Same Room, Different Maps
On what happens when good people doing good jobs produce outcomes nobody intended
I want to start with a hearing that went well.
I was observing a domestic violence problem solving court — a specialized docket where families work through DV-related cases with a consistent team of providers, advocates, and a judge who has built relationships with everyone in the room over years of shared work. The GAL was present. Two providers were attending virtually. A victim’s advocate sat nearby. The DCS caseworker was there. The judge ran the hearing.
What struck me was the quality of the room.
Not the outcomes specifically — though those were good, a client moving toward program graduation, the team celebrating their progress together. What struck me was the ease of the translation. Each person spoke from inside their own institutional role — the provider offered clinical observations, the GAL offered her recommendation, the judge elicited and synthesized — and each of those contributions landed. Nobody had to fight for their perspective to be heard. Nobody’s script crowded out anyone else’s. The team moved.
One of the advocates said, about the client: they’ve gone leaps and bounds. My vote is hell yes they’re going to phase three.
The judge smiled. The room agreed. They scheduled the phase advancement.
I have been thinking about that room ever since — not because it was exceptional, but because it clarified something about what makes the other rooms harder.
—
The people in that hearing were not unusually gifted or specially trained in collaboration. They were ordinarily good people doing their jobs. The GAL was doing what GALs do. The providers were doing what providers do. The judge was doing what judges do. The DCS worker was doing what DCS workers do.
What made the room work was not the individual quality of any one person’s performance. It was the accumulated social capital of a team that had been working together long enough to know each other’s scripts. Not to agree with them. Not to abandon their own institutional obligations in favor of someone else’s. But to know them well enough that translation happened almost automatically — the provider’s clinical language landed with the judge, the GAL’s recommendation was legible to the caseworker, the family’s experience was visible to everyone rather than only to whoever happened to be asking the right questions.
Anthropologists Aradhana Sharma and Akhil Gupta have argued that the state is not a unified apparatus of power operating according to a coherent design. It is a set of everyday practices — the forms, the hearings, the case notes, the staffings — performed by ordinary people who are, like everyone else, mostly in between the good and the bad, doing their jobs under pressure, shaped by their institutional scripts without being reducible to them. Power asymmetry is real in these rooms. Coercion is real. But its agents are not villains. They are people with their own formation, their own constraints, their own sense of what the job requires.
That framing changes what we are looking at when a room doesn’t work. Not bad actors. Not institutional malice. Orthogonal everyday practices that were never quite designed to talk to each other — and the specific conditions that allow translation to happen in some rooms and not others.
—
I want to be precise about what orthogonal means here, because it is not the same as conflicting.
Conflicting maps would mean that the GAL and the caseworker want different things for the family — that their institutional obligations put them in direct opposition. That happens, but it is not the primary dynamic I am describing.
Orthogonal maps mean that the GAL and the caseworker are both oriented toward the family’s wellbeing, both doing their jobs competently, and both operating from institutional frameworks that were developed independently, that use different vocabularies, that weight different kinds of evidence, and that produce different — not contradictory but differently angled — pictures of the same family’s situation.
The GAL’s picture is built from the child’s perspective, from provider reports and case manager statements when they get them. The picture is also built from legal frameworks, relational experiences operating within the courtrooms, and the specific language that drives results in the legal context. The caseworker’s picture is built from case documentation, service engagement, court compliance, monthly visits with family, and the longitudinal knowledge of a case’s trajectory over months or years. The judge’s picture is built from the legal record, the testimony, and the specific framework of what the court is authorized to address.
Each of these pictures is accurate. None of them is complete. And the family sitting in the room is the only party who holds all of them simultaneously — because they are living the case rather than managing it — but they are also the party with the least institutional vocabulary to say so.
This is the structural condition at the center of child welfare practice. The family is not one voice among several. But sadly and ironically their voice is the only continuous thread running through every room, every hearing, every handoff, every caseworker transition. Theirs is the map the system never quite reads. And when the practitioners in a room cannot translate between their orthogonal maps, it is the family that navigates the gap — not because they chose to, but because they have no choice. Nobody else is living the whole case.
The domestic violence court worked because years of shared history had made the orthogonal maps mutually legible. The client in that room did not have to carry the translation alone. Their experience was visible to everyone in the room without requiring them to speak the institutional language of each person present.
Most rooms do not work that way. And the consequences fall where they always fall. The social capital that closes the gap between orthogonal maps — the mutual legibility built through shared history and repeated encounter — exists in some rooms and not in others. Where it is absent, the family carries what no family should have to carry alone.
—
I have been sitting in on Child and Family Team Meetings — the mandated gatherings that bring together families, caseworkers, providers, and other stakeholders at regular intervals throughout a case — long enough to have a clear picture of what goes wrong.
