Explorations into community governance — observing primates in the wild

Zarinah Agnew
Jun 6, 2016 · 11 min read

Experimenting with new ways of being, reclaiming the tribe and testing which new cultures support different types of behaviour — these are the kinds of things that get us leaping out of bed in the morning at the Embassy Network. We often ask ourselves, which are the kinds of things that a group of motivated humans can do by living together that would otherwise be extremely difficult? Residential social experimentation in the wild is one. In 2013, as part of this larger goal, the Embassy SF, our flagship location in San Francisco embarked on an eight month long series of experiments into governance and decision making.

Somewhat unexpectedly to me at least, all 14 members of our intentional community, opted to take part in a live-in experiment whereby the community operated under four different governance structures. External individuals were brought in to implement a range of measures to assess Satisfaction, Productivity, House Function and Hours Contributed to the community. We had our phones and laptops tracked and weekly meetings recorded for language. What ensued was an amazing, difficult, illuminating and confusing series of experiences. We learnt hard lessons fast, and others took longer to land. We learnt a great deal about the difficulties of social science in general. We blossomed as a community, we formed new bonds and in many ways became closer, we fell apart and then we came back for more.

This is our tale.

Wait, but why?

Community governance refers to the processing for making all the decisions and plans that affect and govern life in the community. For community governance to be effective, it must be about more than simply process, it must also be about getting things done in the community. Why do we care about such trouble and strife? For many of us, we are excited not only by observing human behavior but also iterating on the ways of running communities, or indeed societies. If we don’t ourselves try different ways of operating, how can we know what to improve on, what works and what doesn’t? What type and size of group do different systems work best for?

One weekend in Jan 2014, the community headed up to the Russian River to plan out the year ahead. Over the next 48 hours we plotted together about various things, the governance experiments being a major one. After much deliberation, excitement and some healthy skepticism, the goals were settled on.

Goals of the experiment

The goals were set out as, firstly, to pilot a range of governance structures for a community house, and for other communities to learn from the documentation of this process. Second, to learn about the very process of social experimentation. Finally, to push the boundaries of those within the community, in order to explore the workspace of different governance structures and learn something new about each one.

Between cold beers and dips into the ice cold water of the Russian River, a spreadsheet detailing exactly what each governance structure could rule over was agreed upon. We agreed that each governance structure could change anything that took place in common areas of the house. It could ask for 2 hours a week of our time. It had dominion over house operations, utilities, all behaviour in the house (uniforms — yes!, language change — yes!). We pooled a small amount of money that each system could decide what to do with. Exceptions, were that no governance structure could remove a resident from the house or bring in a new one. These important decisions would remain subject to consensus. We set a budget and crucially, decide to recruit a political scientist to come and oversee the project. The governance experiment was born.

Each governance structure had dominion over the following: -

  • Use of the allotted monthly budget
  • Alcohol and food purchases
  • Utilities (cleaner, gardener, odd home fixing jobs)
  • Use of language in the house
  • Allocation of all shared space in the house
  • Delegation of small chores

Our dependent variables would be:

1. Personal happiness of citizens

2. Productivity of citizens

3. Community strengthening

Making the magic happen

Two months later, we had a volunteer from Stanford, a recent escapee from the political science department who agreed to come and live with us for the duration. We decided collectively on our first governance structure. The community had run previously on a range of different decision making styles that had not yet been formalised. For our first month we formalised our version of doocracy that centered around enabling those who wish to take action, space and ease to do so.

We met weekly for feedback sessions and to collect data on happiness, productivity and community participation. Each of these session was recorded for posterity and hopefully some language processing. To begin with, we all had our laptops track our behaviour, we used the Reporter app to assign random questions throughout the day. We had daily and weekly surveys to complete that probed various aspects of our daily lives. Very quickly, however, we learnt that the biggest difficulty around social science is adherence to data collection strategies. People failed to use the app and to install the software for tracking. Daily surveys were lacking and we soon realised that the weekly surveys would be all the data that we had.

None of these issues are unsolvable, but going forward we would advise taking appropriate measure to ensure that your experiment doesn’t fail simply due to inadequate data collection strategies.

