My Path

from "The Trouble With Elections: Everything We Thought We Knew About Democracy is Wrong," Chapter 1.4

My own path to the concept of sortition did not begin in academia. Although I majored in political science, I never read a thing about sortition in college. My interest in sortition grew from decades of in-the-trenches involvement in electoral politics, two decades as an elected public official, and decades as an election reform policy analyst and activist. I got involved in third party politics while a student at Middlebury College in Vermont. My opposition to the war in Vietnam led me into leftist politics in Vermont’s small Liberty Union Party. I ran for the State Senate and then Lieutenant Governor of Vermont. These were “educational” campaigns (this being the euphemism for campaigns with no hope of victory).

In 1980 I was involved with the founding of the new Citizens Party, which put forward environmentalist Barry Commoner as its presidential candidate. Then, in 1981, the unexpected happened in the Burlington, Vermont municipal election. I was running for a seat on the city council as a Citizens Party candidate against an incumbent Democrat, while a friend, and fellow former Liberty Union Party leader, Bernie Sanders, was running for mayor as an independent. Much to our surprise, we both narrowly won our respective elections. I was the first Citizens Party candidate elected in the country, and Bernie Sanders unseated the incumbent Democratic mayor. After several terms as mayor, Bernie Sanders went on to be elected as Vermont’s Representative in Congress in 1990, and then, to the U.S. Senate in 2006 — all as an independent. After serving a decade on the city council, including a term as Council President, I was eventually elected to the Vermont House of Representatives under the Progressive Party label. I served a decade as a State Representative, stepping down in 2001.

During my time as a legislator, it became obvious to me that the “people’s house” was not very representative of the people who actually lived in Vermont. My epiphany came while dealing with a bill affecting the rights and responsibilities of tenants and landlords in an eviction situation. The cramped committee room had a set of desks for legislators arranged in a rectangle, facing the center, nearly touching, with just enough room in the middle for a tape recorder and microphones. One desk at the head of the rectangle was the chairman’s, and at the other end of the rectangle the witnesses would take a seat one at a time to give their testimony. The committee members were an outgoing and garrulous bunch. Shy wallflowers almost never become legislators. 

During a break between witnesses, one committee member regaled a few of us with a story about how a landlord friend of his had had an apartment trashed by a renter. The story had vivid details about damage to the carpeting and walls. While not directly relevant to the bill under consideration, the story did create a frame with a clear perpetrator and victim. Nobody offered a counter-story in which a tenant had been abused by a bad landlord. While discussing the eviction bill during committee “mark-up” (when committee members propose amendments, etc.) I realized that committee members were relying more on anecdotal information from their own circle of contacts and those of their fellow committee members, than any statistical reports or formal testimony from scheduled witnesses. They also naturally interpreted any testimony through the filters of their own life-experience, giving extra credence to information that corresponded with their preconceived ideas, and re-interpreting or discounting testimony that didn’t fit their world view. Testimony on behalf of tenants seemed to be especially discounted. With hindsight, I recognize that I did the same thing. But being a working class advocate, I was on the opposite side from most of my colleagues. I saw that the experiences and beliefs of legislators shape legislation far more than facts. 

I conducted an informal survey to find out how many of my colleagues on the committee were tenants, vs. home-owners or landlords. Roughly one third of Vermonters were renters, but not one of the committee members was a renter. So I expanded my informal survey to the entire House, and as far as I was able to determine, only one out of 150 members was a tenant (no, it wasn’t me -- my wife and I had recently purchased a house). It was readily apparent that the interests of tenants were disadvantaged in the House, and the resulting statutes reflected that fact. I thought that if the House were genuinely representative of the population, meaning one third of the legislators had been renters, that the nature of the debate, and possibly the outcome, could have been very different. While research indicates that mere diversity of opinion within a group does not necessarily mean a melding of views (the method of deliberation is key, as will be discussed later), one might hope that the debate would have been more robust, with questionable assumptions actually being questioned.

After that experience, I frequently commented that any 150 Vermonters pulled from the phone book would be more representative than the elected House membership. While the transition away from landlines to cell phones makes the proverbial phone book an outdated reference, a genuinely random sample would be more representative in terms of the renter/landlord balance, but also in every other respect, including gender, class, race, age, and political philosophy. 

The intellectual conservative, William F. Buckley, Jr.  made a similar point decades earlier, when he quipped that he would “sooner live in a society governed by the first two thousand names in the Boston telephone directory than in a society governed by the two thousand faculty members of Harvard University.”  Buckley was probably attempting to belittle the elite faculty of Harvard, rather than seriously proposing sortition. However, this sentiment, that random citizens might be better than elites, is actually rather common. A public opinion survey by Rasmussen Reports conducted in January of 2010 found that just thirty-six percent of Americans thought Congress does a better job than a random group of Americans selected from a telephone book could do, while forty-five percent thought the random sample would do a better job. By 2024, just twenty-seven percent thought Congress does a better job, while fifty-four percent thought a random group of ordinary people would do better. This response might just be an expression of disdain, as in “a monkey could do a better job than Congress.” But when made more concrete, support for the concept of sortition actually increases. The Center on Policy Attitudes, associated with the School of Public Affairs of the University of Maryland, did a survey in 1999 which included the following more detailed question:

“Imagine that a group of 500 American citizens was selected from all over the country to be representative of the entire US population. This group then met and were informed on all sides of the policy debate on a number of public policy issues and had a chance to discuss these issues. They were then asked to make decisions on what they thought was the best approach to these issues. Do you think the decisions of such a group would probably be better or worse than the decisions that Congress makes?”

66 percent thought the representative group would be better than Congress, and just 15 percent thought they would be worse. A poll released by a democratic lottery advocacy organization, Of By For, showed almost identical results in 2020. A 2021 survey of over 4,000 Americans conducted by the Pew Research Center (though with less specific question wording) found essentially the same attitudes. My decade of experience serving in the state legislature convinces me that this popular assessment is correct. While there are some exceptional individual legislators (both good and bad), and elected legislators are probably more extroverted and egotistical than the average citizen, I saw no evidence that they are more competent, intelligent, compassionate, nor capable than any random group. However, as I will demonstrate in subsequent chapters, elected policymakers have unique electoral imperatives and psychological biases that make them generally less competent and capable than ordinary people.

I stepped down as a legislator after ten years, and went to work with a national voting reform organization called FairVote: The Center for Voting and Democracy as a policy analyst. For the next decade I primarily worked on proportional representation and a ranked ballot voting system known as instant runoff voting or ranked choice voting. While I still believe such reforms have the potential to make improvements to American elections, I have concluded they don’t address the root problems. 

The ground-breaking use of a democratic lottery in the Canadian province of  British Columbia sparked my interest in this fundamentally different model of democracy. In 2004, 160 randomly selected residents (one man and one woman from each of the province’s 79 legislative districts plus an aboriginal man and woman) were convened as a “Citizens’ Assembly” with a mandate from the provincial government to evaluate the electoral system, and if they deemed it prudent, to propose a new one for the province. Over many months they went through a learning phase, a public hearings phase and a deliberative phase to produce a final referendum proposal. As an election reform expert, I gave testimony to this citizens’ assembly. The quality of their work was impressive, and demonstrated the potential of sortition. This was my aha moment. I came to the stunning realization that “this is what democracy should look like!”  From then on, the primary focus of my political reform work has been the study and advocacy of democratic lotteries as an alternative tool for operating a democracy.