When Child Protection Confirms Its Own Assumptions

Child protection services have become highly skilled at examining families. We need to become equally practised at examining the organizational conditions through which we understand those families, make decisions and respond when things go wrong. My concern is that CPS has become too insular in how it defines problems and recognizes solutions. This is not a claim that every organization operates identically, or that committed practitioners and leaders are absent. It is an argument about what happens when a system relies heavily on its own assumptions, procedures, and measures to determine whether it is working, while experiences that might challenge those assumptions struggle to influence its decisions. An organization can hold meetings, complete audits, and commission reviews while repeatedly returning to the same explanations: families need to engage more, practitioners need to follow procedures more carefully, and supervisors need to monitor compliance more closely. Each explanation directs attention towards changing someone’s behaviour, while the conditions shaping that behaviour remain largely intact.

The local rationality principle offers a useful starting point for understanding this pattern. In human factors research, it means examining how people’s decisions make sense within the information, goals, pressures and organizational expectations surrounding them at the time. Dekker (2002) argues that understanding error requires reconstructing that context rather than treating a disappointing outcome as proof that the earlier decision was obviously unreasonable. For child protection, this raises an uncomfortable question: what does our organization make reasonable for people to do?

A practitioner may value careful inquiry with a child and family but know that overdue recording will attract immediate scrutiny. A supervisor may believe in reflective discussion while being expected to account primarily for deadlines and outstanding tasks. A manager may encourage openness but respond defensively when a concern calls an established practice into question. Under those conditions, prioritizing documentation, shortening reflection or withholding uncertainty can become understandable adaptations. This is my application of local rationality to organizational life in CPS: people learn what receives attention, what attracts criticism and what they must do to get through the working day.

Understanding that adaptation does not remove responsibility; in fact, it gives accountability somewhere more useful to go. Alongside examining an individual decision, we must examine why the organization made that decision more likely. The danger becomes greater when these adaptations begin to confirm the assumptions that produced them. Merton’s (1948) account of the self-fulfilling prophecy describes how an initially mistaken definition of a situation can prompt behaviour that helps bring the expected result into existence. Applied to CPS, this offers a way of examining whether our expectations and interventions sometimes contribute to the difficulties we later present as evidence that those expectations were correct. It is an explanatory lens, not proof that every unsuccessful intervention follows this pattern.

Consider a hypothetical organization that assumes family networks cannot sustain safeguarding without extensive professional control. Professionals consequently develop the arrangements, determine responsibilities, and bring relatives into the process after significant decisions have already been made. Network members have limited opportunity to influence what is expected, explain what they can realistically provide or identify the support they need. When the arrangements struggle, the organization concludes that families cannot be relied upon. That conclusion may overlook a different explanation: the organization expected dependable participation without creating sufficient opportunities for ownership, preparation, support and testing. It may have helped produce the fragility it then attributes entirely to the family.

This does not establish that a particular network is capable or appropriate because capability, willingness and reliability still need to be demonstrated. It does mean that professional conclusions should examine the contribution of organizational practice, including whether promised services and support were actually available. The same pattern can appear within the workforce. Leaders assume staff will not raise concerns responsibly, so they tightly control information. Practitioners learn that challenging an established position carries a personal cost. Fewer concerns reach leadership, and the absence of challenge becomes reassurance that the organization is functioning well. In both examples, the system becomes harder to correct because it interprets the consequences of its own arrangements as confirmation of its original beliefs.

Insularity does not necessarily mean an absence of information; it can also mean that information is admitted only on terms that protect the prevailing explanation. A parent’s account becomes “resistance.” A practitioner’s concern becomes a “training need.” A failed service contribution becomes a family’s “lack of engagement.” The words may sound familiar and professionally acceptable, yet they can close inquiry before the relevant conditions have been understood. This is why changing the vocabulary of practice is insufficient. An organization can adopt the language of participation, reflection and learning while preserving the same distribution of authority, attention and resources.

Munro’s review of child protection in England identified how defensive responses, procedural demands and performance measures could shift attention away from the quality of help children received. Her analysis located these difficulties in how the system operated, including how it responded to uncertainty and serious incidents (Munro, 2011). My concern is that this warning remains useful wherever organizations respond to recurring difficulties mainly by adding requirements. A new form can be completed. A new instruction can be circulated. Neither establishes that practitioners have more time, families have better access to help, nor do leaders receive a more accurate account of everyday work.

