Roche v Secretary of State for Defence

MR JUSTICE DAVIS: The Chemical and Biological Defence Establishment of the Ministry of Defence is better known to the public by the name of its principal location, Porton Down. Its practices and procedures over the years have been the subject of both confidential reports and open debate, and from time to time have been modified: for example, as one instance, following the death of a volunteer soldier in 1953 after tests involving nerve gas agents had been conducted. There have been those over the years, both in Parliament and outside Parliament, who have pressed for epidemiological research into the long-term effects, if any, in relation to chemical and biological warfare testing at Porton Down involving the use of volunteers from the armed services. Until relatively recently, however, those proposals have been rejected, the view stated by the various governments of the day being, in essence, that there was no evidence that volunteers had suffered long-term damage and therefore there was no justification for such research.

The appellant, Thomas Roche, who was born on 14th September 1938, served in the Royal Engineers between February 1954 and 1968, when he was discharged on medical grounds, those grounds being agreed to be wholly unrelated to the present claim.

In 1962 Sapper Roche volunteered to undergo tests of chemical weapons at Porton Down. He was paid extra wages for his time spent there.

What happened is conveniently recorded in the decision of the Pensions Appeal Tribunal, the subject of this present appeal.

It was noted that no documentary evidence existed in respect of the incidents relating to the 1962 visit. However, the evidence of Mr Roche, which was accepted, was that the 1962 testing was for mustard gas. He, together with five or six fellow soldiers, were seated in a small room for about six hours. Patches of cloth were attached to their arms and a liquid applied. At the end of the six hours the cloth patches were removed and the soldiers were allowed to return to their units.

Mr Roche could not recall any side effects either immediately after the tests or in the weeks and months that followed. It was noted that in 1963 he was treated for an acute attack of bronchitis, but it was found that that was not linked to the tests.

In July 1963 Mr Roche returned to Porton Down for a second series of tests. Those tests were partially documented. Mr Roche, in the first testing period, again underwent the mustard gas tests in exactly the same format as described already, the only side effect being a slight reddening of the skin beneath the patches, although he did recall one of his colleagues suffering a small amount of blistering. In addition, however, Mr Roche underwent a GF gas single breath inhalation test, GF being a kind of nerve gas. This was, it appears, in around the middle of July 1963. For the purpose of that testing Mr Roche was required to enter a sealed airtight compartment and was fitted with a face mask, through which a single dose of GF gas was administered, after which he removed the mask and exited the chamber. On that occasion Mr Roche, as he recalled, experienced an immediate reaction of a tightening of his chest and momentary shortness of breath. Apart from those immediate effects, there were no other immediate effects and no other effects noted in the weeks and months that followed. There were no accurate records as to the doses administered in any of the tests, although certain estimates had been prepared.

Following his discharge Mr Roche's medical history continued to be, as it was found, "unremarkable". In 1981 his medical records indicate him as visiting his doctor, complaining of shortness of breath and wheeziness. There was no family history of asthma and Mr Roche had apparently stopped smoking cigarettes in 1969, although he may have continued occasionally to smoke a pipe. He was diagnosed at that stage as suffering mild hypertension and late onset bronchial asthma.

In 1989, however, he was found to have high blood pressure and was diagnosed as suffering from Chronic Obstructive Airways Disease ("COAD"). A doctor suggested that a possible link with the Porton Down tests should be further investigated.

The upshot was that in due course Mr Roche applied on 10th June 1991 for a war disability pension, citing hypertension and breathing problems. That claim was rejected in January 1992. There was then a significant period of delay until Mr Roche appealed against that rejection by an appeal lodged on 8th November 1998. COAD was cited as the condition in respect of which he was claiming.