After 5 weeks, we met to discuss our next structure. We decided to try out a system that some of us had concocted over many years as a thought experiment for lunar societies. We called it the Wiki-ocracy, and under this structure, residents are all invited to collaborate in the creation of proposed policies. Every individual’s voice is heard. A final decision is made by a randomly selected, hopefully statistically representative, jury drawn from the residents.

What we found was surprising — In contrast to the doocracy, this system supported all voices being heard — people reported feeling heard even if their option was not chosen by the jury, and felt more satisfied with outcomes, even if the outcome was not representative of their stance.

Unfortunately, after a streamlined doocracy that required very little input or work from anyone, this system felt like a really high overhead. Suddenly we went from no communication required, to high levels of required communication. The level of emails rose dramatically and for this system to work, you really need buy in from participants. After a few weeks this increased overhead was associated with voter apathy.

Anecdotally, we noticed that participation increased as these experiments started. Residents were excited to be generally collaborating on something together. Our weekly ‘meetings’ were fun, we enjoyed getting together and letting the conversation buzz. We had gone from no house meetings in order to avoid bureaucracy to house meetings that felt exciting and fun. Under this burgeoning project, community and connection thrived!

Three months into the experiment, and we met once again, to decide on our third governance style. Before starting the experiment, we had penciled a list of potential systems, including philosopher kings, and an ignorant dictator. One we had felt certain that we’d try was a dictatorship. Our community is largely based on principles of decentralisation (link to examples), and we were keen to test our own ideas to see whether it might in fact work better under a centralised power. We had two clear candidates for this position, self-selected of course (is there any other way for a dictator to rise?). At the last minute, as new suggestions were floated, our dictatorship was outvoted in favour of a corporate structure.

This was a last minute turn around, a surprise to many of us. It’s an interesting sensation to go to sleep one night under a representative jury system and to wake the next morning with the job title of Executive of House Operations for a group of 14 strong willed individuals, 8 of which are now your ‘employees’.

The job of the CEO was to immediately appoint executives whose job it was to maximize happiness, productivity, and community within their domains. Three executive roles were created — Executive of House Operations’, ‘Executive of Guest Operations’, ‘Executive of Events’, as well as a Chief Financial Officer. The CEO also had to appoint a board which comprised two members, whose role was to ensure that the CEO was operating in the best interests of the company. Of the $200 that we pledged to the general project each month, it was agreed that under this system, this money would be returned in full to each employee, unless they were fired. With cackles of glee, it was decided that executives were to have full control over their employees.

As you can imagine, chaos ensued…

Before you could say ‘hierarchy should be voluntary!’, our language started to change. In our emails to each other we were referring to each other by our job titles instead of using names.

[relating to a complaint made by an executive and non particiation from employees] -“I second this person’s frustration. Lack of employee response is really disheartening and unfair.” — somewhat disturbing examples of language change in email communication

As the person appointed Executive of House Operations, I was quickly confronted on my management style by others in the house who had real experience of management.

“As executive I spend my time just delegating and failing to delegate, then trying to figure out outsourcing, instead of actually getting things done.”

They were right, I was doing a terrible job. I updated my approach. Nonetheless, within a couple of weeks of this structure, I found myself one day, sitting with one of the other Executives (Executive of Events), sorely lamenting of the lack of engagement from our employees. We were doing all the work and ‘for what!?’, we exclaimed. Now, I consider myself to be someone who believes in equality, distributed roles and voluntary contribution. I hold many anarchist principles dear to my heart. Nonetheless, after two weeks operating within this imposed hierarchy, I found myself suggesting to the other executives, that we needed some way of rewarding ourselves for our work, and showing our employees that our roles were special and needed some kind of acknowledgement. ‘What about an executive’s-only champagne bar?’, I proclaimed, with genuine excitement. ‘Yes, in a common area so that others can see that we get a reward’, chimed in one of the others. We stopped for a second to process what we had just suggested; eyes bulged around the room as we realised in that moment, we were manifesting everything we hated about the world — segregation, inequality, punishment, absence of mobility through social roles. What an instrumental sensation this was for me, to realise how quickly my behaviour could change. How strongly influential imposed hierarchy could be on my personal attitude.

