FEDERAL COURT OF AUSTRALIA Read v Military Rehabilitation and Compensation Commission [2018] FCA 848 Appeal from: Read and Military Rehabilitation and Compensation Commission (Compensation) [2017] AATA 1109 File number: QUD 409 of 2017 Judge: LOGAN J Date of judgment: 8 June 2018 Catchwords: ADMINISTRATIVE LAW – military compensation – appeal against decision of the Administrative Appeals Tribunal – whether a want of logicality in Tribunal's reasons – omission of a link in findings of fact as to why applicant ineligible for compensation because he did not suffer from the metal ailment claimed. Held – appeal allowed, decision of the Administrative Appeals Tribunal set aside. PRACTICE AND PROCEDURE – appeal against decision of the Administrative Appeals Tribunal – power of the Court under s 44 of the Administrative Appeals Tribunal Act 1975 (Cth) to make prescriptive remitter order. Legislation: Administrative Appeals Tribunal Act 1975 (Cth) s 44 Commonwealth Employees' Contribution Act 1930 (Cth) Military Rehabilitation and Compensation (Consequential and Transitional Provisions) Act 2004 (Cth) Safety, Rehabilitation and Compensation Act 1988 (Cth) Veterans' Entitlements Act 1986 (Cth) Cases cited: Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321 Collector of Customs v Pozzolanic Enterprises Pty Ltd (1993) 43 FCR 280 Dingwall v Commonwealth of Australia [1994] FCA 1099 Minister for Immigration and Ethnic Affairs v Wu Shan Liang (1996) 185 CLR 259 Muggeridge v Minister for Immigration and Border Protection (2017) 351 ALR 153 Wright v Commonwealth of Australia [2005] VSC 200 Date of hearing: 12 December 2017 Date of last submissions: 12 December 2017 Registry: Queensland Division: General Division National Practice Area: Administrative and Constitutional Law and Human Rights Category: Catchwords Number of paragraphs: 39 Counsel for the Applicant: Mr D Honchin Solicitor for the Applicant: Purcell Taylor Lawyers Counsel for the Respondent: Ms K Slack Solicitor for the Respondent: Australian Government Solicitor ORDERS QUD 409 of 2017 BETWEEN: ALLAN WINFIELD READ Applicant AND: MILITARY REHABILITATION AND COMPENSATION COMMISSION Respondent JUDGE: LOGAN J DATE OF ORDER: 8 June 2018 THE COURT ORDERS THAT: 1. The appeal be allowed. 2. The decision of the Administrative Appeals Tribunal (Tribunal) dated 19 July 2017 be set aside. 3. The matter be remitted to the Tribunal for rehearing with the following directions: (a) on that rehearing, the Tribunal is not to be constituted by the person who made the decision to which the appeal relates; and (b) the rehearing be conducted on the footing that the findings of fact made by the Tribunal at para 97 of the reasons for the decision to which the appeal relates stand as findings of fact on the rehearing. 4. The respondent pay the applicant's costs of and incidental to the appeal, to be taxed if not agreed. Note: Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011. REASONS FOR JUDGMENT LOGAN J: 1 Allan Winfield Read (Applicant) served in the Australian Army for over 42 years, achieving the most senior non-commissioned rank, Warrant Officer Class 1. For the period from 24 January 1966 to 24 January 1986, he was a member of the Regular Army. Thereafter, he served in the Army Reserve, being finally discharged on 19 May 2008. 2 From 24 January 1966 until 29 January 1970, Mr Read was posted to the Army Apprentice School at Balcombe, Victoria. On the evening of 29 August 1968, while Mr Read was serving as an Army Apprentice, an accident occurred at the School. Six Army apprentices were electrocuted when a flag pole they were then carrying came into contact with overhead electricity wires. Two of them, Apprentices P.K. Russell and A.J.T. Pearce, were killed in that incident. They suffered horrific injuries. The survivors sustained serious burns. 3 The Administrative Appeals Tribunal (Tribunal) found that Mr Read, though he did not witness the accident, was one of those at the School who came to the scene in its immediate aftermath. The Tribunal was satisfied that Mr Read: (a) provided assistance in the post-accident recovery phase; (b) attempted to provide assistance to Apprentice Pearce; and (c) helped in the transportation of the body of Apprentice Pearce from the accident scene to the Army ambulance. Mr Read had turned 19 earlier in August 1968 (date of birth 1 August 1949). 4 It was not until April 2010 that Mr Read made a claim for compensation to the respondent Military Rehabilitation and Compensation Commission (Commission) for delayed onset post-traumatic stress disorder (PTSD). That claim was rejected by the Commission both initially and on internal review on the basis of an absence of satisfaction that Mr Read was present in the aftermath of the accident. The Tribunal affirmed the rejection of the claim not on this basis but rather because the Tribunal was not reasonably satisfied that the Applicant had delayed onset PTSD. 5 Mr Read has appealed against the Tribunal's decision pursuant to s 44 of the Administrative Appeals Tribunal Act 1975 (Cth) (AAT Act). 