It appears that in the meantime Mr Roche had been investigating the possibility of a civil claim against the Ministry of Defence. A letter intimating such a claim was sent. In the result a certificate, pursuant to section 10 of the Crown Proceedings Act 1947, was issued on behalf of the Secretary of State on 3rd August 1995. That stated as follows:

One consequence of that certificate was that Mr Roche was thereby, as a matter of English law, prevented from bringing civil proceedings against the Ministry of Defence. It has been suggested that, had he not been so prevented, he probably would have been granted legal aid. That I think, speaking for myself, is very debatable given that both then and now Mr Roche had not, and has not, sufficient evidence to show an arguable case that his ailments were, on the balance of probabilities, due to the tests conducted at Porton Down and so attributable to his military service; and one may query whether legal aid funds would have been advanced to enable inevitably extensive and complex medical and scientific research to be undertaken under the guise of public funding for litigation purposes. It is not, however, necessary for me to say more than that. I was in fact told that a challenge to the validity of the section 10 certificate is due to be heard in the European Court of Justice in the very near future.

However, the provisions of the Naval, Military and Air Forces (Disablement and Death) (Services Pension) 1983 Order were potentially available in appropriate cases to provide compensation for those who might not otherwise be able to recover any compensation. The provisions of the 1983 Order impose a very decidedly lower standard of proof on an applicant than otherwise would ordinarily be the case for civil proceedings commenced in England and Wales.

The relevant provisions of the 1983 Order read as follows. By article 3 it is provided:

Article 5 is the crucial article for present purposes. In the relevant respects it reads as follows:

It was by reference to the 1983 Order that Mr Roche initiated and then pursued his claim for a war disability pension, on the footing that his disablement was due to an injury attributable to his service.

Mr Roche's appeal eventually, after one or more previous hearings, came on for hearing before the Pensions Appeal Tribunal on 23rd October 2003. Adjournments had previously had to be granted. In addition, there was a regrettable muddle as to whether hypertension was also to be the subject of the appeal. The Pensions Appeal Tribunal ultimately, at the hearing of 23rd October 2003, gave permission to Mr Roche to pursue the appeal as to hypertension also out of time. But it adjourned consideration of that aspect to a later date, to enable the relevant evidential material to be considered and collated and responded to by the Ministry of Defence. The hearing and subsequent decision, therefore, was solely by reference to COAD.

By its decision released on 16th January 2004, the Pensions Appeal Tribunal rejected Mr Roche's appeal. It ruled, amongst other things, that there was no reliable evidence to suggest a causal link between the tests on Mr Roche at Porton Down for mustard gas or GF gas and the claimed condition; and that the test of reasonable doubt under article 5 was not met.

Section 6 of the Pensions Appeals Tribunals Act 1943 permits an appeal, with permission, on a point of law. The Pensions Appeal Tribunal subsequently refused permission to appeal; but permission was granted by Newman J after a consideration of the papers on 18th July 2004. It is in those circumstances that the appeal has come on for hearing before me.

The appellant, Mr Roche, was represented by Mr Pilbrow of counsel and the respondent Secretary of State by Mr Clive Lewis of counsel.

Some argument was addressed to me as to the meaning and effect of article 5 of the 1983 Order. As I see it, a number of points are clear enough.

I was referred to the decision of Newman J in the case of Busmer [2004] EWHC 29 Admin, a decision given on 20th January 2004. In the course of his judgment in that case, Newman J, after referring to a number of authorities, said this:

A useful illustration, by reference to its own facts, of the way in which article 5 works in my view can be found in the unreported decision of Drake J in the case of Westcott v Secretary of State for Social Services (11th December 1987). In that particular case the applicant claimed a pension on the ground that he suffered from osteoarthritis of the knees and lumbo-sacral spine said to be attributable to service. It was said that he had experienced a heavy fall while testing a parachute in Norway whilst serving in the 1st Airborne Division, the fall taking place in 1945. A surgeon put in a report which, amongst other things, said that it was absolutely impossible to give a firm indication as to whether Mr Westcott would not have suffered arthritis had he not had his accident during the war, and that in the view of the surgeon he should be given the benefit of the doubt in his case. Drake J indicated that the answer thus given was to the effect, "Yes, Mr Westcott's orthopaedic pathology involving the knees and spine could be related to heavy parachute landings". The Department for Health and Social Security for its part then obtained its own medical reports from its own medical division. Those reports were to the effect that the evidence pointed very strongly indeed to the conclusion that Mr Westcott's condition was entirely unrelated to any parachute landing and, on the contrary, was primary osteoarthritis particularly related to Mr Westcott's considerable obesity. Drake J agreed that the evidence could be said to point strongly against Mr Westcott's claim. But, as he pointed out, there was also some evidence which did support his claim. He noted that the tribunal there had not expressly rejected as unreliable the consultant orthopaedic surgeon's opinion procured by Mr Westcott, and in those circumstances the appeal had to be allowed.

