|[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]|
Upper Tribunal (Administrative Appeals Chamber)
You are here: BAILII >> Databases >> Upper Tribunal (Administrative Appeals Chamber) >> JM v Secretary of State for Defence (AFCS) (Tribunal procedure and practice (including UT) : fair hearing)  UKUT 332 (AAC) (11 June 2015)
Cite as:  UKUT 332 (AAC),  AACR 3
[New search] [Context] [View without highlighting] [Printable RTF version] [Help]
JM v Secretary of State for Defence (AFCS)
 UKUT 332 (AAC)
Mr Justice Charles
Upper Tribunal Judge Rowland
Upper Tribunal Judge Lane
For the Appellant: Elaine Banton instructed by Hogan Lovells International LLP
For the Respondent: Adam Heppinstall instructed by the Treasury Solicitor
The appeal is allowed. The decision of the First-tier Tribunal dated 23 May 2013 is set aside for inadequacy of reasoning and procedural unfairness.
The case is remitted to a differently-constituted panel of the First-tier Tribunal to be re-decided in accordance with guidance given in this decision and the following directions.
1. The appellant must send, so that it is received by the First-tier Tribunal within one month of the date this decision is sent to the parties, a witness statement setting out his detailed account of the circumstances of the two assaults on him and his allegations of the campaign of bullying against him;
2. Upon receiving the appellant’s witness statement, the First-tier Tribunal must send a copy of it to the respondent.
3. The respondent must send, so that it is received by the First-tier Tribunal within one month of the appellant’s statement being sent to him by the First-tier Tribunal, a written submission setting out his case in the light of the appellant’s witness statement.
4. Upon receiving the respondent’s submission, the First-tier Tribunal must send a copy of it to the appellant (unless the respondent has already done so).
5. The First-tier Tribunal may vary any of the above directions.
This decision contains conclusions and guidance on a number of topics. It does so under the following headings.
Background and procedural history
The First-tier Tribunal’s decision
Section 5A(1) of the PAT Act 1943 as amended
The nature of an appeal before a First-tier Tribunal
Inadequacy of reasons
The general approach that should be taken to decision
making and reasoning
The approach to decision making and reasoning
of the First-tier Tribunal in this case
The relevant provisions of the 2005 and 2011 Orders
The underlying purposes of the relevant Articles
Are there useful principles in other areas of law that may assist
in determining whether the test in the AFCS is satisfied?
The case law under the war pensions instruments and its
application to the identification of a service cause
Guidance on bullying
The categorisation exercise
The steps to be taken in the application of the AFCS test
The predominancy test
The present case
What needs to be done at the remitted hearing
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 – 
 –   – 
 – 
 – 
REASONS FOR DECISION
1. This is an appeal against a decision of the War Pensions and Armed Forces Compensation Chamber of the First-tier Tribunal (the War Pensions FTT or the FTT) dated 23 May 2013, for which a statement of reasons was signed on 7 June 2013, (the Decision) on appeals by the appellant against decisions of the respondent (the Secretary of State) under the Armed Forces Compensation Scheme (AFCS) refusing to pay him compensation in respect of (a) damage to his two front teeth, (b) a stab wound to his leg and (c) depression.
2. The decisions of the Secretary of State were made on 29 March 2011 and 3 June 2011 were made respectively under the Armed Forces and Reserve Forces (Compensation Scheme) Order 2005 (SI/2005/439, as amended) and 2011 (SI 2011/517) respectively (the 2005 Order and the 2011 Order). The terms of the relevant articles are set out later in this decision.
3. Permission to appeal was given by the then President of the War Pensions FTT on the grounds that the following arguable questions had been raised by the appellant’s representative:
(a) whether there was cogent circumstantial evidence that the Tribunal failed to take into account;
(b) whether there was a breach of care on the part of the Army because of their failure to remove the appellant from the Unit following his harassment allegations; and
(c) whether the Tribunal should have set out all the causes of the claimed condition before deciding what the predominant cause was.
4. He also gave permission so that the Upper Tribunal could give guidance on the correct approach to be taken in AFCS bullying cases. It was directed that this appeal be heard by a three judge panel for this purpose.
5. Procedural unfairness was not a ground of appeal for which permission was given by the Chamber President of the War Pensions FTT. It was raised by us at the main hearing. No adjournment was sought and we dealt with it at the hearing.
6. When we were writing this decision, points occurred to us that had not been put to the parties. These related to the application of the relevant articles on causation. A further hearing took place to deal with the points we raised. Unfortunately that gave rise to some delay.
Background and procedural history
7. The appellant enlisted in August 2005 and was discharged from the Army in September 2009. He was a private soldier. It was alleged that on 2 September 2008 the appellant head butted his roommate after that roommate had told him to clean up his section of the room. So, in this alleged incident in Kuwait he was alleged to be the aggressor. In the context of the investigation of that incident (and others) the appellant complained that he had been the victim of a campaign of bullying since February 2008 by other members of his troop on Operation Telic 12 and in barracks in Gütersloh.
8. As the appellant had raised this allegation of bullying, in early November 2008 a Captain C was appointed to carry out a Formal Investigation of Harassment of the appellant. Statements taken in the context of this investigation covered the alleged head butting by the appellant as well as his allegation of a campaign of bullying. However, because the appellant had returned to Gütersloh, he was interviewed there by a Captain T after the others had been interviewed.
9. A letter dated 14 November 2008 from Captain T records that at interview the appellant decided to rescind his allegations because he felt that the interview process would be too painful to continue with. That letter annexed a draft withdrawal letter to be signed by the appellant also dated 14 November 2008. On the letter from Captain T, beside the reference to that letter of withdrawal, there is a manuscript note to the effect that this was because of pressure from others. Counsel for the appellant told us that the FTT were told that this was the appellant’s reason for dropping the investigation.
10. A number of statements were taken later in November 2008 from members of the troop. The statements refer (a) to a view that, given the appellant’s state of mind, the victim of the head butting could have been anyone, (b) to the appellant’s upset and difficulties relating to his wife and child leaving him, and (c) to the appellant’s stammer. In one of the statements there is a reference (d) to some of the troop beginning to say from time to time that they “would hammer him when they got back”. In relation to the assault by the appellant in Kuwait, some members of the troop stated that the appellant had previously been told to tidy up his part of the room because he was leaving food there and cockroaches and a mouse had been seen.
11. The report relating to the investigation of the alleged assault is dated 27 November 2008 and its penultimate paragraph (with our emphasis) states:
“The RMP investigation at reference A does not go into detail regarding [the appellant’s] complaint of bullying. During Capt [T’s] interview with [the appellant], no details could be gleaned as to the specific complaint of bullying against him. [The appellant] has deployed with problems with his private life and these have become public knowledge within Force Protection. [The appellant] has been involved in a number of confrontational incidents with members of Force Protection Tp during the tour which, have caused those who initially called him a friend to distance themselves from him. There does not seem to be any campaign of bullying against [the appellant] based on these witnesses who lived and worked closely with him, only sporadic instances of banter which a present NCO rectified the situation.”
12. This account makes it clear that no details of the appellant’s allegations were given at the time. The conclusions reached were therefore based on the statements of others, some of whom could well have been the alleged perpetrators to whom his allegations could not be put and who could not be questioned in the knowledge of what they were.
13. The appellant’s case was that members of the troop were on leave at the end of 2008 and in January 2009 and that he was stabbed in the leg by one of them in February 2009 shortly after their return to Gütersloh, although he was unable to identify the assailant. It was accepted that this assault took place. The other physical injury on which he based a claim (damage to his front teeth) was inflicted later in August 2009 when he was head butted by another soldier. This assault led to a Court Martial. The assailant admitted the offence and was punished. The court accepted that the assailant felt that he had been provoked in an argument about cigarettes he had given to the appellant and which he wanted back. The appellant, however, was not present at the hearing and the papers before us do not contain any account given to the Court Martial of his version of events relating to that incident or any previous bullying.
14. In March 2010, after his discharge from the Army, the appellant made four claims under the AFCS which were based on those two injuries, an allegation that he suffered from depression as a result of bullying in the Army and an allegation that he suffered from hearing loss (which is not material to this case and to which we make no further reference). His claim form includes very limited particulars of his allegation of bullying, and of the circumstances surrounding and leading up to the two assaults on him, although it did name the person who had damaged his front teeth and said that the incident had been reported to the duty sergeant at the barracks and to the Royal Military Police. It does not identify the person who stabbed him, nor contain any explanation about why he believes he was assaulted. Neither does it refer to the incident in which he assaulted another soldier or to whether that incident was connected to the alleged bullying. In his claim form in respect of his claim based on depression he says:
“Depression was caused through constant bullying, verbal abuse, just because I stutter, I am Liverpudlian and that my wife left me a few weeks prior to deployment. Felt intimidated, low self-esteem and although I tried talking to them things didn't change.”
15. The Secretary of State appears to have obtained documents from the Army and some medical evidence but then rejected the claims without inviting the appellant to provide further details of his case beyond asking him on 23 March 2011, in relation only to the claim in respect of depression, whether he had reported the alleged bullying to a senior officer and, if so, whether there had been an official investigation. (It is not clear whether the claimant answered that letter or understood that it asked questions – see pages 40, 45 and 46 of the depression bundle – and it was only long after the decision had been made and the appeal lodged that the records of the investigation mentioned above were obtained by the Secretary of State.)
