Introduction

Anyone notice the heat wave this year (2026) in Europe? We are still failing to deal with global warming even though we have known about the problem for over a hundred years and the science has been clear on the matter since the 1980s. It could be argued, our current socioeconomic system is starting to fail in a number of other ways too. From inequality to incompetent politicians failing to solve problems.

Way back in 2011, we (EOS) proposed in the Design a potential solution; an organisational structure for a future sustainable, moneyless, socioeconomic system based around holons. This is something I had been working on since the 1990s.

Imagine a team that is free to choose how it works and what it works on, yet still works together with other teams. You are imagining holons at work. Holons are autonomous, task orientated, groups or entities that work together; like Lego bricks that can be put together to build something bigger. But, due to their autonomous nature, we didn’t specify in the Design much detail on how holons would work. The idea was that the holons would work that out themselves. However, since then I have been working on the idea of holons in another context. There has also been some interesting developments in this direction. Scientific research, for one, has started to show that this type of approach can offer a viable solution to complex problems. There have also been some interesting real world examples of implementation. From that, I have been rethinking the idea of holons for a sustainable society and how they would work. I have also had questions asking how would things work in a sustainable holonic moneyless society. So, I thought I’d look into holons in more detail and give more of a blueprint describing how they would work. For that, I am going to start with some background information covering some basic concepts in this article. I will start with an overview of the concepts behind the idea, such as what is a holon, the ideas of mission command and leader – leader, as well as a very quick look at the science and some real world examples. In future articles, I will go into more details about an envisioned holonic structure and functionality as well as some ideas for implementation.

If we actually want a sustainable society that balances our needs with those of the planet, then we really need to look at how we do things. Our current system is not sustainable and we need to change. Change to a system that is robust, adaptable, and human centric. Holons are the building blocks for such a society.

Background

Holons

Holons exist to achieve a given goal. They can be temporary, with a end point in time, or they can be more permanent, with no specified end point. To achieve its task, a holon might be divided into a number of sub holons, which in turn, could also be divided into sub holons. The holons itself could also be part of another holon. So we have the idea of holons within holons within holons to form a holonic system. Arthur Koestler coined the word “holon” in his book “Ghost in the Machine” [ghost] in 1967. The word is from the Greek and means “part-whole” and captures this idea of holons with in holons. This idea comes from observations of how nature is organised. The classic example given are cells which then form organs which then form organisms. But another example would be ants. An ant is a holon with a given task. It works together with other ants to form a colony, which is another holon. The colony is itself a holon within an ecosystem, which can be seen as the top most holon. No where in this system is there any central, authoritarian, control. Not even the queen ant gives orders. The idea can be extended to the human world and there are companies that organise in this distributed fashion. Without managers and hierarchies, nature organises effectively to handle complexity.

Mission command; goals and tasks

Mission command [mc] is a concept I have borrowed from the military. In mission command, a commanding officer doesn’t give orders. Instead, the commander would give the goal to be achieved and the commanders intent with the goal. Sub commanders then are free to achieve that goal within the given intent. I’m borrowing the concept of mission command as setting goals is extremely important within holonic systems. The goal defines what the holon is trying to achieve. And as we have a system of holons within holons, there is a cascade of goals as goals are passed down from one holon to its sub holons. One holon could have the goal to provide 100% renewable energy to a given area. The goal could then cascade down to one of its sub holons as 50% wind power and then the goal could cascade down to its sub holon as a goal to build a wind farm of 10 wind turbines. Every holon must have a goal that is compatible with the overall goal of the holonic system. So, our area goal would need to be compatible with the overall goal of “the highest standard of living for the longest time possible”. This is what keeps the whole system together; everything is working towards the same goal but in its own way. “In its own way” because each holon is autonomous. So the goal states what has to be achieved but not how to achieve it.

But to achieve a goal, each holon will have a task to do. The task is why that holon exists and achieving the task fulfils the intent of the above holon in creating the holon. If a task is too complex, it can be broken down into sub tasks. Each sub task then becomes a task for one or more other holons. Those holons will then have there own goals. So, there is self-symmetry in the system like a fractal.

This whole system requires trust. As each holon is autonomous, it achieves its task in whatever way is perceived to work best. That means that the holon above must trust that the sub holons will achieve their task and the sub holon must understand the intent and goals of the above holon so both holons have a common understanding. The sub holon must also have the capability to achieve the goal. This also allows for greater diversity. Different people, different cultures, could achieve a given task in a variety of different ways. So long as the task is achieved, the method is not of interest to the above holon. Each holon is free to act as it sees fit but there are some limits. It can act and take initiative but within the bounds of the overall goal and the intent of the above holon. In other words, it can do as it wishes so long as it stays on track.

In creating goals and tasks, each holon will have a understanding of what it is to achieve and that understanding will be common with other holons that work together to achieve the task of the above holon. So there is a shared understanding [team] which means that holons collaborate with each other to achieve the overall goal.

