How Mutual Aid Groups Use Double-Loop Learning to Settle Member Defalcations Without Involving Police

In a 1977 issue of the Harvard Business Review, organizational theorist Chris Argyris measured how institutions fail by observing a hypothetical thermostat. He noted that a thermostat clicking on at 69 degrees is merely correcting an error—but a thermostat asking why it was set to 69 degrees is redesigning the system.[1]

Today, grassroots mutual aid networks use that exact 1977 framework to solve their most painful modern problem: internal financial theft, or defalcation, without calling the police. When a collective discovers a missing $5,000 meant for emergency rent relief, the actionable takeaway is not just to change the bank passwords.[1][5]

For conventional nonprofits, defalcation triggers an immediate call to law enforcement. But for mutual aid networks, which expanded massively during the 2020 pandemic and often operate on abolitionist principles, inviting the state into their operations costs too much in community trust.[5]

Instead, these collectives apply Argyris’s concept of double-loop learning. By treating financial theft as a structural feedback metric rather than an isolated crime, groups resolve defalcations internally while overhauling the resource-distribution bottlenecks that allowed them to occur.[5]

Beyond Changing the Locks

When funds vanish, the instinctive response is what Argyris termed single-loop learning. In a single-loop response, a group identifies the immediate error and applies a direct, restrictive fix to prevent it from happening again.[1]

In the context of a mutual aid defalcation, a single-loop correction is purely administrative. The group removes the offending member, revokes ledger access, and requires two signatures on every future check to plug the immediate leak.[5]

However, single-loop learning fails to ask why the theft occurred in the first place. It assumes the collective's underlying structure is sound, the disbursement rules are fair, and the problem was entirely the result of one bad actor.[1][5]

For mutual aid groups, this assumption is often fatal to their mission. If the collective's own structures create artificial scarcity or desperation, simply changing the locks guarantees the pressure will eventually break a different window.[5]

The Double-Loop Framework

Double-loop learning requires an organization to step back from the immediate error and interrogate the governing variables that produced it. The collective must ask whether their own policies, communication channels, or workload distributions incentivized the theft.[1]

“A thermostat that automatically turns on the heat whenever the temperature in a room drops below 69°F is a good example of single-loop learning,” Argyris wrote. A double-loop thermostat, he explained, would ask why it was set to 69 degrees to begin with.[1]

Often, this inquiry reveals that the member who took the money was experiencing an acute crisis that the group's formal processes were too slow to address. The theft becomes a harsh diagnostic tool, exposing a gap between the collective's stated values and its actual practices.[5]

If a member stole to pay rent because the official grant approval process took 14 days, the 14-day wait is the root failure. The actionable solution is to redesign the disbursement timeline, not just punish the individual.[5]

Accountability Without the State

By shifting the focus from individual punishment to structural redesign, double-loop learning aligns closely with the principles of transformative justice. Organizations like INCITE! define this as a community-led process that bypasses the criminal legal system entirely.[2][3]

When a mutual aid group applies this framework, the process begins with securing the remaining resources. The collective then initiates a structured accountability process with the member who misappropriated the funds to understand their motivations.[2][5]

This involves acknowledging the impact of the theft and developing a realistic restitution plan. Restitution in these settings rarely looks like a standard repayment schedule dictated by a court or a collection agency.[2]

It might involve the member contributing 40 hours of administrative labor to the collective, returning the funds in $50 increments over two years, or participating in mediated dialogues. Crucially, the member is not permanently exiled if they engage honestly.[2][5]

Structural Transformation

While the accountability process addresses the individual, the double-loop learning process redesigns the collective. The group uses the insights gained from the defalcation to overhaul its resource-distribution models and internal governance.[5]

If the theft occurred because a single treasurer was overwhelmed and isolated, the group decentralizes its financial operations. This typically involves rotating finance committees of three people, transparent public ledgers, or regular collective audits.[5]

If the member stole because they could not access emergency funds through official channels, the collective streamlines its disbursement process. A group might reduce its approval window from 14 days to 48 hours to ensure members can meet basic needs.[5]

These structural changes transform the defalcation from a catastrophic betrayal into a catalyst for organizational maturity. The collective emerges more resilient, with systems that are better aligned with its core mission of mutual support.[5]

The Cost of Internal Resolution

Despite its utility, applying double-loop learning to financial theft requires immense emotional labor. Facilitating an accountability process while managing the trauma of betrayal exhausts volunteer collectives that are already stretched thin.[4][6]

Recent studies on transformative justice interventions, such as a 2024 Nuffield Foundation project involving 15 community workshops, highlight the difficulty of sustaining these models. The process relies entirely on the willingness of the offending member to participate.[4]

If the individual denies the theft, refuses to engage in restitution, or simply abandons the group, the transformative justice framework hits a hard limit. Collectives are left with the financial loss and no formal recourse.[2][4]

In these cases, groups must complete the double-loop learning process without the closure of restitution, focusing solely on internal structural reforms. Yet, by refusing to outsource their conflicts to the state, these networks actively build the infrastructure for community-led resolution.[5][6]