6 Such appeals lie only on a question of law. The notice of appeal is prolix both in its pleading of alleged questions of law and related grounds of appeal. The hearing of the appeal resulted in quite some revision by counsel for Mr Read as to what, truly, was the error of law made by the Tribunal. The error, as I understood it, was said to be that the Tribunal had erred by impermissibly undertaking its own assessment as to whether Mr Read satisfied the diagnostic criteria for PTSD as found in the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5), as opposed to deciding whether, on the material before the Tribunal, Mr Read was entitled to compensation according to governing statutory criteria. 7 Unlike, with respect to the parties in their submissions on the appeal, the Tribunal did at least turn its mind to the existence, at the time of the accident in 1968, of the Commonwealth Employees' Contribution Act 1930 (Cth). Likewise, the Tribunal referred (Reasons, para 141) to a transitional provision, s 124 of the Safety, Rehabilitation and Compensation Act 1988 (Cth) (1988 Act), but did not explore how that became applicable by reference, for example, to the Military Rehabilitation and Compensation (Consequential and Transitional Provisions) Act 2004 (Cth). Instead, seemingly in response to the way in which the case was presented by the parties, the Tribunal came, by reference to the definition of "disease" in s 5B of the 1988 Act, to identify the issue as to whether it was satisfied that Mr Read's condition was "an ailment suffered by an employee, or an aggravation of an ailment, that was contributed 'to a significant degree' by the employee's employment" (Reasons, para 146). Earlier and apart from an issue (issue (a)) as to whether Mr Read was at the accident scene at all, the Tribunal had identified (Reasons, para 28), by reference to the Commission's State of Facts, Issues and Contentions, the issues as for determination as: (b) The correct diagnosis of the Applicant's current psychiatric condition and the date of onset of this condition; and (c) Whether the Applicant's psychiatric condition developed as a result of the 1968 electrocution incident. 8 Each party was legally represented in the hearing before the Tribunal. As is so frequently the case, the Tribunal's reasons were influenced by the way in which the parties came to identify the issues for determination. The Tribunal's reasons need to be read in this light. Neither party submitted on the appeal that the just-quoted formulation of the issues by the Tribunal did not reflect the end result of the application of the compensation legislation in force from time to time. I have not therefore explored the correctness of their mutual assumption. 9 Mr Read's submission was that, in addressing the identified issues, the Tribunal had not in the end addressed which body of medical evidence about Mr Read should be accepted but rather looked to DSM-5 as the touchstone and made its own diagnosis. 10 There was competing specialist medical evidence before the Tribunal as to from what Mr Read suffered. 11 One psychiatrist, Dr Sarah Beaney, who was called on behalf of Mr Read, opined in a report of 29 January 2016, by a detailed and reflective reference to DSM-5 and other specialist literature, notably including a study by Horesh et al of Israeli veterans of the 1982 Lebanon War (the Horesh study) that he suffered from PTSD. 12 Another, Dr Michael Likely, to whom Mr Read had been referred in 2009 by the Department of Veteran's Affairs, opined that Mr Read suffered from an adjustment disorder with anxious mood. He attributed this to the 1968 accident and also to a 2004 incident that had also occurred during the course of Mr Read's military service. He also made an impairment assessment by reference to a document used for the assessment of pensions payable under the Veterans' Entitlements Act 1986 (Cth) (VEA Act), the Guide to the Assessment of Repatriation Pensions (GARP). That impairment assessment had no relevance to the compensation claim. 13 Mr Read was also referred to another psychiatrist, Dr Johan Scheepers, this time at the request of the Commission. He opined, in a report of 29 December 2015, "The diagnosis of posttraumatic stress disorder is specifically not included in Mr Read's diagnostic formulation." In this report, Dr Scheepers expressed this view: Again the onset of symptoms at this stage of Mr Read's life may have as much to do with his extensive physical illness, changes in work environment and changes in the home environment as any long-standing underlying psychological vulnerability. Based on these considerations, it is my opinion that to ascribe the appearance of symptoms of mental ill health late in life to events that happened in 1968 is not a reasonable argument. 