“Representation and Randomness,” Part Three

(See parts one and two in Peter Stone’s review.)

The third paper in the Constellations symposium is “Lot and Democratic Representation: A Modest Proposal,” by Alex Zakaras. This paper has already received some attention here, and so I shall try and approach it from a somewhat different angle.

Zakaras’ “modest proposal” calls for the replacement of national- and state-level senates with randomly-selected legislative bodies. These “citizen legislatures” would not be responsible for drafting legislation. Rather, they would be responsible for approving or vetoing legislation proposed by the second, elected legislative body. They could also compel their elective counterparts to hold floor votes on legislation that is stuck in committee or otherwise stalled. And they would be solely responsible for drawing and redrawing legislative district boundaries (p. 457-458). I particularly like the latter idea. It seems to me that many of the most egregious failings of modern legislatures stem from the fact that they almost invariably get to write their own rules, and enforce those rules upon themselves. That works about as well as most self-policing—it’s better than nothing, but sometimes not by much. I would see redistricting, as well as the creation and enforcement of codes of legislative ethics, as tasks particularly well-suited for a randomly-selected group of ordinary citizens.

Zakaras defends his proposal by arguing that it would advance four vitally important democratic values.—“equal consideration of interests, equal recognition, political autonomy, and deliberation” (p. 459). Before examining his treatment of these values, I should add that Zakaras also compares his proposal to several other recent proposals involving sortition, including ideas by John McCormick, Kevin O’Leary, and David Poulin-Litvak (p. 457). But from his discussion of democratic values, I am uncertain why he believes that his proposal, and not some other, is the right way to protect these values. Others on this site have asked why a randomly-selected legislature could not draft legislation itself, rather than just approving or disapproving legislation drafted for it. I don’t have a strong opinion on this topic myself, but I do wish that Zakaras had explained why he believes that those democratic values are best advanced by a randomly-selected legislature that disposes without proposing.

As for the values themselves, the one that seems to do the most work for Zakaras is equal consideration of interests. The other values seem less connected to the story he wants to tell about democracy. (Indeed, it’s a little hard to tell why “deliberation” should be regarded as a specifically democratic value at all.) This is apparent in his “master story” about democracy, which I gather is supposed to tie together all four values. “Democracy,” Zakaras writes, “presents itself as an answer to the question: how should we make collective decisions? It says, simply, that everyone should have an equal voice in collective decision-making” (p. 459). It’s pretty easy to see how this relates to equal consideration of interests, especially in light of the claim by Robert Putnam (which Zakaras fully accepts) that “The most obvious hypothesis—so obvious that it has rarely been scrutinized carefully—is that decision makers will favor the interests of the social groups from which they come” (quoted on p. 455). And so democracy requires equal consideration of interests. A decision-maker will advance the interests of the group or groups to which he belongs. Therefore, the best way to ensure equal consideration of all interests is to make decisions using groups that mirror the distribution of interests in the population. Random selection does this better than any other procedure can.

This seems to be the heart of Zakaras’ story. It’s also the most plausible part of the argument. Less plausible are his efforts to compare elections with sortition. Zakaras follows Bernard Manin in noting the dual nature of elections. They are democratic, in that everyone has an equal vote, but they are also aristocratic, in that some candidates are much more likely to win than others (p. 456). In Zakaras’ terms, as voters citizens have equal voices with which to protect their interests, but as potential officeholders some citizens will reliably have their voices heard more than others. The result will be less egalitarian than selection by officeholders by lot. This conclusion is unsurprising, but it is unclear from the argument made by Zakaras why any democrat would ever want to elect anyone. Remember that Zakaras’ proposal retains extensive use of elections. I gather that Zakaras wants to endorse elections as a method for ensuring expertise or the like. But that’s a story that’s assumed and not told in this paper.

Finally, I note with some dismay that Zakaras accepts Pettit’s argument that a citizen can be said to exercise control over decisions when those decisions are made by people who think as she does, or by people who advance the same interests she has (pp. 462-463). (I discussed this proposal in part one of my comments on the Constellations symposium.) I remain baffled by the persuasive power of this argument. Presumably Zakaras would not move out of his house because somebody who is very much like him agreed to sell it. There is simply a universe of difference between an agent making a decision that advances your interests and an agent making the decision you wanted him to make, and no amount of philosophical contrivance can shrink it. I fully accept Zakaras’ claim that elections do a poor job of ensuring citizens the ability to exercise control over their representatives (p. 464). But the correct conclusion to draw from this is that in modern societies the idea of literal self-rule is an utter nonstarter. We should abandon it—or at least radically recast it—rather than undergo desperate contortions to save it as a viable democratic ideal.

