Return to Work Corporation of South Australia v Sunman - [2021] SASCA 125

SUPREME COURT OF SOUTH AUSTRALIA
(Court of Appeal: Civil)
RETURN TO WORK CORPORATION OF SOUTH AUSTRALIA v SUNMAN & ORS
[2021] SASCA 125
Judgment of the Court of Appeal
(The Honourable President Livesey, the Honourable Justice Bleby and the Honourable Justice Stanley)
28 October 2021
APPEAL AND NEW TRIAL - APPEAL - GENERAL PRINCIPLES - RIGHT OF APPEAL - WHEN APPEAL LIES - ERROR OF LAW - PARTICULAR CASES INVOLVING ERROR OF LAW - FAILURE TO GIVE REASONS FOR DECISION - ADEQUACY OF REASONS
WORKERS' COMPENSATION - ENTITLEMENT TO COMPENSATION - EMPLOYMENT RELATED INJURY, DISABILITY OR DISEASE - ARISING IN COURSE OF EMPLOYMENT
Application for permission to appeal on a question of law from a decision of the Full Bench of the South Australian Employment Tribunal. The Full Bench, by majority, dismissed an appeal from a decision of a single member of the Tribunal, which had held that a worker had died in compensable circumstances.
The worker was driving a truck in the course of his employment when he suffered a stroke. He died four days later. His widow and children sought compensation for his death under the Return to Work Act 2014 (SA), on the basis that he had died from a compensable injury.
There was conflicting medical evidence on the question of causation of the stroke. The question of law arising on the appeal was whether the trial judge’s reasons were adequate in finding that the worker’s death occurred in compensable circumstances.
Held (by the Court), granting permission to appeal, allowing the appeal and remitting the matter for rehearing before a different presidential member of the Tribunal:
1. The reasons of the trial judge were not adequate.
2. The trial judge was faced with two possibilities on the medical evidence. These were that the worker’s turning of his neck and his onset of symptoms consistent with a stroke, and subsequent death, were causally connected, or their temporal relationship was no more than a coincidence. He found that the medical evidence did not conclusively exclude a possible connection between the turning of the neck and the stroke. In those circumstances, he held that he was entitled to act on his intuitive reasoning. On that basis, he found that the stroke was causally connected to the worker turning his neck while driving.
3. It was not clear on the judge’s reasons whether in doing so, he was relying on positive evidence that a causal connection was possible, or whether he was relying on lay, common sense reasoning given the immediacy of the symptoms that followed the worker turning his neck. The second of these approaches, being that endorsed by the High Court in Adelaide Stevedoring Company Limited v Forst (1940) 64 CLR 538, was not available in this case. Even if he was taking the first approach, being that described in EMI (Australia) Limited v Bes [1970] 2 NSWR 238, he was required to address the competing medical hypotheses. He did not do so adequately.
Return to Work Act 2014 (SA); Workers Rehabilitation and Compensation Act 1986 (SA) ss 30(1)(2); 31(1); 44; 53; South Australian Employment Tribunal Act 2014 (SA) ss 32(1a); 68; Evidence Act 1929 (SA) ss 45A(1); 53(1), referred to.
EMI (Australia) Limited v Bes [1970] 2 NSWR 238; DL v The Queen (2018) 266 CLR 1, applied.
Adelaide Stevedoring Company Limited v Forst (1940) 64 CLR 538, distinguished.
Barrett & Ors v Return to Work SA [2019] SAET 28; Amaba Pty Ltd v Booth [2010] NSWCA 344; Ramsay v Watson (1961) 108 CLR 642; Barnes v Department for Education and Child Development [2014] SAWCT 40; Pham v Workers Rehabilitation & Compensation Corporation (1995) 181 LSJS 241; Australian Iron and Steel Ltd v Connell (1959) 102 CLR 522; Return to Work Corporation of South Australia v Barrett & Others [2021] SAET 60; Archibald v Byron Shire Council [2003] NSWCA 292; 129 LGERA 311; Wiki v Atlantis Relocations (NSW) Pty Ltd [2004] NSWCA 174; 60 NSWLR 127; Taupau v HVAC Constructions (Qld) Pty Ltd [2012] NSWCA 293; Chappell v Hart (1998) 195 CLR 232; Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247; Assad v Eliana Construction & Developing Group Pty Ltd [2015] VSCA 53; Sun Alliance Insurance Ltd v Massoud [1989] VR 8; Flannery v Halifax Estate Agencies Ltd [2000] 1 WLR 377; [2000] 1 All ER 373; Jones v Bradley [2003] NSWCA 81; Mount Lawley Pty Ltd v Western Australian Planning Commission (2004) 29 WAR 273; Waterways Authority v Fitzgibbon (2005) 79 ALJR 1816; 221 ALR 402; Pollard v RRR Corporation Pty Ltd [2009] NSWCA 110; Whisprun Pty Ltd v Dixon (2003) 77 ALJR 1598; 200 ALR 447; Wright v Australian Broadcasting Commission [1977] 1 NSWLR 697; Houlahan v Pitchen [2009] WASCA 104; Pettitt v Dunkley [1971] 1 NSWLR 376; Resi Corporation v Munzer [2016] SASCFC 15; Keith v Gal [2013] NSWCA 339; Re Minister for Immigration and Multicultural and Indigenous Affairs; ex parte Palme (2003) 216 CLR 212; Transport Accident Commission v Kamel [2011] VSCA 110, considered.

RETURN TO WORK CORPORATION OF SOUTH AUSTRALIA v SUNMAN & ORS
[2021] SASCA 125
Court of Appeal - Civil: Livesey P, Bleby JA and Stanley AJA

THE COURT: This is an application for permission to appeal against a decision of the Full Bench of the South Australian Employment Tribunal.
On 26 May 2013, Mr Adrian Barrett was driving a truck in the course of his employment with Lucas Earthmovers Pty Ltd at its Iron Baron mine near Whyalla, when he suffered from an episode relating to his health. He was taken to Whyalla Hospital, from where he was airlifted to the Royal Adelaide Hospital. On 30 May 2013, he died from a multifocal cerebral infarction, that is, a stroke. He was 41 years old.
The first respondent, Ms Hayley Sunman, is Mr Barrett’s widow. The second, third and fourth respondents, Emma Lee Singh, Taylor Barrett and Brooke Barrett are Mr Barrett’s children from a prior marriage. They each sought compensation under the Return to Work Act 2014 (SA), on the basis that Mr Barrett had died from a compensable injury. The essence of their claim was that while Mr Barrett was driving the truck at work, he turned his neck to look in the left rear vision mirror. This action caused the release of an embolus in his brain that resulted in the stroke and his death.