16. Because there was medical evidence of the stab wound, it was accepted that the first assault took place and that “the assailant was probably another serviceman and probably someone in the same Unit as the appellant”, but the Secretary of State did not accept that that injury was caused by service. That decision was made on 29 March 2011.
17. However, apparently because the medical evidence he had obtained did not confirm the damage to the claimant’s front teeth and because he was then unaware of the Court Martial, the Secretary of State initially did not accept that the second assault had ever occurred and his decision, also made on 29 March 2011, was to that effect.
18. As regards the claim in respect of depression, in the light of medical evidence the Secretary of State accepted that the appellant was suffering from depression but considered that, even if there had been bullying, there was also evidence of “personal, domestic and marital stress” so that it was “due to multiple factors and hence not caused predominantly by service”. That decision was made on 3 June 2011.
19. The appellant appealed against the decisions rejecting the claims in respect the injuries incurred in the assaults and of depression, with the support of his grandfather who made further allegations in general terms that the appellant had been bullied and asserted that the Army had allowed the perpetrators to get away with the bullying and had therefore been in breach of a duty of care. The appeals led to the case being reconsidered and to the Secretary of State accepting that the second assault had occurred, but he maintained his position that none of the injuries was caused by service.
The First-tier Tribunal’s decision
20. The three appeals first came before the FTT on 14 August 2012. Not being satisfied with the information provided by the Secretary of State, the FTT directed him to provide further medical evidence and all documents from the appellants ‘P’ file relating to his allegation of bullying. They also consolidated the appeals under a new reference number. The case came back before a differently-constituted panel on 23 May 2013. Both parties were represented, but not (as far as we are aware) by lawyers.
21. The FTT dismissed the appeal. In the statement of reasons, the procedural history of the case, the relevant legislation and detailed summaries of the parties’ contentions are set out. The statement then continues –
“Findings of Facts
28 The appellant was born on 13 February 1988 and served in the army from 11 August 2005 to 11 September 2009.
29 We find that the damage to the appellant's front teeth and the stab wound to his right leg were both isolated incidents, constituting criminal offences, and they were not part of a pattern of bullying by the appellant's colleagues or superiors.
30 In February 2009 the appellant was stabbed in the leg by a masked man in the comer of his dormitory block. We have no further information about the details of this attack, but as there is nothing to suggest that the attack was carried out by an intruder the natural inference we draw is that the assailant was very probably another serviceman and probably someone in the same Unit as the appellant. However we do not accept the appellant's contention that this attack was related to the complaints he had made about bullying as this contention is unsupported by any evidence.
31 In July 2009 the appellant was head butted by a colleague, resulting in damage to his front teeth. This incident was appropriately investigated by the army and the appellant's assailant pleaded guilty at a Court Martial to the offence of assault occasioning actual bodily harm, for which he was sentenced to an SSPO order for 60 days. The attack on the appellant was preceded by an argument between the appellant and the assailant and had no connection with the appellant's complaints about bullying.
32 The allegation by the appellant of a pattern of bullying by his colleagues and superiors is not borne out by the evidence. In October 2008 the appellant claimed that he had been subjected to a campaign of bullying since February 2008 and this complaint resulted in a formal harassment investigation. Statements were taken from a number of witnesses and a Harassment Investigation Report was prepared on 27 November 2008 (see pages 69 and 70 of the general bundle). The conclusion reached was as follows: ‘There does not seem to be any campaign of bullying against [the appellant] based on these witnesses who lived and worked closely with him, only sporadic instances of banter for which a present NCO rectified the situation.’ It is also significant that when interviewed the appellant said that he no longer wanted to pursue the claim of alleged bullying and he is recorded as having signed a statement confirming his withdrawal of any complaint.
33 As to the appellant's depression, we find that this was due to multiple factors and in particular personal, domestic and marital stress. In reaching that conclusion we rely on the totality of the evidence and in particular the consultant psychiatrist’s report dated 24 February 2010 (on pages 33 and 34 of the depression bundle). We note that in the fourth paragraph of that report the psychiatrist has recorded that the appellant said he was depressed ‘because of his marriage breakdown and because of leaving the Army’. There are references in the report to allegations of bullying but, for the reasons set out above we find those allegations to be unsubstantiated insofar as they suggest a general pattern of bullying.
34 We have noted the view of the Combat Stress psychiatrist, in the discharge care plan assessment notes dated 21 February 2011, that the appellant had ‘PTSD related to bullying in the Army’. However the psychiatric report of 24 February 2010 did not find PTSD and Dr Fletcher was wholly dependent on the appellant's version of events and symptomatology. Dr Fletcher described bullying as the cause of the appellant's PTSD; no other cause was mentioned. As already stated, we do not accept that the alleged general pattern of bullying had taken place. We note as well that Dr Fletcher considered the appellant's mental illness to be ‘partially related to service’ rather than ‘predominantly related to service.’
35 We accept and endorse the views expressed by the SPVA medical adviser in the first five paragraphs of his advice dated 11 February 2011 (on page 37 of the depression bundle). In the fifth paragraph the medical adviser concluded that the appellant's depression was due to multiple factors and not caused predominantly by AFCS service.
36 In the light of our findings as set out above, we are satisfied that the appellant has failed to establish on the balance of probabilities that any of the three conditions which are the subject of this appeal were caused solely or predominantly by service.
37 The two attacks on the appellant which resulted in the damage to his front teeth and the wound to his right leg both constituted isolated criminal offences. The sole cause of the injury in each instance was the criminal act of the assailant. In each case service merely provided the setting for the incident during which the injury occurred; it was not the cause of the injury. We do not consider that Service was a cause at all of either injury, but if it was, it was certainly not the predominant cause.
38 The two passages from Lord Hoffman's speech in the Empress Car Co case which Mrs Davies sought to rely on are not, in our view, of any assistance to the appellant. We note that in Upper Tribunal Judge Mesher's decision in CAF/316/2010 he did not consider that those passages assisted the appellant in that appeal.
39 We would also comment that the emphasis by the appellant, and by his grandfather on his behalf, on the alleged breach of care on the part of the army is misplaced. The primary issue for us is not ‘Has there been a breach of duty of care towards the appellant?’, but ‘What is the predominant cause of each of the appellant's injuries?’
40 For the reason set out above, we have found that the appellant's depression was due to multiple factors and was not predominantly caused by service.
41 The appeal is accordingly dismissed in respect of all three conditions.”
22. The appellant now appeals with permission granted by the FTT for the reasons set out in paragraph 3 above. He argues that, on the evidence before the FTT, they were bound to find that his injuries arose out of a pattern of bullying of which the Army was aware and which it failed to stop and so the injuries were predominantly caused by service.
23. The Secretary of State argues that the FTT was entitled to find that there had been no pattern of bullying and that, in any event, there was no evidence linking the two assaults to any such pattern and that therefore the FTT did not err in law.
Section 5A(1) of the Pensions Appeals Tribunals Act 1943
24. Section 5A(1) of the Pensions Appeal Tribunals Act 1943 applies to this appeal to the War Pensions FTT. It provides that:
(1) Where, in the case of a claim to which this section applies, the Minister makes a specified decision
a. he shall notify the claimant of the decision specifying the ground on which it is made, and
b. thereupon an appeal against the decision shall lie to the appropriate tribunal on the issue whether the appeal was rightly made on that ground
25. Section 5A(1)(b) is cited in paragraph 5 of the Decision. No further express mention is made of the ground or grounds on which the Secretary of State made his decision. However, it is clear that the FTT proceeded on the basis that the ground was that the appellant was not entitled to an award under the relevant provisions of the AFCS (and the reasons given for that conclusion were not the ground or grounds). In our view, the FTT was correct to do so. We agree with paragraph 12 of the decision of Commissioner Bano in CAF/656/2006 where he says:
“The terms of section 5A of the 1943 Act follow those of section 1, which confers a right of appeal in respect of a decision that an injury is neither attributable to nor aggravated by service. The Minister must notify the claimant of the decision, specifying "that it is made on that ground", and an appeal lies "on the issue whether the claim was rightly rejected on that ground". It seems to me to be clear that the "ground" of a decision under section 1 is a finding that the statutory grounds of entitlement are not satisfied, and I would similarly hold in this particular statutory context that the "ground" of a decision under section 5 is a finding that the statutory conditions of entitlement to an award are not satisfied, rather than the reasons for that conclusion.”
The nature of an appeal before the FTT
26. We need to restate the nature of an appeal to the War Pensions FTT because of a general submission by Mr Heppinstall on behalf of the Secretary of State that it was open to, and appropriate for, the FTT simply to accept the conclusion of an internal Army investigation (without hearing from the appellant on the issue) because the investigators had seen the witnesses. In support of that general submission it was argued that it would be wholly disproportionate for a War Pensions FTT to hold a trial relating to historical allegations at which rival witnesses were called and cross examined.