In one meeting I attended, a Burmese family was in the room: a father, his partner, an uncle, an aunt, a grandmother. Few of them spoke English. We were working through a phone interpreter, translating line by line. The caseworker — a year or so into their role — held the meeting agenda tool three inches from their face and read from it. When it came time to discuss strengths, they asked for three. They asked again. They kept asking until I stepped in to redirect the question.
What the tool calls strengths are not a compliance item. They are the family’s existing capacity to transform the situation that brought the case. A family that has kept a grandmother in the home to help a mother suffering from major mental disorder, that has an aunt and uncle present, that showed up to this meeting despite every language and cultural barrier in the room — that family has strengths. But they do not have the institutional vocabulary to name them on demand, in English, through a phone interpreter, for a worker who has not yet told them why the question matters.
The family kept returning to the problems. There was no dialogue. The conversation never turned toward what the family already knew how to do, because nobody in the room made space for it. At one point, an ambiguity surfaced about whether the grandmother and aunt — the placement — had been told they were permitted to supervise visits with the mother. The courts had specifically stated they could not supervise. The caseworker did not know what he had communicated. The case manager had not been in genuine dialogue with this family long enough to know what they understood. I clarified for the family when the stakes of ambiguity couldn’t be higher.
In a second meeting, I watched a different caseworker orient their body toward the licensed foster parent — a stranger to this family — for the entirety of the meeting. The foster parent sat on one side of the room. The father sat on the other. The caseworker never crossed the floor. Eye contact with the father was incidental. The room’s geography became the case plan’s geography. The family’s perspective was structurally unavailable before anyone had said a word.
Neither of these workers was malicious. One was undertrained and overwhelmed. The other had learned, somewhere, that the path of least resistance runs through the most institutionally legible person in the room. Both were doing something recognizable as their job. And in both rooms, the family was left holding the translation alone. Not because anyone decided it should be that way. Because the social capital required to distribute that weight had never been built.
—
In my county, a small team of supervisors and I have been attending these meetings to observe and model.
The observation is straightforward: we watch what happens when the people in the room try to produce a shared picture of a family’s situation from their orthogonal maps. We watch whether the family’s voice enters that production or gets organized out by the institutional agenda. We watch where the translation fails — where clinical language doesn’t land, where the family’s description of their own needs doesn’t fit the case plan’s categories, where the orthogonal maps produce a gap that nobody fills.
The modeling is more delicate. When the translation fails in a specific moment — when a conversation closes down, when the family’s voice gets crowded out, when the room is producing friction rather than coherence — we offer a different move. Not by taking over. By demonstrating, in real time, what it looks like to bring the family’s map into the room as legitimate knowledge.
As our practice model consultant once said, “This is not about doing anything new; it’s about doing what we already do better.” This is likewise not about asking practitioners to care more, or to try harder, or to be better people. It’s about asking them to locate the family as the continuous thread the room keeps losing and to understand every stakeholder as a contributor to the social capital that makes timely permanency — and epistemic justice for families — possible. To understand that when the worker reads from the agenda without looking up, or orients toward the foster parent rather than the father, what gets lost is not just rapport — it is the only knowledge in the room that belongs to no institutional map and is answerable to no professional obligation except the truth of this family’s situation.
That knowledge does not disappear when it goes unread. It gets carried home by the family, alone, to be navigated without institutional support.
The domestic violence court produced something for its client because the room had accumulated enough shared history that nobody had to carry the translation alone. What the CFTM is designed to produce, in every case, is a version of that condition — not through years of shared history, but through the deliberate, practiced act of making the family’s map visible in real time.
—
I do not know if the intervention will close the gap. The theory of change is coherent — social capital among practitioners produces better translation, better translation produces more meaningful participation for families, more meaningful participation produces better outcomes. But the causal chain is long and the measurement is hard and the system is under constant pressure to count what is easy to count.
What I know is that the family in that Burmese household, and the father sitting alone on the other side of that room, needed someone to make their map legible to the room. Not to replace their map with an institutional one. To bring it in.
That is what the work is for. And the work has a name.
That is the next piece.
—
Sources
Freire, P. (1970). Pedagogy of the oppressed (30th anniversary ed.). Continuum.
Fricker, M. (2007). Epistemic injustice: Power and the ethics of knowing. Oxford University Press.
Putnam, R. D. (2020). Bowling alone: The collapse and revival of American community (revised and updated ed.). Simon & Schuster.
Sharma, A., & Gupta, A. (Eds.). (2006). The anthropology of the state: A reader. Blackwell.
—
The Social Construction Worker has worked in child welfare for six years and has recently completed an MSW at an accredited institution. This Substack explores the nos framework — a theory of how social reality gets produced, maintained, and disrupted — developed at the intersection of child welfare practice, social construction theory, and a deep suspicion of any map that mistakes itself for the territory.