Freedman J’s decision in Mather v Ministry of Defence [2021] EWHC 811 (QB) – which can be read here – is interesting for (at least) two reasons. First, it demonstrates the court’s reluctance to order a preliminary issue trial in a complex and novel claim. Secondly, it foreshadows what promises to be a significant disease trial in 2022, with some fascinating arguments on causation that have the potential to end up in the Supreme Court.

The Claim in Outline

The Claimant worked for the Defendant as a painter and finisher with the RAF between 1989 and 2003. He alleges that his use of organic solvents during that employment has caused multiple sclerosis (“MS”) and psychiatric injury, and that these injuries were caused by the Defendant’s negligence and/or breach of statutory duty under the Control of Substances Hazardous to Health (“COSHH”) Regulations. The Defendant contends that the claim is statute-barred and, in any event, denies breach of duty, causation and loss.

Unsurprisingly, the question whether the Claimant can show that his MS was caused by his work for the Defendant is a crucial one. As is common in cases of chemical exposure, there are two aspects to this question viz. (1) whether exposure to the organic solvents in question can cause MS (so-called ‘generic causation’), and, if so, (2) whether that exposure did cause the Claimant’s MS (so-called ‘individual causation’).

The Defendant’s Application

The Defendant, confident that it will succeed on causation (thus rendering the trial of other issues otiose), applied to have causation tried as a preliminary issue in one of two ways:

  • by defining a preliminary issue on causation with the Claimant’s best case on exposure being the assumed facts, and limiting oral evidence to medical causation; or
  • by formally opening the trial, limiting oral evidence to that relevant to medical causation, and then making a ruling on causation, with other issues to follow if necessary.

The Claimant’s position, in terms, was that a preliminary issue trial on causation would increase rather than decrease complexity and expense.

Foreshadowing the Arguments as to Causation

The parties’ respective arguments as to causation were adumbrated by Freedman J at [12]-[19].

In short, the Defendant’s case is that proof of generic causation – that the “exogenous environmental factor” in question is capable of causing MS – requires “proof that the exogenous factor more than doubles the risk of the index condition”: see [12]. The Defendant further contends that MS is an indivisible disease, such that the ‘material contribution’ approach to causation does not apply: see [15].

The Claimant takes issue with the Defendant’s approach for the reasons recorded at [18]:

The Claimant submits that the law on causation is less clear than that advanced on behalf of the Defendant. In particular, he raises the following issues, namely:

(1) Whether MS is a divisible or indivisible condition. The Defendant’s case is that it is indivisible and the Claimant’s answer to this is that it is often not simple to say whether a given disease is or is not indivisible. Attention is drawn to Bonnington which was assumed to be a divisible disease case (pneumoconiosis), but the Privy Council in Williams v Bermuda [2016] AC 888 stated at [32] that there was no suggestion that it was divisible in Bonnington. It is therefore submitted that there is scope for uncertainty as to what is divisible and what is not.

(2) Whether the psychiatric injury in this case is divisible even if the MS is indivisible. Generally psychiatric injury has been treated as divisible. Despite this, it is difficult to understand if in fact the psychiatric injury is a consequence of the MS, how it could be treated as anything other than the designation of the MS. If the MS were then treated as indivisible, it may be difficult for the Claimant to show that the psychiatric injury was divisible.

(3) If in fact MS is indivisible, there is said still to be scope for an argument that a material contribution test may be applied to an indivisible case. This appears to be contrary to the understanding of the Court of Appeal in Heneghan at para. 23. However, in Williams v Bermuda, there is scope for this argument. There have been subsequent authorities doubting the approach of Williams v Bermuda, but the authorities have not all spoken with one voice.

(4) Alternatively, if MS is indivisible and even if there is no scope for the material contribution test, an issue might arise as to whether this case nonetheless comes within the third category of Heneghan such that causation may be established if it is proved that the exposure to solvents materially increased the risk of the Claimant contracting an indivisible disease. This would involve the extension of the Fairchild exception to a case of MS.

(5) The Claimant submits in respect of various diseases the Court has not required the doubling of risk to be proven e.g. mesothelioma: see Sienkiewicz v Greif [2011] UKSC 10; palmar arch disease: see Transco v Griggs [2003] EWCA Civ 564 per Hale LJ at para. 35; dermatitis; see McGhee v National Coal Board [1973] 1 WLR 1. Whether it is required in connection with lung cancer is questionable: yes, per Smith LJ in the CA, but obiter no, per the Supreme Court at para. 35. A potentially important case of a claimant succeeding against the Ministry of Defence even although doubling of risk was not proven is the case of Wood v Ministry of Defence [2011] EWCA Civ 792 (a case about exposure to organic solvents and Parkinson’s disease). This is therefore a controversial area which provides a focus for a debate as to whether a claimant is required to prove doubling of risk in the context of contracting MS.

