Specialization and Taking Turns Governing

Adam Cronkright
  • Heimilisfang: Kingston, Ontario, Canada
  • Skráð: 24.06.2011 11:45



My two suggestions:


1) Government representatives should specialize by function or area of responsibility


Why is each member of parliament asked to make decisions on such diverse issues like budgeting, foreign policy, schools, the economy, healthcare, roads, waste management, and energy?  It is possible for an individual to make informed and prudent decisions about issues in one of those big arenas, maybe even a couple.  But as the quantity and diversity of the decisions that a public official is asked to make increases, what happens to the quality of those decisions?  How much time can be devoted to each topic?  How deeply can each issue be investigated?  How much input can the public have?  How much deliberation and reflection can take place?

Parliaments and congresses around the world recognize this problem, and have responded by forming committees and sub-committees to specialize in certain areas.  One problem with this approach is that the formation of these committees is not democratic (ordinary citizens cannot decide which elected representatives form the budget review committee, etc.).  Another problem is that committee recommendations still rely on approval by the rest of parliament/congress before any legislation can be passed.  Members not on the committee cannot be as well informed about the issue, but their vote counts equally.  This part of the process is often hijacked by party politics.

And why are we sometimes forced to vote for a candidate whose views on healthcare or foreign policy we strongly disagree with, but we like their approach to the economy and we simply dislike all the other candidates even more?  Is it reasonable for me to expect any one person to represent my interests across the broad range of important issues that affect me as a community member?

In almost all types of work, specialization is seen as a key to efficiency.  Why should community decision-making be any different?

Iceland could have small independent governing bodies that are each in charge of a specific function or area of responsibility.  The complexity of the function would determine the level of specialization (i.e. “transportation” could be broken up into several bodies, one governing “marine transportation”, another “road transportation”, etc.).

These independent governing bodies would coordinate with each other on issues affecting multiple functions or falling under multiple areas of responsibility.

Iceland could also have “second-level” bodies that would not deal with any specific community responsibility, but would instead specialize in administrative policies.  One body might be in charge of budgeting, another making procedural changes to “first-level” bodies, and still another might arbitrate conflicts between “first-level” bodies on broad issues.  It would be important that these “second-level” bodies would have no say over substantive issues.  For example, a “second-level” body might decide that a new “first-level” body is needed to govern “recycling”, but the “second-level” body could not make actual policy on recycling.  See the web link for a diagram helps illustrate these divisions.



Arguments for this approach


By allowing “first-level” governing bodies independence and the ability to coordinate themselves, this structure is not bureaucratic.

Having only one area of responsibility should give representatives:

  • More time for public input
  • More time to investigate issues and seek expert opinions
  • More time to deliberate with representatives who hold opposing views
  • The opportunity to grow in their understanding of their area of responsibility over the course of their term

Restricting representatives to only one area of responsibility should also:

- Restrict campaigns to focusing on the specific issues of one functional governing body, instead of producing vague promises about all kinds of broad issues

- Make it easier for voters to evaluate their performance and hold them accountable

Breaking up power between “first-level” bodies (substantive issues) and “second-level” bodies (administrative issues) can prevent conflicts of interest that are common in current democratic structures (like when members of Parliament or Congress – the winners of their respective political campaigns – are asked to decide on campaign finance reform).


Possible arguments against this approach:

Representatives who spend their entire term/career dealing with one area of responsibility might be more vulnerable to the influence of special interest groups.

Even if each independent body was limited to just 10 members, this structure would definitely require more representatives than currently proposed in the draft constitution.  What happens when Icelanders go from voting for 63 positions in Parliament to filling maybe 30 independent governing bodies with say 10 members each?

  • Voting becomes less informed – how can people learn as much about all the hundreds of candidates running for those 300 positions?
  • Citizens could become fatigued by all the campaigning and voting
  • Citizens may resort to simply voting along party lines

These challenges highlight one of the major problems for any electoral democracy:  A structure with fewer elected representatives tends to ask too much of the representatives.  A structure with more elected representatives tends to ask too much of the voters.  Most governments respond to this problem by creating bureaucratic agencies to handle different functions and lighten the load on the elected representatives.  But bureaucracies tend to be inefficient, inflexible, and undemocratic.