On 23 September 2014, an authorised representative of the appellant, Employers Mutual, rejected the respondents’ claims, pursuant to ss 30(1)(2), 31(1), 44 and 53 of the (now-repealed) Workers Rehabilitation and Compensation Act 1986 (SA) (‘WRC Act’). On 17 October 2014, the respondents filed a Notice of Dispute in the Workers Compensation Tribunal. On 31 October 2014, the compensating authority confirmed the disputed decision.
A Deputy President of the Tribunal conducted a hearing on liability in 2018. At issue was whether there was a causal connection between Mr Barrett driving the truck on 26 May 2013, and the stroke that resulted in his death. On 21 February 2019, the judge found that on the balance of probabilities, a causal connection had been established. He set aside the determination, finding that Mr Barrett died in compensable circumstances such that his wife and children were entitled to compensation under the WRC Act.
The appellant appealed to the Full Bench of the Tribunal. A majority of the Full Bench dismissed the appeal.
Pursuant to s 68 of the South Australian Employment Tribunal Act 2014 (SA) (‘the SAET Act’), an appeal lies to this Court on a question of law against a decision of the Full Bench. An appeal cannot be commenced except with the permission of the Court of Appeal. On 30 April 2021, Doyle JA referred the question of permission to be heard together with the appeal.
The Notice of Appeal raises the following issues:
·whether the reasons of the trial judge were adequate (Ground 1);
·whether the trial judge erred in failing to engage with and resolve conflicts in the medical evidence (Ground 2);
·whether the trial judge erred in proceeding on the basis of common experience as to a temporal connection between Mr Barrett moving his head and suffering the onset of symptoms of the stroke (Ground 3); and
·whether the trial judge erred by deciding the matter on the basis of intuitive reasoning, in isolation from the medical evidence (Ground 4).
Whether reasons for judgment are adequate to discharge the obligation to give reasons is a question of law on which an appeal may be brought pursuant to s 68 of the SAET Act. The second, third and fourth grounds set out above are complaints of error that support the question of law articulated as Ground 1, being whether the reasons were adequate.
Background
The evidence of events prior to Mr Barrett’s stroke
At approximately 8:30 pm on 26 May 2013, Mr Barrett’s supervisor, Mr Paul Merritt, received a call from Mr Barrett requesting that he come and see him. Upon arriving, Mr Merritt saw Mr Barrett leaning on handrails near the cabin of the truck. Mr Barrett told him that as he was backing the truck, he looked in the side mirror in the loading area and heard a ‘click’ in his neck. He then noticed ‘weird things happening in his peripheral vision’. He parked the truck at a safe location and called Mr Merritt. Mr Barrett also said that he was experiencing ‘wobbly boots’, that he had a headache, and that he had just vomited. Mr Merritt saw evidence of that vomit, by the truck.
The trial judge admitted the evidence of Mr Merritt on the basis that it fell within the res gestae exception to the rule against hearsay. In the alternative, he determined that in any event he would modify the rules of evidence pursuant to s 32(1a) of the SAET Act and admit the evidence. This was on the basis that the words Mr Barrett spoke to his supervisor were natural and spontaneous, evoked by the event itself and made in circumstances where they were likely to be true.
Two of Mr Barrett’s colleagues, Mr Craig Hoffman and Mr Malcolm Williams, drove Mr Barrett to the Whyalla Hospital. Mr Barrett’s condition deteriorated shortly after they arrived at the hospital. He was subsequently airlifted to the Royal Adelaide Hospital, where he died four days later.
The judge excluded the evidence of Mr Hoffmann and Mr Williams about Mr Barrett’s account to them of what had happened. He also excluded the evidence of Ms Sunman relating to Mr Barrett’s work activities and his prior state of health. He admitted the notes of the Whyalla Hospital in respect of Mr Barrett’s admission, expressing this to be under s 45a of the Evidence Act 1929 (SA). This appears to have been a reference to the former s 45A(1) of that Act, which provided for the admission of business records in evidence. At the time of the trial, the relevant section was s 53(1).
The judge’s reference to the former section was a slip of no consequence. In circumstances where the author of the notes on admission was not called, and did not otherwise provide a statement, the judge directed himself to approach the record in the notes of Mr Barrett’s account with caution. He took the same approach with respect to the history recorded by the examining doctors.
The consequence of these exclusions and directions was that as to the events leading up to Mr Barrett’s stroke, the only admitted evidence was Mr Merritt’s account of what Mr Barrett had said to him.
The medical evidence
Four medical experts provided expert reports and gave oral evidence. The first respondent called Dr Cheryl Charlwood, a forensic pathologist. The second, third, and fourth respondents called Associate Professor Andrew Lee, a specialist neurologist. The appellant called Professor Richard Burns, a consultant neurologist, and Associate Professor Tim Kleinig, a specialist neurologist.
The competing hypotheses of causation raised by the medical evidence were complex, to say the least.
Dr Charlwood conducted the autopsy on Mr Barrett. She noted that a thrombus of a few weeks’ or months’ duration was located within the right vertebral artery of Mr Barrett’s brain. She found that Mr Barrett had died as the result of multifocal cerebral infarction; thromboemboli of the basilar, left posterior cerebral and posterior inferior cerebellar arteries (‘PICA’); and right vertebral artery thrombosis. There was no significant dispute over these findings.
Dr Charlwood hypothesised that when Mr Barrett turned his neck to look into the rear mirror, this caused vessel stretching sufficient to disrupt the organising thrombus on the right side. That disruption produced a thrombus in the basilar artery and a secondary thrombus of the left posterior cerebral artery. While Dr Charlwood could not exclude the possibility that the thromboembolism may have been caused from an unknown source, she was of the opinion that the head turning movement was ‘almost certainly’ the causal event, given the ‘extreme strong temporal relationship’ between the symptoms experienced at work and his subsequent progression to death.
Professor Burns provided his expert opinion with the advantage of having read Dr Charlwood’s report. He disagreed with Dr Charlwood’s hypothesis. He accepted that neck movement can trigger acute dissection of the vertebral or carotid artery. However, in the absence of Mr Barrett having made any complaint of pain, or any vertebral artery dissection having been identified at autopsy, he did not accept the hypothesis that the head turning had caused disruption of the existing thrombus in the right vertebral artery.
Professor Burns considered that there were only two plausible explanations. The first was that given the existence of an unexplained pre-existing thrombus in the right vertebral artery, Mr Barrett may have had a clotting tendency, which caused a thrombus to form spontaneously. The second possibility was that the fatal stroke was the result of an embolus arising from a pre-existing thrombus in the left atrium. He noted however, that there was no thrombus identified in the left atrium at the time of autopsy.
Professor Burns thought that the sequence of events, with the development of symptoms at the time that Mr Barrett had turned his head, was coincidental. In response, Dr Charlwood confirmed her view of the importance of the temporal relationship, which caused her to consider that Mr Barrett turning his head had almost certainly precipitated the events leading to his death.