Advertisements

9 Responses

  1. > But the correct conclusion to draw from this is that in modern societies the idea of literal self-rule is an utter nonstarter. We should abandon it—or at least radically recast it—rather than undergo desperate contortions to save it as a viable democratic ideal.

    Good point. The Athenians had no problem defining democracy as ruling-and-being-ruled-in-turn. This is certainly more attainable in practice than the modern myths of ruling by proxy.

    In addition to being completely unrealistic, however, the modern self-rule conception is an extremely individualist conception of politics. It is an every-man-to-himself conception where unless I personally have my say I cannot expect my interests represented. This has negative implications on several levels. First, of course, if one takes the every-man-to-himself idea literally, politics becomes a zero-sum game which means no cooperation is possible. All collective action and interaction in general is useless. Secondly, if you are disadvantaged, you can only blame yourself, since the only possible reason for being disadvantaged is that you did not represent yourself well (after all, no one else can represent you). Thirdly, life becomes an exhausting exercise of vigilance in which you can trust no one and you must always show up for every decision. This is exemplified by the advice often made on the Left that the voters must exert pressure on elected officials in order to get their agenda promoted. (This may be true, but if so, why bother with elections at all?)

    Like

  2. Yoram: “The Athenians had no problem defining democracy as ruling-and-being-ruled-in-turn. This is certainly more attainable in practice than the modern myths of ruling by proxy.”

    How so in a large-scale modern state, without some extremely radical decentralisation and/or secession? Marcus Schmidt has proposed something along these lines for Denmark (reported in Hansen, 2005), but Denmark has only three million citizens, so an annual allotment of 70,000 would give everyone a good probability of serving once in a lifetime. Not so for Britain or the USA.

    I think we would all agree that the modern self-rule conception is a mirage (that’s why the natural right notion of consent has not survived the transition from the age of the ‘politique corporation’), but Pettit is right to argue that consent by proxy is the least-worst alternative. Peter’s example of selling his house does not reflect the nature of the legislative process, which deals with general questions like what is the appropriate level of taxation for property transactions (which would only kick in if *you* took the decision to sell your house). Such in-principle decisions are perfectly well made by proxy — I really don’t see the problem (or, in practice, the alternative).

    Keith

    Like

  3. > How so in a large-scale modern state

    Naturally, if the population is large enough, your turn to sit on an allotted body may not come within your lifetime. As I see it, this really makes no difference. After all, even in a group in which you get a single one year term in a lifetime, it is hard to believe that it is the decisions made during that single year that really matter in the representation of your interests. It is the process of rotation, rather than its particular instantiation, that matters.

    Like

  4. > Peter’s example of selling his house does not reflect the nature of the legislative process, which deals with general questions like what is the appropriate level of taxation for property transactions (which would only kick in if *you* took the decision to sell your house). Such in-principle decisions are perfectly well made by proxy — I really don’t see the problem (or, in practice, the alternative).

    Keith, I’m not sure I see your point. There are a great many political decisions, both general and specific, that get made by officials that I cannot be said to have authorized to act for me. I don’t see any point is saying that I make those decisions “by proxy,” at least if you take the latter term to involve consent.

    I should add that I am not objecting to the proposal that sometimes decisions should be made by people who are sufficiently “like me,” and that those decisions might even be normatively binding on me (though not always, as in my house example). I think that such decisions might well advance my interests well. I just think that such processes don’t have any necessary connection to consent. It’s a good thing, other things equal, for decisions to be made that reflect our interests. Ir’s another good thing, other things equal, for decisions to be made by agents that we have authorized. The latter might be good for both intrinsic and instrumental reasons (i.e., authorized agents might be likely to advance our interests). But these are two different good things. I think Isaiah Berlin is a dramatically overrated political theorist, but I think he is right to say that political theorists make a huge error when they try to reduce all our values to one “supreme” value. I think consent theorists are often guilty of this.

    Finally, I wholeheartedly agree with Yoram’s idea that a focus on consent may make us shift the blame for bad governmental decisions onto the victims of those decisions. At the extreme, the myth of “government by the consent of the governed” can become positively Faustian. Anyone who objects to the rotten treatment they get from the government gets met with the retort, “Well, isn’t that your signature on the dotted line?” That’s not helpful, to put it mildly.

    Like

  5. The point that I’m trying to make is that the passing of laws (unlike the decision whether or not to sell your house) involves the exercise general judgment and democratic norms require that those judgments should reflect the ‘general will’. By contrast whether or not you sell your house is nobody’s business but your own.