14 In a supplementary report of 18 May 2016, prepared by Dr Scheepers after having had the benefit of considering Dr Beaney's report, he allowed that, "The argument that PTSD may develop and present itself late in life many years after the initial, triggering event appears to be a clinical reality" but opined, "whether this condition may be applied to Mr Read remains an open question". He considered that Dr Likely's 2009 diagnosis of adjustment disorder with anxiety symptoms to be "a more accurate reflection of Mr Read's presentation than [PTSD]". What Dr Scheepers did not go on to explain in this report was how this preference was to be reconciled with Dr Likely's view that the condition so diagnosed on this 2009 presentation was related to the events in 1968 and 2004, whereas he had related it to different and non-defence service related causes in his initial report. 15 The Tribunal also had evidence concerning Mr Read before it (and referred in its reasons to this) from Mr Neil Graham McLeod, a registered nurse practising at Townsville Private Mental Health Service. Mr McLeod's evidence was favourable to Mr Read. The Tribunal discounted (Reasons, para 104) this evidence because his "observations are those of a layman and were not of particular assistance … in determining the nature of the Applicant's psychiatric condition. Ultimately, the evidence of the three abovementioned psychiatrists has been critical in the determination of this matter." As pleaded in his notice of appeal, there was some complaint by Mr Read about this discounting of Mr McLeod's evidence. 16 I did not understand this complaint to be pressed on the hearing of the appeal on behalf of Mr Read on the basis that, on reflection, it was accepted that no question of law was entailed in the complaint. In any event, in Minister for Immigration and Ethnic Affairs v Wu Shan Liang (1996) 185 CLR 259 (Wu Shan Liang) the High Court cited with approval a passage from a judgment of the Full Court of the Federal Court in Collector of Customs v Pozzolanic Enterprises Pty Ltd (1993) 43 FCR 280 at 287 that the reasons of an administrative decision-maker were "not to be construed minutely and finely with an eye keenly attuned to the perception of error". Approaching the Tribunal's reasons in this way and reading the passage concerned in context, the Tribunal's description of Mr McLeod as a "layman" is nothing more than a way of emphasising that he was not a medical practitioner at all, let alone one specialist in psychiatry, that it was to these specialist views that the Tribunal would give decisive weight and that it was necessary for the Tribunal to determine which of these was to be preferred. The explanation as to why the Tribunal discounted Mr McLeod's evidence was, with respect, rational. Such evaluative judgements on the factual material before it are for the Tribunal. The discounting of Mr McLeod's views entailed no error of law. Further, Mr McLeod's views were not ignored by Dr Beaney or Dr Scheepers. His views formed part of the material which informed the opinions they expressed about Mr Read. 17 The initial part of the Tribunal's reasons, dealing with whether Mr Read was present at the accident scene in 1968 is, with respect, a model of thorough and reasoned analysis, eloquently expressed. Reflecting upon this and upon what was so emphatically approved in Wu Shan Liang about the reasons of administrative decision-makers, I have been much troubled by whether the balance of the Tribunal's reasons evidences, as the Commission submitted, nothing more than a permissibly reasoned preference for one body of specialist medical evidence over another or whether the Tribunal has indeed, as was submitted for Mr Read, impermissibly elevated DSM-5 to something akin to a statutorily mandated statement of principles under the VEA. 18 Addressing this ground of challenge does require an analysis of the Tribunal's reasons to do justice both to the parties as well as, for that matter, to the Tribunal. But I have undertaken this task firmly bearing in mind the observations approved in Wu Shan Liang. 19 It was common ground on the appeal, as it necessarily had to be, that DSM-5 has no legislative sanction as the touchstone for whether a compensation claimant is suffering from a mental ailment. Yet it is given great prominence in the Tribunal's reasons, overtly responsive to a like prominence in the submissions made to the Tribunal on behalf of the Commission. The following passage from the Tribunal's reasons (paras 195 to 205), which includes the Tribunal's own choice of headings ("Criterion C" – read "G" – is taken from the DSM-5 diagnostic criteria), highlights this prominence: Application of DSM-5 195. Ms Slack comprehensively addressed the diagnostic criteria outlined in DSM-5 in her written submissions – Exhibit 18 paras 41 – 57. 