Fabrication Fits Better

What the Court Can Actually Do

A family court is an adjudicatory instrument, not an investigative one. It hears claims, examines what is put in front of it, and applies a body of law to facts as presented. For most of the questions it is asked to resolve — what someone is like as a parent, what really went on inside a household, who is telling the truth about events no one outside the family witnessed — the underlying reality is not directly accessible to the court at all. What the court has is accounts: testimony, reports, observations filtered through professional frames, contemporaneous documents whose meaning depends on context the court was never in. The court's job, in cases of this kind, is to choose between accounts.

Choosing between accounts is not the same operation as finding out what happened. It is a distinct, narrower activity, and the distinction matters because the instruments built for one are not the instruments built for the other. The court can test an account for internal consistency, for fit with available documentation, for plausibility against the framework the law expects to apply. It cannot reach behind any of these things to the events themselves. Whatever happened in the home is not a witness the court can call.

The Decision Arrives Regardless

The court cannot pause until it knows. A child has to live somewhere. A schedule has to be set. An order has to issue. The case will close, on a date already fixed, with or without a satisfactory grasp of what actually occurred. So the question is not whether the court is in a position to determine the truth. The question is what it does when it is not — because the proceeding does not stop on the day the truth becomes unreachable. It continues, and produces a result, in exactly the same form it would produce a result on a case where the truth was knowable.

That continuity is the part worth holding onto. The decision the court issues looks the same whether it rests on a verified account, on the best of two unverifiable accounts, or on something the court chose because the choosing was unavoidable. The reader of the order cannot tell which it was. The court itself often cannot tell, because the question of what kind of basis the decision actually had is not a question the court is structured to ask of its own outputs.

The Account That Wins

When the underlying truth is unreachable, the account that wins is the one that fits the instrument. Fit is a real property. An account fits well when it is internally consistent, when it aligns with the available documentation, when it is corroborated by professional witnesses speaking in the register the court is used to hearing, when it tracks the legal framework the case is being decided under, when it requires no awkward digressions to explain its own gaps. A well-fitting account hands the court the materials it needs to write its decision. A poorly fitting one does not.

The mistake is to assume fit tracks truth. Sometimes it does. Often it does not, because fit is a property the account has in relation to the instrument, not in relation to reality, and the two properties can be optimised separately. An account engineered to fit the instrument can be assembled by anyone with the time, the resources, and the willingness to construct it. An account that is true to what happened, by contrast, is not engineered. It is reported. It has the shape of the events it describes, not the shape the instrument is built to receive.

Lived Experience Makes Poor Material

A lived account is irregular by nature. It has the texture of memory: vivid where the event was vivid, faint where it was not, ordered by significance rather than by chronology, full of things that were obviously true at the time but cannot now be evidenced because no one thought to evidence them. The parent describing what their home life was actually like over a long period is reporting something that does not exist anywhere else. Whole years of routine that left no record. Conversations no one transcribed. Patterns visible only across many small incidents, none of them documented individually because each was, at the time, just life.

These accounts have specific failure modes when fed to an adjudicatory instrument. They contain incidents whose context the listener does not share. They drift in tense as the speaker moves between specific memories and the general picture they are trying to convey. They contain gaps where the speaker simply does not remember, and the gaps look the same as evasion to a listener who cannot tell the difference. They sometimes contradict the contemporaneous documentation because the documentation, written by someone with a different frame, was always going to read the same events differently. None of this is dishonesty. It is what truthful reporting of an ordinary life sounds like when the listener was not present for the life.

Fabrication Makes Excellent Material

A constructed account does not have these problems. It can be polished. It can be made internally consistent across every retelling, because someone designed it to be. It can be aligned with the supporting documentation, because the documentation was selected or generated to align with it. It can be delivered without gaps, because gaps are filled in deliberately rather than left as honest blanks. It can be calibrated to the legal framework, because the constructor knew what framework was coming. It can be corroborated by witnesses primed for the account, because corroboration is something that can be arranged when the account is the product rather than the report.

Engineering and fabrication are not the same thing — a well-resourced and coached truthful party can produce an engineered account too, and many do — but fabrication is engineering at its most fully realised. It is the form in which every property the instrument rewards can be optimised, because nothing has to remain faithful to anything outside the construction. Engineered truth is engineered against the friction of what actually happened; engineered untruth has no such friction. So while the structural advantage belongs to engineering generally, fabrication is the version of engineering with no upper bound on fit, and in close cases that upper bound is the thing that decides.

A well-built fabrication is legible in exactly the ways the court is built to read. It is the account the court was, in a sense, asking for — the one whose shape matches the shape of the receiving instrument. Lived experience cannot be engineered for the venue. It is what it is, and what it is, in the venue, is poor material.

The Conclusion That Was Already Forming

There is one more factor, and once it is added the mechanism is complete. By the time most family-court cases reach the deciding hearing, a tentative conclusion has been forming for some time — in the welfare officer's first impression, in a recommendation written before the most contested evidence was heard, in the inherited file from a prior hearing, in the framing of the report the court will rely on. The conclusion is not yet final, but it has shape, and the shape exerts a pull on the materials that follow. New information is fitted to it. Information that does not fit is treated as anomalous, or not weighted, or politely set aside.

In that environment, the engineered account has a second advantage. It can be built to the forming conclusion. Whoever is building it can hear which way the proceedings are leaning and shape the account so that it slots into the gap the conclusion has reserved for it. The lived-experience party cannot do this, because lived experience is not adjustable to the venue's predictions. They go on reporting what happened, even when what happened does not align with where the matter is heading. From the court's point of view, one party is producing materials that increasingly cohere with the picture forming, and the other is producing materials that do not. That difference is, again, treated as evidence — and again, it is a property of the instrument's relationship to the accounts, not of the accounts' relationship to reality.

What Survives

The party telling the truth about their own life loses, not because the court preferred a lie, but because the form their truth takes is not the form the court can process. The other party wins, not necessarily because they intended a fabrication, but because what they produced was engineered, by accident or design, for the venue that received it. The court issues an order. The order looks like a finding. From inside, the court appears to have weighed accounts. From outside — from the vantage of anyone who knows what actually happened in the household — the court has weighed legibility, and legibility was a property of the engineered account in a way it could not be of the lived one.

The mechanism does not need anyone to fail. It needs only the architectural condition that the instrument cannot verify what happened, must decide anyway, and selects in those cases by fit with materials whose fit was the product the engineered party brought to the room. The welfare officer, the report writer, the judge can each be careful, well-trained, conscientious, and the result holds — because the result is generated by the structure they are working inside, not by their conduct on the day. The lived-experience party brought their life. The other party brought a document. A venue that cannot tell which is which, and must choose between them, will choose the one its instruments are built to read.

Naming this is what makes it visible. Without a name, the party whose lived experience was excluded is left with the unrebuttable feeling that they were not heard — a feeling the court, looking at its own record of evidence considered and witnesses examined, will sincerely deny. They are both right. The party was not heard, because what they were reporting was not in a form the court could hear. The court did consider the materials, and the materials it considered were the legible ones. The asymmetry is not in the court's good faith. It is in the structural mismatch between the instrument and the kind of account a truthful life produces — and as long as that mismatch is unnamed, the result it generates will keep being mistaken, by the court and by everyone reading the court, for the result a fact-finding exercise would have reached.