Blurred lines in limitation

March 2026

Martin Jensen

The recent Court of Appeal case of Ellen Kay v Martineau Johnson (A Firm) [2026] EWCA Civ 224 again highlights the complexity surrounding the application of section 14A of the Limitation Act 1980 in professional negligence claims. In this article we examine the divergence of views in the Court of Appeal as to issue of constructive knowledge as it applies to claims against solicitors.

The Facts

In 2007 the claimant, Ms Kay, engaged the defendant solicitors to act in her ancillary relief claim against her husband. In September 2008, a clean-break settlement agreement was concluded, notwithstanding reservations as to the husband’s financial transparency. Shortly after the settlement, the claimant queried with the defendant whether it could be reopened. It could not, the defendant advised, and they repeated that advice in May 2009, before ceasing to act.

As matters played out the claimant did not emerge from her divorce feeling flush, whereas as the years went by in 2017 she came to notice conspicuous displays of wealth by her ex-husband. This led to the claimant requesting her file from the defendant in 2018. Some two years later in March 2020 she sought legal advice from a barrister, that advice being funded by her new partner. Again the advice was that the settlement could not be reopened, but that there might be a claim against the defendant for their advice at the time it was entered.

Almost another three years went by before the claimant issued a negligence claim against the defendant in March 2023. She contended that in 2008 the firm had failed to advise her properly on investigating her ex-husband’s means and had failed to advise of the possibility of obtaining an NSMO (a nominal spousal maintenance order). Unlike the clean-break settlement the claimant was saddled with, obtaining an NSMO would have allowed her to revisit matters in the event her husband’s financial circumstances improved (and the defendant conceded it was likely an NSMO would have been made if the ancillary relief application reached a final hearing).

The defendant solicitors denied negligence, but also relied on a limitation defence, with the result that there was a trial on a preliminary issue as to whether the claim was statute-barred.

The Preliminary Issue

Section 14A sets out the relevant knowledge (whether actual or constructive) a claimant must have before the three-year period will star to run under that section. Amongst other matters, this must include knowledge both:

of the material facts about the damage in respect of which damages are claimed – subsection 6(a)
that the damage was attributable in whole or in part to the act or omission which is alleged to constitute negligence – subsection 8(a)

For the purposes of subsection 8(a) – what we refer to as ‘attribution knowledge’ – a claimant must have some reason to believe that there has been something remiss with the advice the defendant provided, although a claimant need not know that the acts or omissions constitute negligence.

However, it is unnecessary for a claimant to have actual knowledge of the relevant matters. Constructive knowledge will suffice. Therefore, a claimant’s knowledge of relevant matters will include additional knowledge they “might reasonably have been expected to acquire”:

from facts observable of ascertainable by them – subsection 10(a), or;
from facts ascertainable by them with the help of appropriate expert advice which it is reasonable for him to seek– subsection 10(b).

If it would have been objectively reasonable to seek expert input as to whether an adviser might have given incorrect advice, a claimant will be fixed with the knowledge that such expert advice would have imparted. The exception to that (referred to in the case as the ‘proviso’ to subsection 10(b)) is that a claimant will not be fixed with such constructive knowledge as they would have acquired from an expert provided that they have taken “all reasonable steps” to obtain and act on that advice.

In this case, the claimant plainly had knowledge of the damage around the time of entering the settlement agreement, since she believed at that time that her husband had been less than candid about his financial affairs. However, she claimed she lacked the attribution knowledge required under subsection 8(a) – namely that her being short-changed in the ancillary relief claim was in some way attributable to a shortcoming in the advice she had received from her solicitors.

The claimant’s case on attribution knowledge was that she had asked the defendant solicitors in 2008 and 2009 whether it was possible to reopen the settlement and on both occasions was told that it was not. At that stage, she had no reason to believe there had been fault on her solicitors’ part. It was not until she obtained counsel’s opinion many years later that she had cause to suspect the defendant and acquired the relevant attribution knowledge, and her claim was brought within three years of that.

In October 2024, the Judge rejected the claimant’s arguments. He found that the claimant had acquired actual attribution knowledge by the end of 2009, having consulted the defendant at that time. The claimant had retained the defendant to assist her in obtaining what she regarded as a “fair deal” in terms of the distribution of wealth in her ancillary relief claim. She realised in 2009 that this had not been achieved. As the Judge found, “she would or should have asked herself: ‘How is it that, with the supposed benefit of legal advice, I have found myself in this position?’.” He went on to find that, “The very nature of the later advice [i.e. in May 2009] (discouraging as it was in relation to the likelihood of success on any attempt to re-visit the Settlement) was such that it raised a suspicion about the soundness of the earlier advice.” If that did not amount to actual knowledge, the Judge held it was sufficient to trigger the requirement to seek expert input at that stage, thereby fixing the claimant with constructive attribution knowledge in any event.