196. Her submissions comport with the factual matrix outlined above are soundly based and each is accepted by the Tribunal. 197. The evidence presented supports the diagnosis of Dr Scheepers. The following observation of Dr Scheepers flows logically from the evidence – Exhibit 11 p.3: "Mr Read evidently has access to enough resilience to contain the effects of the incident to the extent that he continued to function adequately. He may well have been irritable with his wife or started drinking more alcohol, but such observations would not qualify him for a diagnosis. My opinion remains the same, that neither in 1968 after the events nor in 2009, 40 years after the event, does Mr Read's behaviour or clinical presentation present a convincing picture of posttraumatic stress disorder. I have assessed many members of the ADF who have significant PTSD and Mr Read's history does not conform to that experience." 198. However, it is important to highlight the most significant feature of this matter which weighs heavily against a finding of delayed PTSD. 199. Criterion C requires that a person suffering PTSD persistently avoid stimuli associated with the traumatic event. This can entail avoiding distressing memories or thoughts or avoiding external reminders that arouse distressing memories. 200. The Applicant continued serving in the Australian Army for more than 17 years after the 1968 incident, and then enlisted in the Army Reserves for another extended period. After the 1968 incident he underwent combat training in the SAS and rose through the ranks, reaching the rank of Warrant Officer 1. If not for his assault of a Major in circa 1979, his evidence is that he would have continued to be promoted. All of the evidence suggests he thoroughly enjoyed Army life and was very good at the tasks required of him. Far from avoiding service life after the 1968 accident, the Applicant embraced it with vigour and enthusiasm. 201. After leaving the Army the Applicant went into the very stressful industrial relations area, focusing on workplace health and safety issues. He was confronted on a regular basis with people being injured, and on a few occasions being, killed. He was an inspector, and would have been engaged in many confronting and confrontational situations. 202. Throughout this period, he was actively playing contact sports (rugby) and engaging in vigorous and potentially dangerous social activities (hunting wild pigs). 203. None of this comports with a diagnosis of avoidance behaviour as mandated by Criterion C. [sic] 204. Moreover, it cannot be said that the Applicant meets Criterion G, namely his condition caused significant distress and impairment in social, occupational or other areas of functioning. 205. There simply is no convincing evidence that from 1968 until at least 2008 that the Applicant was in any significant way suffering from an impairment of any of those types. On the contrary, the evidence suggests that the Applicant was a social individual who excelled in social and professional activities. He rose through the Army ranks, he excelled in the public service, he was an active Reservist, he enjoyed fishing, camping and pigging. [The heading and other reference to "Criterion C" are an obvious typographic error and should be "Criterion G".] [Ms Slack appeared for the Commission both before the Tribunal and on the appeal.] 20 The above-quoted passage followed an earlier reference by the Tribunal (at para 98), unremarkable in itself, to a need to make a finding as to the condition from which Mr Read suffered: 98. Ms Slack submits (Exhibit 18 para 11) that unlike in other psychological ailment cases, determining a precise diagnosis of the Applicant's condition is critical because of the historical nature of the claimed causative event and need for the Applicant to be diagnosed with a condition that can have a delayed onset. 21 Dr Scheepers does not make explicit mention of DSM-5 in his initial report at all. It is Dr Beaney who introduces DSM-5 in her report, assessing from what condition he suffers by overt reference to its various criteria and also other specialist knowledge and literature, including, as mentioned already, the Horesh study. In so doing, she makes particular reference to why, in her view, Criterion G is met in Mr Read's circumstances (voicing an understanding that there is a difference between her and Dr Scheepers in this regard). Dr Scheepers makes a passing, responsive reference to DSM-5 in his supplementary report, noting that delayed onset PTSD is not included in DSM-5, "except as being with onset after 6 months". 