Suffice to say, this structure damaged social connections, the way we communicated and so on. The house infrastructure didn’t fall apart — food got ordered, beds were slept in - but to some degree the community did. This was a hard time for us and it took us a while to recover. We wrapped up this system eager to move on. As a direct response to try and repair some of the social damage that our corporate structure wreaked on the tribe, we designed a system that aimed to maximise appreciation of each other..

This system was loosely based on a Holocracy, a few tweaks and twists. It prioritized empathy and community building with the emphasis on inspired rather than equal contribution. To kick things off, we all volunteered to create teams that would run different parts of the house and anyone could sign up for those teams. We put into place a social reinforcement strategy whereby for each 15 minutes that anyone put towards the community, the house got paid (from a dedicated fund) $15. This way if we didn’t announce how we were contributing, we would be losing the house money. We created a GroupMe channel for this community oriented bragging, and we called wait for it.. BILBO BRAGGINS.

Our current Bilbo Braggins channel- now featuring sideways brags also

This was wonderful, over the next few weeks, we remembered how each of us were contributing in different ways. It rendered visible all the silent contributions that people were making. Those who quietly took the bins out, or dealt with insurance admin were suddenly brought into the limelight and celebrated for their contributions. We bragged fully and with open hearts, know that bragging was essential if we were to earn this money back. The house ran nicely, and we valued each other, harmony was gently restored.

At the end of it all

For reasons external to the experiment, we had to cut our explorations short, and the next governance structure was cancelled. I won’t ruin the surprise here, as we will be embarking on our next set of experiments in the next 6 months and we will be sure to include this in round II of embassy does governance. Suffice to say, it will be a good one.

We had a wonderful time collaborating on a large project like this together. It made us realise that we could do interesting things as a group, that we couldn’t do if we didn’t live together, if we didn’t trust each other, if we didn’t as a group value social experimentation. Our quantified findings are listed elsewhere, but in short all measured were significantly modulated by governance structure, and metrics were not always correlated with each other. Overall, doocracy reigned down on high, as the structure that maximised all metrics. We returned back to doocracy at the end of it all, but we kept aspects of other systems in place. We still use parts of the wikiocracy, and Bilbo Braggins remains a firm fixture.

Our general learnings and questions:

  1. There appears to be some generic benefit of change: each change in governance system seemed to be accompanied by a burst of increased enthusiasm and participation. Is there something inherently motivating and/or empowering about change? If so, is it worth the cost of change?
  2. One of the things that these experiments did for us, was continuously shift and redistribute the various hierarchies that tend to form in groups. This was helpful in pointing out
  3. The benefit of being aware of what each person may be doing behind the scenes. The benefit of honest and clear communication
  4. Despite considerable levels of buy-in and commitment to this experiment from all individual residents/citizens (remember each housemate pledged $200 a month to put towards this), we experienced much difficulty in active data collection. We experimented with various phone apps that collected data from us at random times of the day, we tried daily, weekly and monthly surveys, we used software on our laptops that monitored our productivity. Nonetheless we found it hard to maintain even these forms of data collection. So what’s the solution? It may be that we need to develop better passive measures, better technology.

Our measures (in case you’re interested..)

Download and install RescueTime on laptops, smartphone, anything you want your productivity measured on.

Download Reporter app — this is a truly wonderful app for tracking anything. It allows you to change the types of questions/prompts, the timing, the types of answers and so on.

Change Alerts per day (at least 1)

Included default questions “How did you sleep?”, “Where are you?” and “Who are you with?”


1. Did you have sex yesterday? Yes/No

2. How productive were you today? Multi-choice : Options: 1- Very unproductive, 2, 3, 4, 5, 6, 7 — Very productive

3. Did you exercise today? Yes/No

4. Did you interact with a resident today? Multi-choice- No interaction, Light conversation, deep conversation or support, received useful information, imparted useful information, collaborated on work/project

5. Did you interact with a guest today? — same as 4

6. What did you do for the house today? Note

7. How happy do you feel? Multi-choice: Options: 1 — very unhappy, 2, 3, 4, 5, 6, 7 — very happy

8. If you are with people, how much are you interacting with them? Multi-choice: N/A, 0 — no direct interaction, 2 — some interaction, 3 — fully engaged/collaborating

Summary : 2 questions upon waking, 4 at some point during the day, and 5 at night.

Everyone was free to add others to the list.

Zarinah Agnew

Written by

got to get it out somewhere