These challenges bring us to my second suggestion for the Constituent Council…


2) Community decisions should be made by (everyday) community members.  Functional governing bodies should each be made up of a cross-section of the general population.


In an ideal political process, every community member could and would:

  • Become well informed on all the issues affecting their community
  • Deliberate with other community members who hold different views
  • Be equally involved in coming to an agreement on each issue

This would insure that the interests of all community members are properly represented.  However, this is neither practical nor possible.

But what if Iceland filled each small “first-level” governing body with a cross-sectional sample of the population?  These 10-20 citizens could become well informed on a specific set of community issues, listen to expert and lay testimony, deliberate with each other at length, and work to come to a consensus decision on each issue in their small area of responsibility.  Since, statistically, they they are a mini-version of the whole community, wouldn’t their decisions be very similar to the decisions made in the ideal but impossible process described above?

How might each governing body be filled with a cross-section of the population?

1) Citizens could be selected randomly from the entire population as with jury duty in many countries

  • The main advantages of this approach is that it produces a near perfect cross-sectional sample of the community and it is easy to administer
  • The main disadvantage is that it is compulsory and some citizens might lack motivation in their new temporary role as a public official

2) Citizens could volunteer for public office and members would be randomly selected from the pool of volunteers.  Stratified sampling techniques would be used to keep governing bodies representative of all identifiable demographics in the population (i.e. make up for the fact that certain demographics are less likely to volunteer for public office).

  • The main advantages of this approach are that there is no compulsion and it ensures that governing body members are at least somewhat motivated to serve the public
  • The main disadvantage is that it requires more administrative work to achieve a near cross-section of the community (polling, census data)

There would be fewer “second-level” bodies.  These could be filled by randomly selecting from volunteers who have served at least once on a “first-level” governing body.

Terms for both first- and second- level members could be strictly limited (maybe 1 year), and rotation of office staggered so that incumbents could pass on important knowledge and skills to new members.

Like with jury duty or maternity leave in many countries, citizens who are selected to hold public office could have their jobs held for them during their leave of public service.  Alternatively, by creating more first-level governing bodies, each body becomes more specialized, the workload spreads, and policy-making could become a part-time commitment for the relatively few Icelanders selected to hold office at any time.


Minimum age requirements and exclusions for citizens with certain types of criminal records could be put in place by the “second-level” procedural bodies.

Note: The combination of these two “suggestions” creates a democratic government structure known as Demarchy. It was first proposed by the Australian Professor John Burnheim in Is Democracy Possible? (1985).


Arguments for this approach:

This democratic structure is less vulnerable to corruption.  Any individuals or groups looking to corrupt the political process in democracies around the world are provided a fairly stable group of career politicians to target.  In this alternative approach, everyday citizens are randomly selected to govern specific functions and they are regularly swapped with new randomly selected citizens.  Each new member is an unknown entity.  They might blow the whistle on any threat or bribe attempt. Or, if they are corrupt, they cannot make any policy without the consent of all other group members, their term is short, and they cannot serve on that particular governing body again.

Since there are no voters, the cross-section approach eliminates the problems of voter ignorance, voter fatigue and media manipulation during elections.

Randomly selecting policy-makers from the population or from all volunteers means that campaigning could not improve your chances of being selected.  So…

  • No money is spent campaigning
  • Special interest groups / corporations cannot influence policy-makers (governing body members) by providing campaign finances or threatening to withhold future contributions
  • Political parties cannot manipulate voters (there would be no parties and no voters)
  • Policy-makers would be free to change their views, or view some issues conservatively and some issues liberally without being disciplined by their political party
  • Policy-makers would not owe anyone any political favors

Just like one cannot make a career as a juror, in this type of government one could not make a career as a politician.  Policy-makers could not be re-elected, so they would not be influenced by re-election or career concerns.  Instead they can focus on trying to make the right decisions, whether or not those decisions are easy or popular at the time

Once elected, there is no good way to insure politicians actually represent their constituents.  But with this structure, each selected citizen represents their own interests and, by being part of the cross-sectional sample, represent all others in the population with similar interests and views.

These governing bodies would be far more representative of the whole population than a parliament of elected officials.  Elections tend to favor higher educational and income levels, members of majority populations, and men.  Random selection does not discriminate.  If you are a part of the community, you can have an equal chance to be involved in community decisions.  Icelanders would simply take turns governing.