On behalf of the appellant in the present case, Mr Pilbrow advanced, in summary, the following grounds of appeal. First, he submitted that the Pensions Appeal Tribunal had failed to apply the correct test under article 5, in that it had inadmissibly raised the evidential burden faced by Mr Roche and had paid too much attention to the concept of "reliable evidence" and had acted contrary to the decision of Newman J in Busmer. Secondly, it was submitted that, rather than asking whether a reasonable doubt had been raised as to whether all the conditions of article 5 had been fulfilled, the Pensions Appeal Tribunal had, as a matter of wrong approach, weighed, or at least appeared to weigh, certain evidence against other evidence. Thirdly, it was submitted that the Pensions Appeal Tribunal had misinterpreted the evidence of Professor Alistair Hay, the court appointed expert, which, on a proper analysis and when viewed as a whole and in the context of all the other evidence, did raise a reasonable doubt. Finally, it was submitted that the Pensions Appeal Tribunal had failed to understand that the lack of any epidemiological follow-up study went to the issue of whether or not a reasonable doubt had been raised on all the evidence.

On behalf of the respondent, Mr Lewis submitted that the Pensions Appeal Tribunal was required to consider the evidence before it in its totality; that it was entitled to attach the weight that it did attach to the unchallenged written evidence of Professor Hay, which evidence was addressed specifically to the case of Mr Roche; and was entitled to conclude on the evidence that no reasonable doubt was shown.

There was a considerable quantity of evidence before the Pensions Appeal Tribunal. This included a medical opinion dated 28th January 1999 from an army medical doctor. This stated an opinion that, in the light of the lapse of time between the Porton Down test and the manifestation of Mr Roche's symptoms and for other reasons there set out, it was contended as follows:

For his part, Mr Roche had already enlisted the support of Professor Hay (who I might add is now the Professor of Environmental Toxicology at the University of Leeds). Professor Hay had in fact provided to Mr Roche over the years, it would appear, a number of advices and reports. Latterly a series of questions were drafted by Mr Roche's advisers addressed to Professor Hay; this in fact being at the suggestion of the Pensions Appeal Tribunal after a previous hearing before it on 3rd October 2002.

The questions posed to Professor Hay in this way and his answers are of very considerable importance for the purposes of this appeal. The questions started by asking Professor Hay to consider the various documents in the light of the proposition that Mr Roche was attempting to establish symptoms of COAD, hypertension (with associated epilepsy) and depression. Thus it is to be noted that the questions were directed to wider conditions than simply COAD. Question 6 is phrased in this way:

The answers of Professor Hay were provided by a letter addressed to the Tribunal dated 19th February 2003. The answer to question A6 was in these terms:

Professor Hay went on to give further answers (in an interim passage having stated his own personal regret that no detailed study had ever been undertaken by the relevant Secretary of State). In answer to the question at B4, his answer was this:

The Secretary of State then commented on the various answers given by Professor Hay by his own statement, which is dated 14th October 2003. The Secretary of State accepted that the comments of Professor Hay had been impartial and evidence-based, and in terms stated that he was broadly in agreement with Professor Hay's arguments and conclusions within the limits of existing evidence. In paragraph 4 of that response, it may be noted, the view was indicated that in each of sections A, B and C of the questions and answers Professor Hay had gone on to discuss why, in the light of the case specific evidence, "... any such causal link is however unlikely in his expert view". In paragraph 6 this was stated on behalf of the Secretary of State:

It is perhaps to be noted that Professor Hay was not asked any further questions, nor was he asked to give any oral evidence at the hearing before the Pensions Appeal Tribunal in which he might have expanded or further clarified his answers; nor was Professor Hay, as it happened, present when Mr Roche gave his evidence orally before the Pensions Appeal Tribunal.