Information

For this all to work, there also needs information to be shared. Information sharing between holons helps each holon make the right decision. Information could be shared via meetings, emails, or common displays, for example. It also helps with the shared understanding between holons. A holon will need to know if its actions will negatively effect other holons, will its actions make other holons obsolete, lessons learnt might help other holons, and feedback can help a holon improve itself. Also, one holon may make an innovation that could be useful to another holon. Information is the glue that holds the system together, helping to prevent duplication, conflicts, and to avoid failures. To achieve information flow, there needs to be lines of communications between holons. Lines of communication could be a simple as having a representative of one holon sit in on meetings of another holon. Or it could be a common web page that provides the latest news for all to read like a modern newspaper. Or a knowledge base like Gitlab.

Distributed

As the system is composed of autonomous holons within holons, there is no central command. So the system is a distributed system. Such systems can exhibit emergent behaviour that are not explicit in the rules governing the system. An ant colony is such an example. It results from the interactions of numerous ants rather than centrally planned. It results for the cooperation within the system so that the sum of the parts are greater than the whole. Distributed systems also have other properties. They are scalable. New holons can be added without breaking the system. Like adding a new computer to the Internet. So, if we get the system working on a small scale we can scale the system to a much larger size. It is also robust. An ant dies but the colony survives. We lose cells in our bodies but we carry on. As different holons can work different ways new ways of working can develop. If they fail, the whole system doesn’t fail. If they succeed, the lessens learnt can be passed on to other holons. This difference and distributed nature means the system is flexible and adaptable. As things change, the system can adjust by experimenting, creating new holons, or removing old holons as needed.

Leader – Leader

The focus on building a holonic organisation is the tasks to be achieved and we organise around those tasks, not around people as we traditionally organise. However, the bottom level of our proposed holonic organisation are individuals. Each individual is an autonomous person with skills and expertise. They work on a given task to achieve a given goal and intent. “Leader – Leader” is a model that was put forward by David Marquet in his book “Turn the Ship Around” [ship] and the idea defines how individuals should behave within a distributed or holonic system. The focus is on achieving excellence, which means you can tolerate some degree of error as that would allow growth and learning.

For this to work, there must be, first, a common understanding of what is to be achieved, what Marquet calls “clarity”. Everyone understands the goal and what the intent is. Individuals are then given control of their task so they can act. Someone who is an expert in a given area doesn’t need to be told how to do things. They can make the decisions. They own their own projects. They then build the skills to achieve mastery. If a person has the skill to do a task and an understanding of the task, they can be trusted to achieve that task. They become their own leader.

Holacracy

The book “Holacracy” by Brian Robertson [hol] gives a more detailed structure to how a holonic system work. This provides more details than we originally proposed in the Design. In the book, holons are referred to as circles. Each circle has a task but within each circle there are roles, which define the jobs to be done but not who does what job. One person can have many roles in multiple holons. How each holon connects with other holons or forms a holonic system can change over time. Roles within each holon can also change. “Holacracy” has a built in system for handling these changes in the form of governance meetings. Governance meetings looks at the structure and any needs for change can be proposed and acted on. They define new roles, add or subtract holons, and define new policies as needed. There are also internal technical meetings to focus on tasks, which could also mean changes have to be proposed to the governance meeting in reaction to real world events. Technical meetings define actions and metrics. Strategy in a holacracy is more like rules of thumb and act as an overall guiding principle for making decisions. Another concept in Holacracy that I found interesting is the idea of “tension”. A tension is either a problem or an opportunity. In Holacracy, they are both handled the same way. Whether it is a problem or an opportunity, there is a difference between where we are today and where we want to be tomorrow. Another important aspect to Holacracy is the documentation. What holons exists and who has what role can easily be found in the documentation. There is also a constitution that outlines how the system works so everyone knows what to expect and how the system works.

Problems

No system is ever perfect, so, in this section, I want to look at some of the problems associated with holonic systems.

Tragedy of the commons

This is where a common resource gets over used such as overfishing in the North Sea. In a distributed system, it is possible that too many holons end up using a resource and that resource gets over used or destroyed. In our future holonic system we do have a resource management system called Energy Accounting. Through Energy Accounting, resources are tracked and monitored in real time and there are mechanisms that control the use of resources to prevent a tragedy happening. There are holons with the task of managing the resource allocation system.

Loss of direction

The system works if all the holons are working towards the same goal. Without a common understanding, holons can drift away from the overall goal. The system can then fall apart and fail. To prevent this, goals cascade down through the system but each holon has the task of assuring that its sub holons have goals that do align with the overall goal. Governance meetings can be used to address any drift.

Coordination problems

It is possible that two or more holons end up doing the same job or tasks are so complex that one holon has too many sub holons to work with. This is where information sharing comes in. Knowledge bases could be used to share experiences and what has already been done or learnt. Current news update informs holons what others are doing. But there also links between holons that allow information flow so that holons know what other holons are doing and what tasks they are working on. Also, we can take the ideas of governance meetings and tensions to identify the problem and restructure the holonic system as needed.