Professor Kleinig provided a hypothesis similar to Professor Burns’ first hypothesis. That is, Mr Barrett had an underlying tendency to arterial thromboembolism, which caused a thrombus to form from an undetermined source. He said that this hypothesis was supported by Mr Barrett’s pre-existing right vertebral artery occlusion, and the absence of an embolic source at autopsy.
Professor Kleinig noted that the imaging taken before Mr Barrett’s death revealed an infarction within the left PICA. He explained that as the blood flowed ‘downstream’, from the left vertebral artery to the left PICA, the clot that lodged in the left PICA, the basilar artery, and branches of the basilar, must have originated from the left vertebral artery, the aortic arch, or the left side of the heart. The thrombus in the right vertebral artery could not have been the source of this infarction.
AP Kleinig’s evidence was that in the 6,000 cases he had seen, he had not known one in which the mere turning of the neck had caused an existing thrombus in the vertebral artery to be liberated. He described this as an ‘unprovable assertion, a remote possibility’.
Dr Charlwood said, in response, that there was no evidence on autopsy to suggest that Mr Barrett had a tendency to arterial thromboembolism. She further explained that she was not suggesting that the noise or sensation that Mr Barrett felt in his neck was anything other than a consequence of ligamentous or skeletal movement. That is, the ‘click’ in his neck was not the embolus being liberated. However, she maintained the importance of the sequence of events, in that Mr Barrett’s symptoms commenced following the moving of his neck.
In response to Professor Kleinig’s concern with regard to the thrombus in the left PICA, Dr Charlwood explained that one of two things could have happened: either the thrombus was present as a result of a ‘terminal event’, that is, not present until just before death, or as a result of consequential flow turbulence or stagnation, which had the potential to disrupt or reduce blood flow.
Professor Lee, who was called by Mr Barrett’s children, explained that a close temporal relationship was required to establish an association between an event and a condition. To this end, he relied on what he understood to have been Mr Barrett experiencing pain when he turned his head. He thought the temporal relationship was clear. He later accepted that he had interpreted Mr Barrett feeling a ‘crick’ in his neck as reflecting pain.
Professor Lee offered another possible explanation for the left PICA infarction. This was that on turning his head, Mr Barrett occluded his left vertebral artery, and given that the right vertebral artery was already compromised, this movement caused left PICA ischaemia. This disruption of blood flow and the deceased’s tendency to clot caused the infarction in the left PICA. At the same time, the movement also liberated the clot from the pre-existing thrombus from the right vertebral artery, causing the basilar infarction.
The approach of the trial judge
The trial judge found that there was a close temporal association between Mr Barrett turning his head and the onset of symptoms that culminated in his death.[1] Turning to causation, he first reminded himself that legal proof based on the balance of probabilities is less rigorous and demanding than proof of a fact for scientific purposes.[2] He found that each expert medical witness was doing their best to assist in making sense of the medical conundrum.
[1] [2019] SAET 28 at [67].
[2] [2019] SAET 28 at [68], referring to Amaba Pty Ltd v Booth [2010] NSWCA 344 at [116].
The judge then made the following findings and observations about the expert medical evidence. First, he found that some of the assumptions made by Dr Charlwood and Professor Lee had not been established. He identified their understanding that Mr Barrett’s neck movement was sudden or painful, which he was not prepared to find. Nonetheless, he continued:[3]
However even without such findings, the temporal connection between neck movement and the onset of symptoms that they both heavy [sic] relied on in support of their hypotheses, remains.
[3] [2019] SAET 28 at [70].
The judge found that Mr Barrett had a pre-existing thrombus in his right vertebral artery. His next steps were critical to his ultimate findings on causation. First, he expressed reservations about the hypotheses of Dr Charlwood and Professor Lee:[4]
I did not find the hypothesis put forward by Dr Charlwood and Professor Lee as to how that thrombus was involved in the fatal stroke compelling. I felt the same way about Professor Lee’s alternate hypothesis based on evidence of ischaemia in the left posterior inferior cerebellar artery.
[4] [2019] SAET 28 at [71].
This did not, however, amount to a rejection of those hypotheses outright:
That said, although the evidence of Professor Burns and Professor Kleinig gave me cause have [sic] some doubts about the hypotheses put forward by Dr Charlwood and Professor Lee, I do not regard those hypotheses as irrational or implausible. I was left with the sense that they were possible explanations.
The judge then reminded himself that it was not decisive that the biological processes by which events might be causally connected are not properly understood.[5] The question then, of course, is by what legal reasoning might a conclusion be reached in the absence of a proper understanding? The judge’s approach contains two strands of reasoning that require careful delineation.
[5] [2019] SAET 28 at [72], referring to Amaba Pty Ltd v Booth [2010] NSWCA 344 at [116].
The first strand of reasoning: Bes

First, the judge invoked the statement of Herron CJ in EMI (Australia) Limited v Bes,[6] as to the circumstances when a court may rely on ‘intuitive reasoning’ to reach a conclusion of causation between an event and a medical episode, in that case, death:[7]
[I]t is not incumbent upon the applicant, upon whom the onus rests, to produce evidence from medical witnesses which proves to demonstration that the applicant’s contention is correct. Medical science may say in individual cases that there is no possible connexion between the events and the death, in which case, of course, if the facts stand outside an area in which common experience can be the touchstone, then the judge cannot act as if there were such a connexion. But if medical science is prepared to say that it is a possible view, then, in my opinion, the judge after examining the lay evidence may decide that it is probable. It is only when medical science denies that there is any connexion that the judge is not entitled in such a case to act on his own intuitive reasoning.
(Emphasis added)
[6] [1970] 2 NSWR 238.
[7] EMI (Australia) Limited v Bes [1970] 2 NSWR 238 at 242 (Herron CJ).
We make three observations about this passage. First, Herron CJ’s articulation of this concept of ‘intuitive reasoning’, available in the event of medical science conceding the possibility of a causal relationship, called in aid the reasoning of the High Court in Ramsay v Watson.[8] The respondent in that case was a former employee of the Government printing office. He suffered from Bright’s disease and alleged that this had been caused by lead poisoning contracted during that employment. He succeeded at trial in negligence against his former employer, and then on appeal. The High Court dismissed the appeal. In the course of doing so, it articulated reasoning that was open to the jury given the conflicting expert reports:[9]
It was said, however, that there was no real risk, or no foreseeable risk, that persons working in the parts of the Printing Office where the respondent had worked would be injured by lead. Both sides conducted tests at the Government Printing Office to ascertain the extent of lead in the atmosphere. The results, and the interpretation of the results by the experts, were conflicting. There was much dust lying in some places. This was laden with lead particles. This dust might sometimes be stirred up and thus breathed by workers on the premises. And the melting of lead in the making of monotype castings produced fumes bearing lead particles, so that persons habitually in that part of the printery could not avoid inhaling more lead than do persons whose activities do not bring them into regular contact with lead. The respondent was thus brought into contact with lead at his place of work. There is no evidence of any other source from which he could have got such an accumulation of lead as was in his body.