    The $64,000 question is how to uncover the general will — advocates of sortition arguing that an allotted assembly is the best way of implementing Condorcet’s proposals to do that. As to whether that also involves the operationalisation of consent strikes me as a semantic quibble. It’s certainly no worse than claiming that elections indicate consent (Manin’s argument).

    As Fishkin puts it ‘consulting the public’s considered judgments is a bit like seeking its collective informed consent’ (2009, p.195). I’m afraid “a bit like” is about as good as it gets.

    Keith

    PS As to the further case of the particular decisions that are taken by public officials, the best one can do is to rehearse Rousseau’s argument on delegated powers — we ‘consent’ to those decisions in so far as they are faithful to the letter and spirit of the sovereign legislature.

    Like

  6. I received, via Twitter and Gmail an invitation to participate at this site in discussions relating to sortition. Having read a small sample of what is here, it does indeed seem an interesting place. I propose to give something of a synopsis of the ideas that underpin my take on sortition and its rationale. I apologise in advance if in doing so, I fail to address substantially discussed questions on this site, but noting as I do the metalanguage here, it struck me that trying to search the site to avoid this might be unproductive.

    For me the key question of representative governance turns on the legitimacy question. In what senses, if at all, is the exercise of executive power a bona fide expression of the attempt to meet all of the legitimate and contested claims of the community the sovereign ostensibly serves?

    To qualify in this way, it seems to me that the legislature, in its composition, must be and be seen to be rather like the population as a whole in its composition. All of us are far more inclined to suppose that people who are socially like us are more likly to resist doing things we’d fundamentally object to and to be predisposed to serving the interests we see as valuable. That the legislature is like the populace as a whole isn’t a guarantee against them acting recklessly or like tyrants, but it makes it less likely.

    On the other hand, we surely know that large sections of the populace aren’t highly across policy, even in a big picture sense. This is one of the contextual factors that subverts good policy because career politicians can exploit this ignorance (or complain that it constrains them) to do things that amount to very poor policy. In my view, sortition (or any proposed systemt of governance) should foster inclusivity and empowerment. We ought to want a better informed and more engaged citizenry. My outline below aims at ensuring that over time, the pool of people who are engaged with policy and have the skills to analyse and develop good policy grwos.

    We know that whatever similarities communities have they remain complex and socially heterogenous. Even at the most basic levels of service there are going to be conflicts between stakeholders over how to value public goods and to set priorities in the deployment of scarce resources. How does one define an externality or restrain a collective action problem?

    Although politics is rarely discussed in such express terms, these questions, in ostensibly democratic polities, have tended to fall within the rather woolly notion that self-selected delegates of some kind, subsequently approved at a plebiscite counted on the basis of some voting system acquire a kind of power of attorney to adjudge the best interests of the nation as a whole. Edmund Burke (IIRC in his letter to the constituents of Bristol in 1774) many years ago tried to reconcile the notion of delegate authority with service to the notion of the nation as a whole according to conscience and reason. It wasn’t all that convincing or robust, in my opinion, and essentially amounted to a warrant for arbitrary exercise of power.

    One may excuse Burke somewhat of course. In 1774, only a minority of the public was literate and numerate. And even in the case of those who were, there were few vehicles for them to express themselves on matters of policy. It wasn’t entirtely clear which community the people were part of and of course by contemporary standards everyone was massively ignorant of many of the things one would want to be aware of to begin to make rational public policy. It’s by no means clear what by the consent of the governed would mean, if we assume, as we do in matters of contract, that consent must be informed to be genuine.

    Today, at least in the advanced industrial economies, we have excellent education and communication systems. Most adults are at a minimum, literate and numerate and at least in principle, capable of making rational decisions on their own behalf. One may question whether representative governance is superior to, for example, some form of direct democracy. Today, we have highly secure communications vehicles for people to participate directly. We could, in theory, establish a kind of meeting quorum threshhold, specify times for people to participate, have mirrors reconciling all of the data and so on and so forth. Eventually, a vote on a matter could be taken.

    One suspects that in practice, such a system would be very unwieldy, and that there would be objections from aggrieved parties about the integrity of the process, particularly if there were any serious technology based system failure. A voter pool of 1400 might work well this way, but a voter pool of 14 million (as is the case in my country, Australia) would seriously test the schedule and technological constraints of such a system.

