Family Justice at a Crossroads: Five 2026 Judgments on Evidence, Fairness and Children’s Welfare

Five judgments, five distinct legal questions

These five decisions were reported together in a late-October Today’s Family Lawyer roundup. They do not constitute a single legal development, a new policy or a change in the burden of proof. They involve five different tasks: a finding-of-fact assessment, a discretionary costs award, an appeal arising from a change in care, a summary international-return application, and permission to appeal a placement order.

Those distinctions matter. A fact-finding judgment tells us what the court considered proved on the evidence before it; it does not always decide the final child arrangements. A costs judgment is not a declaration that every disputed allegation was fabricated. A Hague return order determines the appropriate country for the next stage of welfare litigation; it does not dictate final residence. An appeal is a challenge to an identifiable error, not a fresh welfare contest.

We have published separate case analyses, linked above. The purpose of this article is different: to ask what these decisions collectively reveal about the quality of the family justice process, the evidence it relies upon, and the barriers faced by parents without specialist legal assistance.

The child’s voice must be traceable, not merely invoked

In QR v ST, the court was presented with a substantial bundle—368 pages—but the underlying first-hand accounts and source material relevant to what the children had said were a central evidential concern. The judgment did not establish the allegations of abuse against the children, rejected the mother’s allegations that the father had coached or alienated them, and made a finding of abuse by the mother against the father. These conclusions must not be recast as proof that the children lied. The judgment was concerned with fact-finding and subsequent welfare steps remained to be determined.

The systemic question is uncomfortable: what does it mean to say the child’s voice has been heard if an important account appears only in an adult’s interpretation, a professional summary, or an incomplete disclosure trail? The court is entitled to evaluate hearsay and secondary accounts, but their origin, context and reliability are crucial. A chronology that repeats a summary with confidence does not supply the missing original note. Nor is a 368-page bundle inherently better evidence than a small, well-authenticated set of primary documents.

The practical solution is not for parents to question children repeatedly or turn them into witnesses of adult conflict. It is for adults and professionals to preserve contemporaneous accounts, identify what has been withheld or lost, seek proportionate lawful disclosure and distinguish wishes and feelings from matters a child could not fairly be asked to decide. A child’s welfare is not protected by volume of paperwork alone.

For litigants in person, a useful evidential record specifies the source, author, date, exact words where available, context, any later correction and the issue to which the evidence relates. Our disclosure and records guide sets out ways of organising such records without confusing subject access, court-directed disclosure and the evidence admitted at a hearing.

Safeguarding allegations and the costs trap: reject the false binary

The headline figure in FAZ v MAZ & Anor (Costs) is £27,109.75. It is tempting to turn that into a simple warning against making allegations in children proceedings. That would be legally careless and potentially damaging to safeguarding.

The father had claimed £256,667.80 following a hearing at which sexual-abuse allegations were not proved. Mr Justice Poole carefully assessed the reasonable and proportionate costs related to the fact-finding exercise at £108,439 and awarded 25% of that amount. Crucially, the court accepted that the mother genuinely believed that abuse had occurred, while finding that she had acted unreasonably and dishonestly in aspects of how the case was advanced. The decision turned on those particular conduct findings, the scope of the hearing and the judge’s discretion.

It follows that two propositions can both be true: a parent may genuinely fear abuse and need the protection of a responsive safeguarding system; and a court must protect procedural fairness against misleading, embellished or unsustainable allegations. The answer cannot be to stigmatise every disclosure, nor to dispense with proper testing because a concern is sincerely held.

JSH Law’s position is that evidential integrity is safeguarding. Exact language matters: ‘the child told me’, ‘the school record states’, ‘the police investigated’, ‘the professional assessed’, and ‘the judge found’ are five different propositions. A police decision to take no further action is not itself a family court finding that abuse did not happen. Equally, a report of suspected harm is not a court finding that it did. The Family Court must work with these distinctions rather than collapse them into easy labels.

At a DRA, the consequences may be immediate

In B v G, the High Court considered the mother’s appeal against an order transferring a nine-year-old boy’s care to his father at a dispute resolution appointment. The case did not establish a presumption that transfers are justified after disrupted contact. It examined whether the Recorder had considered the child’s welfare and acted fairly, including in relation to his wishes, sibling relationship, parental capacities, mental health and a proposed transition.

Mrs Justice Lieven rejected the proposition that a DRA is categorically unable to produce a transfer-of-care order. The social worker had been cross-examined, the possibility of transfer was known to the mother’s representatives and the court had undertaken a welfare analysis. The High Court regarded the departure from a proposed transition plan as arguable but dismissed the appeal on that point. The transfer was framed as a welfare decision, not a punishment for a parent’s behaviour.

For a represented party, that may be a sobering reminder. For a litigant in person, it is more. Court lists and hearing labels are often opaque. Someone who expects ‘only a review’ may fail to identify that a social worker recommends the most significant possible change to the child’s everyday life. The ability to participate fairly depends on notice that is meaningful in practice: knowing what the court might do, what evidence it can rely on and what procedural objection or alternative order may need to be raised.

That does not mean every DRA must become a fully contested trial. It means the appropriate process depends on the decision being contemplated, the real factual dispute, the available evidence and the child’s welfare. The right safeguards can reduce the risk of injustice without turning delay into an end in itself.

The Hague Convention asks a narrower question than ‘where is the child safest forever?’