Becoming a learning organization requires arrangements through which experience can change decisions. Senge (2006) places systems thinking, examination of mental models and shared learning at the centre of that work. The implication for CPS is that we need to examine the assumptions behind our response, as well as whether the response was implemented. Psychological safety matters within those arrangements. Edmondson’s (1999) research found an association between team psychological safety and learning behaviour in organizational teams. Although that study was not conducted in child protection, it provides relevant support for examining whether people can raise uncertainty, acknowledge mistakes and ask for help without expecting interpersonal punishment. An invitation to speak is therefore only a beginning. What happens after someone speaks determines whether the invitation becomes credible.

This is the purpose behind Aligning Together, our organizational framework at ATA Consultancy. It helps organizations establish the systems, leadership practices and everyday routines needed to notice early errors and emerging difficulties, learn from experience and make timely corrections. Its purpose includes learning from successful adaptations and understanding what makes good work possible. The framework can be used independently, alongside our other ATA frameworks or within an organization using another practice model. Its organizational learning method can be carried through facilitated conversations, working tools and approved records, with the application supporting its use. Responsibility for inquiry, decisions, and organizational change remains with people.

Aligning Together examines seven connected domains:

• Purpose, governance and accountability.

• Leadership, culture and psychological safety.

• Workforce capability, capacity and support.

• Operational design, coordination and decision authority.

• Information, weak signals and operational awareness.

• Organizational learning and system improvement.

• Resilience, continuity and sustained alignment.

These domains help locate why an expected practice may be difficult to deliver. A recurring missed supervision session, for example, may involve staffing capacity, competing operational demands, an unapproved resource decision and a reporting arrangement that treats each cancellation as an isolated diary problem. Asking the practitioner to manage their time better would leave most of those conditions untouched.

The framework also draws on the five high reliability principles described by Weick and Sutcliffe (2001): preoccupation with failure, reluctance to simplify interpretations, sensitivity to operations, commitment to resilience and deference to expertise. These principles encourage attention to small departures, competing explanations, actual working conditions, recovery capacity and relevant knowledge regardless of organizational rank. Their use in other sectors informs the framework; it does not establish its effectiveness in child protection (UC Davis PSNet Editorial Team, 2024).

Nine adapted audits help organizations begin this inquiry by asking questions about the conditions supporting practice, retaining evidence and differing perspectives alongside judgments. They are not a certification exercise, and their ratings do not establish that an organization is reliable or that a child is safeguarded. Completing every audit is not a prerequisite for taking useful action. The continuing work takes place through a process called 4A, defined by Analyze, Action, Assess and Adjust.

Analyze investigates what was noticed and the conditions contributing to it. People examine workload, assumptions, cultural expectations, coordination and decision authority. They ask what would challenge the preferred explanation.

Action turns that understanding into an authorized and resourced change. It identifies what will happen differently, who is responsible, who can approve the necessary resources and what difference is expected.

Assess examines whether the change was delivered and whether it helped. It includes the experience of those affected, contradictory evidence and unintended consequences.

Adjust uses that evidence to retain, revise, or stop the change; explain the decision to contributors; and establish what is needed to sustain any improvement.

Listening needs to lead to changes that children, families, communities, and staff can hear and see based on their contributions. Leaders remain responsible for organizational repair; participation must not become a way of transferring unresolved service responsibilities to families or already stretched practitioners. Aligning Together must be open to the same scrutiny, where its research-informed foundations do not establish that the complete framework improves outcomes. Its usefulness, burden, equity and effects need to be examined locally. If its routines create more recording, discourage candour or become another way of seeking reassuring scores, those routines need to change.

CPS will remain vulnerable to a self-fulfilling prophecy for as long as it examines families and practitioners more readily than it examines the conditions it creates around them. A system that expects compliance, rewards reassurance and discounts contradictory experience can keep finding reasons to do more of what it already does. Local rationality helps us understand how people adapt to that system. Organizational learning gives us a disciplined way to change it. The practical test is whether an early concern can reach someone with authority, lead to a reasoned and resourced correction, and be followed by evidence that the correction helped. That is the capacity Aligning Together is intended to build.

References

Dekker, S. W. A. (2002). Reconstructing human contributions to accidents: The new view on error and performance. Journal of Safety Research, 33(3), 371–385.

Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383.

Merton, R. K. (1948). The self-fulfilling prophecy. The Antioch Review, 8(2), 193–210.

Munro, E. (2011). The Munro review of child protection: Final report—A child-centred system. Department for Education.

Senge, P. M. (2006). The fifth discipline: The art and practice of the learning organization (Rev. ed.). Currency.

UC Davis PSNet Editorial Team. (2024, September 15). High reliability. Agency for Healthcare Research and Quality, Patient Safety Network.

Weick, K. E., & Sutcliffe, K. M. (2001). Managing the unexpected: Assuring high performance in an age of complexity. Jossey-Bass.