    It seems to me that Direct Democracy, while unworkable for the day to day business of government, might well be very good for the big picture stuff. It might well be that at a scheduled point every year, the populace as a whole would affirm or amend a set of national policy goals for the next 1, 3, 10, 20 even 50 years. Throughout each year, working groups within the legislature would make proposals on various matters that would be publicised within the populace as a whole. Studies would be commissioned, analysis done submissions considered and grouped and proposals refined. Those who dissented could mount counter-campaigns and if these acquired significant support these too could be put at the annual consideration of the national plan.

    Once a plan was agreed, the task of the legislature would be to implement it as effectively and efficiently as possible and as above, to propose amendments where it was still thought to be inadequate.

    It would be the legislature that would be chosen by sortition. In my view, half way through the term of the legislature (perhaps 2 years in) one would draw from those who had expressed an interest in becoming representatives and were not excluded on one ground or another (serious criminal offences, mental incapacity etc) a long list (perhaps 20-25) per district. The candidates would then be assessed for their skills and offered training in areas they deemed pertinent to their areas of primary interest — (eg surveying, engineering, finance, social policy etc). They would be provided with support staff and the resources needed to communicate with their constituency. 12 months into the process a deliberative poll would be held. Each candidate would submit up to 10 policy areas in which they had a vision for the country, and the responding constituency would weight these areas for significance (proportions of 100%) and for their support (scaled from 1-10). One could choose of course to vote simply to reject the candidate.

    At the conclusion of the poll each candidate would be given the data from poll and this would be publicised. The candidate could revise policies withdraw or continue as he saw fit. This process would be iterated in six months and then finally just before the final sort.

    Ultimately all the values from all the polls would form the basis for a weighted pool and the final selection would be conducted like a lotto competition. The candidate would be selected along with two alternates, in the event the selected person was unable to serve out his or her term.

    It seems to me that this would result in a legislature that would substantially resemble in its composition the polity as a whole, though in practice it would still be skewed somewhat up the socio-economic scale, in favour of those with relatively advanced policy making and communication skills. Yet it would not guarantee that those with the most consensual views of what was good policy had exclusive access to power. Some with more controversial ideas would benefit from anomalies in the selection, and ensure that the legislature was not simply a ‘hive mind’.

    In this system political parties could no longer be vehicles for the promotion of candidatures. Their attempts to influence the pool could only be indirect, since even a majority of support would not gurantee a candidate would succeed. Their role would be to attempt to create a consensus around certain policy goals by making a rational public interest case.

    The strength of this system, it seems to me, is that a much wider slice of the population would see being a member of the legislature as plausible, and thus be more enegaged with the legislature’s policy-making behaviour. Within 20 years, pretty much everyone would know someone who had been considered for office and the idea that one might be called upon to serve might moderate some of the more reckless claims people sometimes make, knowing they will never be called on their claims. In short, we would get a better talent pool to draw upon, which in the longer run, is perhaps even more important than the structure of governance.

    I take note of the problem raised above about whether one would be happy to have someone like you decide whether your house should be sold.

    It seems to me that one of the indispensible features of democracies concerns the reach of state authority. The purpose of the surrender of individual sovereignty to states is to make it possible to resolve conflict over public goods or social obligations to one’s community. What one does that is regarded by the community as within one’s private discretion is not something to be determined by the soveriegn power, even if they really are like you. What the soverign should do is set the rules for public goods as a whole, specifying claims and obligations that are legitimate and denying those that are not.

    Like

  7. Hi Fran,

    Welcome and thank you for your thoughts. There are quite a few points in your proposal that are worth discussing in some more detail – I am turning this comment into a post, aiming to have a discussion in the comment thread to that post.

    Like

  8. […] post was originally posted as a comment by Fran […]

    Like

  9. […] “Representation and Randomness,” Part Three « Equality by lot Quote:"Zakaras’ “modest proposal” calls for the replacement of national- and state-level senates with randomly-selected legislative bodies. These “citizen legislatures” would not be responsible for drafting legislation. Rather, they would be responsible for approving or vetoing legislation proposed by the second, elected legislative body. They could also compel their elective counterparts to hold floor votes on legislation that is stuck in committee or otherwise stalled. And they would be solely responsible for drawing and redrawing legislative district boundaries (p. 457-458). I particularly like the latter idea. It seems to me that many of the most egregious failings of modern legislatures stem from the fact that they almost invariably get to write their own rules, and enforce those rules upon themselves. That works about as well as most self-policing—it’s better than nothing, but sometimes not by much. I would see redistricting, as well as the creation and enforcement of codes of le (tags: democracy monarchism philosophy weekendreading) […]

    Like

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

%d bloggers like this: