Dahl’s arguments against an allotted legislature

Posted on March 22, 2010 by Yoram Gat

In his 1970 book, After the Revolution?, Robert A. Dahl suggests appointing by lot advisory committees for the president of the U.S., for congresspeople, for state governors and for large city mayors^[1]:

Let us imagine that the membership of each advisory council were to consist of several hundred constituents picked by the same procedures used to ensure randomness in modern sample surveys; that the citizen selected would be required to serve […]; that suitable provisions would insure against hardships arising from the obligation to serve – for example, the citizen selected would not only have all relevant expenses taken care of but if he (or she) were poor or unemployed he (or she) might receive a stipend, while an employed person would continue to receive his (or her) regular pay; that one would serve for a year and be ineligible for a second term; that a council might meet at intervals for a total of several weeks in the course of a year; that it would have its own presiding officer (and a professional parliamentarian); that it would invite the elected official to meet with it, to answer questions, hear the debate and discussion…

A timid thinker would have focused on making the argument supporting such an idea against “conventional critics” who would argue that “the proposal goes too far”. Dahl, however, dismisses those critics with one paragraph, and spends the rest of the discussion (pp. 150-153) arguing against “a less conventional critic” who argues that the proposal “does not go far enough” because it does not suggest using the lot to replace elections for selecting government officials.

What makes Dahl’s argument even more interesting is that he goes beyond the classical Socratic argument (the ship of state must be controlled by the most qualified pilot rather than by a random group of laymen). Dahl presents two lines of argument against applying sortition to select the legislature. In the first argument he makes the following points:

To achieve adequate understanding of public policy, terms for the allotted delegates would have to be fairly long – as long as those of current elected delegates.
Obligating people to serve such long terms is “inconceivable”.
But on the other hand: “If service were optional, the system no longer being random would lose much of its merit.”

Dahl concludes the discussion with his second argument:

In Athens, after all, the officials chosen by lot were subject to the Assembly. In the absence of elections, to whom would a legislature chosen by lot be subject? Since – as anyone familiar with the laws of probability knows – the chances are by no means negligible that a sample of five hundred might deviate by a considerable margin from the mean of the whole population, occasionally we might find ourselves with a highly unrepresentative legislature subject to no authority except the next lottery. I cannot think of a better way to discredit the idea and democracy itself.

Both of those arguments are, to my mind, not very strong. It is intriguing that Dahl was content to leave the matter at that. Others were not, however, and the 1980’s have seen a number of authors proposing various forms of sortition that are not merely advisory (Mulgan, Callenbach and Phillips, Burnheim).

[1] This proposal to use the lot to distribute political power may be the first such proposal by a well known political philosopher in a very long time. Richard Mulgen, in his 1984 paper “Lot as a Democratic Device of Selection”, asserts that “[t]he last major political theorist to have taken the lot seriously was Rousseau.” Peter Stone (who pointed me at After the Revolution?) points to Robert Paul Wolff’s book In Defense of Anarchism (1970 as well) as another modern instance of reference to the lot as a political tool. Wolff, however, uses the consideration of the lot mainly as a tool for examining justifications for majoritarianism and argues, briefly, that it should not be used as a tool to reach political decisions.

Let us imagine that the membership of each advisory council were to consist of
several hundred constituents picked by the same procedures used to ensure
randomness in modern sample surveys….

Filed under: Sortition |

11 Responses

Terry Bouricius, on March 22, 2010 at 5:40 pm said:

For those unfamiliar with Robert Dahl, he is considered by many as the Dean of American Political Science, and is very highly respected. He discusses sortition more recently in his 1989 book “Democracy and Its Critics,” which concludes with a proposal for the creation of a modern democracy through what he terms “minipopulous consisting of perhaps a thousand citizens randomly selected out of the entire demos.” He imagines them meeting using modern telecommunications, and that each might deliberate for a year on a given issue. There would be numerous versions of these bodies, each dealing with a particular issue, with another minipopulous creating an agenda for these issue bodies.

LikeLike
Dahl’s arguments against an allotted legislature « Pro Bono Statistics, on March 23, 2010 at 5:28 pm said:

[…] 23, 2010 A post on Equality-by-Lot. Posted in Uncategorized | 1 Comment […]

LikeLike
Yoram Gat, on March 23, 2010 at 7:11 pm said:

He imagines them meeting using modern telecommunications, and that each might deliberate for a year on a given issue.

Those minipopuli are still conceived as having merely advisory power. There are two obvious problems with Dahl’s proposal: (1) that the decision makers might well ignore the advice of such bodies (2) that working part time over a year it is very difficult to reach an informed opinion on complex matters.

Both of those issue interact and become more severe when the matter of motivation is considered. When there is a good chance that your hard work will be ignored, there is little motivation to put in that hard work, which of course reduces the chance of reaching an informed opinion even further, which of course serves as a good justification for ignoring the advice to begin with. Dahl did not address any of those issues, neither in 1970 nor in 1989.

