In the world of organizations, leaders often operate on a simple, intuitive belief: if you change how a team works, the results will change too. If a company introduces self-managing teams or encourages people to share what they know, performance should go up. This seems logical, but it hides a deep mystery. When a team improves, it is rarely clear why it improved. Was it because everyone learned more? Was it because they felt more trusted? Or was it simply that the right person happened to be in the right room at the right time? Conversely, when a new method fails, managers struggle to pinpoint the missing piece. They cannot see the invisible gears turning inside the machine. For decades, researchers have tried to map these connections, often relying on surveys and statistics that show correlations but cannot prove the underlying mechanics. The question remains: does giving a team freedom actually make them better, or does it just make the outcome harder to predict?

A new study by Franny Philos Sophia of the Elanare Institute cuts through this fog by treating an organization not as a collection of people, but as a logical system that can be tested with absolute precision. The researcher built a formal theory called RCAS, which stands for Reference–Choice Adaptive System. Instead of guessing how practices lead to results, the study breaks the process down into explicit, intermediate steps, mapping exactly how a specific action changes a team's internal state. To ensure there were no errors in this complex logic, the entire theory was translated into a computer language used for mathematical proofs. A computer program then checked every single step of the reasoning, verifying that the conclusions followed inevitably from the starting assumptions. This approach is rare in the study of human groups, where results are usually based on observation rather than proof. The goal was not to tell companies what to do, but to understand exactly what is possible and what is impossible within the logic of teamwork.

The study begins by defining what it means for a team to have genuine choice. Imagine a team given a clear goal and a set of boundaries, but allowed to decide how to reach that goal. This is different from being told exactly what to do, and it is different from having no rules at all. The researchers found that when a team operates with this kind of structured freedom, they can achieve the same high-quality results while using different methods. One person might solve a problem quickly, while another takes a more detailed approach, yet both succeed. This proves that standardizing the outcome does not require standardizing the method. However, the study also revealed a critical vulnerability: having layers of rules does not automatically mean the system works. It is possible to design a structure where the rules at the top allow something that the rules at the bottom forbid. For a system to be sound, the boundaries must be actively maintained; they do not fix themselves just because the structure exists.

One of the most significant findings concerns the difference between sharing what you know and creating something new. The study distinguishes between two types of capability growth. The first is diffusion, where knowledge spreads from a skilled person to others. The second is creation, where someone develops a skill that exceeds what anyone in the group currently possesses. The computer-verified results show a clear limit: sharing knowledge can make a team stronger in terms of volume and resilience, but it cannot raise the ceiling of what the team is capable of achieving. If a team has one expert who can solve the hardest problems, teaching everyone else to be competent will allow the team to do more work and survive the loss of a member, but it will not allow them to solve a problem that the expert could not solve alone. To break through that ceiling, someone must create new knowledge that goes beyond the current maximum. This distinction is vital because it explains why some teams that share information endlessly still hit a performance wall.

The researchers also broke down the concept of "performance" into three distinct parts, rather than treating it as a single number. First, there is the frontier, which represents the highest quality or complexity the team can currently achieve. Second, there is the volume, or how much work the team can produce in parallel. Third, there is resilience, which measures how many members the team can lose before it falls below a required standard. The study proves that knowledge sharing boosts volume and resilience. As more people learn the skills of the expert, the team can produce more work simultaneously and can withstand the departure of individuals without collapsing. However, the frontier—the peak of what is possible—only moves if someone creates something new. This means that a team can become incredibly efficient and robust without ever becoming more innovative.

Perhaps the most surprising result challenges a popular idea in modern management: that self-managing teams automatically lead to better performance. The study constructed a specific scenario, verified by the computer, where a team had all the right structural features. They had genuine choice, they covered all the necessary skills, and they updated their own rules to improve over time. Yet, in this scenario, their performance was zero. This counterexample demonstrates that having the right structure does not guarantee success. The conditions for self-management are necessary, but they are not sufficient. Whether a team actually performs well depends on factors outside the formal logic of the system, such as whether the goals were set correctly or if the environment is favorable. The theory proves that you cannot deduce success solely from the presence of autonomy.

The study also looked at how teams handle rules and constraints. It found that for a set of rules to be effective, every single rule must be necessary. If a rule can be removed without hurting the team's ability to function, it should be removed. This aligns with the idea of lean management, where unnecessary bureaucracy is stripped away. The researchers showed that if a team has a rule that adds no value, the system is not in its most efficient state. Furthermore, they found that teams can improve their rules over time, but only if they have a mechanism to keep the current rules if no better option is found. Without this "stay" condition, a team might accidentally make their rules worse by constantly changing them in search of improvement.

In the end, this work provides a clear map of the territory between practice and performance. It separates what is logically guaranteed from what is merely hoped for. It confirms that sharing knowledge makes teams more capable of doing more work and surviving setbacks, but it cannot make them more innovative unless someone creates something new. It shows that self-management is a powerful tool, but it is not a magic wand that ensures success. The study leaves the final link to the real world open for testing, acknowledging that while the logic is sound, the application depends on the messy reality of human organizations. By proving exactly where the mechanisms work and where they fail, the research gives leaders a way to diagnose their teams. If a team is struggling with volume, they need to share knowledge. If they are struggling with innovation, they need to create new capabilities. If they are struggling with resilience, they need to build redundancy. The paper does not offer a one-size-fits-all solution, but it offers something more valuable: a precise understanding of the gears that turn the machine of an organization.