27. When, as here, the conclusion of the relevant earlier investigation is not agreed, we do not accept that general submission. Indeed, in our view, its adoption would, in effect, undermine the exercise by the FTT of its independent judgment on the value of evidence presented to it.
28. Like many other Chambers of the First-tier Tribunal, the War Pensions FTT is a fact finder and decision maker. This role is recognised by limiting appeals from it to material errors of law. Its duty is to decide entitlement under the claims de novo: Barratt v Minster of Pensions (1948) 1 WPA 1225. In Barratt, Denning J held that the tribunal had been wrong to base its decision on the view that there was no evidence before it to upset the decision of the Ministry. The Tribunal’s function was to examine the evidence afresh and see whether the finding is justified.
29. Issues of practicality and proportionality will inevitably arise when dealing with problems of this sort where a re-investigation of the original complaints, for example by calling witnesses, is not practicable. We accept, as counsel submitted, that something equivalent to a state or criminal trial is not appropriate in the circumstances of Tribunal proceedings in this jurisdiction. On the other hand, the War Pensions FTT must not abrogate its function as the fact finder. That does not mean that having reviewed and assessed all the relevant material the War Pensions FTT cannot accept the view reached by an investigator or the Secretary of State. It does mean that before accepting or adopting that evidence and those conclusions a War Pensions Tribunal will have to consider the manner in which the investigation was conducted and thus, for example, whether it met the requirements of procedural fairness.
30. Mr Heppinstall’s submission is disconcerting because the Secretary of State is effectively a constant presence before the War Pensions FTT and if he is advancing such an approach, he is pointing the War Pensions FTT in a direction that is wrong and is likely to promote unfairness, particularly when questions arise as to whether the investigation has been appropriately thorough.
31. Tribunals and courts are regularly faced with problems such as these but are nevertheless required to analyse the relevant investigation or decision making process and assess the weight which they are to be given. Where the conclusion of a prior investigation is to be preferred to a claimant’s version of events, it is generally an essential ingredient of a fair procedural process for a tribunal to give the claimant the opportunity of dealing with points that may lead the tribunal to reject his version. This is obviously so where the case hinges on findings on credibility. Even though others who can give primary evidence on a relevant dispute are not present for questioning, a tribunal must grapple with how the competing versions stack up. It should only rarely be necessary to resort to the burden of proof.
32. War Pensions appeals are, in addition, cooperative rather than adversarial proceedings to which the principles in Kerr v Department for Social Development  UKHL 23,  1 WLR 1372, R 1/04(SF) apply. The Secretary of State acknowledged this to be so at the second hearing of the appeal. He accepted that, even though the claimant expressly bears the burden of proof under the AFCS (see Articles 50 and 51 of the 2005 Order and Articles 60 and 61 the 2011 Order), this -
i) did not mean that his officers should not take reasonable steps to investigate the claims made under the AFCS, and
ii) did not have any material impact on the ground of appeal we had raised, namely procedural fairness.
As to point (i) he submitted and we accept that his officers routinely investigate issues giving rise to claims and the claims themselves, raise question of claimants, obtain medical records, commission medical examinations and reports and seek information from military units and other sources to investigate relevant incidents and claims (see Articles 52 and 53 of the 2005 Order and Articles 62 and 63 of the 2011 Order).
33. The Secretary of State (like other public authority decision makers) must act fairly as the decision maker under the AFCS.
34. The War Pensions FTT (like other tribunals) must also act fairly. It is charged with making decisions in an appropriately informal setting and manner which is proportionate to the issues.
35. It is well recognised that what is fair and procedurally appropriate varies as between different decision making settings.
36. Each of the legal, medical and service members of the War Pensions FTT brings their experience and expertise to the tribunal. The experience and expertise of the service and medical members will often go to the heart of factual and causative issues to be decided. The experience and expertise of the legal member are particularly relevant to procedural fairness and decision writing.
37. Although what fairness requires is dependent on the context of the decision, a basic constituent is that a person knows the case he has to answer or address, and is given an opportunity to make representations thereon (see for example R(Doody) v Home Secretary  1 AC 531 at 560 D/G). So, for example, both:
i) a decision maker on behalf of the Secretary of State, and
ii) members of a tribunal
need to ensure that they are appropriately informed on the bases upon which a claim is being advanced and that a claimant is appropriately alerted to points that they consider to be relevant to the determination of the claim and given an opportunity to answer them.
38. At the second hearing, as at the first, the Secretary of State submitted that:
i) his officers had carried out an appropriate, proportionate and fair investigation of the original allegation,
ii) when the claims had been made his decision making officers had carried out an appropriate, proportionate and fair investigation of them, and
iii) the FTT had acted appropriately, proportionately and fairly and had done the best they could on the evidence before them.
39. In our view, the basis of the appellant’s case on bullying as the cause of his injuries cried out for clarification and particularisation at two stages:
i) when his claims under the AFCS were made, investigated and decided by the Secretary of State, and
ii) when they were decided by the FTT.
While we are concerned with the latter, it is obvious that further investigation of the claims by the Secretary of State would not only have usefully informed his decision but have provided a centrally important opportunity to the claimant to explain his claims after he had left the Army and was no longer at risk from those he alleged had bullied him. It would also have provided centrally important information to the FTT.
40. The notes of proceedings taken by the medical member and the judge indicate that the appellant did not give an account of either the assaults or the asserted bullying he suffered in the Army. Instead, he made a few interjections and some comments to the FTT. Whether they were in response to questions is unclear.
41. The notes indicate that at no stage did the FTT, or the parties’ representatives (the appellant was represented by someone from RBL) seek any further explanation or detail from the appellant of the asserted bullying and why he said it was connected to the assaults and his depression – other than that set out in paragraphs 15 and 16 of the FTT’s decision, namely:
According to [the appellant’s representative] the appellant felt that it was a result of his asking for an investigation into harassment to take place that the subsequent events, i.e. the stabbing and head-butting incidents, followed. She submitted that therefore service must be the predominant cause of those two incidents.
The appellant himself made the point to us that if someone makes an allegation of harassment they should be taken away from the people against whom the allegation was made, whereas he was forced to stay alongside the same group of people.
42. That puts the case on the basis of common ground that the appellant had made the allegation and asked for an investigation. That is different from his case that he had been the victim of a campaign of bullying. There may, of course, be a connection between the two. This is because a conclusion that the appellant had been the victim of bullying, a campaign of bullying or isolated assaults which resulted in injuries might have affected the evidence given during the investigation by witnesses who could be implicated in the asserted events.
43. Although the papers before it did not contain anything like a detailed account of the appellant’s allegations, which were based on both a campaign of bullying and the fact that he had made a complaint that he was the victim of bullying, the FTT did not ask the appellant or his representative to explain why the appellant contended that the assaults were part of the asserted earlier or continuing campaign of bullying or a reaction to his complaint of bullying, rather than isolated incidents.
44. As the appellant’s case lacked clarity and particularisation, the representatives of both parties should have made efforts to identify the true nature of the appellant’s case and to explain and address the issues mentioned in the last paragraph.
45. But in all the circumstances, we consider that these points were so obvious and central, that the FTT acted unfairly and so erred in law by failing:
i) to take steps to ensure that they - and the parties – understood what the case was really about by giving the appellant an opportunity to particularise his allegations and explain why he withdrew them, and
(ii) to establish the bases on which the appellant was making his claim.
46. Had these failures not taken place, there was a real prospect that relevant matters would have been identified.
47. In our view, these failures had the result that the FTT (and the Secretary of State before them) took an adverse view of the appellant’s credibility without a sufficient investigation and understanding of his case and without putting to the appellant the central factual and other issues that were troubling them and thereby giving him an opportunity to address them. The FTT’s procedural unfairness is of itself a sufficient basis for allowing the appeal.
Inadequacy of reasons
48. In his written submissions counsel for the Secretary of State reminded us that we must approach decisions of the War Pensions F-tT with ‘considerable respect’, citing a well known passage from Secretary of State for Defence v Duncan  EWCA Civ 1034,  AACR 5 per Carnwath LJ at -.
49. These submissions on behalf of the Secretary of State were a version of submissions made on his behalf by Mr Heppinstall in the ionisation appeals heard by Mr Justice Charles in Abdale v Secretary of State for Defence (WP)  UKUT 477 (AAC). Paragraphs - of that decision deal with those points and we do not repeat them here. Those paragraphs of Abdale also elaborate on the duty of the War Pensions FTT to give adequate reasons and the approach that should be taken to determining whether it has erred in law by failing to do so.
50. They show that there are limits to respect that can be shown even to a specialist tribunal. A line must be drawn between giving a generous reading to its decisions and condoning manifest errors of law. For example, it is likely that it will be found that an FTT has erred in law if an appropriately generous reading of its decision shows that (a) it has failed to set out its understanding or approach to the application of the relevant test, (b) it has set out an account of evidence and assertion made by or on behalf of the parties with insufficient linkage between them and its conclusion, (c) it has made findings without giving any or any adequate express reasons for them, or (d) it has stated its conclusion without giving any or any adequate express reasons on why the evidence and submissions it has recorded and the findings it has made found the conclusion reached on the relevant test.