LikeLike
Terry Bouricius, on March 23, 2010 at 11:57 pm said:

Well, yes, Dahl does say he imagines these sortition bodies would not be a substitute for elected legislative bodies, but rather a complement (“supplement, not replace”). However, this does not necessarily imply they would be merely advisory. Under his description certain kinds of societal decisions might be transferred to sortition bodies, while others are retained by elected bodies (or markets, or whatever). I am more interested in a total substitution, but such a change might best happen by such a step-by-step transition. The fact that the Dean of American Political Science has written favorably about sortition bodies, is an important badge of credibility when presenting the concept to thoughtful people.

LikeLike
Yoram Gat, on March 25, 2010 at 5:21 pm said:

I agree that Dahl’s endorsement is a feather in the cap of the idea of sortition. It is however a decades-old feather.

More importantly, when taking a step-by-step approach, it is important not to take steps whose results will discredit the ultimate goal. I suspect that, in view of the problems I described above, allotted advisory committees would be easy to target as being amateurish sideshows driven by cranks, and this would be portrayed as indicating that the idea of sortition is radically flawed.

In that sense, policy juries (i.e., allotted groups that pick between alternatives designed by elites) along the lines suggested by Keith Sutherland or by Ethan Leib are a better small-step change toward sortition.

LikeLike
Ronald Mercer, on April 7, 2010 at 11:46 am said:

Sorry.
Sortition makes a fantastic “suggestion box”.
No oligarchy is “reigned in” without an empowered sortition chamber.

LikeLike
links for 2010-05-17 « Marty Andrade, on May 18, 2010 at 3:04 am said:

[…] Dahl’s arguments against an allotted legislature « Equality by lot Quote:"as anyone familiar with the laws of probability knows – the chances are by no means negligible that a sample of five hundred might deviate by a considerable margin from the mean of the whole population, occasionally we might find ourselves with a highly unrepresentative legislature subject to no authority except the next lottery. I cannot think of a better way to discredit the idea and democracy itself. […]

LikeLike
Early advocate of sortition in government Robert Dahl has died | Equality by lot, on February 10, 2014 at 5:20 am said:

[…] Dahl was a prominent political scientist and an early advocate of using sortition in government. He proposed advisory allotted bodies in his 1970 book After the Revolution and made a similar proposal […]

LikeLike
Common Lot Sortitionist, on February 10, 2014 at 2:48 pm said:

[…] the chances are by no means negligible that a sample of five hundred might deviate by a considerable margin from the mean of the whole population, […]

Yoram, I recall a series of posts in which you explained that the Law of Large Numbers would, as I recall, assure chances of a deviation as great as — for instance, 40/60 of men and women — something like 1 out of 100,000.
So who has the math right?

+++++++++

[…]But on the other hand: “If service were optional, the system no longer being random would lose much of its merit.”[…]

If the populace approves service being optional — recognizing that the allotted body will then be less statistically representative — I don’t think that would damage its merit.
Most people would, I believe, prefer to avoid or ignore most policy decisions that do not have an immediate and direct effect upon their lives. ‘Giving these over’ to a group willing to ‘devote service’, for the good of the nation, would be fully acceptable… especially given the riposte: “Well, if you don’t like what they are doing, then put yourself in the allotment pool.”

LikeLike
Yoram Gat, on February 10, 2014 at 5:57 pm said:

> So who has the math right?

See here: https://equalitybylot.wordpress.com/2010/05/13/stratified-sortition/.

> If the populace approves service being optional — recognizing that the allotted body will then be less statistically representative — I don’t think that would damage its merit.

I think that when people turn down allotted seats in a sortition-based system (or, indeed, accept the nominations but are unwilling or unable to participate effectively in policy making) the system does become less democratic. The solution, however, is not to make service mandatory but to reduce the hurdles and increase the incentives for (effective) service. This is a non-trivial task that would have to be addressed on an ongoing basis by the democratic system itself, but it is not an insurmountable problem.

LikeLike
keithsutherland, on February 10, 2014 at 6:19 pm said:

David

>Most people would, I believe, prefer to avoid or ignore most policy decisions that do not have an immediate and direct effect upon their lives. ‘Giving these over’ to a group willing to ‘devote service’, for the good of the nation, would be fully acceptable.

That’s an accurate paraphrase of Benjamin Constant’s argument for “modern liberty” (electoral representation). All you need to add is that if you don’t want to devote service yourself then you can choose someone else to do it for you and, hey presto, you end up with professionals and political parties. Sortition does presuppose some kind of republican ethos (“ancient liberty” in Constant’s parlance) and Yoram is right that how to achieve it is a non-trivial problem. I agree that financial incentives are important but that all smacks a bit of possessive individualism. Public service should be seen as a duty that all citizens should aspire to for intrinsic reasons and that requires more than bribes.

LikeLike