“Not what we normally do”

“Holonic organisation”, “task focused”, and “tensions” are not the way most people experience organisations. Today, most organisations are hierarchical and people have less autonomy than they would do in a holonic organisation. Some people also enjoy the thrill of being in control of others. This problem comes down to training so people understand how things work but there should also be monitoring so that the tension can be identified and corrected at a governance meetings.

“Do whatever we like”

Sometimes the idea of autonomy can be interpreted as “do whatever we like” as if there were no rules. People can end up abusing the freedom. If that was the case, then the system could fragment and chaos would reign. In a holonic system, that is not how things work. The idea of mission command and intent sets directions and bounds on the autonomy to keep the whole system together and working in the same direction. Actions and behaviours must align with the goals and the intent so we have clear boundaries to autonomy. Documentation, information flow, and governance meetings helps to keep the whole process transparent and to allow holons to self-correct if any deviation from the goal or intent is detected. On an individual level, the leader – leader model helps to guide behaviour. So we have freedom with limits and autonomy without the anarchy.

The Science Behind the Idea

Imagine a workplace with no managers? Could that work? How would that work? There is a strong body of scientific evidence from many decades of study across multiple domains that show that it is actually possible to have such a workplace. Not only possible but they show that such a workplace can outperform a traditional work place with managers. In fact, one can argue that the evidence overwhelmingly show distributed organisations can outperform traditional hierarchical approaches in complex environments. These studies are then backed up by real world examples. In this section, I will just touch, very briefly, on just a small sample of the research and then give a few real world examples.

Distributed systems and societies have been researched for many years. Their application domain is complex systems that are non-linear and difficult, if not impossible, to predict. Distributed multi-agent or holonic systems have been shown to work effectively and efficiently in such domains. In [HMAS], for example, holonic systems were found to emerge naturally in resource restricted domains. Their fractal nature allowed scalable organisation and autonomy combined with cooperation was found to be optimal for cooperation. Research into such systems has shown how such systems can mimic nature [swarm] and have evolutionary advantages as their dynamic nature allows them to adapt rapidly to changes [found]. Simple agents were shown to be able to solve complex problems using local rules, which demonstrates that distributed decision making in complex domains can work. As an example, ants follow two simple rules to find the shortest path to the colony. That can map on to our holons, which are focused on a local tasks but can achieve a global task like sustainability.

Trust and communications have also been shown to be vital for distributed systems to work. In [vt], distributed virtual teams were shown to out perform co-located teams. Trust and communication was an important part of the distributed teams success.

There is a body of research that shows the ideas can successfully be applied to human organisations such as shown in [smo], which reviewed 75 research papers on self managing organisations and looked at the emergence of human resource management in a distributed system without any hierarchy. Successful companies such as Morning Star, with 400 employees, Buurtzog, with 15,000 employees, and Haier, with 80,000 employees, were given as example of self managing organisations (the Teal Landscape Report actually gives over 300 such companies [teal]). However, although such systems can contribute to improved productivity within companies and a better working environment for people they can have problems if tensions are not resolved [prob], highlighting the importance of communication and conflict resolution structures.

As a specific example, I will look at SEMCO. The Brazilian company SEMCO [semco], with over 3,000 employees, shows one example of distributed systems working well. SEMCO uses participatory management and built an environment of trust which resulted in a company that self organised. Employees could choose their work hours, set their own schedule, and job titles. As a result, the company has a low turn over, the employees innovated, and the company saw an increase in profits.

Valve, with over 400 employees, is another company [Valve] that also exemplifies distributed ideas. It has a flat management structure with no formal managers formed in a holonic like organisation. Employees choose their own projects and have a high degree of autonomy. Decision making uses a consensus based system with peer feedback. The system, however does run into problems with conflict resolution which, it can be argued, results from poor goal setting. Despite some problems, Valve is still a highly effective company that encourages innovation and creativity.

My final example is the Joint Special Operations Task Force, which, in the early 2000s, was failing to keep up with Al Qaeda in Iraq. Information flowed up through the chain of command and orders flowed down, which was slow and inefficient. Information was “need to know” and sealed in in silos which hindered effective decision making. General McChrystal realised this was an organisational problem and converted the traditional, military, hierarchical command and control organisation into a decentralised, distributed, network of teams. What McChrystal calls a “team of teams” [team]. The result was an organisation that could handle the complexities of the war in Iraq far more effectively. McChrystal’s method emphasised a “shared consciousness” so everyone knew the goal and situation, communications and information sharing between teams, “empowered execution”, which gave autonomy to sub commanders, as well as a “eyes on, hands off” approach from senior commanders, who acted, as McChrystal puts it, more like gardeners, creating the right environment for the teams to operate in.

What these examples have in common is that they show distributed systems can and do work. Especially in complex and dynamic environments. Trust, information sharing, autonomy, a shared understanding, are important parts to achieving success. However, autonomy needs to be bound with clear goals and a common understanding of the tasks to be performed. For that, information needs to flow.