· 2026-11-02T21:11:32+0000 · 951 posts

Citation: Boleno & Newcombe [2026] FedCFamC1A 191

Predictive Risk Assessment Stands: Supervised Time Until Age 10 Upheld on Appeal, with Costs Against the Father

In Boleno & Newcombe [2026] FedCFamC1A 191, Altobelli, Christie and Parker JJ heard the father's appeal from final parenting orders made on 29 May 2026 ([1]). The primary judge found that unsupervised time with the father posed an unacceptable risk of psychological harm to the child, and ordered professionally supervised time building to once per month until the child turns 10, then unsupervised monthly time ([15]–[16]). The father pressed four grounds. All failed ([38], [51], [69], [86]). The appeal was dismissed ([87]). The father was ordered to pay the mother's costs of $14,879.78 and the Independent Children's Lawyer's (ICL) costs of $5,662.80, which he had conceded if the appeal failed ([88]–[89]). The important point is that an Isles & Nelissen predictive risk assessment can rest on unchallenged past conduct plus present insight and capacity for change; a self-executing age trigger supported by Family Report evidence is not the same as "manufacturing" expert evidence; and the ease of a future s 65DAAA variation application is not a consideration the primary judge must weigh. All paragraph references below are to the judgment.

Why it matters: Risk findings are evaluative, not discretionary, and they stand or fall on the correctness standard ([20]–[21]). This appeal shows how a primary judge can lawfully move from historical conduct to a future risk conclusion without inventing a crystal ball: past events, present insight and demonstrated capacity for change supply the predictive foundation ([31]–[34]). It also draws a clean line under Lainhart & Ellinson: fixing unsupervised time to a known future event (turning 10) on existing expert evidence is not the same as ordering therapy so that an unknown psychologist can later decide ([81]–[82]). And s 65DAAA is about protecting children from unnecessary litigation, not about designing orders to make a parent's future variation application easier ([85]).

🧩 Facts and Issues

Background

The parties married in 2014. The mother lived in Australia from 2016; the father joined her in 2019. They separated on 30 September 2021 when the child, born in 2020, was an infant, and remained separated under one roof until January 2023. The child is their only child ([3]).
On 20 June 2022, before physical separation, the parties entered consent orders. The child lived with the mother and spent substantial time with the father each week and for one third of school holidays ([4]).
In late December 2024 and early January 2025, the father alleged that the child had made disclosures of sexual and physical abuse in the mother's household. He made video recordings of discussions with the child about the alleged disclosures, reported the matter to police and the Department of Communities and Justice (DCJ), and withheld the child from the mother ([5]). The mother denied the allegations ([5(f)]).
Both police and DCJ completed thorough investigations and concluded that the child was safe in the mother's care. The father was advised to return the child in accordance with the consent orders. He did not ([5(g)]–[5(h)]).
The mother obtained a recovery order pathway. On 17 January 2025, after a contested interim hearing, the child was returned to her care and time with the father resumed under the earlier consent orders ([6]). The father maintained the allegations ([7]). On 2 July 2025, consent orders provided for supervised time ([8]).
The trial ran over five days in November and December 2025 ([9]). At trial the father did not seek a finding that the child had been sexually abused in the mother's care, though he expressed some ambivalence about how the alleged disclosures had arisen ([10]).

What each side sought

At trial ([11]):

the father sought unsupervised time building to three consecutive nights each alternate weekend;
the mother sought professionally supervised time once per month until the child turned 13, then time in accordance with the child's wishes;
the ICL proposed professionally supervised time transitioning to unsupervised time once the child reached 10.

The Family Report writer opined that the father posed a serious and ongoing risk of psychological harm if he had unsupervised time, that time should remain supervised until age 10, and that he had demonstrated minimal insight into the impact of his actions ([12]–[13]). By age 10, the writer said, most children have a clearer sense of right and wrong, making them harder to influence into false allegations ([13]).