In Re AL, the High Court ordered a five-year-old child’s return to Spain, where she was habitually resident, notwithstanding the mother’s allegations of domestic abuse and serious concerns about the impact of return on her mental health. The judge did not find the allegations to be fabricated. Instead, he applied Article 13(b) of the 1980 Hague Convention and examined whether a grave risk would arise in the actual circumstances of return.

The detail mattered: separate accommodation; funds for rent, maintenance and therapy; undertakings; arrangements designed to limit contact between the parents; and evidence about healthcare and language accessibility in Spain. The judge considered psychiatric evidence closely, including the distinction between likely deterioration in wellbeing and a sufficiently evidenced risk of serious consequences for the child. The Article 13(b) exception was not established on the facts.

This illuminates a recurring justice problem. Parties may understandably describe the whole history of an abusive relationship. But the judge in a summary return case is deciding a narrower legal question, not conducting the full final custody assessment. A procedural system is more accessible when litigants understand the distinction before filing evidence. Those resisting return need prompt, specialist advice, accurate medical evidence where appropriate, and a practical evaluation of protective measures. Those seeking return should not assume that formal undertakings are beyond challenge: their effectiveness and enforceability are part of the analysis.

In both domestic and international proceedings, courts must resist treating risk as a box ticked by reciting the word ‘safeguards’. The inquiry should remain concrete: who will provide the protection, when, with what enforceable mechanism, and what happens if it fails?

Appeals expose the gap between lived unfairness and appellate error

The mother in The Mother v The Local Authority opposed the placement order and adoption plan, although she had agreed to the care order. At reconsideration of permission to appeal she explained that distinction. HHJ Hesford expressly considered it but refused permission, having reviewed the trial judgment and concluded that the trial judge had already evaluated the options, the mother’s recent progress and the child’s welfare timescales.

The distinction between later progress and an appealable error is legally significant. A family member’s circumstances may genuinely improve after the order. That does not automatically demonstrate that the order was wrong on the evidence available to the trial judge. An appeal must identify an error of law, an unjust decision arising from serious procedural irregularity, or some other ground within the governing appellate test. A parent may feel unheard and still struggle to articulate the kind of error that Part 30 requires.

This is not a reason to discourage legitimate appeals. It is an argument for proper transcripts, prompt advice, realistic grounds and understandable explanations of the court’s reasoning. In public law care and placement cases, where adoption can permanently alter legal relationships, specialist regulated representation and potential legal aid must be considered urgently.

As legal technology becomes more accessible, there is a further danger: a plausible AI-drafted ‘appeal ground’ can confuse a disagreement with the result for an error in the decision. Every proposed authority, quotation and transcript reference requires human checking. A fluent document is not the same as a legally arguable one.

What these cases tell us about the justice system

It would be unjustified to present these five judgments as proof that the family courts always fail children or parents. The decisions demonstrate judges grappling with difficult evidence, nuanced legal standards and welfare consequences. Several explain their reasoning in considerable detail. That is a strength worth acknowledging.

Yet the cases also show how easily process can become inaccessible. An original account may not be in the bundle. A contested allegation may acquire unwarranted certainty through repetition. A DRA can carry consequences that a parent did not anticipate. An international application can be governed by a legal question markedly different from an ordinary welfare case. An appeal can fail because the party cannot connect their objections to an error recognised by the appellate court.

These are practical access-to-justice problems, not merely questions about courtroom manners. The strongest procedural reform would improve the integrity and usability of the evidential record: source references, concise and meaningful issues lists, prompt disclosure where lawfully available, clear hearing notices and decisions that explain the child welfare reasoning and the next lawful steps.

Safeguarding and fairness should not be positioned as competing ideologies. The integrity of an allegation matters to the person reporting possible harm, to the parent accused, and above all to the child whose circumstances may be changed. An evidence-led approach must remain trauma-informed without treating trauma as proof of a particular factual allegation; equally, it must not equate gaps in a frightened person’s account with automatic dishonesty.

Six questions every litigant in person should answer

The following is JSH Law’s practical preparation framework, not a statutory test.

1. Which decision is actually before the court? Fact-finding, temporary care, costs, international return, permission to appeal and final welfare are not interchangeable.
2. What is alleged and what is established? Separate a disclosed concern, documentary report, professional conclusion, admission and judicial finding.
3. Where is the original evidence? Identify its author, date, context and any missing underlying records; preserve originals.
4. What does the child need from this particular decision? Address emotional and physical safety, relationships, continuity, wishes and realistic alternatives.
5. What procedural fairness is required? Consider notice, oral evidence, special measures, disclosure, an interpreter or other justified direction.
6. What precise order is being sought? Make the proposed timetable, responsibilities, safeguards and review mechanism intelligible.

If you use AI to organise documents, verify every quote, record, date and legal authority against the original. Family Court confidentiality and children’s protected information remain binding concerns. See our Family Court Disclosure and Records Guide.

Sources, further reading and JSH Law support

Primary judgments: QR v ST [2026] EWFC 300 (B); FAZ v MAZ & Anor (Costs) [2026] EWFC 301; B v G [2026] EWHC 2668 (Fam); Re AL [2026] EWHC 2705 (Fam); The Mother v The Local Authority [2026] EWFC 305 (B).

Legal framework: Children Act 1989; Family Procedure Rules Part 30; Practice Direction 12J. See also the 29 October 2026 judgment roundup which prompted this comparative analysis.

Published family judgments may contain reporting restrictions. No individual or family should be identified by combining anonymised details with other sources. The cases above are discussed only to the extent permitted by their published terms.