct election of key financial magistracies.


  43. Let us say there is a proposal to build a bridge from New York to Paris. There are thousands of assemblies around the country, composed of citizens who, by virtue of their citizenship, are empowered to attend and debate. Experts could be invited in, Skyped in, arguing for one or the other side. At each local assembly there could be a debate on stage, before the assembly. Members of the assembly would be able to contribute their thoughts. At some point a vote is taken. These votes taken nation wide would determine the outcome. Alternatively, delegates from each assembly could be chosen by sortition to attend a central meeting, where the issue is again debated and the vote finalized.


  44. Keith,

    H.D.F. Kitto, in “The Greeks,” addresses the issue at length. There was an overall social decline, loss of morale, subsequent to the plague, which decimated the population, taking Pericles, among others, and defeat in the Peloponnesian War. Athenians began withdrawing from public life into private life. There was a change in art, theatre and architecture and life style and not surprisingly a change in attitude towards political life.

    P.S. Key financial magistrates were elected in the 5th century as well.


  45. Hansen’s view is that the 4th century reforms were a conservative backlash against assembly-based democracy. He also claims that 5th century magistrates were primarily strategoi and the lack of elected magistrates was one cause (along with the vicissitudes of assembly decision making) of the emptying of the state coffers. At the risk of repeating myself, this is the WRONG FORUM to argue the case for direct democracy


  46. Strategeoi, yes, and those with a high degree of financial responsibility. Financial irresponsibility was dealt with severely. In one case from the fifth century B.C., the ten treasurers of the Delian League were accused of misappropriation of funds. Put on trial, they were condemned and executed one by one until—before the trial of the tenth and last—an error of accounting was discovered, allowing him to go free.


  47. That’s a well-known case, but it dealt with (alleged) criminality not incompetence. It also illustrates the danger of decision making by large numbers of people, another reason why there was an increased emphasis on decision-making by randomly-selected juries.



    There you go again, Keith, trying to shut people up (and getting your caps key stuck while you are at it).

    No, it makes a lot of sense to discuss “direct democracy” in this forum. Like a sortition-based system, “direct democracy” is presented by its supporters as a system of government that is an alternative to an elections-based system. Analyzing this proposal and comparing it to sortition is certainly of interest.


  49. Yoram,

    Thank you for your support. Mighty Keith is a poor host for it was He who invited me onto this forum and is now dismissing me before the pheasant under glass has even arrived.


  50. > These votes taken nation wide would determine the outcome.

    So I would have to become informed about the various aspects associated building every bridge that I might use or be affected by and every factory whose products I might use or be employed by or be affected by in some other way? And if I have an idea that I think should be discussed, how would I be able to put it up for discussion when the discussion involves thousands, and maybe millions of people? This is completely unworkable and inevitably controlled by various elites.

    > Alternatively, delegates from each assembly could be chosen by sortition to attend a central meeting, where the issue is again debated and the vote finalized.

    But didn’t you claim that this is an oligarchical arrangement?


  51. Well, Keith has a disproportional presence here – ignoring him is often the best policy, IMO.


  52. Yoram,

    Thank you for your thoughtful comment. The key issues that need to be legislated by all of us at the national level are budget and war vs. peace. Many issues are local issues to be handled on a local basis.

    You are right. If this alternative were chosen it would be oligarchic on top of a solid democratic base. And further if the delegates, emerging from an open democratic assembly and chosen by sortition, were to diverge from the sense of their assembly of origin they would be required to return to that assembly for further instruction.

    I highly recommend Stephen Shalom, “A Political System For A Good Society,” which appears on Shalom considers various forms of government and then opts for what he is calling a system of nested councils, which is pretty much what I am proposing.


Leave a Reply

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

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

Twitter picture

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

Facebook photo

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

Connecting to %s

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

%d bloggers like this: