Chapter 1

The playroom has been designed by someone who read a manual about therapeutic play environments and followed it exactly. The stuffed animals are arranged by size. The crayons are sorted by color in a caddy that rotates, so a child can find the preferred shade without the added variable of frustration. The low table has rounded corners. The rug is washable. The shelves are labeled with both words and pictures, because predictability is considered regulating and because there is a literature on this, which means there are citations, which means there are people in offices who have justified the maple shelving in Room 302 with peer-reviewed evidence.

The one-way mirror is also framed in maple. A study from 2004 found that adult subjects described maple as warmer than oak and more professional than pine, which is useful if your business model depends on making adversarial parents feel they are in a space both child-centered and court-defensible. The children do not care about the wood finish. The children perceive, with the unnerving accuracy children reserve for things adults pretend are invisible, that someone is watching them from behind glass.

I know this because I was one of those children once, though in my case the wood was oak, the year was 1993, and the evaluator's name was Dr. Ruth Alder, who is now retired and, according to the licensing board database, living in Tucson.

Room 302 belongs to Central Ohio Behavioral Associates, third floor, East Broad Street, Columbus, Ohio. Converted medical building. Beige hallway. Commercial carpeting in a shade best described as legally prudent. My office is across from the observation room, and between them is a supply cabinet containing the tools of my profession: standardized forms, sharpened pencils, test manuals, consent packets, scoring sheets, and the Fisher-Price Clinical Play Assessment Kit, which costs $847 per set and is replaced annually because the toys are considered assessment materials and assessment materials must be current. Eight hundred forty-seven dollars would cover a great deal of cereal and pasta and winter socks. It buys, in this context, a miniature plastic kitchen, a doll family with neutral expressions, and a wooden train no child has ever touched without immediately using it in a way the manual did not anticipate.

On my desk are the forms for the Landon evaluation, stacked in the order I will need them. Petition. Temporary orders. Attorneys' letters. Intake summaries. Parenting Stress Index, fourth edition. Child Behavior Checklist. Parenting Alliance Measure. Behavioral observation forms with forty-two items and no blank space. The absence of blank space is not an accident. Blank space invites language, and language invites trouble.

I sit down, align the stack against the edge of the desk, and open the file.

Landon, Thomas v. Landon, Claire. Franklin County Domestic Relations Court. Two children: Sophie, age nine. James, age six. Motion for allocation of parental rights and responsibilities pending final hearing. Guardian ad litem appointed. I have met the guardian once, in another case. He wore a navy suit and billed in six-minute increments.

I read Thomas Landon's declaration first because it is on top, not because I prefer fathers, though angry mothers often assume I do and frightened fathers often assume I don't. The declaration is articulate, measured, and composed in the register of a man who has spent his adult life explaining things to worried people. He is a pediatrician at Columbus Children's Hospital. He writes that he is seeking a stable parenting arrangement in the best interest of Sophie and James. He writes that he remains concerned about Claire's increasing emotional volatility, difficulty maintaining routine, and tendency to involve the children in adult conflict. He writes that he hopes this process will provide clarity and structure for the family moving forward.

Declared purpose. Concern for the children. Hope for clarity. The phrase best interest of the child appears twice in three paragraphs, as it always does. It is the system's liturgy. You say the words and everyone is absolved in advance.

Thomas's sentences are efficient. He uses no exclamation points. He does not overstate. He does not accuse. He expresses concern. Concern is an excellent posture in family court. Anger suggests grievance. Concern suggests stability.

I turn to Claire's declaration. It is longer. It circles. The first paragraph begins in one place and ends somewhere adjacent. She writes that Thomas is controlling, that he arranges everything, that he makes the children nervous without ever raising his voice. She writes that he has made her look unstable for years by staying calm while she reacts to things no one else seems to see. She writes, He doesn't hit them. He just—arranges them.

I stop there.