[8] (1961) 108 CLR 642.
[9] Ramsay v Watson (1961) 108 CLR 642 at 645-646.
Ramsay v Watson concerned a jury trial. There was therefore no capacity to examine the reasoning of the fact finder. However, it was clear that certain of the evidence allowed for a hypothesis of causation.
The second observation relates to the competing medical evidence in Bes. The deceased worker in that case was killed when he suffered a syncope, or loss of consciousness, while driving. His car swerved off the road and struck a telegraph pole. His widow’s claim that the death was work-caused was based on the fact that twelve days earlier, a ladder had fallen onto the deceased’s head at work. He reported the accident and claimed he had a headache. The evidence was that in the days following that incident, he continually complained of a headache, took a number of Bex powders, a form of analgesic, lost his appetite and appeared to experience some functional disorders that in turn affected his mood and outlook on life. There was evidence that on the day of the accident, he appeared vague and slow in his reactions. He had previously been a healthy man with no history of headaches, head injury, loss of consciousness or collapse.
The trial judge found that it could be inferred as a reasonable probability, rather than as mere conjecture, that the deceased’s syncopal attack was largely, if not wholly, due to the intake of analgesics consequent on the ladder accident. He consequently found that the death was work-caused.
The employer had led evidence from an expert medical witness hypothesising that the deceased’s syncope was caused by a pre-existing vascular condition. The trial judge had rejected that hypothesis on the basis of the post‑mortem examination, which had found no evidence of any such condition. Having rejected that hypothesis, the medical evidence that the judge did accept went no higher than that the deceased’s symptoms were evidence of a post‑concussional state and that it was at least possible that this state was the cause of the syncope and, hence, death.
This was the reasoning that Herron CJ approved when observing that ‘if medical science is prepared to say that it is a possible view, then, … the judge after examining the lay evidence may decide that it is probable’. Importantly for the present case, however, Herron CJ only reasoned to this effect on the premise that competing medical hypotheses had been excluded:[10]
Before he could come to this conclusion he had to reject certain other postulates that were placed before him, certain other hypotheses that were put before him, by other medical witnesses.
[10] EMI (Australia) Limited v Bes [1970] 2 NSWR 238 at 240 (Herron CJ).
The competing hypothesis in that case was falsifiable. That was not the case in the present matter. In either situation, a trial judge is required first to address any completing hypothesis before engaging in a process of ‘intuitive reasoning’ based on a possibility of causation raised in the evidence.
The third observation from Bes is that Herron CJ also observed that the trial judge:[11]
was not entitled to determine the issues on a basis of intuitive reasoning based on common experience of mankind, his Honour being bound in a case such as this to look for some assistance from the medical scientists who were called before him to support his ultimate decision.
[11] EMI (Australia) Limited v Bes [1970] 2 NSWR 238 at 239 (Herron CJ).
This appears to have been a reference to the facility of reasoning described in Adelaide Stevedoring Company Limited v Forst, discussed below.[12] In context, Herron CJ was distinguishing the process of intuitive reasoning on the premise that expert medical evidence admitted of the possibility of causation, from the reasoning process described in Forst.
[12] (1940) 64 CLR 538.
The second strand of reasoning: Forst
Following his reference to, and quoting of, Bes, the judge cautioned himself against falling into the fallacy of post hoc, ergo propter hoc.[13] However, he also said that ‘sometimes the characteristics surrounding events that are temporally connected are such that the connection cannot be ignored’. He then quoted a passage from an earlier judgment of his in Barnes v Department for Education and Child Development:[14]
…“it is obvious that there must be occasions when post hoc, ergo propter hoc is a fair working hypothesis”. The application of that hypothesis requires caution because coming to a conclusion based solely on the order of events can lead to error. Just because something follows an earlier event in time it does not necessarily mean that the earlier event must be the cause of what follows. There will, however, be occasions when the trier of fact will be “greatly impressed by the sequence of events.” Sometimes the association in time and place is so close and dramatic as to enable a tribunal of fact to draw the inference of causation. In such cases the trier of fact is permitted to begin the assessment of the evidence “from the standpoint of the presumptive inference which this sequence of events would naturally inspire in the mind of any common-sense person uninstructed in pathology.” If the medical evidence does no more than raise the possibility that the lay common-sense presumption is wrong the trier of fact is permitted to rely upon his or her own intuitive inferences. (footnotes omitted)
(Emphasis in original)
[13] [2019] SAET 28 at [74].
[14] [2014] SAWCT 40 at [94] (Gilchrist DPJ), quoted in [2019] SAET 28 at [74].
This passage is, in terms, an exposition of the reasoning process described by Rich ACJ in Forst. The footnotes omitted from the quote by the trial judge confirm this to be so.
Forst concerned a worker who performed two tasks involving exertion, following which he collapsed and died. The medical evidence showed a conflict of opinion. The arbitrator found that the death was due to coronary thrombosis. However, he held that the worker’s exertion was not a satisfactory explanation for his death. On an appeal by way of rehearing, the Supreme Court held that while the evidence was not conclusive, it showed that physical exertion was commonly a cause of coronary thrombosis. Having regard to the probabilities, the proper conclusion was that the exertion was the cause.
A majority of the High Court dismissed the appeal. Acting Chief Justice Rich expressed himself to be ‘greatly impressed by the sequence of events’. He then reasoned:[15]
I do not see why a court should not begin its investigation, i.e., before hearing any medical testimony, from the standpoint of the presumptive inference which this sequence of events would naturally inspire in the mind of any common-sense person uninstructed in pathology.
[15] Adelaide Stevedoring Company Limited v Forst (1940) 64 CLR 538 at 563 (Rich ACJ).
He went on to find that the current state of medical opinion was insufficient to connect coronary thrombosis with effort, but that this was not enough to rebut the presumption that flowed from the observed sequence of events.[16]
[16] Adelaide Stevedoring Company Limited v Forst (1940) 64 CLR 538 at 564 (Rich ACJ).
Whatever the practical scope of this principle over eighty years later, it does not establish any principle of presumption as to cause on the basis of a mere temporal connection between incident and injury. As this Court noted in Pham v Workers Rehabilitation & Compensation Corporation,[17] the reference to a ‘presumptive inference’ simply describes a process of reasoning from proven facts. That process of reasoning was held to be appropriate in the circumstances in Forst.[18]
[17] (1995) 181 LSJS 241.
[18] Pham v Workers Rehabilitation & Compensation Corporation (1995) 181 LSJS 241 at 253 (Mullighan J, Cox J agreeing).
Forst provided a common law solution to a seemingly intractable problem on the state of the existing medical science. It continues to state a principle of the common law. Nevertheless, there is a temptation to look askance at the confidence in ‘common sense’ inherent in Rich ACJ’s approach. That is not simply a function of the passage of time since the judgment: it is expressed in the dissent of Dixon J:[19]
[W]hether an inference can or should be drawn from the fact that in the present case the collapse of the deceased occurred after unusual exertion seems to me to depend on the answer first given to the pathological question whether there is any natural connection between exertion and the formation of a thrombus.
