The conduct rules of the Australian Bars all impose a “sitting out” period on former judges before they may appear as counsel in their former court. The conventional view has been that compliance with the prescribed period is the relevant safeguard: serve the time, return to practice. Brasch J’s decision in Tanit & Tanit [2026] FedCFamC1F 171 puts that view under real pressure. Compliance with the cooling-off rule, Her Honour holds, is necessary but not sufficient; the apprehended-bias test does the heavier work, and a presiding judge may need to recuse on their own motion notwithstanding that the bar has been formally cleared.
What Tanit decides
The proceeding was a case-management hearing in BFA-set-aside litigation under Pt VIIIA of the Family Law Act 1975 (Cth). A retired Family Court judge — anonymised in the published reasons as “Mr B” — announced his appearance as senior counsel for the husband. Brasch J recused herself on apprehended-bias grounds, applying the orthodox Ebner formulation: a judge is disqualified where a fair-minded lay observer might reasonably apprehend that the judge might not bring an impartial mind to the resolution of the question to be decided: Ebner v Official Trustee in Bankruptcy (2000) 205 CLR 337, [6]. The “double might” test was reiterated in Burgess v Minister for Immigration and Border Protection (2018) 259 FCR 197, [36] (Charlesworth J); and the two-step structure — identifying the matter said to skew the decision-maker, then articulating a logical connection to the feared departure — was reaffirmed by the Appeal Division in Henley & Bestari (2024) FLC 94-169, [25]–[26].
Brasch J located the issue within the Webb v The Queen (1994) 181 CLR 41 “conduct” and “association” categories: Mr B had conducted himself as a judge of the same court for many years and now proposed to conduct himself as senior counsel in it. Section 8 of the Federal Circuit and Family Court of Australia Act 2021 (Cth) was deployed to close off any argument that the abolished Family Court was a separate institution: it is continued in existence as Division 1.
The pivotal move is Her Honour’s treatment of the conduct-rule cooling-off period. Mr B had “sat out his time” under the Bar Association of Queensland’s Barristers’ Conduct Rules and the cognate uniform rules. Her Honour accepted that, but treated compliance as necessary, not sufficient: “the proper administration of justice must triumph over time out periods”. The AIJA Guide to Judicial Conduct (4th ed, 2026) was marshalled for the proposition that those timelines are “minimum standards only” — a floor, not a safe harbour. Brasch J was also alive to two symmetrical apprehensions: that the husband had “the inside running” through his choice of counsel, or, conversely, that the court might overcorrect against the former judge’s client to avoid the appearance of favour. Neither is consistent with justice being seen to be done. The Appleby and Blackham article Her Honour relied on — Gabrielle Appleby and Alysia Blackham, ‘The Growing Imperative to Reform Ethical Regulation of Former Judges’ (2018) 67(3) International and Comparative Law Quarterly 505 — adds the further dimension of former judges making submissions on precedents they themselves decided, and the subterranean optics of return to practice while drawing the judicial pension.
The conduct-rules landscape
The uniform Legal Profession Uniform Conduct (Barristers) Rules (adopted in NSW, Victoria and WA, with cognate provisions in Queensland) impose “sitting out” periods on former judges before appearing in the court on which they served. The proposition Tanit drives home is that conformity with whatever the rule prescribes does not pre-empt an apprehended-bias challenge — and on Ebner, may not stop a presiding judge concluding, on the court’s own initiative, that recusal is required.
Does Tanit reach below the judicial tier?
A natural follow-up for practitioners considering registrar or judicial-registrar appointments is whether the decision applies to quasi-judicial offices. The textual answer is no: a Deputy Judicial Registrar exercises delegated power and the office is not co-extensive with judicial appointment. The specific concerns Brasch J articulated, counsel pressing precedents they themselves authored, the judicial pension complicating the optics, “subconscious” deference between former colleagues, apply with markedly diminished force, if at all, to registrar-tier roles. The Appleby and Blackham analysis is expressly directed to judicial return to practice. That said, Ebner is fact-sensitive.The category matters; the analysis does not stop there. This much is my own inference rather than authority: the principle in Tanit should be read as a doctrinal one, not a status-based one, and its reach down the registry tier will be calibrated to the strength of the association in each case.
The live points
Settled: the Ebner test; that the cooling-off rules in conduct codes are not determinative of the apprehended-bias question (a straightforward application of Ebner‘s primacy over a discretionary professional standard); and that “sitting out” is a regulatory minimum, not a ceiling, as the AIJA Guide expressly states.
Live: whether Tanit will be followed at first instance with any consistency — particularly where the parties consent, as the wife arguably did in Tanit itself, and whether the conduct rules will be reformed to align with the apprehended-bias floor rather than sit beneath it. A respectable contrary view exists: the wife did not object, the husband had selected his counsel, and Ebner itself cautions against treating apprehended bias as a vehicle for judicial squeamishness or for displacing a litigant’s choice of representation without good cause: see Western Australia v Ward (1997) 76 FCR 492, 498; Grimwade v Meagher [1995] 1 VR 446, 450. Equally, Brasch J’s reasoning is squarely orthodox; an appellate court asked to review a recusal of this kind would be unlikely to interfere.