Then I read the line again, because whatever else Claire Landon is, she is not inarticulate. The problem is not that she cannot describe what she means. The problem is that what she means does not correspond neatly to a scoring rubric.

I set the declaration down and make no note, because there is nowhere to put the note that matters. Arranges them. The phrase would not survive contact with cross-examination. It would sound subjective. Pejorative. Insufficiently operationalized. An attorney would ask, Dr. Cade, can you define arranged in clinically measurable terms? And the honest answer would be no, not in a way the court would accept.

I read the attorneys' letters next. Thomas's counsel describes him as a devoted, involved father who has maintained continuity in the children's medical care, school routines, and extracurricular schedules. Claire's counsel describes her as the children's primary emotional caregiver and warns against mistaking polish for safety. Attorneys, unlike evaluators, are permitted adjectives when they favor their client.

By nine-thirty I have a legal pad half full of the usual distinctions: father cooperative, mother distressed; father routine-oriented, mother affectively reactive; both deny alienation; each accuses the other of undermining. Every contested custody file contains, in some form, two narratives and a demand that an evaluator convert them into one.

Outside my office, someone laughs in the waiting room. Down the hall, the copier starts and then jams with the passive-aggressive whine office machines use when they know you cannot abandon them permanently. I close the file and look through the narrow window in my office door toward Room 302.

The playroom sits empty under fluorescent lights carefully selected to imitate natural warmth. The dollhouse is intact. The plastic food has been returned to its basket. The miniature family—mother, father, boy, girl—stands upright in the living room as if waiting to be interpreted. The room is supposed to invite spontaneity. It does, but only within parameters. Everything in this profession is an attempt to create conditions under which people will reveal themselves while also ensuring that what they reveal can be defended in court.

The mirror reflects nothing from my angle. That is the point.

I stand, take my coffee mug to the supply cabinet, and set it on the shelf while I pull out a fresh packet of drawing paper. Standard letter size. Bright white. Smooth enough for crayons, sturdy enough not to tear under pressure from a child who bears down too hard. I place the packet on my desk beside the forms and remove the top sheet.

Draw a picture of your family doing something together.

The prompt has existed in some version for decades. It appears in assessment manuals, training seminars, continuing education modules. It is considered useful because it offers information about attachment, perceived relationships, developmental level, emotional tone, organizational style. Depending on which article one cites, it may also provide data on conflict, role perception, anxiety, and self-concept. A sufficiently motivated professional can extract a tremendous amount from a child's drawing, provided the extraction is accompanied by cautionary language and enough conditional verbs.

What the prompt actually measures is less often discussed.

Not the family, exactly. Not even the child's feelings about the family, though that is the polite fiction. The prompt measures, first, the child's understanding of what a family is supposed to look like when an adult with a clipboard asks the question. It measures competence in expectation. It measures whether the child has learned the shape of an acceptable answer.

I hold the paper by one corner and think, not for the first time, that there is no protocol for asking what a child did not draw.

No item for omission. No score for erasure. No line on the form that says: subject appeared to understand the assignment too well.

I slide the blank page into the top drawer of my desk with the others. On the desk, the Landon file remains open to Claire's declaration. He just—arranges them. The dash does more work than the words. It marks the place where language failed and she kept going anyway.

At ten o'clock, I will meet Thomas Landon for the first parent interview. He will arrive on time or a few minutes early. He will shake my hand. He will answer the questions I ask and several I have not yet asked. Claire will come later this week and sit in the same chair and make the room mean something different simply by being angry in it. Then the children will come, and I will hand them crayons and white paper and ask them to show me their family doing something together.

This is a normal beginning.

That is one of the more efficient lies my profession tells. Normal beginning. Neutral setting. Standardized instruments. As if any family arrives in this room unaccompanied by the months or years that shaped them into the versions of themselves I am about to score.

I straighten the forms again. Forty-two items. No blank space.

The file waits on my desk. The room waits behind the mirror. In the drawer, the paper stays white.