[19] Adelaide Stevedoring Company Limited v Forst (1940) 64 CLR 538 at 570 (Dixon J).
In any event, the scope of application of the Forst principle is necessarily narrow. The process of reasoning it permits does not require a medical foundation as a starting point. For that very reason, it depends on lay experience of sufficient notoriety to justify departure from the discipline of pathology.[20]
[20] See Australian Iron and Steel Ltd v Connell (1959) 102 CLR 522 at 535-536 (Menzies J).
The trial judge had, by this point, described two different potential processes of reasoning. Both of those processes will be excluded by the acceptance of medical evidence contrary to the hypothesis in question. However, in the absence of such contrary medical evidence being accepted, they describe different things. Forst describes a process of reasoning by a ‘common-sense person uninstructed in pathology’. By contrast, the ‘intuitive reasoning’ described in Bes is premised on the existence of medical evidence admitting of the possibility of causation. Before engaging in this process of reasoning, the trier of fact must have rejected any competing hypotheses.
The trial judge’s reasoning
It is necessary to set out the judge’s application of the principles he described, in full:[21]
Returning to the facts in this case, in the end I am left with one of two possibilities. Mr Barrett’s turning of his neck and his onset of symptoms consistent with a stroke are causally connected or their temporal relationship is no more than a mere coincidence.
I think the former is more likely. The sequence of events described by Mr Merritt, that is: Mr Barrett turning head, then noticing weird things happening in his peripheral vision, feeling the need to park his truck and get out the vehicle, experiencing wobbly boots, having a headache, vomiting, and feeling the need to call his supervisor to request that he come to see him, is to my mind impressive. I think this is a case where it is appropriate for me to apply the presumptive inference. My intuitive reasoning tells me that Mr Barrett’s turning of his neck and his stroke are connected.
What is significant about the medical evidence is that it does not conclusively exclude a possible connection between Mr Barrett’s turning of his neck, his stroke and death. That being so, I am permitted to act upon my intuitive reasoning.
[21] [2019] SAET 28 at [75]-[77].
The judge’s reference to the presumptive inference in the second of these paragraphs suggests that he was applying the reasoning described in Forst. However, his reference to his ‘intuitive reasoning’ in the same paragraph evokes the reasoning process described in Bes. This impression is fortified by the judge’s earlier exposition of both principles, as set out above.
The third of these paragraphs does not clear up the matter. It could be a reference to either process, in that it is premised on the medical evidence not conclusively excluding a causal connection. If such a causal connection was conclusively excluded, neither process of reasoning would be available. Again, however, Bes relies on positive evidence that a causal connection is possible, together with the rejection of, or accounting for, competing medical hypotheses. Forst relies on the availability, through notoriety, of lay, ‘common-sense’ reasoning where medical science is unable to explain causation.
The process of reasoning the judge deployed is, with respect, not clear. The difficulty arising from this observation is that whatever the scope for application of Forst, it is not the case that a ‘common-sense person uninstructed in pathology’ would be in a position to reason that Mr Barrett turning his head was causative of his stroke, simply on account of the close temporal connection. There is nothing notorious about such a putative connection.[22] The process of reasoning described in Forst was unavailable.
[22] Australian Iron and Steel Ltd v Connell (1959) 102 CLR 522 at 535-536 (Menzies J).
By contrast, the process of reasoning described in Bes may have been available. However, this possibility raises its own complications on the trial judge’s reasons, which can be described by reference to the judgment of the Full Bench on appeal.
The Full Bench
The question of law raised on this appeal is whether the reasons of the trial judge were adequate. The first ground of appeal, that the Full Bench erred in concluding that those reasons were adequate, supports that question. The reasons of the respective members of the Full Bench provide helpful platforms of analysis from which to assess the adequacy of the trial judge’s reasons and, in that assessment, to identify whether the Full Bench did err.
The appellant’s complaint about the adequacy of the judge’s reasons reflects the reasoning of Hannon AuJ in dissent on the Full Bench appeal. Auxiliary Judge Hannon considered the approach of the trial judge in stages. First, he reached the view that he considered the hypotheses of Dr Charlwood and Professor Lee to provide possible explanations, notwithstanding that he had some doubts about those explanations.[23] He then characterised the approach of the judge to be an application of the reasoning process described in Bes.[24] He considered that the judge had not engaged in Forst reasoning, and that it would not have been appropriate to do so.[25]
[23] [2021] SAET 60 at [224].
[24] [2021] SAET 60 at [225].
[25] [2021] SAET 60 at [226].
However, as Hannon AuJ recognised, in order to reason permissibly from the starting point that Dr Charlwood and Professor Lee provided possible explanations, it was necessary to address the evidence of Professor Kleinig, certainly to the extent that it contradicted or qualified this evidence as to possibility. To this end, Professor Kleinig’s reports were emphatic that there was no conceivable way in which Mr Barrett having turned his head could have caused his stroke without vessel dissection. Further, the images of infarction in the left PICA required explanation, in that they could not have been caused by the thrombus in the right vertebral artery.
In cross-examination, Professor Kleinig said that it was ‘beyond the realms of possibility’ that the basilar artery thrombus could have emanated from the right vertebral artery, given the existence of the embolus in the left PICA. He then conceded that there was a theoretical possibility that a clot could break off from the right vertebral artery thrombus. Auxiliary Judge Hannon then observed as follows:[26]
A notable feature of the evidence, not addressed by the Judge, was that Professor Kleinig’s opinion as to the significance of the presence of the left PICA infarction was initially considered by Professor Lee to be highly persuasive, and to support almost beyond reasonable doubt that a clot had traversed along the left vertebral artery to subsequently impact upon the basilar artery. Importantly, Professor Lee formed this initial view even though he accepted that there was a strong temporal relationship between events at work and the subsequent stroke, and that it was possible that a clot could have been released from the right vertebral artery despite absence of evidence of arterial dissection. Professor Lee went on to develop an alternative postulation, based on his opinion that there was a right PICA infarction as well. This was the critical factor which, in his opinion, undermined the significance attributed by Professor Kleinig to the left PICA infarction, a view with which Professor Kleinig disagreed.
The point is that, absent acceptance of his alternative postulation, Professor Lee accepted the force of Professor Kleinig’s opinion that the existence and longevity of the left PICA infarction excluded stroke resulting from a clot released from the right vertebral artery, even though Professor lee accepted there was a theoretical possibility of such a clot release despite no evidence of dissection.
[26] [2021] SAET 60 at [230]-[231].