The representativeness of this approach gives legitimacy to the decisions of the governing bodies.  Citizens know that there is a good chance that someone on each governing body:

  • Comes from a similar background as them and has similar interests at stake
  • Has been given the time/resources to become well informed about the issue
  • Agreed to the decision by consensus and did not simply get outvoted.

Possible arguments against this approach:

This system demands too much participation by citizens who are used to the occasional trip to the ballot box.  This is true, but this type of participation would be more engaging than a trip to the ballot box and genuine political involvement can produce politically motivated citizens.  Terms could be short, and policy-making could easily be made into a part-time commitment.

Random selection does not weed out people who might make poor community decisions.  While this is true, these individuals have to convince the other 9(+) members in their functional governing body to go along with their poor policy proposals.  And elections don’t do a great job of weeding out terrible policy-makers either.

Consensus decision making cannot handle deep-seated conflict.  What fair democratic decision making process can?  At least with short terms and rules against serving twice on the same governing body, individuals with strong views on an issue might worry that if they do not reach some sort of compromise agreement, the group will…once their term is up.  Alternatively, governing bodies dealing with highly contentious issues could operate with some majority voting system.  But movements away from consensus should be considered with caution.

This type of government structure is untested.  True. The same argument was probably made against the parliamentary system back in 930 CE, but fortunately Iceland went ahead and founded the Alþingi.  And systems based on similar principles have operated with notable effectiveness:

  • In Canada, the British Columbia and Ontario governments randomly selected citizens to form Citizens’ Assemblies to investigate and make recommendations for electoral reform.
  • The jury system, particularly the American jury system, relies on randomly selecting everyday citizens to make important and often complex community decisions on both criminal and civil disputes.  It continues to prove vital to the American justice system, even though politics has historically prevented juries from being even partial cross-section samples of their communities (notoriously excluding women and African-Americans).

  • Random selection (also know as lot or sortition) was also extensively used in the first democratic state government – Ancient Athens:
  • The Assembly was open to all citizens (although just free men qualified for citizenship)
  • 500 citizens over the age of 30 were randomly selected to form the Council
  • The person who presided over the Council and the Assembly was randomly selected each day (to keep groups from influencing the agenda beforehand)
  • Most public offices were filled by randomly selecting citizens for 1-year terms (10 city magistrates, 10 market magistrates, 11 people to run jails, 5 road supervisors, etc.)
  • Elections were only used to fill the few positions where possessing special skills were seen as vital (like generals).  - Otherwise, elections were purposefully avoided.


Thanks for reading, look forward to your comments.


Further Reading:





1 See The Jury and Democracy (2010) by Gastil, Deess, Weiser, and Simmons

2 Carson, Lyn and Brian Martin. Random Selection in Politics (1999), p. 31


Störf Stjórnlagaráðs fara fram fyrir opnum tjöldum og við bjóðum öllum sem fylgja samskiptasáttmálanum að setja inn ummæli á vefinn.

Markmið umræðukerfisins er að bjóða upp á málefnalega og uppbyggjandi umræðu. Ærumeiðandi og óvægin ummæli í garð einstaklinga og þrálátar rangfærslur verða fjarlægð.

Samskiptasáttmáli Stjórnlagaráðs
  1. Við fögnum uppbyggilegri rökræðu en líðum ekki persónulegar árásir né misnotkun á umræðukerfinu.
  2. Við hvetjum til gagnrýnnar og upplýsandi umræðu forðumst gífuryrði og leggjum ekki öðrum orð í munn.
  3. Við notum einföld en skýr skilaboð og virðum skoðanir annarra.
  4. Allar athugasemdir sem endurspegla fordóma í garð einstaklings eða hópa verða fjarlægðar.
  5. Við áskiljum okkur rétt til að loka á notendur eftir tvær viðvaranir og þriðja brot á þessum sáttmála.

Þessi vefur er sameign okkar sem höfum áhuga á umræðu um málefni stjórnarskrár Íslands. Það er nýmæli að bjóða upp á opna umræðu af þessu tagi við mótun opinberra skjala og við vonumst til að þessi vefur verði skref í rétta átt í þróun umræðuhefðar á netinu.

Vinsamlega tilkynnið brot á sáttamálanum til umraeda@stjornlagarad.is.