Thursday, February 23, 2006

A Democratic Alternative to Elections

A Democratic Alternative to Elections

Legislative bodies are selected through elections in all modern democratic countries. In view of this rule, to which there is not even a single exception, one might presume that elections are the only democratic method to select a legislative assembly. Such a presumption would be wrong. There is another democratic method to select legislators, a method that was used widely in the past (and was in fact considered more democratic than elections), but is almost completely forgotten today.

The alternative method is called sortition [1]. Sortition is the method of selecting the legislators by drawing lots among all the citizens. Those citizens whose names are drawn serve as legislators for a fixed period, as is the common practice with elected legislators. [2]

My goal here is to describe briefly the characteristics of the sortition method and to compare it to the method of elections. But first, due to the possibility that the idea of sortition seems fantastical to the modern reader, here are two historical facts:

1. The legislative work in the Athenian democracy was carried out by two bodies – the Assembly which consisted of all Athenian citizens, and the Council, which consisted of a few hundred citizens chosen for a period of one year through sortition. In Athens, the use of elections was reserved for a small number of positions, such as generals – positions that were considered as requiring specialized skills or talents. Sortition was used in other Greek democracies as well and in a more limited fashion in the Italian republics of Florence and Venice. [3]

2. Philosophers of politics, from antiquity (Aristotle) to the 18th century (Rousseau, Montesquieu) discussed sortition. When compared to elections, sortition was considered the more democratic of the two methods and elections was considered the more oligarchic. As they saw things, sortition tends to put more governing power into the hands of the average citizen, while elections tend to concentrate power in the hands of a small set of distinguished people. [4]


The essential characteristic of a legislative body selected through sortition is that it has a statistical resemblance to the population from which it was drawn: the proportion of women, lawyers, minorities, homosexuals, millionaires and day laborers in the body will be similar to their proportion in the population. In general, the proportion of members who belong to a certain group, who subscribe to a certain political opinion, or who follow a certain pattern of behavior will be similar to the proportion in the population. The rule is the same one used in opinion polling – a sample drawn at random from a population is unlikely to be substantially different from the entire population regarding with respect to any given criterion.

This characteristic – the distributive resemblance – assures that the legislative body will be a reflection of the people: if the people are mostly reasonably honest and skillful, so will be the legislators. On the other hand, if the people are mostly corrupt or incompetent, the legislators will not be any better.

The trouble with elections

According to the ideal elections scenario, honest and competent people become well known for being honest and competent, and, as such, are elected by the public as its legislators. In such an ideal situation, the legislative body is better than the public as a whole, and despite the fact that the legislators are unlike most of the public (they are, if nothing else, more honest and competent than normal people), the legislators are able – due to their superior wisdom – to understand the needs and wishes of the public and to promote them [5]. In other words, according to the ideal scenario, elections do indeed tend to concentrate the governing power in the hands of an oligarchy (exactly as the classical philosophers described), but since it is an oligarchy of honesty and competency (as opposed to oligarchy of wealth, nobility or access to the means of violence), then it is an oligarchy that serves the public interest. Such an oligarchy, it is argued, should be embraced. [6]

This ideal scenario may materialize, at least in an approximate form, as long as the number of voters is small – no more than a few dozen people. In such a small group all the voters know each other well and can therefore perceive the character and faculties of each candidate fairly accurately. In such a situation the decision who to vote for is rational and may bring about positive outcomes.

However, when the number of voters is large things are very different. When there are hundreds, thousands or millions of voters, elections, even those elections which would be recognized as free and fair, tend to generate legislators who are rich, well connected and devious. Not only are the skills and honesty of the elected not better than those of the average citizen, they are actually far worse.

There are several reasons for this poor state of affairs. Some of those are due to imperfections of the present implementation of the elections method and may be remedied. Others, however, are inherent to the elections method itself. Essentially, the two characteristics of the method which cause the election of people who do disservice to the public are:

1. In order to become a credible candidate a person has to be well-known. Few people, if any, would vote for a person of whom they have never heard, while there are quite a number of people who would be willing to vote for a person about whom they know very little apart from his or her name. Therefore, a candidate who is not known by a large part of the public cannot be elected.

The number of well-known people is limited, and they tend to be people with unusual skills and background: as mentioned above, they are usually richer than the average citizen, and they are better than average at self-promotion. They have also been unusually successful socially and politically. The group of potentially credible candidates is therefore small and restricted to people who are likely to be different than most of the pubic both in their objective situation and in their world view.

2. When the group of voters is large, most voters must vote for someone whom they never met, and certainly do not know personally. Under such circumstances, how can they know what are the candidates’ skills and characters? The voters learn about the candidates using only information transmitted via the mass media – either in political ads or in articles and interviews. Therefore, the ability of most voters to truly know the candidates is extremely limited and their view of the candidates is determined crucially by the way they are portrayed by partisan operators. The door is wide open for people with control over media content (either through the purchase of ad spots, or through influence on journalists and editors) to attract votes by creating a false appearance of competency and honesty.