Auxiliary Judge Hannon concluded that the trial judge was required to engage in a critical analysis of the evidence of Professor Kleinig on the impact of the left PICA infarction on the Charlwood and Lee hypotheses. Both Professor Kleinig and Professor Lee emphasised the importance of this issue as being ‘key’.[27] In the circumstances of Professor Kleinig’s reports and then his limited concession, the judge was first required to determine whether ‘there remained a controversy as to whether medical science denied any possible connection between the right vertebral artery thrombus and the basilar artery stroke’.[28] If so, he was required to resolve that controversy by ‘rational examination and analysis of the competing views, and adequate reasons for his preference for one opinion over another’.[29] If he concluded that Professor Kleinig’s concession did allow for the possibility, that conclusion needed to be supported by adequate reasons.
[27] [2021] SAET 60 at [232].
[28] [2021] SAET 60 at [233].
[29] [2021] SAET 60 at [233].

His Honour concluded that there was no analysis or examination of Professor Kleinig’s evidence in this respect, and that the brief reference to Professor Kleinig’s placing significance on the left PICA infarction did not summarise this position adequately. There was no analysis of the competing aspects of the hypotheses against this issue. These had to be resolved before engaging in Bes reasoning based on accepting the possibility of a connection between Mr Barrett turning his head and suffering the stroke. For this reason, Hannon AuJ held that the judge’s reasons were inadequate and lacked transparency. He would have allowed the appeal.
Deputy President Judge Rossi reached the opposite conclusion. He accepted that while the reports and some statements of Professor Kleinig indicated an opinion that the hypotheses of Dr Charlwood and Professor Lee were impossible, other statements indicated that he accepted that there was a remote possibility of a causal link.[30] Indeed, he found that Professor Kleinig’s evidence ‘overall’ permitted such a finding.[31]
[30] [2021] SAET 60 at [76]-[77].
[31] [2021] SAET 60 at [87].
He also had regard to certain exchanges between the judge and counsel for the Corporation, which he interpreted as indicating that the Corporation was not advancing a case that the causal link advanced in those hypotheses was scientifically and medically impossible. To this end, he relied on a statement by counsel for the Corporation in submissions accepting that the Court could find, on the balance of probabilities, that a medically unlikely event had occurred.[32] He pointed to counsel’s acceptance of the proposition that the question of legal causation was a matter for the judge.[33]
[32] [2021] SAET 60 at [83].
[33] [2021] SAET 60 at [79]; [84].
Against that background, he observed that there were cases where a court was in a position to accept one body of medical opinion over another. He continued:[34]
In such cases there is a duty to not only identify the preferred expert opinion, but also to engage with the issues canvassed and to explain why the opinion of one expert is accepted over the other.[35] However, in this case, after a detailed consideration of complex expert medical evidence, his Honour did not clearly prefer one hypothesis over another.
(Footnote in original)
[34] [2021] SAET 60 at [89].
[35] Archibald v Byron Shire Council [2003] NSWCA 292; 129 LGERA 311 at [54] per Sheller JA; Wiki v Atlantis Relocations (NSW) Pty Ltd [2004] NSWCA 174; 60 NSWLR 127; Taupau v HVAC Constructions (Qld) Pty Ltd [2012] NSWCA 293 at [133] per Beazley JA.
That being the case, he held that a process of presumptive inference was available to the judge. He described that process by reference to Forst.[36] For the reasons explained above, that process was not available in this case. His conclusions must be seen in that light, although in terms they also evoke the process described in Bes:[37]
His Honour acknowledged the hypothesis advanced by Professor Kleinig and Professor Burns, of essentially infarction by an embolus from an unknown proximal source but that, at the same time, the evidence did not rule out the possibility of a causal link between the turning of the head and the sequence of events which led to the death of Mr Barrett.
The trial judge was correct to find that the absence of the medical evidence establishing a compelling relationship between the turning of the head and the subsequent death did not rule out a causal relationship.
Having found that the medical evidence allowed for the possibility of a causal link, the trial judge adopted the correct approach in having regard to the presumptive inference of a causal link and the adoption of intuitive reasoning in relation to the evidence received.
(Footnote omitted)
[36] [2021] SAET 60 at [95]-[112].
[37] [2021] SAET 60 at [115]-[117].
Deputy President Judge Rossi then referred to statements in Chappell v Hart[38] to the effect that the question of causation at common law in Australia is to be resolved as a matter of common sense. The conclusion is often necessarily reached intuitively; too detailed a description of the steps involved in the reasoning process is apt to mislead.[39] He found that the trial judge had availed himself of an available approach and had not erred in his reasoning process.
[38] (1998) 195 CLR 232.
[39] (1998) 195 CLR 232 at [93] (Kirby J), [148] (Hayne J).
Deputy President Judge Calligeros essentially agreed with the reasons of Rossi DPJ. Critically, he found that it could be inferred that ‘the Judge took the appellant to not contend that the existence of a causal relationship between the infarction and employment was impossible or medically implausible, and that he did so fairly. On that basis, he concluded that the judge dealt with the evidence of Professor Kleinig appropriately.[40]
[40] [2021] SAET 60 at [18].
Adequacy of reasons: matters in contest
These judgments on appeal map out the issues relevant to this Court’s consideration of whether the trial judge’s reasons were adequate. It is necessary to understand the position the appellant adopted at trial with respect to the evidence of Professor Kleinig, either by words or conduct, in order to identify what was required of the trial judge’s reasons for accepting the causal link between Mr Barrett turning his head and suffering the stroke. In DL v The Queen, Nettle J, while dissenting in the result, articulated clearly the principle relevant to the complaint in this matter:[41]
Since parties must be able to see the extent to which their cases have been understood and accepted, a trial judge will ordinarily be expected to expose his or her reasoning on points critical to the contest between the parties.[42] This applies both to evidence and to argument. If a party relies on relevant and cogent evidence which the judge rejects, the judge should provide a reasoned explanation for the rejection of that evidence.[43] If the parties advance conflicting evidence on a matter significant to the outcome, both sets of evidence should be referred to and reasons provided for why the judge prefers one set of evidence to the other.[44] Similarly, while a judge is not required to deal with every argument and issue that might arise in the course of a trial,[45] if a party raises a substantial argument which the judge rejects, the judge should refer to it and assign reasons for its rejection. And in providing reasons, the judge is required to make apparent the steps he or she has taken in reaching the conclusion expressed, for reasons are not intelligible if they leave the reader to speculate as to which of a number of possible paths of reasoning the judge may have taken to that conclusion.[46] Failure sufficiently to expose the path of reasoning is therefore an error of law.[47]
(Footnotes in original)
[41] DL v The Queen (2018) 266 CLR 1 at [131].
[42] See Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247 at 279 per McHugh JA; Assad v Eliana Construction & Developing Group Pty Ltd [2015] VSCA 53 at [34].
[43] See Sun Alliance Insurance Ltd v Massoud [1989] VR 8 at 18 per Gray J (Fullagar J and Tadgell J agreeing at 20).