Unchallenged findings that drove the risk conclusion

None of the primary judge's key findings about the father's past conduct were challenged on appeal ([14]). In summary, the primary judge found that:

the father's video recordings were "troubling" and showed directive and leading questions to which the child adapted her answers ([14(a)]);
there was incongruence between the father's descriptions of the child's demeanour and what the recordings showed ([14(b)]);
the father falsely told police and DCJ that there were no family law orders in place when he withheld the child ([14(c)]);
his narrative of detailed disclosures contrasted "starkly" with the short phrases in the video evidence, and that evidence was rejected ([14(d)]);
the child made no disclosure to police or DCJ in the father's absence despite multiple interviews ([14(e)]);
filming the child when, on his own evidence, she was distressed, without evidence of attempts to comfort her, was "a failure of protective parenting" ([14(f)]);
the allegations "had not emanated spontaneously from the child" ([14(g)]);
aspects of his account of returning the child and delaying a police report were "implausible" ([14(h)]); and
his repeated questioning, filming and keeping the child from her primary carer exposed her to psychological harm and to the risk of desensitising future protective responses ([14(i)]).

Ultimately the primary judge found that the father's willingness to co-opt the child to pursue the allegations meant the child would be at unacceptable risk of psychological harm if she spent unsupervised time with him ([15]).

The final orders under appeal

In summary, the mother has sole responsibility for major long-term decisions; the child lives with her; and the child spends time with the father as follows ([16]):

for two months from the final orders: four hours on the second Sunday of each month, professionally supervised at a contact centre;
thereafter: four hours on the second Sunday of each month, professionally supervised in the community (or at a contact centre if no agreed community supervisor); and
from the commencement of 2031 (when the child will have turned 10): as agreed in writing, and failing agreement, four hours unsupervised on the second Sunday of each month.

Order 19 also permits the father, on seven days' written notice, to attend school functions and events to which parents are normally invited, subject to school policy ([17]). Supervision is at the father's expense. Other orders deal with language during supervised time, remote communication, and injunctions not directly relevant to the appeal ([18]).

Issues

Ground 1: did the primary judge fail to apply Isles & Nelissen by not evaluating the magnitude and likelihood of the identified risk ([25]–[38])?
Ground 2: were the reasons for rejecting the father's proposed restraints and protective measures as an alternative to supervision inadequate ([39]–[51])?
Ground 3: was the inference that, in mid-2025, the father still told his psychologist he believed the child's alleged reports, not reasonably open ([52]–[69])?
Ground 4: did the primary judge err in determining that professional supervision was the mechanism to ameliorate the risk, including by fixing supervision until age 10 ([70]–[86])?
If the appeal failed, what costs orders should follow ([88]–[89])?

⚖️ Applicable Law – Legislation, Regulations, Rules

Pt VII / s 60CC of the Family Law Act 1975 (Cth): parenting orders are made by reference to the child's best interests. The evidence-based unacceptable-risk finding is one material consideration within that framework ([20]; [85]).
s 65DAAA of the Act: a party seeking to vary final parenting orders must show a significant change in circumstances and that it is in the child's best interests for the orders to be revisited. The ease of overcoming that threshold in a hypothetical future application is not a consideration the primary judge must weigh ([83]–[85]).
s 114UB of the Act: costs. The father conceded that if the appeal was dismissed, costs should be ordered. Having regard to the s 114UB(3) factors, the Court was satisfied the circumstances justified costs in favour of the mother and the ICL ([88]–[89]).
s 114Q of the Act / Part XIVB: publication is under the pseudonym Boleno & Newcombe.
Schedule 3 to the Federal Circuit and Family Court of Australia (Family Law) Rules 2021 (Cth): the mother's costs were calculated on the scale; the ICL's by reference to applicable legal aid rates ([88]).

📌 Precedents Relied On

The Court relied on these authorities:

Isles & Nelissen (2022) FLC 94-092: assessment of risk is an evidence-based evaluative conclusion, not discretionary; the overall parenting judgment is discretionary and is influenced by the s 60CC considerations, of which the risk finding is one ([20]). Risks of harm are postulated from known historical facts and present circumstances ([32]). Evaluation of unacceptable risk entails magnitude of harm and prospect of occurrence ([26]).
M v M (1988) 166 CLR 69: the foundational High Court framework for risk within the wider best-interests inquiry ([26]).
Napier & Hepburn (2006) FLC 93-303: magnitude and prospect of the postulated harm ([26]).
Bant & Clayton (2019) FLC 93-924: risk is assessed within best interests ([26]); do not "atomise" the evidence ([67]).
Warren v Coombes (1979) 142 CLR 531 and GLJ v Trustees of the Roman Catholic Church for the Diocese of Lismore (2023) 280 CLR 442: the correctness standard for evaluative findings; the appellate court decides the correct outcome while allowing for trial advantages ([21], [54]).
House v The King (1936) 55 CLR 499: the categories of appellate intervention in a discretionary decision ([22]).
Australian Coal & Shale Employees' Federation v The Commonwealth (1953) 94 CLR 621: a discretionary decision is strongly presumed to be correct ([23]).
Minister for Immigration & Ethnic Affairs v Guo Wei Rong (1997) 191 CLR 559: past events often provide a reliable basis for assessing the probability of recurrence ([31]).
Fitzwater & Fitzwater (2019) 60 Fam LR 212 at [138] (Austin J, dissenting, later endorsed): the predictive exercise of risk assessment is influenced by factual findings about past events ([32]).
Bennett & Bennett (1991) FLC 92-191 and Sun Alliance Insurance Ltd v Massoud [1989] VR 8: reasons are inadequate if the appellate court cannot ascertain the path to the result, or justice is not seen to have been done ([40]).
AMS & AIF (1999) 199 CLR 160 at [150]: avoid an "overly critical, or pernickety, analysis" of reasons ([40]).
Pollard v RRR Corporation Pty Ltd [2009] NSWCA 110: reasons must expose the resolution of points critical to the contest and do justice to the parties' cases ([41]).
DL v The Queen (2018) 266 CLR 1 at [130]: the adequate-reasons requirement applies to material aspects of the determination ([42]).
Bradshaw v McEwans Pty Ltd (1951) 217 ALR 1; Holloway v McFeeters (1956) 94 CLR 470; G v H (1994) 181 CLR 387: civil inferences must rise above equal competing conjectures; an inference is an exercise of ordinary human reason ([55]–[56]).
Eastley & Eastley (2022) FLC 94-094 at [33]: assess risk on the whole of the evidence; strength lies in cumulative effect ([67]).
Metwally v University of Wollongong (1984) 159 CLR 447: a party cannot argue on appeal for an order they did not seek at trial ([72]).
Horwitz & Tavares (2024) FLC 94-174 at [20]: the primary judge is not obliged to depart from the parameters of the parties' proposals ([73]).
Akston & Boyle (2010) FLC 93-436; Treloar & Nepean (2009) FLC 93-417: it is well open to give significant weight to Family Report evidence ([74]).
Grella & Jamieson [2017] FamCAFC 21 at [18]; CDJ v VAJ (1998) 197 CLR 172 at [151]: parenting orders involve prediction that cannot be scientifically guaranteed ([78]).
Lainhart & Ellinson (2023) FLC 94-166: distinguished; that case concerned orders designed to procure or manufacture future expert evidence so an unknown psychologist could later decide ([81]–[82]).
Radecki & Radecki (2024) FLC 94-232: the s 65DAAA threshold for revisiting final parenting orders ([83]).

🧠 Analysis

Issue

When a primary judge finds unacceptable risk of psychological harm from unsupervised time, and orders professionally supervised time until a self-executing age of 10, what must the appellant show to overturn that conclusion? Does Isles & Nelissen require a freestanding quantitative assessment of magnitude and likelihood beyond the path from past conduct, present insight and capacity for change? When are reasons for rejecting proposed restraints adequate? When is an inference about ongoing belief in contested allegations open? And does fixing a future unsupervised transition to a known age, on existing Family Report evidence, offend Lainhart & Ellinson or s 65DAAA?

Rule

An unacceptable-risk finding is evaluative and is reviewed on the correctness standard ([20]–[21]). The overall parenting orders are discretionary and attract House v The King error ([22]–[23]). Risk is postulated from known historical facts and present circumstances; past events inform the probability of recurrence ([31]–[32]). Reasons must expose the path to resolving points critical to the contest, without a pernickety audit of every argument ([40]–[42]). Inferences in civil cases need a reasonable and definite inference on the balance of probabilities, not equal competing conjectures ([55]–[56]). Risk is assessed on the cumulative effect of the evidence, not atomised strand by strand ([67]). A party cannot complain on appeal about the failure to make an order they did not seek at trial ([72]). A self-executing progression supported by existing expert evidence is not the same as manufacturing future expert evidence ([81]–[82]). The ease of a future s 65DAAA application is not a mandatory consideration ([85]).

