Mazur in the Court of Appeal — who is really “conducting litigation”, and how the Court is likely to decide

Friday 27 February 2026

The Mazur appeal has become a proxy battle over the operating model of modern litigation in England and Wales. On one reading it is a narrow point of statutory construction under the Legal Services Act 2007. On another it is a question about what the public is entitled to assume when she instructs a regulated firm — and whether Parliament meant the regulated solicitor to be the human “gate” through which every procedural act must pass.

The Court and the parties

The appeal (CA-2025-002754) is being heard by a three-judge Court of Appeal panel comprising:

• The Master of the Rolls, Sir Geoffrey Vos

• The Chancellor of the High Court, Sir Colin Birss

• Lady Justice Andrews 

The order published by the judiciary also sets out the alignment of the parties:

• Appellant: Chartered Institute of Legal Executives (CILEX). 

• Respondents in substance: Julia Mazur and Jerome Stuart (the defendants in the original debt claim who challenged the use of a non-authorised fee earner), and Charles Russell Speechlys LLP (the firm pursuing the underlying fees via instructed solicitors). 

• Interveners opposing CILEX’s appeal: the Solicitors Regulation Authority (SRA), the Law Society of England and Wales, and the Legal Services Board (LSB). 

• Interveners supporting CILEX’s appeal: the Association of Personal Injury Lawyers (APIL) and the Law Centres Network (the Law Centres Federation). 

That coalition tells its own story. CILEX, APIL and the Law Centres Network are defending a labour structure that has grown up inside litigation over decades — teams where authorised litigators supervise, but do not personally execute, a very large proportion of day-to-day procedural work. The SRA, the Law Society and the LSB are defending the conceptual clarity of “reserved legal activity” — and the consumer-protection instinct that the person who does the reserved act should, in some meaningful sense, be a person the regulatory system can directly discipline. 

The legal kernel — what Sheldon J actually decided

The High Court decision under appeal (Mazur v Charles Russell Speechlys [2025] EWHC 2341 (KB)) arose from an everyday-looking debt recovery claim — except that the Particulars of Claim were signed by a “Senior Litigation Executive” who was not a solicitor. The defendants argued that this was the unlawful conduct of litigation by a non-authorised person. 

Sheldon J’s key conclusions, as summarised by subsequent practitioner commentary, were these:

• Supervision does not confer authorisation — a non-authorised individual may not conduct litigation merely because he or she acts under the supervision of an authorised person. 

• Parliament created express supervised-delegation carve-outs for certain reserved activities (for example, limited advocacy rights in chambers under Schedule 3) but did not create an equivalent carve-out for the general “conduct of litigation”. The absence was treated as deliberate. 

• The judgment did not finally map every task onto one side or the other of the line — it distinguished “supporting” litigation (permitted) from “conducting” litigation under supervision (impermissible), but left the most operationally important question — where that line sits in practice — only partially answered. 

This last point is why the appeal has detonated across the profession. If the forbidden category is “conducting litigation”, but the statute’s definition remains high level, then uncertainty becomes contagious — and uncertainty in litigation becomes a costs weapon.

The arguments — and their relative strengths

1) CILEX’s case: Parliament did not mean to break settled practice

CILEX’s “core submission”, as reported in legal media coverage, is essentially historical and purposive:

• There has been long-standing custom and practice of solicitors conducting litigation through supervised non-solicitor employees — for “decades, if not centuries” — and the 2007 Act was not intended to criminalise that embedded reality by accident. 

• If a properly supervised employee does work for the purpose of conducting litigation, CILEX says the employee is not his or herself “carrying on” the conduct of litigation — rather, the authorised litigator is doing so through him or her, taking responsibility in law. 

The strength of this argument is practical legitimacy. Courts are reluctant to discover, late in the day, that ordinary litigation across the country has been structurally unlawful — particularly where Parliament’s broad reform aim in 2007 included liberalising legal services markets, not tightening them into a brittle guild. That intuitive reluctance will matter to a panel led by Vos MR, who is acutely alive to systemic consequences.