22 The Tribunal expressly recognised (at para 99) that Dr Beaney had made reference to DSM-5 in her report and that she considered that he met the criteria for PTSD specified in it. 23 Immediately after this recognition, the Tribunal remarks (at para 100) of DSM-5 that, "The various editions of this Manual have been cited and relied upon in numerous matters in all Australian jurisdictions." The Tribunal then refers (ibid) to a passage concerning an earlier edition of this Manual in one of the HMAS Voyager damages claim cases, Wright v Commonwealth of Australia [2005] VSC 200 (Ashley J) (Wright), without identifying that it was common ground in that case (Wright at [55]) that, "PTSD is a diagnosis made by reference most often to criteria set out in the Diagnostic and Statistical Manual of Mental Disorders which is published by the American Psychiatric Association. It is now in its fourth edition, its title being shortened to 'DSM-IV'." 24 Nowhere does the Tribunal make a positive finding either that it was likewise common ground in the present case or that there is a consensus amongst the specialist psychiatrists that, unless a claimed PTSD condition conforms with the criteria in DSM-5, it cannot be accepted as PTSD or even express a preference for the view of a specialist who had ascribed such a status to DSM-5. Nor even does the Tribunal make a finding that specialist evidence supports a view that, ordinarily, PTSD ought to display features described in DSM-5 but that there is a respectable and emerging view in the specialty that, in particular circumstances, delayed onset PTSD might not. Yet that looks to me to be the unarticulated premise upon which the Tribunal's discussion and related conclusions under the headings "DSM-5" and "Criterion C" (read "G") are based. 25 As it did not have statutory recognition, it was not for the Tribunal to approach DSM-5 as if it were a checklist in the nature of a statement of principles under the VEA which did not, for this reason, require an expressly adopted, supporting evidentiary foundation for its forensic application. Incongruously with the absence of such express recognition in its reasons, the Tribunal was, patently, aware of this in an exchange in the course of submissions with Mr Read's counsel, Mr Honchin (who also appeared for Mr Read on the appeal), as the following passage (transcript, 14 June 2017, pp 164-165) reveals: SENIOR MEMBER: You're confusing me now, Mr Honchin, that was not my question. The statement of principles aren't worth a pinch of salt in these proceedings, I know that … . What I'm asking is a much more pertinent question to these proceedings. To what extent is this document to guide me in reaching a conclusion? MR HONCHIN: In my submission, where you have medical opinions before you, it is not. You are entitled to rely upon the medical opinions of doctors before you to make your decision based on balance of probabilities in relation to those doctors. SENIOR MEMBER: I mean, for example, both of the doctors were quite clear that unless the particular criteria outlined in a particular condition of this particular document were met, then I couldn't find that a person had PTSD. Now, quite frankly, if anyone was to say that a document published by a psychiatric association in the United States which has reached its fifth edition, had reached such a stage of biblical inerrancy that I would have to say that I will follow it with apodictic certainty and if anything, even a semicolon wasn't met and I couldn't make a decision I think that would be regarded as an abdication of my discretionary power, wouldn't it? MR HONCHIN: Absolutely. In my submission, what - - - SENIOR MEMBER: So that's what I was asking for your submission about. MR HONCHIN: In my submission, you are entitled to rely upon medical evidence, a diagnoses by a doctor that, in effect, there was a - just one second. If there's expert opinion which depends on the reliable body of knowledge and experience to underpin that expert opinion … [Emphasis added] In this passage, "this document" is DSM-5. 26 It is apparent from the cross-examination of Dr Scheepers before the Tribunal in relation to DSM-5 that one end to which it was directed was the securing of recognition by him that its diagnostic criteria were neither immutable over time nor wholly prescriptive nor diagnostically restrictive. Dr Scheepers certainly conceded the first of these propositions. 27 Paragraph 197 in the Tribunal's reasons, quoted above, displays reasoning which, read in isolation, looks to be nothing more than a reasonable and rational explanation for why it is that Dr Scheepers' view is preferred by the Tribunal – nothing more than it better conforms with the overall factual picture revealed by Mr Read's military and civilian career and domestic circumstances after 1968. That would be fatal to concluding that the error of law asserted by Mr Read was present. A difficulty with dismissing that assertion that it appears that the Tribunal has bolstered and explained its preference by conducting its own category G assessment. 