Application

Past conduct plus present insight is a lawful predictive foundation

Ground 1 failed. The father's unchallenged past conduct was carefully reviewed ([27], [14]). The Court rejected the characterisation of the risk findings as confined to a "discrete period" ([28]–[29]). Ceasing harmful conduct only after time was suspended and then supervised does not mean the risk has ceased ([30]). Predictive assessment cannot be done other than by reference to past events ([31]). The father's past conduct, current level of insight, lack of compassion and limited demonstrated capacity for change provided a "strong foundation" for the predictive exercise ([33]). The primary judge's conclusion that there was a risk of psychological harm and that the risk was unacceptable ([15], [34], citing primary reasons at [193]) already embodied an assessment that the risk was likely to continue ([34]).

School-event attendance is not inconsistent with supervised time

The father argued that Order 19 (attendance at ordinary school events on notice) was inconsistent with a finding that professional supervision was needed ([35]–[36]). The Court disagreed. Without a demonstrated basis for concluding that attendance at school events, with other adults and staff present, would give the father the opportunity to expose the child to the identified risks, no inconsistency was established ([37]). Ground 1 failed ([38]).

Reasons for rejecting restraints were clear and cumulative

Ground 2 failed. The primary judge expressly addressed the submission that restraints on the father's conduct would mitigate the risk, and explained why she rejected it: she was not persuaded he had properly understood the impact of his conduct, or that his change of position was sincere (primary reasons [194], summarised at [44]). Further reasons at primary [195] took into account uncritical acceptance of the alleged disclosures, delay in reporting, apparent indifference while making recordings, lack of insight or compassion, and past willingness to act contrary to court orders even after police and DCJ told him to comply ([45]). Those paragraphs alone would dispose of the adequacy complaint ([46]). In any event they sit within detailed reasons as a whole, including the unchallenged findings at [14] ([47]). The primary judge concluded she could not have confidence that restraints would ameliorate the risk, but was satisfied supervision would ([48]). Compliance with an undertaking in the lead-up to the final hearing under the court's overview did not leave the Court "guessing" ([49]). Taken as a whole, the reasons explain the rejection of the father's amelioration proposal ([50]). Ground 2 was not established ([51]).

The psychologist-notes inference was open, and not material in any event

Ground 3 failed. The contested inference was that, at the time of the psychologist notes, the father still communicated a belief that the child's alleged reports were truthful ([52]–[53]). The notes recording those matters as relayed by him, without any recorded disavowal or doubt, gave rise to a strong inference that he presented them as factual ([58]). He could have called the psychologist to say otherwise. He did not ([59]). Cross-examination acknowledgments bolstered the inference ([61]). Later denials did not bind the primary judge, who had found other key parts of his evidence implausible ([62]). The inference was open and correctly drawn ([63]).

Any date error (June rather than September 2025) favoured the father ([57]). The inference was not material in any event: on his own evidence he maintained the allegations were true until on or after 19 May 2025, well after police and DCJ had concluded the child was safe ([64]). Timing of the asserted change of position was only one of many factors ([65]–[66]). It is not appropriate to atomise the evidence; risk is assessed on the cumulative effect ([67]–[68]). Ground 3 had no merit ([69]).

Age-10 supervision was within the dispute parameters and supported by the Family Report

Ground 4 failed. The father's appeal submissions drifted toward complaining that supervision should have been indefinite rather than until 10 — an outcome in his favour that he had not sought at trial, where he argued that if supervision were ordered it should be "significantly shorter" than the mother and ICL proposed ([71]–[72]). Indefinite supervision was outside the parameters of the dispute and was not open ([73]).

The age-10 proposition was strongly and directly supported by the Family Report writer's recommendation, which the primary judge accepted and gave significant weight ([74]). Although the writer appropriately conceded that her recommendation drew on general developmental considerations, she did not resile from recommending supervision until this child turned 10, and she did not endorse the father's proposal ([75]–[76]). Parenting orders always involve prediction that cannot be scientifically guaranteed ([78]). Lack of certainty does not require indefinite protective measures ([79]). To the extent the expert evidence was "qualified", it was to say the risk might not have abated by 10 — not that it would abate earlier — and that reading favoured the father ([80]).