51. It does not reflect well on the War Pensions FTT that the Secretary of State should feel it is necessary to emphasise the respect that should be accorded to a specialist Tribunal’s decision when he is trying to persuade the Upper Tribunal or the Court of Appeal:
i) that the War Pensions FTT has taken the right approach in law (both substantively and procedurally), and
ii) that the absence of sufficient express reasoning should be cured by a generous reading between the lines.
The general approach that should be taken to decision making and reasoning
52. It was argued on behalf of the Secretary of State that, as the test for payment under the AFCS is one of fact, the FTT (and the Secretary of State, who is the other decision maker under the Articles) should be left to reach conclusions of fact unburdened by critical examination from a superior court or tribunal.
53. We accept, of course, that there are limits to the extent to which a superior court or tribunal is entitled to interfere with a lower tribunal’s decision. The Upper Tribunal cannot, for example, interfere with an FTT’s decision unless they failed to recognise the proper legal test or, having recognised it, made a decision as to which side of the line the case fell on that was outside the bounds of reasonable judgment (R(DLA) 5/05 see in particular paragraph 3 of the headnote and   of the Decision; Moyna v Secretary of State for Pensions  UKHL 44,  1 WLR 1929, R(DLA) 7/03). The headnote in Horsfall v The Minister of Pensions (1944) 1 WPA 7 provides a summary to the same effect:
“provided a Tribunal applies a proper test, it is a question of fact for them whether an injury or disease is attributable to war service.”
54. As is often the case in appeals before a FTT the relevant legal test uses ordinary English words which may or may not be defined, and the conclusion on its application is an issue of fact.
55. In R v Monopolies and Merger Commission ex parte South Yorkshire Transport  1 WLR 23 at 32F to 33A Lord Mustill examines the tensions that arise when courts or tribunals at different levels are confronted with imprecise words to which a range of meanings may be given:
“Once the criterion for a judgment has been properly understood, the fact that it was formerly part of a range of possible criteria from which it was difficult to choose and on which opinions might legitimately differ becomes a matter of history. The judgment now proceeds unequivocally on the basis of the criterion as ascertained. So far, no room for controversy. But this clear-cut approach cannot be applied to every case, for the criterion so established may itself be so imprecise that different decision makers, acting rationally, might reach differing conclusions when applying it to the facts of a given case. In such a case the court is entitled to substitute its own opinion to that of the person to whom the decision has been entrusted only if the decision is so aberrant that they cannot be classed as rational: Edwards v Bairstow  AC 14. The present is such a case. Even after eliminating inappropriate senses of "substantial" one is still left with a meaning broad enough to call for the exercise of judgment rather than an exact quantitative measurement. Approaching the matter in this light I am quite satisfied that there is no ground for interference by the court, since the conclusion of which the commission arrived was well within the permissible field of judgement.”
56. Courts and tribunals are repeatedly warned against the dangers of ‘taking an inherently imprecise word, and by redefining it thrusting on it a degree of spurious precision’ - South Yorkshire Transport, per Lord Mustill at 29C. The correct approach is to construe the words by reference to their ordinary meaning, their statutory context and purpose (see in particular the headnote and paragraphs 7 to 12).
57. The problem in giving guidance to fact finders in reaching and explaining their conclusions is that they cannot re-write the test other than by writing in definitions contained in the relevant statutory provisions. In this appeal, only the word ‘predominant’ is defined (article 2 of each Order).
58. It follows that in construing and applying the relevant test, its underlying purpose is an important and often determinative factor to be taken into account in deciding whether on the facts of a given case it is satisfied.
59. To summarise, the points made above show that:
i) the proper construction of the test (which involves construction of the legislation) is a matter of law,
ii) once the test is established its application will be based on a broad approach to the evidence and the conclusions reached on it. This involves fact finding followed by reasoning and analysis based on those findings,
iii) there are dangers in rewriting or reformulating the test other than by writing in definitions,
iv) different decision makers taking the correct approach in law to the application of a statutory test can lawfully reach different decisions on its application in a given case,
v) the reasoning of the FTT should adequately demonstrate their understanding of the test and how they have applied it, and
vi) the purposes of the legislation are relevant to the determination of the meaning and application of a statutory test.
60. Accordingly, in their decision making process it is important that a War Pensions FTT:
i) applies the proper test in the right way when considering whether and the extent to which service caused an injury, and
ii) demonstrates through their reasoning that they have done so.
61. This means they must reach appropriately reasoned conclusions on the evidence, and then explain why the facts lead to the decision reached on the application of the causation test set by the Articles. It requires the tribunal to address the “why” or the “because” in contrast to expressing their conclusions as bare assertions.
62. So under the AFCS the fact finding exercise will involve:
i) identifying the possible causes and making and explaining findings on their existence and so for example, if it be the case, why an asserted cause (e.g. a campaign of bullying) is found not to exist, and then
ii) explaining why those factors and their effects have caused the War Pensions FTT to conclude that the causation test set by the Articles has been satisfied or not satisfied.
In most cases it is likely that the War Pensions FTT can best demonstrate and explain its understanding and application of the test by setting out why the factors identified have caused them to conclude that the test is or is not satisfied.
63. As stated in Secretary of State for Work and Pensions v Nelson  UKUT 525 (AAC) at paragraph 23, this approach reflects the old adage that it is difficult to define an “elephant” but we know one when we see one and so we can explain why we think we have seen one by describing what we have seen.
64. We acknowledge that the War Pensions FTT will often have to make difficult value judgments and findings in relation to provisions in the AFCS Orders whose language contains words of degree and which will apply to a wide range of circumstances. We hope that this guidance will help them to do this and to explain why they have reached their conclusions.
The approach to decision making and reasoning of the FTT in this case
65. The FTT set out the respective submissions and matters raised on behalf of the appellant and the Secretary of State in paragraphs 10 to 18 and 19 to 27 respectively.
66. An example of error can be seen in the way the FTT dealt with the Secretary of State’s submission that an injury may only be accepted as due to service
“… where the claimant can clearly be shown to have suffered as a direct result of both the abuser/s and the victim (claimant) acting under a compulsion of service or in pursuance of the service’s legitimate objectives".
With regard to both assaults the Secretary of State maintained that service only provided the setting for the incidents and the resultant injuries cannot be construed as due to service.
67. The FTT made a finding that service merely provided the setting for and not the cause of, the injury (see paragraph 37) but they did not explain why this was so. Nor did they comment on the Secretary of State’s formulation of the test which, as a matter of law, is too narrow and is wrong. Indeed, Mr Heppenstall accepted that the formulation does not correctly encapsulate the effect of case law and following consultation with those instructing him, confirmed that the Secretary of State would not in the future use that test or include it in submissions to the War Pensions FTT.
68. If the basis of the FTT’s finding was the Secretary of State’s incorrect formulation of the test, their finding is also tainted. If the FTT did not rely on the Secretary of State’s formulation, its error lay in a failure to explain why they rejected it and to set out the approach they applied. Either reading establishes an error of law because either the FTT adopted the wrong test or they failed to explain the test they were applying.
69. We pick out a few other errors affecting the decision which lead us to overturn it. Under the heading “Findings of Facts” the FTT make the following errors of law:
i) they make findings effectively without any reasons being given for them,
ii) taken in the round and as already mentioned, the findings are a rejection of the appellant’s case on the facts without giving him a fair opportunity to deal with the points that troubled the FTT on the papers,
iii) in paragraph 29, the FTT find that the injuries to the appellant’s front teeth and the stab wound were isolated incidents but nowhere do they explain why that finding is relevant or determinative. (An isolated incident could be caused by service and could amount to bullying),
iv) in paragraph 30, the FTT state that they do not accept the appellant’s contention that the stab wound was related to the complaint he made about bullying “as this contention is unsupported by any evidence”. There plainly was evidence upon which the connection could be made either to the complaint itself or to the subject matter of the complaint (and the FTT do not say to which they are referring). Not only was this the appellant’s evidence, but the FTT might have drawn an inference from the timing of these events that they were connected to bullying,
v) in paragraph 31, the FTT deal separately with the head butting. They seem to accept the version given by the assailant to the Court Martial without knowing the appellant’s case or explaining why the incident was not linked to the pattern of bullying alleged or the complaint made about it,
vi) in paragraph 32, the FTT find that the pattern of bullying is not borne out by the evidence (a) without addressing the possibility that a victim of allegations such as those made by the appellant might withdraw them for legitimate reasons and, it seems, (b) without considering the significance of him having later repeated the allegation to Dr Fletcher and, it seems, (c) in reliance on an investigation by the Army that did not seek to probe the appellant on the nature of the bullying alleged and proceeded only on the statements of members of his troop who had been interviewed without details of the appellant’s allegations having first been obtained so that they could be put to them,
vii) so in paragraphs 29 to 32, in addition to not giving the appellant a proper opportunity to address the points concerning them and so to expand on and explain his case, the FTT make their findings without taking into account relevant circumstantial and other evidence in favour of the appellant’s case and/or without explaining how they reached their conclusions having regard to that evidence, and
viii) in paragraphs 33 to 35, where the FTT deal with the claim based on depression, it finds that it was due to multiple factors and not caused predominantly by service. In paragraph 35, they appear to accept the Secretary of State’s erroneous formulation and they do not explain that conclusion by, for example, a reference to the other factors and the relative weight to be given to them, nor to Dr Fletcher’s view (recorded in paragraph 34) that the appellant’s mental illness was “partially related to service”. They comment that Dr Fletcher’s view is wholly dependent on the appellant’s version of events but that is not necessarily a reason for rejecting a medical opinion. The FTT’s view is also ironic, given that they accept the accounts of others, such as the assailant who was court-martialled, without the same critical concern and, indeed, without hearing from the appellant on the central points.