[44] See Flannery v Halifax Estate Agencies Ltd [2000] 1 WLR 377 at 382; [2000] 1 All ER 373 at 378; Jones v Bradley [2003] NSWCA 81 at [131] per Santow JA (Meagher JA and Beazley JA agreeing at [1], [2]); Mount Lawley Pty Ltd vWestern Australian Planning Commission (2004) 29 WAR 273 at 283 [28]; Waterways Authority v Fitzgibbon (2005) 79 ALJR 1816 at 1835-1836 [131]; 221 ALR 402 at 428-429 per Hayne J; Pollard v RRR Corporation Pty Ltd [2009] NSWCA 110 at [66] per McColl JA (Ipp JA and Bryson A-JA agreeing at [1], [85]); Assad [2015] VSCA 53 at [36].
[45] See Whisprun Pty Ltd v Dixon (2003) 77 ALJR 1598 at 1610 [62]; 200 ALR 447 at 464 per Gleeson CJ, McHugh and Gummow JJ.
[46] See Wright v Australian Broadcasting Commission [1977] 1 NSWLR 697 at 701 per Moffitt P (Glass JA agreeing at 713); Soulemezis (1987) 10 NSWLR 247 at 280 per McHugh JA; Houlahan v Pitchen [2009] WASCA 104 at [94] per Newnes JA (Pullin JA and Miller JA agreeing at [1], [2]).
[47] Pettitt [1971] 1 NSWLR 376 at 382 per Asprey JA.
It therefore becomes critical to understand what is in issue between the parties. In Resi Corporation v Munzer,[48] Lovell J distilled several propositions from the authorities, the first being:[49]
“The extent and content of reasons will depend upon the particular case under consideration and the matters in issue”.[50] While a judge is not obliged to spell out every detail of the process of reasoning to a finding it is essential to expose the reasons for resolving a point critical to the contest between the parties.[51]
(Footnotes in original)
[48] [2016] SASCFC 15.
[49] Resi Corporation v Munzer [2016] SASCFC 15 at [71] (Lovell J, Sulan and Stanley JJ agreeing), quoted in [2021] SAET 60 at [17].
[50] Keith v Gal [2013] NSWCA 339 [110].
[51] Re Minister for Immigration and Multicultural and Indigenous Affairs; ex parte Palme (2003) 216 CLR 212 [40].
As Rossi DPJ observed, however, the penultimate principle Lovell J identified was:[52]
Reasons for decision are to be read fairly and in the context of the manner in which the trial was conducted. Reasons may appear by necessary inference from what is stated expressly.[53]
[52] Resi Corporation v Munzer [2016] SASCFC 15 at [71] (Lovell J, Sulan and Stanley JJ agreeing).
[53] Transport Accident Commission v Kamel [2011] VSCA 110; Assad v Eliana Construction & Developing Group Pty Ltd [2015] VSCA 53.
Munzer concerned the adequacy of reasons for a judge’s acceptance of medical evidence that a plaintiff’s chest pain was caused by pleural plaques. The medical opinions were diametrically opposed as to the biological plausibility of the plaintiff’s pleural plaques being responsible for his chest pain. The Corporation’s case was that the pain was caused by psychological factors.
The trial judge had found that the plaintiff was a credible witness. He made a specific finding as to when the plaintiff had complained to his doctor about chest pain. That finding was inconsistent with the Corporation’s case that the pain was psychologically based. Accepting that the development of pleural plaques consequent on the entry of asbestos fibres into the lungs was physical damage, the judge then found that it was ‘likely that for organic reasons that damage is part of the cause of Mr Munzer’s chest pain’.[54] That was the extent of the reasons given for the finding of causation.
[54] Resi Corporation v Munzer [2016] SASCFC 15 at [83].
The Full Court held that the reasons were adequate and, indeed, that there was little more that could be said. Faced with the diametrically opposed opinions, but with the finding of complaint that undermined one set of opinions:[55]
The Judge resolved the issue by finding that the respondent was genuine and that the timing of his complaints was inconsistent with the appellant’s case namely, that psychological factors could account for the pain. The only question that then remained was whether the opinions of Professor Allen and Professor Cramond satisfied him on the balance of probabilities that there was an “organic” basis. He was so satisfied and there was abundant evidence to enable him to make such a finding.
[55] Resi Corporation v Munzer [2016] SASCFC 15 at [89] (Lovell J, Sulan and Stanley JJ agreeing).
The reasons of the trial judge as to causation in that case can be understood as demonstrating an application of the reasoning process described in Bes. That was available because the judge had made findings as to credit and complaint that undermined the competing hypothesis of a psychological cause of the pain. In the event, the Court held:[56]
[O]n a fair reading of the Judge’s reasons, the broad outline and constituent facts of the reasoning on which he has acted can be seen. A fair reading of the Judge’s reasons exposes his reasoning process in a way that the losing party can understand.
[56] Resi Corporation v Munzer [2016] SASCFC 15 at [91] (Lovell J, Sulan and Stanley JJ agreeing).
In the present matter, the complaint about the failure to address the evidence of Professor Kleinig depends in the first instance on what had been left in issue.
The appellant’s case at trial with respect to the evidence of Professor Kleinig
The first respondent submitted that whatever might have been the appellant’s case at the commencement of the trial, by the end of the trial the appellant no longer maintained it to be impossible that Mr Barrett turning his neck could have caused a pre-existing embolus in his neck to be released without dissection. It is to be recalled that Professor Kleinig had raised this thesis of impossibility.
Professor Lee identified evidence of infarctions in the right cerebellar hemisphere, as well as the left, and said that the infarctions were of about the same age. Professor Kleinig accepted in cross-examination that infarctions existed in the right cerebellar hemisphere. He disagreed with Professor Lee that a possible source of those infarctions was the right PICA. However, the respondent says that this challenge was not put to Professor Lee, in that all that had been put to Professor Lee was that the right cerebellar infarctions were of ‘incidental significance’.
We pause to make the following observations about this evidence. When it was put to Professor Lee that the infarctions in the right cerebellar hemisphere were of ‘incidental significance’, it was clear in context that this meant they carried no evidential weight in assessing causation. Professor Lee clearly interpreted the question as such, commencing his answer with, ‘No’, upon which he then elaborated.
Professor Kleinig’s dispute that the infarctions were in the right PICA territories occurred in answer to a question in cross-examination, when it was put to him that that was where they were located. He agreed with the statement in Professor Lee’s report that there was ‘evidence of restricted diffusion or an acute CVA involving the right cerebellar hemisphere’. He then disagreed with counsel’s proposition that sought to locate the infarctions more precisely, in the right PICA. There was no unfairness in the evidence having been so adduced.
In any event, after Professor Kleinig gave evidence, the first respondent applied to call Professor Lee in rebuttal to explain his reliance on the MRI evidence for his conclusion that the right cerebellar infarctions were supplied by the right PICA. She also sought the opportunity to engage a neuroradiologist to interpret the MRIs to identify the source of the infarctions.