Also included in the voluminous evidence submitted to the Pensions Appeal Tribunal were numerous research papers and studies relating to biological and chemical weapons. For example, a report on the "Medical Effects of Chemical Warfare Agents" published by the Working Party on Chemical and Biological Warfare and produced, as I was told, in around 1991, dealt with delayed and long-term effects relating to the use, amongst other things, of mustard gas and nerve gas. In the course of that report this, amongst other things, was said:

In a report prepared by a panel operating in the United States of America on the "Possible Long-Term Health Effects of Short-Term Exposure to Chemical Agents", produced in 1984, detailed comments were also made. Amongst other things, it stated in the course of the report that with regard to mustard gas:

Another report to which my attention was briefly drawn was a report on "Delayed Toxic Effects of Chemical Warfare Agents", produced by the Stockholm International Peace Research Institute. As I understand it, that principally was concerned to stress the point that insufficient material and case studies were available to advance any firm views one way or the other.

Another report, prepared by Veterans at Risk and headed "The Health effects of Mustard Gas and Lewisite", produced in 1993, stated, amongst other things, in the course of that report that the evidence found indicated a causal relationship between exposure and a number of health conditions, including chronic respiratory diseases. It was also stated in the course of that report that indirect evidence, based on a review of the relationships between acute and chronic effects caused by other substances, suggested that the likelihood of long-term respiratory effects may not necessarily be linked to the presence of an acute respiratory response.

It seems, on the whole, fair to say, and as the Pensions Appeal Tribunal itself pointed out, that, speaking generally, the reports were generalised reports; and throughout there seems to have been a constant refrain as to the insufficiency of data and case studies, which restricted the findings or observations that could usefully be made.

I turn then to the actual decision of the Pensions Appeal Tribunal itself. This set out the background facts, much of which I have already recorded. The Tribunal expressly found, I should add, that Mr Roche was an honest and truthful witness, albeit his recollections were understandably affected by the lapse of time.

The Tribunal went on, at page 4 of its report, to refer to the testing which Mr Roche underwent and found that the research papers, some of which I have mentioned, contained generalised conclusions. The Tribunal then referred to the evidence of Professor Hay and stated:

The Tribunal went on to summarise Mr Roche's case to the effect that the long-term effects of the exposure had caused or contributed to his current respiratory problems. The Tribunal went on to say this in the course of summarising the case:

The Tribunal then went on to summarise the Secretary of State's contentions that there was simply no evidence to support Mr Roche's claim. The Tribunal then proceeded to direct itself as to the law. It referred to article 5 and stated as follows:

The Tribunal then went on to deal with a further submission made by counsel then appearing for Mr Roche about the lack of testing evidence, which I do not think it necessary to set out in full here.

The Tribunal then came to its findings, which can be found at page 6 of its decision. It found expressly that Mr Roche suffered no long-term respiratory effects from skin contact with mustard gas following the tests in 1962 and 1963. It found expressly that Mr Roche was administered only small doses of mustard gas and GF gas, which would have resulted in minimal exposure to mustard gas by off-gassing and a limited and transitory reaction to the GF gas. It then found in paragraph 3 in these terms:

Then in its ultimate conclusions, under the heading "Law", this was said by the Tribunal at paragraph 3 of those conclusions:

The first ground of appeal raised by Mr Pilbrow on behalf of Mr Roche is that the Tribunal, in effect, wrongly overstated the emphasis to be placed on the words "reliable evidence" as set out in article 5(4) and thereby were led into error.

I am bound to say I can see nothing whatsoever in that particular point. The Tribunal's statement that if there was no evidence or if it was unreliable evidence it followed that a reasonable doubt cannot be raised, in my view, is a perfectly acceptable paraphrase of what article 5(4) is aimed at. So I reject that ground of appeal.

But the essence of Mr Pilbrow's complaints is to be found set out in his second and third points. Mr Pilbrow frankly accepted that the totality of the evidence no doubt showed a probability, even a strong probability, but a causal link was not shown. But that is not the test under Article 5. In essence, what Mr Pilbrow says is that on the approach that this particular Pensions Appeal Tribunal chose to adopt (that is to say by placing prime emphasis on Professor Hay's written evidence) it simply was not entitled to reach its finding of fact, as set out in paragraph 3, or to reach the conclusion, as summarised in paragraph 4 of its conclusions as to law. In that regard Mr Pilbrow accepted that if that finding and that conclusion was a finding and conclusion properly open to the Pensions Appeal Tribunal, then, as he put it, his appeal was "dead in the water".

Mr Lewis, for his part, says that there were, in essence, three aspects which had to be considered. First, was it possible that the administration of the mustard gas caused, by off-gassing, Mr Roche's respiratory problems; that is, that they were due to his service? Second, was it possible that the single administration of GF gas caused Mr Roche's respiratory problems; that is, that they were due to his service? Third, was it possible that a combination of those administrations caused Mr Roche's respiratory problems; that is, that they were due to his service?

So far as the GF gas is concerned, says Mr Lewis, Professor Hay is unequivocal. He says in his written answers that "this would rule out exposure to GF as a cause of his COAD"; and elsewhere he said "I can find no evidence to relate his exposure to GF with the conditions of which he now complains". As to mustard gas, Mr Lewis was disposed to agree, for the purposes of this case, that the various studies and papers arguably show that exposure to mustard gas in sufficient doses possibly may lead to respiratory diseases. But these papers were, as he pointed out, and as the Pensions Appeal Tribunal itself found, generalised: whereas in the case of Professor Hay his answers were, as again the Pensions Appeal Tribunal noted, specific to the case of Mr Roche. Professor Hay roundly said, stressed Mr Lewis, "there is no merit in the argument that he would have directly inhaled a cumulative dosage of mustard gas tested on each of the men"; and elsewhere: "any inhalation of mustard gas, would, in my opinion, have been trivial, and not a cause of his obstructive airways disease". In other words, says Mr Lewis, even though possibly there may be cases where exposure to mustard gas may cause respiratory diseases, here, as he submits, Professor Hay was in the case of Mr Roche flatly rejecting that as a possibility in that the inhalation and exposure was too insignificant to give rise to that as a possibility.

Accordingly, goes on Mr Lewis' argument, reading the answers of Professor Hay as a whole, and taking the evidence as a whole (as he stressed, and rightly in my view, one must) and reading also the Pension Appeal Tribunal's decision as a whole, the Tribunal was justified in concluding that no reasonable doubt was shown.

I do not agree. As it seems to me, it is precisely because one must read the evidence as a whole that the position, on the approach adopted by the Pensions Appeal Tribunal, focusing, as it did, in particular on the evidence of Professor Hay in the Tribunal's determination, that the position is not so clear as Mr Lewis would have it.

It is to be observed that there are the repeated observations by Professor Hay in his written answers in answer to the questions -- and one must accept, rather leading questions -- that he could not exclude the possibility of the various points put forward. For example, and it is only by way of example, he says, in answer to question A6 with regard to the percentage inhibition figures, that they did not, in his view, exclude the possibility that inhalation of GF could have caused the medical conditions of which Mr Roche was complaining and only an epidemiological assessment could categorically rule that out one way or the other. Mr Pilbrow agreed that that answer, and answers to similar effect elsewhere, could not in themselves show a reasonable doubt. He acknowledged that in the light of the way in which the leading questions were put, and given understandable and ordinary scientific prudence, it might have been difficult for Professor Hay to have answered such a question otherwise; but Mr Pilbrow states that that is at least a reasonable starting point for a case that a reasonable doubt had been raised.

What Mr Pilbrow stresses, however, and in my view with force, is that elsewhere Professor Hay in fact does give somewhat more equivocal answers than in other parts he gives. It is certainly true that Mr Lewis can point to passages in those answers where Professor Hay seems to be wholly unequivocal in rejecting the causal link between the Porton Down tests and Mr Roche's present respiratory illnesses. But elsewhere that is not so. For example, in the answer to question 9 Professor Hay talks about there being no strong evidence to suggest that exposure to GF would be a cause of Mr Roche's symptoms. That is a rather different thing from saying there is no evidence at all of such a kind. Further, on a number of occasions in his report Professor Hay seems to qualify the previous answers that he has given by stating words to the effect that it was unlikely that Mr Roche's condition was caused by his exposure to the chemical testing which he underwent in 1963 (I add that Professor Hay seems to have been focusing in particular on the testing in that year). That is repeated in a number of respects in his written answers.