We acknowledge that some of the errors of law identified in this paragraph overlap, and in particular that matters which inform the FTT’s ‘multiple factors’ approach to the ‘depression’ claim can be identified from the papers before the FTT. We also accept the Secretary of State’s point that in any event, once the FTT had found there was no bullying, the issues relating to multiple factors or causes evaporated.
70. Nonetheless, for the reasons we have set out this appeal succeeds on grounds (a) and (c) referred to in paragraph 3 above as well on the procedural fairness ground. We will deal with ground (b) when we consider the correct approach to be taken to “bullying cases”.
71. At one stage during the hearing it seemed that counsel for the appellant was asking us to make causation findings on the basis that a FTT properly directing itself could only, and so inevitably would, find in favour of the appellant. In our view, that is far from being the case and a proper investigation into the facts followed by an appropriately reasoned decision could found a decision in favour of either of the parties. It follows, as was accepted at the hearing that this appeal must be remitted to a differently constituted War Pensions FTT to determine pursuant to the guidance given by this decision.
The relevant provisions of the 2005 and 2011 Orders
72. Article 7 of the 2005 Order provides as follows:
(1) Subject to articles 15 (1A), 15A (1) and 15 (B)(1) benefit is payable in accordance with this Order to or in respect of a member or former member of the forces by reason of an injury which is caused (wholly or partly) by service where the cause of the injury occurred or after 6 April 2005.
(2) Where injury is not wholly caused by service, benefit is only payable if service is the predominant cause of the injury.
Article 8 of the 2011 Order provides as follows:
(1) Subject to articles 11 and 12, benefit is payable to or in respect of a member or former member by reason of an injury which is caused (wholly or partly) by service where the cause of the injury occurred on or after 6 April 2005.
(2) Where injury is partly caused by service, benefit is only payable if service is the predominant cause of the injury.
73. The key provisions are those which are emphasised in the quotations. It was not suggested that there is a material difference between the two articles and we agree that there is none.
74. Articles 6(2) and 7(2) respectively of the 2005 and 2011 Orders provide (subject in each case to paragraph (3), which is not relevant here) that “the rules of the Schemes are to be construed without reference to any other scheme applicable to the armed forces”. However, we agree with the parties that this provision of the AFCS Orders does not mean that the war pensions instruments that they replaced and the cases relating to them are not a part of the background that assists in the identification of the underlying purposes of the AFCS Orders and thus their construction and application.
75. The AFCS Orders introduce both:
i) a test “by reason of an injury which is caused (wholly or partly) by service”, and
ii) the predominant cause test “where injury is not wholly [or is partly] caused by service”.
“Predominant” is defined in both the 2005 and 2011 Orders as meaning “more than 50 per cent”.
These tests are not confined to claims based on “injury” but by articles 9 and 10 of the 2005 and 2011 Orders respectively they also apply to claims based on death.
76. The 2005 and 2011 Orders were drafted against the background of the war pensions instruments under which an injury was attributable to service if one of its causes was service even though there might be other and more powerful causes: see Minister of Pensions v Chennell  KB 250, (1946) 1 WPAR 253 (a claim under the civilian scheme made under the Personal Injuries (Emergency Provisions) Act 1939 and thus in respect of an injury “caused by” the discharge of a missile (and other matters)) and Marshall v Minister of Pensions  1 KB 106 (a case under a service instrument), both decisions of Denning J as he then was. In Marshall at page 109 Denning J made the points (with our emphasis) that:
“The task of the Minister and the tribunal is to ascertain what are the causes of the arising of the disease, not to assess their relative potency. If one of the causes is war service the disease is attributable to war service, even though there may be other causes.”
77. In that case, one of the causes of the claimant’s hernia was a weakness in his abdominal wall. Denning J concluded, however, (page 110) that another cause of the hernia was a cough produced in the claimant by his conditions of service and which caused his stomach wall to rupture. Even though the weakness of the stomach wall may have been the predominant cause of the hernia, the cough caused by service was also a cause. On the wording of the war pension’s instrument, this meant that the injury was properly attributable to service.
78. So it is apparent that the introduction of the “predominant cause test” and thus the need to assess whether an injury is wholly or partly caused by service introduces something new.
79. Applying the approach set out earlier, the ordinary language and the limited definitions used in setting the key elements of the test show that:
i) “injury’ is only defined to the extent that it includes ‘illness’,
ii) “predominant” is defined,
iii) the injury must be “caused” wholly or partly by “service”, and
iv) if the injury is not caused wholly by service benefit is only payable if “service” is the predominant “cause”.
80. “Cause” is a word with many overtones. It may refer to an event that immediately brings about an outcome or one that leads to it more remotely. It can also be used to mean attribution, viz. that something is capable of bringing about an outcome, or can be regarded as bringing it about, or can explain an outcome. Whether something is capable of, or regarded as bringing about a particular result involves a degree of judgment which is not generally required in straightforward cases of physical cause and effect; for example, where A punches B on the nose which then bleeds.
81. Also the language of the test identifies “service” as the cause or predominant cause. But, like “negligence” or “employment”, “service” is an abstract concept whilst “injury” is caused by one or more events or processes acting on the body or mind.
82. So in identifying the abstract cause of an injury it is necessary, as a matter of language and concept, to identify the events or processes – which we shall call the “process cause or causes” of the injury – and then to ask whether it is, or they are, sufficiently linked to service to satisfy the test that the injury due to each process cause is caused by service (or, using a shorthand, that that process cause is a service cause). Our use of the description “process cause or causes” is merely that and nothing else should be read into it.
83. Deciding whether a process cause is a service cause is an exercise of attribution, and so, of categorisation.
84. After that identification of causes and categorisation has taken place, the AFCS Orders introduce a new and further step. This is the “predominancy test”.
The underlying purposes of the relevant Articles
85. It was common ground that the purpose of the AFCS was broadly the same as the purpose of the war pensions scheme. Since the main hearing we have been referred to the Boyce Review (Cm 7798 - February 2010) and the AFCS statement of policy (August 2012) which support the view that the underlying purpose of the AFCS is the same as that relating to the schemes relating to war pensions that it replaced (as shown by a White Paper, Changes to War Pensions (Cmnd 6459), presented to Parliament in July 1943). There is, of course, an added requirement in the AFCS that compensation will only be payable if service was the predominant cause of an injury.
86. We accordingly accept the submission of the Secretary of State that the underlying purpose and intention of the two schemes were and remain (notwithstanding the huge changes since 1943 in a number of areas including for example the nature of service, health and safety and employment):
i) to establish an entitlement to benefit based on cause, as opposed to breach of duty or fault, for those who sign up to serve the nation
with the consequence that
ii) if service is not the cause (or sufficiently the cause) of the relevant injury the claimant is left to pursue other claims for compensation or support.
87. It was also common ground that the change in wording from ‘attributable’ in the war pensions instruments to ‘caused by’ in the AFCS did not mark a change in the attributive, and hence categorisation, exercise involved in causation in the latter.
88. We agree and we therefore consider that the principles in the old case law relating to ‘attributable to service’ remain relevant because they give guidance on the link that is required between the process cause and service to make it a service cause and so to satisfy the test that the injury be caused wholly or partly by service.
89. It follows that they provide assistance in the exercise of categorisation of process causes that is involved.
90. However and importantly we repeat they do not address the new “predominacy test”.
Are there useful principles in other areas of law that may assist in determining whether the test by the AFCS is satisfied?
91. During these proceedings, we were urged by the Appellant to import principles from other areas of law into the AFCS in deciding this appeal. In particular, the law of negligence and employment law were argued to be relevant. We do not accept that the principles of liability in other areas of law are of assistance in determining whether the test set by the AFCS is satisfied.
92. In the Marshall and Chennell cases Denning J referred to parallels in workmen’s compensation cases and to parallels in criminal law and the law of tort. In our view, by so doing he was not importing an analysis of liability by reference to those areas of law but was referring to and reasoning from established common law notions of causation where there were parallels with the notion of causation in the war pensions instruments.
“… does not depend on the injury having been intended or foreseen by the Ministry of Defence, neither does it depend on any breach of duty. There is simply a public law entitlement to compensation when the terms of the Order are met. The injury having been predominantly caused by service is one of those tests. One is not then looking at a question of causation for the purpose of attributing responsibility under some rule, but looking at the meaning of causation as part of the rule as to entitlement and responsibility for the payment of compensation”.