The judge refused the application. Prior to doing so, he observed that he was unlikely to be able to determine the cause of the stroke as a matter of scientific proof, and that the first respondent’s case relied on the presumptive inference. Counsel for the first respondent acceded to that but added it appeared from Professor Kleinig’s evidence as to the impossibility of the right PICA being the source, that biological plausibility was in issue.
Later in the course of argument, the judge asked counsel for the Corporation as to how it was framing its case:
HIS HONOUR: All right. Look, I might ask the other side. Are you going to be contending that the scientific explanation put in support of the applicant's case is scientifically impossible, or is your case simply going to be there's a temporal connection but when you weigh up all the medical evidence there might be a connection but you're just not going to be satisfied to the requisite degree that they've got over the line.
MR COPPOLA: I understand your Honour's proposition and, with respect, the distinction between the two alternatives is not perhaps quite as bright as your Honour has outlined. We say that there is a fundamental misconception of a mistake of correlation causality, to use the scientific term. We accept fully that your Honour is not in the same position as a scientific fact finder, that's plain from cases like Wanders at first instance, and that medical and legal causation are two separate tests. We accept that.
The respondent submits that the last sentence of this submission was only meaningful if the appellant’s case was not one of impossibility.
Counsel for the respondent on the appeal then relied on the following further statements of counsel for the Corporation in submissions:
·Professor Burns had said of the Charlwood hypothesis, ‘it just isn’t likely’;
·Professor Burns later said he could not exclude the possibility of it occurring, but he ‘just can’t picture it’;
·Professor Kleinig had said that he had treated over 6,000 stroke patients and had never seen occur what was postulated;
·a submission characterising the effect of the opinions:
Fourthly, your Honour, my learned friend Mr Cole said on the one hand you've got a probably not and a maybe. I paraphrase, but that's the effect of his submission. And on the other hand you have two more likely than nots. With great respect, one has to examine the deductive process by which the witnesses arrived at those conclusions, and I've dealt with that. In relation to ‑ ‑ ‑
HIS HONOUR: He also said that it was my call, not the doctors' call.
MR COPPOLA: Yes, your Honour, that's right.
The first part of this exchange simply characterises the submission of counsel for the children. This was a submission in reply, that it was necessary to examine the deductive processes of the witnesses. Counsel for the first respondent submitted that the apparent embrace of the observation that it was the judge’s call, rather than the doctors’ call, meant that ‘impossibility’ was not part of the Corporation’s case.
There are a number of observations to be made about the submission that the appellant had abandoned any thesis of impossibility. First, to say this was the effect of the submission on the last point identified above, reads too much into counsel’s submission at trial. The submission that the judge first had to examine the deductive process spoke against any concession that it was permissible to put aside the competing hypotheses on the basis that the possibility had been conceded. The inference for which the first respondent contends places considerable weight on the precise language of the appellant’s medical experts.
Secondly, counsel’s acquiescence in the proposition that it was the ‘judge’s call’ was not, in our view, an invitation to proceed directly to a process of intuitive reasoning. The judge’s determination of causation at common law had more than one possible path. If the judge had accepted that the Charlwood and Lee hypotheses were impossible and found against the respondents accordingly, that would not have amounted to ceding his responsibility of fact-finding to Professor Lee and Professor Kleinig.
For that reason, the first respondent’s submission also overstates the importance of the exchange, set out above, between counsel and the judge on the application to call rebuttal evidence. While counsel’s response could have been clearer, he maintained that the respondents’ case involved ‘a fundamental misconception of a mistake of correlation causality’ and did not accept the proposition that his case involved accepting that ‘there might be a connection’.
Thirdly, the submissions must be seen in context. Counsel for the appellant at trial had also submitted as follows:
But, in any event, Associate Professor Kleinig told you that, “I do not believe the turning of the head is causally related.” He says, “We have the gold standard” - his term – “of the pathological examination,” and he says, “I just can't conceive of a way that the neck movement could cause this.” He says to your Honour at page 140:
His death was due to an embolic stroke of undetermined source to the vertebrobasilar system via his left vertebral artery causing the right occipital thalamic infarction.
He provides an explanation for that at page 143, 145, 146, 147 of the book and he says, “Clots cannot swim upstream. They flow like sticks in the river.” Page 210, your Honour, he sets out how the MRI and angiogram footage came into his possession. Then at page 211 he says this:
Detailed evaluation of the image is the key. The case turns on this point, if the thrombus in the left PICA was present in the pre‑agonal phase, then this proves that the source of embolus was not the right vertebral but either the left vertebral artery, the aortic arch or the heart, that is, it was an embolic stroke of undetermined source.

Counsel for the appellant was here clearly maintaining not simply an alternative thesis, but a submission that the neck movement could not have caused the stroke.
Finally, in that context, the fact that the experts tended in oral evidence to speak in terms of likelihood, probabilities, and not being able to exclude the causation thesis, was something that required consideration by the finder of fact. These are the words of scientists. Reliance on them on appeal in aid of a submission that the possibility of causation was conceded begs the question. In the trial context, and in light of the Corporation’s submissions in address, acceptance in evidence of a theoretical possibility of causation was insufficient to permit direct progression to reasoning by way of a process of intuitive inference. At the very least, it was necessary to determine the impact of those words and any similar words of potential concession to the hypotheses of Dr Charlwood and Professor Lee.
Inadequacy of reasons
We therefore agree with Hannon AuJ’s conclusion that it was first necessary to determine, having regard to what are said to have been concessions made by Professor Kleinig during the course of cross-examination, whether it remained in issue that a causal connection between the right vertebral artery thrombus and the basilar artery stroke was impossible. If the judge found that this issue did remain, he was required to determine it. If he considered that Professor Kleinig had allowed for a possible connection, he was required to explain that conclusion.
As Hannon AuJ observed, the judge did not explain whether he engaged in this process. On the assumption that he deployed the process of reasoning described in Bes, the necessary premise that the Charlwood and Lee hypotheses provided possible medical explanations could not have been established without addressing and analysing the evidence of Professor Kleinig.
Further, however, it is not clear whether the judge did follow the Bes process of reasoning. He indicated an intention to take that approach. However, his invocation of Barnes v Department for Education and Child Development[57] and the balance of his reasoning tend to indicate that he either equated that approach with, or allowed his reasoning to be influenced by, the approach described by Rich ACJ in Forst. That approach was not available in this case. For this reason also, the judge’s reasoning was not adequately explained.
[57] [2014] SAWCT 40 at [94] (Gilchrist DPJ).
Conclusion
It follows that we answer the question of law articulated in Ground 1 by saying that the reasons of the trial judge were not adequate. We grant permission to appeal, allow the appeal and order that the matter be remitted for rehearing before a different presidential member of the Tribunal.