28 That impression is reinforced by the following passage at the conclusion of that part of the Tribunal's reasons (paras 216 and 217) which addresses "Category C" (read "G") in DSM-5: 216. When she gave oral evidence Mrs Read testified that only on three or four occasions each year would she need to change the sheets because they were damp. Accordingly the frequency of the night sweats was limited to one night every three months. Even noting that these events were uncommon, it is illuminating that during a period of almost 40 years the Applicant did not discuss with his wife the content of his sleeping disturbances. Or if he did, on apparently no occasion did he mention the August 1968 fatal accident. 217. One of the earlier Federal Court decisions on PTSD is Dingwall v Commonwealth of Australia [1994] FCA 1099 (Dingwall). The case concerned a veteran who witnessed the atomic tests at Maralinga, South Australia in 1957. In his proceedings against the Commonwealth, Mr Dingwall claimed he was suffering from PTSD, but had not reported any symptoms to doctors for many years after leaving the Army and never discussed his recurrent nightmares with his wife. Foster J made these observations which are relevant to the present matter ([133] – [134]): "133 It must be borne in mind that the symptoms under discussion are not mere errant, unbidden recollections or hazy dreams. They are emotionally charged and highly distressing experiences symptomatic of psychiatric disorder. They would quite clearly be matters of concern to anyone experiencing them and an eminently fit subject to be reported to that person's treating medical practitioner, whether general practitioner or psychiatrist. It is therefore, in my view, of great significance that, as I have found, no reports were made of these matters to any medical practitioners until Dingwall was involved in the series of medical examinations in late 1992, which were held for the purpose of this case ... 134 I have not failed to take into account Mrs Dingwall's evidence that, after returning from Maralinga, Dingwall's sleeping patterns appeared to alter, in that he became restless in his sleep and at times complained of nightmares. If these nightmares were of the type relevant to a diagnosis of PTSD relating to Maralinga experiences, then I would have expected him to have said something about the content to Mrs Dingwall ... I cannot envisage any impediments to his describing the contents of his dreams to his wife if, in fact, they were a distressing experience and related to the atomic explosions." 29 The Tribunal's reference to Dingwall v Commonwealth of Australia [1994] FCA 1099 (Dingwall) was not gratuitous. It was cited in support of the conclusions adverse to Mr Read to which the Tribunal came. But the passages quoted formed part of a judgement given in a negligence action in which there was evidence, on which Foster J acted (reasons for judgement, para 121) that an assessment of PTSD could be made by reference to the criteria in the Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association (DSMIII), a predecessor of DSM-5. His Honour was then measuring the evidence given by the wife of the applicant in that case against criteria the relevance of which was supported by specialist opinion evidence he expressly accepted via a related finding of fact to this effect. The Tribunal's citation of Dingwall, absent any such express finding of fact, underscores the impression created on the face of the reasons that it is conducting its own evaluation of the DSM-5 criteria as if recourse could be had to them without such a finding. 30 The following paragraphs in the concluding portion of the Tribunal's reasons send mixed signals as to whether this is just a case of preferment of one body of specialist opinion over another or an error of law of the kind asserted: 218. I accept that as a clinical reality, late onset PTSD can be a valid diagnosis. In particular, I accept that with the refinement of psychiatric analysis and diagnosis, it is also open to the Tribunal to find that late onset PTSD can occur at any particular time in an individual's life. There is no arbitrary time limit for the incubation of this ailment. 219. However, as Dr Scheepers pointed out in the Israeli study relied on by Dr Beaney, the various veterans who presented with late onset PTSD all demonstrated symptoms before the eventual diagnosis. In short, although the full impact of PTSD may develop over a long period, a person suffering from this condition will manifest symptoms from the onset of the original traumatic triggering event. 