Lainhart & Ellinson does not outlaw a self-executing age trigger

Lainhart & Ellinson concerned conditions to be met to a psychologist's satisfaction before unsupervised time — orders designed to procure or manufacture future expert evidence ([81]). Here the primary judge fixed a timeframe to a known future event (turning 10) based on existing expert evidence. That is materially different. Lainhart does not say a future progression cannot be built into orders where the evidence supports it ([82]).

s 65DAAA ease-of-variation is not a required consideration

The father argued the self-executing cessation created an "additional hurdle" for any future variation application ([83]). The Court disagreed. Turning 10 simply lifts supervision without a further application, which could hardly be a "problem" for him ([84]). The circumstances for any future s 65DAAA application still include the risk and insight findings, whatever the order's terms ([84]). In any event, as he conceded, the ease of overcoming s 65DAAA is not a consideration the primary judge had to weigh. Building in a pathway to further litigation would cut against s 65DAAA's purpose of minimising children's unnecessary exposure to conflictual litigation. The mandate was the child's best interests on the evidence and s 60CC, not the father's litigation convenience ([85]). Ground 4 failed ([86]).

Costs followed the conceded outcome

The appeal was dismissed ([87]). The mother sought $14,879.78 and the ICL $5,662.80. The father conceded that if the appeal was dismissed, costs should be ordered, and he did not dispute quantum. Costs orders were made accordingly ([88]–[89]).

Conclusion

All four grounds failed. The unacceptable-risk finding was a lawful predictive assessment grounded in unchallenged past conduct and present insight ([33]–[34], [38]). The reasons for preferring supervision over proposed restraints were adequate ([46], [50]–[51]). The contested inference was open and, in any event, not material ([63]–[64], [69]). Professionally supervised time until age 10 was within the dispute parameters, supported by Family Report evidence, distinguishable from Lainhart & Ellinson, and not undermined by s 65DAAA ([73]–[86]). The appeal was dismissed ([87]). The father must pay the mother's costs of $14,879.78 and the ICL's costs of $5,662.80 ([88]–[89]).

🧠 Take-Home Lesson

Risk is evaluative; orders are discretionary. Challenge the risk finding on the correctness standard, and the orders on House v The King. Do not blur the two ([20]–[23]).
Past conduct + present insight = predictive foundation. You do not need a freestanding likelihood spreadsheet. Unchallenged historical findings, current insight and capacity for change can lawfully ground a future risk conclusion ([31]–[34]).
Stopping after the opportunity is removed does not prove the risk has gone. Ceasing the conduct once time is suspended or supervised is not a finding that the risk has ceased ([30]).
Do not atomise the evidence. Attack the cumulative rope, not each strand in isolation ([67]–[68]).
Reasons must show the path, not answer every submission. Express rejection of proposed restraints, tied to insight and sincerity findings, will usually be enough ([44]–[46], [50]).
Inferences: call the witness or live with the note. Psychologist notes recording allegations without recorded doubt can support an inference of ongoing belief; not calling the psychologist leaves that inference open ([58]–[59]).
Materiality still matters. Even an open inference may not carry the appeal if the same conclusion rests on other findings ([64]–[66]).
You cannot appeal for an order you did not seek. Complaining that supervision should have been indefinite, after arguing at trial for shorter supervision, goes nowhere ([71]–[72]).
Age triggers ≠ manufactured expert evidence. A self-executing transition at a known age, supported by existing Family Report evidence, is not Lainhart & Ellinson ([81]–[82]).
s 65DAAA is for the child, not the appellant's litigation strategy. Ease of a future variation application is not a mandatory consideration ([85]).
Conceding costs if you lose means costs if you lose. Quantum on scale / legal aid rates was not disputed ([88]–[89]).
Support workers and McKenzie friends: when a parent pivots from "I believed the disclosures" to "I accept the investigation", expect the Court to test sincerity against recordings, reporting delays, compliance history and any treating notes — and to prefer supervision if confidence is lacking ([44]–[48]).

"The assessment of risk is an evidence-based conclusion and is not discretionary. … It is only the overall judgment, expressed in the form of orders made in the children's best interests, which entails an exercise of discretion." ([20], quoting Isles & Nelissen)

"Past events are not a certain guide to the future, but they often supply the only lawful foundation for predicting whether the risk will continue. Insight and capacity for change are part of that prediction — not optional extras." (FLAST takeaway)

Comments (0)