94. While Judge Mesher did say at  that, in applying the test in the AFCS, some assistance could be gained from considering the time honoured formula “personal injury by accident arising out of and in the course of employment” still current in the Social Contributions and Benefits Act 1992, he was careful to point out that, while there might be a parallel between the causation test in the AFCS and the “arising out of employment” part of the formula, there was no equivalent in the AFCS to the “in the course of employment” part of the formula.
95. In our view, it is important to stress that there is a difference between finding assistance in general principles of causation developed in other areas of law and in importing into the Orders an analysis of liability taken from fields such as employment law, social security law or tort. There is no justification for importing into the field of war pensions principles developed to establish liability in those areas (e.g. Jones v Tower Boot Company  2 All ER 406 on “course of employment” and Burton v De Vere Hotels Ltd  ICR 1 on the effect an employer’s ability to control what happened). Put broadly, those fields required constructs to extend liability to those who would not otherwise be liable. Under the AFCS Orders there is no need to extend liability: it is a given, so long as service wholly or predominantly caused the injury. Accordingly, their introduction into the principles to be applied under the test set by the AFCS is inappropriate and likely to add to rather than lessen the difficulties in its application.
96. That does not mean that changing views on what would be a fair and just result, having regard to the underlying purposes of the AFCS, are not relevant in difficult or borderline cases, but analogies with employment (or other) law are not necessary to introduce that approach to determining what, having regard to those purposes, the cause or predominant cause of an injury is.
97. Accordingly, we are of the view that the FTT did not err in law in not considering whether there was a breach of a duty of care by the Army in not removing the appellant from his unit following his allegations of harassment (ground (b) of the grounds on which permission to appeal was given by the Chamber President of the War Pensions FTT). But, as appears below, this does not mean that the fact that he was returned to the same unit following his allegations of harassment is not, or cannot be, a factor to be taken into account in deciding whether any injury due to a subsequent assault or bullying was caused by service.
The case law under the war pensions instruments and its application to the identification of a service cause
98. The war pensions cases show that not all injuries suffered by a serviceman when he is on duty are caused or sufficiently linked to - and so attributable to - service. On the other hand, the case law also shows that some injuries suffered when the serviceman is off duty are so attributable and so are categorised as a service cause. Examples of the earlier cases are:
i) Horsfall, where Tucker J considered that “attributable to” service has a different meaning to “in the course of” service. This led him to uphold a decision that a fatal heart attack that occurred during a squash game being played for his own amusement on Air Ministry property by an officer formerly responsible for service squash was not attributable to war service. The judge considered that the Pensions Appeal Tribunal had not misdirected itself in holding that the scope of the words “by service” or "attributable to service” embraced “only acts or conditions or events performed or undergone owing to and in compliance with the general or special obligations of service, together with all acts and conditions or events necessarily or reasonably incidental to these obligations as distinct from acts and conditions or events normally incidental to civil life". Tucker J stressed that: "The words used are 'attributable to,' and I think they have a different significance from 'in the course of.'" and also stated unsurprisingly that it must often be extremely difficult to put a particular case on one side or other of the dividing line.
ii) Marshall, where Denning J said: “The essential justification for a finding of attributability is that war service should be one of the causes of the disease. As I explained in Chennell's case, however, it must be a cause as distinct from being part of the circumstances in or on which the cause operates. Cases often occur when the disease would have arisen in any event, war service or no war service. In such cases it is not attributable to war service. They can be best illustrated by a metaphor. If a rope is weak and on that account breaks when it is carrying a normal or less than normal load, the cause of the break is not the load but the weakness of the rope. If, however, the rope is weak and breaks when carrying an abnormal load when it might have stood a normal load, there are two causes, one the weakness of the rope and the other the abnormally heavy load.” Denning J then distinguished the circumstances in which a cause operates from the event or stressor itself. He said: “The schizophrenia cases afford a good illustration. If schizophrenia arises in war service without any special stress or strain, it is not attributable to war service; but if there is severe war stress or strain immediately preceding the onset of symptoms, then it is. There are parallels in workmen's compensation cases, such as death in an epileptic fit in normal conditions – Lander' s case (1933) 102 LJ (KB) 768 – and in abnormal conditions – Wicks v. Dowell & Co., Ld  2 KB 225) and Wilson v. Chatterton  KB 360, 363.”
iii) Wedderspoon v Minister of Pensions  KB 562, (1947) 1 WPA 347, where Denning J held that a naval surgeon who had administered to himself an overdose of a drug had done so in his personal capacity. He said: “The cases show that when the cause of the death or disablement lies in the man’s own personal or domestic sphere, and the war service does no more than provide the circumstances in which the cause operated, it is not attributable to war service.”
iv) Monaghan v Minister of Pensions (1947) 1 WPA 971, where Denning J similarly concluded, in a case where a serviceman on active service died as a result of inhaling his own vomit after drinking, with other servicemen, raw spirit abandoned by the enemy, that although war service gave the opportunity for the drinking the real cause of the death was entirely the personal action of the serviceman.
v) Gaffney v Minister of Pensions (1952) 5 WPA 97, where Ormerod J concluded that a soldier’s claim based on epilepsy caused as a result of injuries to his head when he was attacked, probably on his way back to barracks after a night off, was not attributable to service because he was “on his own business”. He followed Horsfall and rejected the argument that the Royal Warrant applied because the attack would not have happened if the victim had not been in service because he would not otherwise have been in the place where the attack took place.
vi) Giles v Minister of Pensions and National Insurance (1955) 5 WPA 97, where the Appellant was on a day’s leave but wearing his uniform, as he was under a duty to do, when, while sitting by a clump of trees, he was shot in the back by a wad of blank cartridge fired by a cadet on an exercise. Although the Minister conceded the appeal, his counsel invited Ormerod J to expand on what he said in Gaffney. Ormerod J said: “I held in that case that the man was on leave and whatever had happened to him happened while he was in his own personal sphere of action, and that, I think, followed the decisions of Lord Justice Denning and Lord Justice Tucker (as he then was). That, of course is clearly the established rule and a rule which must prevail, but there may be circumstances - and each case, of course, depends upon its own circumstances - when it may be right to say that what happened to the man did not happen entirely "within his own personal sphere", but happened to him for some reason due to the compulsion of his service, and, under those circumstances, it appears to me, as, indeed it has appeared to the Ministry in this case, that he should be entitled to a pension. It is impossible to lay down any rule as to where that line should be drawn because, quite clearly, each case must depend upon its own facts, but in this case, the Appellant was compelled to wear uniform, and, because he was wearing uniform at that particular time although he was on leave, it is highly probable that he was singled out by the cadets or one of them as the target. The cadet was probably under the impression that he was one of the opposing band concerned in the exercise. Therefore I am satisfied that this appeal should be allowed and that the Ministry have acted properly in conceding it, not because it is an injury which occurred to a man when he was "within his own personal sphere", but because it was an accident which occurred due, in part at least, to the compulsions of his service”.
vii) Richards v Minister of Pensions and National Insurance (1956) 5 WPA 631, where Ormerod J again approached the issue by considering whether on the whole of the facts the appellant was engaged on some personal enterprise unconnected with any duty or compulsion of service. He thereupon dismissed an appeal where the claimant was injured by putting his fist through a window after a fight with Private “A” in the latter’s hut arising out of an argument between them following a dance. The exact way in which he so injured his hand and wrist after the incident in the hut was not established but Ormerod J concluded that: “… it is quite impossible to disassociate it in any way from that incident; and, clearly, for the Appellant to go into the hut, as he did, where he had no authority to be at all, for no other reason than to attack Private "A" in the way that he did was a purely personal matter and something which in no way it could be said was due to any compulsion of service.”
viii) Blakemore v Secretary of State for Social Security (14 March 1997, unreported) where Alliott J dismissed an appeal by a claimant who fell from a window and suffered injuries. The evidence was that he was sitting on the windowsill and urinating out of the window. He made no attempt to get back in and fell forward. It was accepted that if that was how his accident happened, it would not be attributable to service, and that is what a tribunal found in 1963. In 1990 the claimant submitted further evidence that he had been pushed or thrown from the window by his platoon sergeant and a colleague. The tribunal rejected his claim again. The focus of the appeal was on their approach to the standard of proof and the appeal was dismissed on the basis that the tribunal had come to the only proper factual conclusion open to it. However, Alliott J went on to say: “While it forms no part of my ratio decidendi I should deal briefly with one interesting point raised before me. If there was or might be substance in the appellant's later evidence, that is to say that if he was the subject of a revenge attack by someone he amongst others had reported for dereliction of duty, and who had subsequently been demoted, was the injury attributable to service? My provisional view, subject to argument in any case in which the point is more immediately at issue, is that it was. The attack was allegedly made on him as a result of what he had done, properly on his own account, as a soldier. In Mr Methuen’s phrase: “An injury suffered by a soldier because he is a soldier or because of what he did as a soldier is attributable to service". See Giles v The Ministry of Pensions and National Insurance”
99. The principle (or per Ormerod J in Giles the established rule) to be derived from these cases is that a claim must fail where the injury was sustained while the claimant was engaged on some personal enterprise unconnected with any duty or compulsion of service; or (as Denning J put it in Monaghan) service only gave the opportunity, or provided the setting, for the injury to occur.