The weakness is that the statutory language is penal. Where an Act creates a criminal offence for unentitled reserved activity, appellate courts tend to be cautious about reading in implied exceptions which Parliament did not write down — especially when Parliament demonstrably wrote down other exceptions elsewhere. Sheldon J’s reasoning has a clean textual spine. 

2) The Law Society and the SRA: the point of reservation is personal entitlement, not merely firm-level responsibility

The Law Society’s publicly reported position is crisp:

• Conduct of litigation is a reserved legal activity which “may only be carried out by authorised persons” — even inside regulated entities — while non-authorised staff may provide extensive assistance if the authorised person retains responsibility, actively supervises, and key decisions and formal steps are referred back for professional judgment. 

The SRA’s guidance, while careful not to present itself as determinative, pushes in the same direction — stressing “substance over form”, and asking who truly assumed responsibility and exercised judgment over the progression of the matter. 

The strength here is statutory clarity aligned with consumer protection. If the reserved activity can be “done” by an unlimited number of non-authorised staff so long as one authorised person exists somewhere above them, then the idea of reservation risks becoming a badge, not a constraint — and the defendants’ point (reported in legal press coverage) about circumvention starts to bite. 

The weakness is operational realism. The Law Society’s own formulation — “extensive assistance” but not “conduct” — is exactly where the fog lives. Without a workable, litigation-proof test, the strict reading becomes an invitation to satellite disputes, opportunistic strike-out points, and costs challenges. The Court will not like a rule that turns case management into an argument about job titles.

3) The access-to-justice interveners: law centres and mass claims cannot function under a maximalist reading

The Law Centres Network has framed the problem as definitional — asking for a concept of “conduct” that is workable and proportionate, and highlighting the difficulty of distinguishing “conduct” from “support” in daily practice. 

APIL’s support, as described in legal reporting, is rooted in the economics of volume litigation — if the decision is upheld without clarification, the cost base of personal injury and similar work shifts upwards sharply, and capacity tightens. 

This line of argument is not, strictly, about what the statute “means”. It is about what kind of legal system Parliament can realistically have wanted — and about the collateral damage of a judicially created bottleneck in a jurisdiction already struggling with delay.

Prediction — what the Court of Appeal is most likely to do

The Court of Appeal is unlikely to leave Mazur exactly as it stood after Sheldon J — not because the textual argument for upholding it is weak, but because the judgment’s practical incompleteness is corrosive. The Court has a strong incentive to produce a clarifying rule that reduces downstream litigation about the meaning of “conduct”.

The most probable outcome is therefore a split-level result:

• The Court will probably preserve the core principle that “authorisation” is personal — supervision alone does not magically make an unauthorised person entitled in his or her own right.

• But the Court is also likely to hold that many activities performed by supervised employees are, in law, done on behalf of — and as part of — the authorised litigator’s conduct of litigation, provided the authorised person truly directs the case, makes the key decisions, and takes responsibility in a meaningful way.

In other words the Court is likely to aim for a practical accommodation — a rule that prevents sham supervision, but does not force every case to be run as though only the authorised person may touch the file. That approach is consistent with the profession’s need for certainty and with the judiciary’s institutional dislike of rules that generate procedural guerrilla warfare.

If the Court goes this way, the judgment will probably include:

• A more concrete list of tasks that are strongly indicative of “conduct” (issuing, signing statements of truth where required, filing steps that commit the party, notices of appeal, formal service decisions) — and a safer harbour for preparatory work done under documented supervision.

• A warning that nominal supervision is not enough — firms will need demonstrable systems, not simply a name at the top of a matter.

That would allow the Court to say — in the careful language appellate courts prefer — that Sheldon J was right about authorisation not arising by osmosis, but wrong to the extent his reasoning is read as criminalising the ordinary mechanics of supervised litigation teams.

Why that prediction fits this bench

This is a formidable panel, and it has been convened precisely because the system needs an answer, not merely a victory. The Master of the Rolls will be alive to the downstream disruption risk, Birss LJ to coherent statutory interpretation that still works in practice, and Andrews LJ — who has deep experience with procedural realities — to the difference between genuine responsibility and a paper fiction. The judiciary’s own decision to manage access to the hearing as a matter of “great public interest” is a small institutional indicator — the Court knows it is writing for the whole market, not only for the parties.