In the further alternative, the Judge found that if the claimant did not have actual or constructive knowledge in 2009, she acquired constructive knowledge by late 2018, in that she ought to have obtained expert advice by that stage – upon realising her husband’s vastly improved means and having requested the defendant’s file – which would have identified the possibility that the defendant had been negligent.

The Judge accordingly held that the claim was statute-barred, prompting the claimant’s appeal.

The Court of Appeal

The Court disagreed with the Judge’s finding that the claimant had acquired actual attribution knowledge in 2009, after consulting the defendant about the possibility of revisiting the settlement.

Newey LJ reviewed the authorities and noted in the context of professional advice that it is not enough that a claimant knows that they have acted on the basis of professional advice and have suffered loss. Where the defective nature of the advice is not apparent on its face, something more is needed to put the client on notice that they might not have received proper advice before they can be expected to investigate matters further. That being the case, the Judge had failed to identify what that ‘something more’ was in this case. It did not necessarily follow from the claimant not having ultimately obtained what she considered to be a ‘fair deal’ that this was the result of any negligence, and the claimant herself did not actually draw such inference. That was fatal to any argument as to actual knowledge and the Court therefore rejected the Judge’s finding that the claimant had actual attribution knowledge in 2009.

As Males LJ cautioned, it is easy to blur the distinction between actual and constructive knowledge and important to keep them distinct when considering the application of section 14A. Whilst the claimant was aware that things had not turned out as anticipated, that did not amount to actual knowledge that her predicament was attributable to an act or omission which it is alleged to have been negligent. She did not then have actual knowledge such as would justify “embarking on the preliminaries to the issue of proceedings, such as submitting a claim” (per Lord Nichols in Haward v Fawcett [2006] UKHL 9).

As to then whether the claimant had constructive attribution knowledge in 2009, there was a divergence of views in the Court of Appeal. Here, as Lewis LJ observed, the point of disagreement between Newey LJ and Males LJ (with whom Lewis LJ agreed) was as to the application of the proviso in subsection 10(b) and whether it could be said that the claimant had nevertheless taken all reasonable steps to obtain expert advice by consulting with the defendant at the outset.

Here, Newey LJ observed that the Judge had not expressly considered the proviso’s application in circumstances where the claimant had in fact specifically asked the defendant in 2009 if the settlement could be reopened and was advised that it could not. Whilst he found the issue not an easy one to determine, he concluded that the claimant’s approach to the defendant at that time satisfied the proviso’s requirements of taking all reasonable steps to obtain legal advice. She had taken those steps but had still not acquired actual attribution knowledge, and therefore she should not be taken to have acquired such knowledge constructively.

Males LJ and Lewis LJ disagreed. Whilst it was reasonable for the claimant to have accepted the defendant’s advice that the settlement could not be reopened, that did not address the issue of whether finding herself in that predicament was due to bad advice; in fact, confirmation that she was stuck with a bad deal should have raised suspicion about the soundness of the earlier advice, as the Judge had observed. That was sufficient to make it reasonable to then begin investigating whether indeed the defendant had in some way been at fault. Here, crucially, the test is objective. Even though the claimant herself did not suspect fault, objectively there were grounds for suspicion and so the claimant was fixed with the attribution knowledge she would have acquired had she investigated at that point.

Whilst the majority were therefore in agreement that the claimant had constructive attribution knowledge in 2009, the Court was unanimous on the alternative (and by then academic) question as to when she might later be fixed with constructive attribution knowledge after obtaining the defendant’s file in 2018. Here it was the claimant’s case that her delay in obtaining legal advice at that stage was simply because she could not afford it. However, the Court was satisfied that a claimant’s personal characteristics are generally not relevant to the question of constructive attribution knowledge, with Newey LJ concluding that it was questionable whether a claimant’s lack of resources can ever be relevant justification for delay in obtaining expert advice once it is objectively reasonable to do so (but if it was, a claimant would have to adduce compelling evidence that a lack of means genuinely prevented them from seeking out the advice they would have been expected to acquire).

Comment

The decision provides useful guidance in respect of claims involving the application of s14A and is particularly helpful to those seeking to advance a limitation defence against them. It makes clear that the objective test for constructive knowledge places the onus squarely on a claimant to begin investigating negligence as soon as it becomes apparent that a transaction has not resulted in the outcome expected, even though the claimant may not at that time suspect fault on the part of their advisers. Moreover, consulting with one’s former advisers may not necessarily displace the requirement on a claimant to seek expert advice elsewhere (particularly since those who have given negligent advice might be thought less likely to recognise that possibility), and asserting a lack of means to justify a delay in doing so is unlikely to assist.

Defendants will therefore want to scrutinise such claims with particular care to ascertain not merely when the claimant themselves first suspected that something had gone wrong with the advice received but when suspicion should have first been aroused.

If you have any questions in relation to the information discussed in this article, please contact Martin Jensen.

Download PDF