Point (3) is particularly interesting. It was certainly once thought that the material contribution test could only operate in cases where the injury was divisible: see B v Ministry of Defence (The Atomic Veterans’ Litigation) [2010] EWCA Civ 1317 at [150] per Smith LJ for the court and Heneghan as cited above. This was doubted by Lord Phillips of Worth Matravers in Sienkiewicz at [90], where his Lordship focussed on the tortious contribution “to the cause” of the indivisible disease. The point was seemingly resolved by the Privy Council in Williams at [32]-[33] per Lord Toulson (citing Lord Phillips’ observations), but it is right to say that the issue remains controversial with creditable arguments both ways. There is also an argument that Lord Toulson’s observations in Williams were obiter (though this was rejected by HHJ Auerbach in Davies v Frimley Health NHS Foundation Trust [2021] EWHC 169 (QB) at [207]).

Point (4) is similarly interesting. Following Heneghan, it is possible to argue for the extension of the Fairchild exception but there has been notable judicial reluctance so to do. In a sense, policy considerations (see e.g. Lord Brown in Sienkiewicz at [187]) point in one direction, whereas a mechanistic application of the Fairchild criteria (see Lord Dyson in Heneghan at [47]) points in the other.

In this vein, Freedman J noted at [22] that:

Ms Harrison QC [for the Defendant] recognised entirely properly and realistically that there are areas in this case of a tertiary nature, in other words, capable of going not only to a second court (a first appeal), but even to a third court (a second appeal). To that end, she recognised the scope of the third and fourth points made by the Claimant at paragraph 18 above.

Refusal of the Defendant’s Application

Against the above background, Freedman J refused to order a preliminary issue trial of causation. His reasons were summarised at [35]:

Applying the criteria considered by David Steele J in McLoughlin v Grovers (A Firm) to the instant case and using the numbering therein, the following conclusions arise:

(1) This is a case where a finding in favour of the Defendant on causation, whether generic or individual (as referred to in paragraph 16 above), would be decisive. However, a finding in favour of the Claimant on causation would involve going back to have to consider the nature of the duty and breach of duty, and then having to visit causation all over again to assess whether the specific breaches (if any) of the duties (to the extent that any were found) caused the loss. This would bring with it the possibility of the Judge trying stage 2 having a different view of the evidence from those found at stage 1. There would then be issues as to how far those findings bound the court at stage 2. This would create especial difficulties if stages 1 and 2 were tried by different judges but might even be challenging if they were tried by the same judge.

(2) The issue on causation involves questions of law. However, it also involves complex factual and scientific issues. In the end, it is not a crisp issue to be decided, but it is multi-factorial, where the factors are a mixture of law and fact.

(3) This then gives rise to the difficulty of agreeing facts for a preliminary issue. There are issues regarding the period of the exposure to take into account and in particular the issues as to whether there should be disregarded (a) any exposure after 1994, and (b) any exposure after 2000. There is the issue as to whether the evidence of the 25 factual witnesses is required and the occupational hygiene experts in the first trial. At lowest, there is great effort in identifying the relevant facts. There is no reasonable prospect of agreeing them.

(4) A split trial between causation and other liability issues would be likely to cause significant delay as a result of a likely appeal. The prospects of an appeal at the end of stage 1 are significant, and the effect of that on the trial issues is considerable and possibly intolerable delay. Further, if the Claimant succeeds at stage 1, there are the difficulties of managing the case at stage 2 (see point (i) above), especially if that follows a long delay following an appeal.

(5) Master Thornett was right to reserve this matter to a case management conference. Many issues have been thrown up. A decision not to order a trial of a preliminary issue does not mean that nearer trial matters cannot be revisited if there are changes in circumstances. However, it would have to take into account the matters set out in this judgment.”

Freedman J went on to note at [39] that “[i]t does seem sensible to separate quantum if that can be done easily, but before making such an order, the Court would wish to hear more about how that would work in practice … because it did not arise on [the Defendant’s] application.”


This case is sure to give rise to some fascinating arguments – at first instance, and possibly on first or even second appeal – not only in relation to causation but also breach of duty and the COSHH Regulations. Those questions of liability will be considered together as opposed to separately. It is suggested that this is the right approach in a case as complex and novel as this. As Lord Scarman observed in Tilling v Whiteman [1980] AC 1, 25: “Preliminary points … are too often treacherous short cuts.”