Indeed, it is hard to detect in the reality of modern democratic countries any resemblance to the ideal scenario of elections. Much of the public has noticed that the elected are mostly busy promoting their own interests and those of people like them (and since most of the public does not fall into this category, the majority’s interests are neglected.) [7] The supporters of the elections system would tell the frustrated pubic: “If this is so, throw out the corrupt politicians and take others in their stead.” However, there is no person whom most of the public knows well enough to be able to vote for with a rational expectation of having that person serve the public well. While every citizen probably knows at least one or two people they appreciate enough to trust with a seat in the legislating body, being appreciated by your acquaintances does not get a person significantly closer to being a legislator since no one is acquainted with tens or hundreds of thousands of people, and having a lower number of supporters is not sufficient to obtain a legislator seat.


In the 18th century, when the first modern democracies were created, it was common among educated people to consider the average person as being senseless, and it was therefore accepted that one of the functions of government is to prevent the exercise of the will of that senseless majority. Some of the framers of the American constitution did not hesitate to state such ideas on public record. Under these assumptions, elections are the natural procedure to select legislators: the 18th century constitutions, which relied on elections, were taking only one step away from the prevailing state (oligarchy or monarchy) toward a democracy, a step that was carefully designed not to place governing power in the hands of the average citizen [8]. These days, on the other hand, the assumption has changed – at least in official and public discourse – and it is customary to attribute rationality to the public and maintain that the government’s role is precisely to carry out the will of the public. Sortition is the natural representation method stemming from these present-day professed beliefs.

Sortition gives up the dream of having legislators who are better than the public but gains in return a legislating body that reflects the public in its composition and whose members understand the needs and wishes of the public. With such a body there is a possibility, and arguably a likelihood, that most of its members would do their best to promote the public’s well-being. It is sortition, rather than elections, that gives the people the rulers they deserve – rulers in their own image.


1. The only modern book (as far as I know) that deals with sortition in detail is Bernard Manin’s The Principles of Representative Government (Cambridge University Press, 1997). Many of the ideas and the data here originate in that book. Another book that deals with sortition – it is, in fact, a call to use sortition for selecting the members of the US house of representatives – is Calenbach and Philips’s Citizen Legislature (Banyan Tree Books / Cedar Glass, 1985).
2. Sortition can be viewed as a variation on the idea of rotation. In the rotation system every group member serves in office for a fixed time period. When the group is large, the rotation cannot be completed. Those members whose names were drawn in the lottery are those whose luck it was to be first in the rotation.
3. See Manin, Principles of Representative Government. Aristotle describes the use of sortition in Athens in his book The Athenian Constitution.
4. Aristotle, Politics, book IV, section 9. Montesquieu, The spirit of the laws, book II, section 2. Rousseau, The Social Contract, book IV, section 3.
5. The ideal scenario is present in the background when democratic elections are discussed, but it is difficult to find an explicit answer in those discussions to the critical question of what is exactly the process by which skillful and honest people become credible candidates able to draw large numbers of votes. It is implied that this occurs spontaneously. The Federalist Papers, for example, touch briefly on this topic: “Who are to be the objects of popular choice? Every citizen whose merit may recommend him to the esteem and confidence of his country. No qualification of wealth, of birth, of religious faith, or of civil profession is permitted to fetter the judgement or disappoint the inclination of the people.” (The Federalist Papers #57) A description of the mechanism by which the members of the public are to learn about the merits of their fellow citizens is omitted.
6. Thomas Jefferson wrote in a famous paragraph from a 1813 letter to John Adams: “[T]here is a natural aristocracy among men. The grounds of this are virtue and talents. [...] May we not even say that that form of government is the best which provides the most effectually for a pure selection of these natural aristoi into the offices of government?”
7. In the United States, for example, a February 2006 poll showed that about 80% of the public thinks that bribery is common practice in Congress. Opinion polls also show that congress rarely has the approval of a majority of the population, and quite frequently has its disapproval rating higher than its approval rating. See
8. See, for example, Alexander Hamilton in Farrand, Records of the Federal Convention, Vol I, page 299, and James Madison, in the same volume, page 431. Edmund Burk expressed similar thoughts in Reflections on the French Revolution, paragraphs 80-81, as did Emmanuel-Joseph Sieyes in Reasoned Exposition of the Rights of Man and Citizen, with his notions of “active” and “passive” citizens.

Can't we just call "sortition" government by grand jury?
Post a Comment

Subscribe to Post Comments [Atom]

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]