100. The approach taken in the cases under the war pensions instruments:
i) shows that the attributability test was construed and applied as a causation test,
ii) demonstrates that a cause is distinct from the circumstances in or on which the cause operates. Cases often occur where the injury or disease would have arisen in any event, service or no service. In such cases it is not attributable to service unless caused by some special or particular characteristic, stress or strain attributable to service.
iii) recognises that in some cases it will be very hard to draw the line between when an injury is sufficiently caused by or linked to service to be “attributable” to it and so a “service cause” and when it is not,
iv) does not espouse a rule that a “one off incident” will not satisfy a causation test for payment, (and as we have mentioned this is recognized in the Secretary of State’s statement of policy),
v) demonstrates, as accepted before us, that the approach advanced by the Secretary of State before the FTT in this case (see paragraph 67) is too narrow and is accordingly wrong,
vi) whether a serviceman is on duty may be relevant but is not determinative, and
vii) recognises that injuries caused by assaults, whether by civilians or by other servicemen, may be due to service or may not: it all depends on the circumstances.
101. However, because the test under the war pensions instruments did not include the concept of predominant cause, these cases do not address how a decision maker should determine whether an injury (or death) was caused wholly or partly by service or what the predominant cause is.
102. Importantly they also confirm that there is no short cut to deciding the question whether an injury was caused by service by way of a simple formula that can be adopted in place of the statutory language. On the other hand, an appellate court or tribunal will not interfere with a finding of the FTT that an injury was caused by service if it has applied the proper test and the FTT has not in any other respect erred in law by, for instance, acting unfairly. These are points we made at paragraphs 52 - 64.
Guidance on bullying
103. The Secretary of State’s statement of policy correctly recognizes that the complex nature of service life can give rise to circumstances where it is difficult to establish whether an injury is caused by service or not and reflects issues raised in earlier case law. The policy states:
“2.7. Most cases are relatively straightforward in terms of establishing whether the injury claimed was caused by service. Given the potentially complex nature of service life, there can be instances where it is difficult to establish whether an injury is caused by service or not. In these cases, the decision maker will be required to make a judgment, having taken account of all the facts of the case, as to whether the injury was caused wholly or partly by service or not. Where the case is difficult or borderline the decision maker may need to engage with the MOD policy team. This can mean it takes a little longer than usual to arrive at a final decision.
2.8. Factors that may mean an injury was more likely to be caused by service include, but are not confined to: acting on orders, responding to a service-related emergency, being on operations or exercise, participation in service-approved and recognised sport, being in a hazardous environment as a result of service, or performing an activity specified in the individual’s job description.
2.9. Factors that may mean an injury was less likely to be caused by service include, but are not confined to: disobeying orders, being at a social event, using alcohol or drugs, or participation in a sporting activity that has not received prior approval from the Defence Council.”
104. None of the old cases was concerned with bullying as such but in principle it clearly needs to be approached in the same way as assaults, having regard to the detailed circumstances of each case. But bullying often operates as a lengthy process rather than a single event or even a series of events and, while it may form part of the background to one or more events that cause specific injuries, the combined effect of the process of bullying may itself give rise to mental injury, as is claimed in the present case, even in a case where no significant physical injury is caused.
105. In our view, communal living in barracks or other communal settings with members of a serviceman’s unit or group is a normal, and often inevitable, part of what a serviceman signs up to do. This brings with it the rules, discipline and policy set by the relevant branch of the armed forces.
106. The armed forces recognise that such a compulsory setting can give rise to bullying and we accept, as was submitted to us, that there is a strong policy to prevent and discourage bullying in that and other service settings. In our view, that policy and its implementation is an aspect of the control of, and requirements relating to, the compulsory communal living of a serviceman as such. Nevertheless, the existence of a policy, even a strong one, does not mean that bullying in the forces cannot exist.
107. The difficulty, in our view, lies in establishing that any bullying is to be categorised as a service cause of any injury that might result from it. This is a matter of judgment and so of reasoning from the findings of fact made in each case.
108. If individual or group acts of that nature are committed by, or are condoned or encouraged by, a more senior serviceman this, too, would be a factor tending to link them and their consequences to service.
109. On the other hand, if the attack happened because of a matter that was not directly connected to service (e.g. a grudge, rivalry or an argument over a personal matter) that would be a factor against that conclusion. Such matters can be part of the everyday interaction with others.
110. Similarly, personal characteristics and factors that are not directly connected to service (such as, in this case, the appellant’s stutter, his Liverpudlian background and the problems relating to his wife and child) may lead to banter and teasing amongst individuals living communally in close and regulated circumstances. It will be a question of fact and degree whether the teasing has crossed the line between servicemen having to learn to live with people who rub them the wrong way and bullying or a campaign of bullying.
111. Breach of the armed forces policy to prevent bullying of servicemen in situations they signed up to or in settings imposed on them because they are servicemen, may provide the necessary link to service. In this context issues relating to the implementation of the policy and the protection of a victim would be relevant as pointers to the extent of the linkage between the events leading to the injury and service.
112. When an injury occurs as a result of an attack by one serviceman on another who has made a complaint about bullying which put those he has complained about at risk of punishment or led to their punishment an analogy can be drawn with the example given by Alliott J in Blakemore and it will have to be considered (a) whether that attack was a continuation of a campaign of bullying and so linked to that bullying and service, and further or alternatively (b) whether that attack is linked to service because of the risk of punishment under service discipline arising from the making of the complaint (as to which the truth of the underlying allegation of bullying is not a necessary, although it may well be a relevant, ingredient).
113. In difficult cases the decision maker can in our view ask himself whether in all the circumstances of the case his reasoning founds the conclusion that to promote the underlying purposes of the AFCS it would be fair and just to conclude that injury caused by bullying was caused by service.
114. We are very aware that the above description of factors and their effects, like the factors and effects listed in paragraphs 2.8 and 2.9 of the Secretary of State’s policy statement, cannot be anything other than examples of relevant or potentially relevant matters. And, we agree with the Secretary of State that it would be wrong to be prescriptive about factors or features and their effects.
115. This leads back to the key points for decision makers on such issues that it is important that they identify their reasoning process and that by so doing they will show why they have reached their conclusion they have and thus how they have understood and applied the test governing their decision.
The categorisation exercise
116. Whether an injury should be categorised as caused by service is a matter that can change over time by reference to changing circumstances, societal or legal norms. So, for example, changes in the requirements of service, policies, terms and conditions and disciplinary regimes may introduce new ideas of what can fairly be considered a sufficient connection to service so that compensation is payable in line with the underlying purpose of the AFCS.
117. It is also inevitable that the tests in the Articles will fall to be applied to a wide range of different circumstances and that within the range of those circumstances will be:
i) issues that relate to the circumstances in which the incident or incidents or course of conduct that are said to have caused or contributed to the relevant injury (or death) and to connect them to the service of the claimant, and
ii) matters that do not relate to such circumstances and which existed before or came into existence after the service of the claimant and so which engage the “predominancy test”.
The steps to be taken in the application of the AFCS test
118. The analysis we have set out founds the conclusion that the correct approach to the issues of cause and predominant cause under the AFCS is:
i) First identify the potential process cause or causes (i.e. the events or processes operating on the body or mind that have caused the injury);
ii) Secondly, discount potential process causes that are too remote or uncertain to be regarded as a relevant process cause;
iii) Thirdly, categorise the relevant process cause or causes by deciding whether the circumstances in which each process cause operated were service or non-service causes. It is at this stage that a consideration of those circumstances comes into play and the old cases on the identification of a service cause applying the old attributability test provide guidance.
iv) Fourthly, if all of the relevant process causes are not categorised as service causes, apply the predominancy test.
119. This staged process accords with the approach advanced by the parties at the second hearing before us.
The predominancy test
120. In this case this is only relevant in respect of the claim based on depression.
121. Judge Mesher summarises the kernel of his judgment in EW in SV v Secretary of State for Defence (AFCS)  UKUT 541 (AAC) at , saying:
“There [in EW] I held that the fundamental test under the AFCS Order 2005 was not whether a claimant was in service or was on duty at the time of the incident in question, but whether the injury was caused and predominantly caused by service.”
122. Based on that passage it was submitted to us on behalf of the Secretary of State that:
i) where service competes with other causes of the injury it needs to predominate, and
ii) to predominate it needs to contribute conceptually more than one half of the causative stimulus.
This submission of the Secretary of State was not challenged. We agree with both points and the way in which this description of the tests is divided. The “more than one half” is based on the definition of predominant in the Orders.
123. We would, however, observe that that in both EW and SV there was only one cause of the claimant’s injury: – being struck by a car in one case and diving into a sandbank in the other – and the issue was simply whether that cause was a service cause. We would therefore have expressed the reasoning differently because we would not have reached the fourth stage set out in paragraph 118 above and thus a consideration of “predominant cause”. That would not of itself have affected the outcome in either case but it would have made paragraph 40 of SV unnecessary. It is to be noted that both cases were decided under the 2005 Order and involved consideration of article 10, the unsatisfactory language of which (“predominant cause” appearing in circumstances where, on our analysis, the phrase should strictly be “only or predominant cause”) may explain the references to “predominant cause” in the reasoning in the decisions. (The linguistic difficulties arising from the use of the phrase “predominant cause” in the 2005 Order may also explain why article 8 of the 2011 Order is in different terms from article 7 of the 2005 Order, albeit having the same effect.)