In consequence, as it seems to me, reading Professor Hay's evidence as a whole, there is a degree of doubt arguably raised by those answers. I do not think that one can conclude that Professor Hay was necessarily saying that he was sure that the causal link between Mr Roche's respiratory problems and service was excluded. It seems to me that although one can certainly cherry pick from Professor Hay's answers which are unequivocal in that regard, equally one can select answers which significantly modify such an unequivocal position.

That being so, I do not think that the Pensions Appeal Tribunal here had the basis (on the approach which it had chosen to adopt) to make the findings that it purported to make at paragraph 3 of the findings of fact and paragraph 4 in its conclusions of law; and it erred in law in doing so. That is particularly so, as it seems to me, when a number of other considerations are also taken into account.

First, the generalised evidence contained in the various studies and reports, which the Tribunal expressly did not reject as unreliable, lent some support, at least potentially, to Mr Roche's case, at least with regard to mustard gas; and at all events those reports were not inconsistent with Mr Roche's case.

Secondly, there are various references in the decision of the Pensions Appeal Tribunal itself which perhaps indicate that on occasions the Tribunal may have been engaging in a balancing exercise. For example, it refers to the "significant probative weight" which it attached to Professor Hay's report. It said that it "preferred" that evidence to the other evidence. It said, by way of example, that the "compelling weight" of the evidence was that Mr Roche had not received dosages likely to have long-term effects. It had referred to the evidence of Professor Hay to the effect that he deemed it "unlikely" that mustard gas was a cause. These points perhaps would not in themselves suffice to invalidate the decision of the Tribunal. But they do seem to me to lend at least some further support for the view that, quite simply, the wrong answer may have been reached given the approach that needed to be applied under article 5(4). In that regard I would reiterate the point raised by reference to paragraph 5 of the conclusions: if the recording of the evidence was that mustard gas was "unlikely" to be a cause, then that would seem to give rise to an acceptance that there was a possibility that it was the cause. It may be, as Mr Lewis said, that that particular answer is to be read in the answer previously given in paragraph 3 of the findings of fact. But, again, as it seems to me, there is some doubt as to that; and at all events my fundamental reason for allowing this appeal is that, on the approach adopted by the Pensions Appeal Tribunal, with the particular emphasis on the answers of Professor Hay, it simply was not entitled to reach the finding of fact set out in paragraph 3 or the finding in paragraph 4 of its conclusions, as it did.

In those circumstances, I need not deal any further with Mr Pilbrow's other ground of appeal, which I have to say did not impress me very much.

In the result, therefore, it seems to me that I must allow this appeal. Mr Pilbrow invited me to allow it without more and not to refer the matter back to the Pensions Appeal Tribunal. However, I allow this appeal not on the footing that the Pensions Appeal Tribunal reached a conclusion which it simply was not open to it to reach; rather, I allow it on the ground that, given the approach that this particular Tribunal chose to adopt, it was not thereafter entitled to reach the conclusion that it did in fact purport to reach. I have not myself been taken through the entirety of the evidence that was before the Tribunal and it is for the Tribunal to assess the entirety of that evidence. Accordingly, I allow this appeal and refer the matter back to the Tribunal for a further hearing.

MR JUSTICE DAVIS: Mr Pilbrow, is there anything you wish to add or say?

MR PILBROW: My Lord, no. As noted yesterday, costs are provided for in rule 28 of the Pensions Appeal Tribunal Rules so they do not arise. My Lord, as far as I am concerned, there is nothing further to add.

MR JUSTICE DAVIS: So the matter will then have to go back.

MR PILBROW: Indeed, my Lord, and it may be, as a matter of practicality, it can be reacquainted with the hypertension appeal.

MR JUSTICE DAVIS: There might be some advantage in that. Obviously it has to be done soon rather than later.

Ms Stone, is there anything you wish to add?

MS STONE: No, I have nothing to add.

MR JUSTICE DAVIS: Thank you both.