220. The Tribunal has determined, based on the material presented, that PTSD is not the correct or preferable diagnosis because not all Criteria B to H of DSM-5 have been met, and, accordingly, a diagnosis of PTSD is not open to the Tribunal. 221. Having reached this conclusion, the causative threshold cannot be achieved. The subject matter of these proceedings cannot proceed to the next step. [Emphasis added] 31 Again, read in isolation, paras 218 and 219 of the Tribunal's reasons look like an unremarkable, logically reasoned, explanation as to why an acknowledged and accepted body of specialist opinion about the phenomenon of late onset PTSD lacks a supporting factual foundation on the findings which the Tribunal has made on the evidence. I have used the adjective, "unremarkable" because, in both judicial and administrative proceedings, it is an often encountered, reasonably open basis for discounting particular expert opinion evidence that the supporting factual foundation is not made out. Yet these paragraphs are immediately followed by what looks to be another conclusion by the Tribunal based on its own assessment of the DSM-5 criteria without any supporting finding of fact in relation to the applicability of those criteria. In other words, in this passage of the reasons also there is an unarticulated premise. 32 Muggeridge v Minister for Immigration and Border Protection (2017) 351 ALR 153 (Muggeridge) offers a recent example of a case in which the decision of an administrator, in that case a Minister of State, was held to be affected by legal unreasonableness constituted by a lack of an evident and intelligible justification, because of the absence of a critical step in the reasoning process, as revealed by the statement of reasons for the decision. Muggeridge was a case concerning a challenge on judicial review to the cancellation of a visa on character grounds. Justice Charlesworth (with whom Flick and Perry JJ agreed) stated, at [55]-[56]: 55. Although the Minister was not required to evaluate the risk of Mr Muggeridge re-offending in any particular way, the Minister did in fact embark upon an evaluation of Mr Muggeridge's prospects of re-offending in a way that was acutely fact dependent. The reasoning adopted by the Minister in this case logically required an assessment to be made of the likelihood that Mr Muggeridge would resume contact with an outlaw motorcycle club and so offend "in a similar fashion" and thus cause "great harm". In light of the Minister's findings concerning Mr Muggeridge's rehabilitation, his debilitating spinal injury and the lack of evidence of any affiliation with any outlaw motorcycle club since his return to Australia, the reasons do not provide any logical basis for concluding that there was a possibility that Mr Muggeridge would resume contact with such a club. 56. In the result, the Minister exercised the discretion in a manner that purported to advance an object of the statute, and yet there is no evident rational connection between that legitimate object and the particular materials upon which the Minister is said to have relied. A rational connection in a case such as the present should not depend upon unnatural implications drawn from the reasons that cannot be reconciled with the express findings concerning Mr Muggeridge's demonstrated rehabilitation, his serious physical debilitation and the absence of evidence that he had had any connections with like motorcycle clubs for more than two decades. 33 One of the authorities cited by Charlesworth J in Muggeridge, at [35](5), is Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321 (ABT v Bond) in which, at 367 — 368, Deane J observed, in relation to the requirement that a statutory tribunal act rationally and reasonably: When the process of decision-making need not be and is not disclosed, there will be a discernible breach of such a duty if a decision of fact is unsupported by probative material. When the process of decision-making is disclosed, there will be a discernible breach of the duty if findings of fact upon which a decision is based are unsupported by probative material and if inferences of fact upon which such a decision is based cannot reasonably be drawn from such findings of fact. 34 Such a form of error is inherently fact specific to the circumstances attending a particular administrative decision, including the reasons given for that decision. 