124. The parties did not elaborate on how the “predominancy test” was to be applied.
125. However, the submission of the Secretary of State reflects:
i) the process of reasoning we have identified, and
ii) an approach that in applying the predominancy test the exercise of judgment is an assessment of the relative potency of process causes (i.e., the events or processes operating on the body or mind that have caused that injury or death) which have been categorised on one hand as service causes and on the other hand as non-service causes.
126. In EW Judge Mesher does not divide up the reasoning process into the stages we have set out and, as we have said, it was not a case where the predominancy test arose. Nonetheless, we consider that at both (a) the categorisation stage and then at (b) the predominancy stage matters of impression arise and that the sophisticated approach envisaged by Judge Mesher in EW at  is appropriate. This is because the exercise of judgment at these stages may well involve the weighing of factors of different types (“apples and pears”) and no formula is available to assist. This approach means that the reasoning that demonstrates how the test has been applied may involve different analyses and so the consideration of different questions according to the different circumstances of the claim.
127. We acknowledge the dangers of giving any prescriptive guidance on the application of the predominancy test. But it seems to us that a need for some guidance on the correct approach to the application of this test is demonstrated by considering whether the claimant in Marshall would have satisfied the predominancy test and receive an award under the AFCS.
128. As we have said, Marshall was concerned with competing causes of injury. The claimant suffered a hernia, which had two process causes: one was the claimant having a weak stomach wall, which was undeniably a constitutional cause, and the other was him having had a cough, in which his conditions of service were accepted as having played a part. In these circumstances, service was a cause of his worsened condition: the service induced cough acted on his hernia. The result was that one of the process causes (the cough) was attributable to service for the purpose of the war pensions instrument.
129. The old war pensions instruments, like the current ones, made no express provision for cases where an injury had more than one process cause except through the concept of “aggravation”. Denning J was at pains to point out that “aggravation” had a limited scope and that the Minister of Pensions had been wrongly interpreting the instruments. He said (with our emphasis):
“If the injurious process exists before war service, the only question is one of aggravation and that depends on whether the injurious process is accelerated or intensified by war service. If the injurious process arises during war service, the question of attributability depends on whether war service was one of the causes of its arising Much confusion surrounded this question before the cases on causation. The medical men advising the Minister used to divide cases into two classes, on the one hand, predominant causes, and on the other hand, contributory causes. They used then to say that the disease was "attributable to" the predominant cause and "aggravated by" the contributory cause. This method of approach was especially noticeable in the predisposition cases. For instance, in hernia, when the predominant cause is the inherent weakness of the wall of the stomach and a contributory cause may be an injury in war service, they used to hold that the hernia was not attributable to war service, but was aggravated by it. The cases in the Court of Session and in this court have amply shown that that approach is wrong. The task of the Minister and of the tribunal is to ascertain what are the causes of the arising of the disease, not to assess their relative potency. If one of the causes is war service the disease is attributable to war service, even though there may be other causes and, it may be, more powerful causes, operating, and to which it is also attributable. There are parallels to be found in other branches of the law. Take manslaughter. I tried a case the other day of a man who hit another on the nose. The blow was not such as to do harm to an ordinary individual, but unknown to all concerned this injured man had a septic antrum. He might have gone on for a long time without its causing him much trouble, but the effect of the blow was to release the poison from the antrum into his system so that he died within a few days. One cause of his death, perhaps the most potent cause, was the septic antrum, but another cause was the unlawful blow. The man who struck him was found guilty of manslaughter because he caused the death. Also, in the case of duodenal ulcer or hernia (in each of which inherent weakness or predisposition is a powerful cause) it has been held that conditions of employment may also be a cause entitling the man to pension or compensation as the case may be: see Huddersfield Corporation v. Watson  KB 842 (duodenal ulcer) and Hughes v. Lancaster Steam Coal Collieries, Ld.  2 All ER 556 (hernia).”
130. It is noteworthy that in all the cases to which Denning J referred there was a constitutional cause or predisposition to injury (and we refer to them as constitutional weaknesses of the claimant) as well as the crime or the stress or strain of employment.
131. We do not consider that Denning J’s discussion of “predominant cause” in Marshall provides persuasive guidance on how to apply the “predominancy test” under the AFCS because he was dealing with a war pensions instrument in which the predominance of a cause was immaterial. Indeed, the main reason he referred to “predominant cause” was to put paid to an error in construing the instrument into which the Minister and his medical advisers had fallen and for which there was no justification in its wording. Further, he gives no indication of the basis for his classification of predominant and lesser causes.
132. In this context we do not see any sign that the intention behind the AFCS is to deprive those with constitutional weaknesses from the protection usually regarded as appropriate in other compensation schemes, that is to say the ‘thin skull’ approach.
133. We acknowledge that, in exercising the judgment between process causes that have been categorised into service and non-service causes of the injury, a literal approach to the language of the test in the 2005 and 2011 Orders could, in an equivalent case to Marshall, found the view expressed by Denning J with the result that the claimant would not get an award because the predominant cause of the injury was the constitutional weakness and the cough was a lesser cause.
134. But in our view the width of the language permits a more sophisticated approach to deciding whether, as the Secretary of State put it, conceptually the service cause contributes more than one half of the causative stimulus for the injury claimed, and thus whether service is the predominant cause in a case where (after the categorisation process) the only competing causes are service and constitutional or other pre-existing weaknesses. In such a case the decision maker generally should firstly consider whether, without the ‘service cause’, the injury would:
a) have occurred at all, or
b) have been less than half as serious.
135. If the answer to the first question is that the injury would not have occurred at all in the absence of the service cause, we consider that this can and generally should found a conclusion that the service cause is the predominant cause of the relevant injury. It seems likely that a claimant in Mr Marshall’s position would succeed on this basis.
136. If however that is not the answer to the first question, the second question will generally found the answer to whether the service cause is the predominant cause of the relevant injury. Thus the second question is likely to be determinative in the present case if it is found that the claimant’s depression was caused both by service and by pre-existing domestic factors.
137. We consider that this approach fits with and promotes the underlying intention of the AFCS to pay compensation for an injury that has more than one process cause that under the categorisation exercise we have described fall to be taken into account as respectively service and non-service causes.
138. We repeat that this is not intended to be prescriptive guidance and that it may need to be modified or abandoned in some cases. For example, we acknowledge that timing issues could cause complications that warrant a departure from it.
The present case
139. In the present case, there was only one process cause of each of the two assaults, i.e., the head butt and the stabbing. In relation to those cases, the only question is therefore whether in each case that cause is a service cause and the claimant’s case on that is based on his assertion that they are sufficiently linked to the bullying he asserts took place and his complaint about it.
140. Accordingly in those cases the fourth stage identified in paragraph 118 above is not reached.
141. However, as regards the depression, there is evidence that it was at least partly caused by the claimant’s reaction to events at home which could not be attributed to service, so that, if it were accepted that another process cause based on his relationship with other soldiers was a service cause, there is the potential for a conclusion at the third stage that the depression was partly caused by service and the fourth stage would have to be considered.
What needs to be done at the remitted hearing
142. The first thing that needs to be done is for the appellant to set out his case clearly. It is not enough simply to allege that a campaign of bullying existed. Having relied throughout these proceedings on a campaign of bullying, he must give particulars of this allegation: who, what, when where, why. Although it has been quite a few years since the events in issue, he will have to do his best with these factual matters. We direct him to provide a witness statement within a month of this decision being sent to him and we direct the Secretary of State to provide a submission in the light of it within a month of being sent it.
143. The appellant may also gather from what we have said in this decision that there are difficult legal issues to be tackled at the rehearing. One of these is whether his physical injuries and his mental health problems were caused by any bullying or his complaint that he was the victim of bullying and if so, whether that cause is to be categorised as a service cause.
144. The parties and the FTT will need to consider whether any further medical evidence is required for the proper application of the predominancy test. We make no direction in that regard.
145. The FTT should take the staged approach set out above and so:
· First identify the potential process cause or causes (and so the events or processes operating on the body or mind that have caused the injury);
· Secondly, discount potential process causes that are too remote or uncertain to be regarded as a relevant process cause;
· Thirdly, categorise the relevant process cause or causes and so decide whether the circumstances in which each process cause operated were such that it is a service cause. It is at this stage that a consideration of those circumstances comes into play and the old cases on the identification of a service cause applying the old attributability test provide guidance.
· Fourthly, if all of the relevant process causes are not categorised as service causes apply the predominancy test.
146. It seems likely that if the FTT find a service cause for the claim based on depression the predominancy test will fall to be applied by the FTT asking itself whether without that service cause the depression would have been less than half as serious.
(Signed on the original)
Mr Justice Charles
Upper Tribunal Judge Rowland
Upper Tribunal Judge Lane
Date: 11 June 2015