35 In my view and with respect to the Tribunal, its reasons for the conclusion that Mr Read does not suffer from PTSD do display the error for which he contends. On the face of those reasons, the Tribunal has elevated DSM-5 to a touchstone against which it has measured whether Mr Read suffers from that condition. It has done so without expressly finding that this is one permissible way in which to determine whether a person suffers from that condition. There was, undoubtedly, evidence before the Tribunal that would have admitted of such a conclusion but the reaching of such a conclusion would necessarily have required the Tribunal also to confront and make related findings concerning the limitations in the use of DSM-5, upon which there was also evidence from the two principal, reporting medical specialists. In support of its conclusions, the Tribunal has, with respect, uncritically, made reference to evaluations in the courts (Wright and Dingwall) of whether the factual foundations for criteria specified in earlier editions of this Manual existed without appreciating that these evaluations were made either against a background, where it was common ground that the Manual could be used in this way or there was an express finding, based on evidence, that it was accepted it could be used by a judge for this purpose. These references reinforce the impression of DSM-5 being used by the Tribunal itself as a touchstone as if no underpinning finding as to its permissible use for that purpose was necessary. 36 Given the understanding voiced by the Tribunal in the course of submissions, having regard to the passage quoted above, it seems inherently likely that the error concerned was unintended. But it is the purpose of reasons, not of observations made in the course of submissions, to inform the persons concerned, both the Commission and Mr Read in this case, why the decision has been made (and the same is true of judicial observations in the course of submissions, as opposed to considered reasons for judgement). The passage quoted from submissions to the Tribunal on behalf of Mr Read also makes plain that it was not common ground that the Tribunal could make its own evaluation of the DSM-5 criteria. They were not promoted by Mr Read's counsel to the Tribunal in this way. He promoted looking to the medical evidence. The Tribunal's reasons disclose an understanding that they were understood to have been promoted by the Commission as a touchstone. Whether or not they could be so used required the Tribunal to rehearse and make findings about their use and any related limitations, based on the specialist medical evidence. Making due allowance for the observations approved in Wu Shan Liang, the reasons disclose that the Tribunal did not do this. Absent a supporting finding of fact about the permissible use of DSM-5, the Tribunal had no lawful warrant to make its own assessment as to whether its criteria were present on the facts it chose to find. The result is a want of logicality, a "missing link", of the kind described by Deane J in ABT v Bond and by the Full Court in Muggeridge. 37 The Tribunal's conclusion, rejecting on the facts found by it, that part of Dr Beaney's evidence grounded in the Horesh study does not, viewed in isolation, exhibit a like want of logicality but this cannot, in my view, be divorced from the want of logicality in the use, adverse to Mr Read, made by the Tribunal of DSM-5. 38 For these reasons, the Tribunal's decision must be set aside and the matter remitted to the Tribunal for rehearing. Ordinarily, such a remitter ought not to be prescriptive as to how the Tribunal is to be constituted on a rehearing. The subject of such constitution is usually regarded as a matter for the President in his administration of the Tribunal. But, given the nature of the error, it would not be appropriate for the Tribunal to be constituted as before. Further, as the error concerned affected only one of the issues for the Tribunal's determination and because the reasons for the conclusion that Mr Read was present in the immediate aftermath of the 1968 accident are impeccable (and not the subject of any separate notice of appeal by the Commission) it would not be appropriate for that issue again to be at large, especially given that almost half a century has now passed since the accident. Indeed, it would be oppressive to the parties and to the persons who gave evidence in relation to that issue to so require. The Court's powers under ss 44(4) and 44(5) of the AAT Act are sufficiently broad to permit the making of a remitter order with related directions on each of these subjects. 39 There is no reason why, on the rehearing, the Tribunal ought not to have regard to the record of the proceeding before the Tribunal prior to this appeal (including a record of any evidence taken in that proceeding). The effect of this is not to limit tendering of such further material, orally or in writing, as the parties may be advised. The costs of the appeal should follow the event. I certify that the preceding thirty-nine (39) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Logan. Associate: Dated: 8 June 2018