JOYCE -v- ANDERSON [2020] WASCA 48 (8 April 2020)

Court of Appeal (WA)|2020-04-08|Before: Mitchell JA

On 23 May 2013, Roger Norman Hill (the deceased) tragically died from injuries he sustained when, at about 3.00 pm on 12 May 2013, a bicycle he was riding collided with a car being driven by the appellant. The collision occurred at the intersection of Eric Street and Broome Street in Cottesloe, which is controlled by a roundabout. Eric Street runs in an east-west direction, while Broome Street runs in a north-south direction.
Just prior to the collision, the deceased was riding his bicycle along Broome Street in a northerly direction, approaching the intersection from the south. He intended to continue along Broome Street after passing through the intersection. The appellant was driving along Eric Street in an easterly direction, approaching the intersection from the west. She intended to continue along Eric Street after passing through the intersection. A give way sign at the entrance to the intersection required the appellant to give way to the deceased. Also approaching the intersection along Eric Street from the east was a car driven by Graham Maddison, with Margaret Ann Staszewski in the front passenger seat.
It is common ground that, after almost coming to a stop at the intersection, the appellant, who had not seen the deceased's bicycle, entered the intersection, blocking the deceased's path. The deceased rode into the driver's side of the appellant's car just in front of the driver's door. He went up onto the bonnet of the car, smashing the windscreen, and rolled off to the side of the car when it came to a stop.
The deceased was taken to Sir Charles Gairdner Hospital (SCGH), where he was treated by medical staff (including a Dr Murphy, who made notes of his cardiothoracic review) in the emergency department. Also while in the emergency department, the deceased spoke to a police officer, Senior Constable Patten, who was investigating the collision. Later, the deceased was admitted to the intensive care unit (ICU) and then the high dependency unit (HDU) of SCGH. In the HDU, the deceased provided a statement to Senior Constable Bedworth, which he said was correct but did not sign before his death.

The trial judge made the following factual findings.

At about 3.00 pm on 12 May 2013 the deceased was riding a bicycle along Broome Street in a northerly direction towards the roundabout intersection with Eric Street in Cottesloe.[1]
The deceased was a recreational cyclist. He was not a serious or expert cyclist. He was wearing shorts and a t-shirt, not lycra. He was wearing joggers, not clip on cycling shoes. He was riding a pushbike, not a racing bike.[2]
At the same time the appellant was driving a BMW sedan in an easterly direction along Eric Street towards the roundabout intersection with Broome Street. The deceased was therefore approaching the intersection from the appellant's right. Give way signs regulated traffic entering into the roundabout.[3]
The speed limit for traffic travelling on Broome and Eric Streets is 50 km/h.[4]
As the appellant approached the roundabout, she slowed her vehicle almost to a stop.[5]
After the deceased rode his bicycle into the roundabout he collided with the appellant's vehicle on the northern side of the roundabout, as a result of which he was knocked from his bicycle.[6]
The deceased was probably travelling at a speed of approximately 15 km/h as he entered the roundabout and no more than 20 km/h.[7] Grounds 5 and 6 specifically challenge this finding as to the deceased's speed.
The deceased was entitled to assume that the appellant had seen him and was either stopped at the give way sign or was stopping at the give way sign to allow him to ride through the roundabout. As he was required to do, the deceased then focussed his attention on Mr Maddison and traffic entering the roundabout from the deceased's right, to which the deceased was required to give way. The deceased was moderating, or adjusting, his speed to allow him to give way to the traffic he observed on his right.[8]
There is nothing in the evidence of Mr Maddison's description of what he observed which leads to a conclusion that the deceased's actions caused Mr Maddison to take any sudden evasive action or that the deceased entered the roundabout in an unsafe manner or at an unsafe point of time.[9]
The appellant drove her vehicle into the roundabout after the deceased had entered the roundabout. She only saw him virtually at the point of collision. Because she was only travelling slowly, at 5 - 10 km/h as she travelled into the roundabout, she was able to brake to a stop almost immediately.[10] Because of the deceased's position in the roundabout when the appellant drove her vehicle into the roundabout there was not sufficient time for him to brake or take other evasive action to avoid the collision.[11]
The finding that the appellant drove her vehicle into the roundabout after the deceased had entered the roundabout is specifically challenged by grounds 6A and 9(b).
Had the appellant been keeping a proper lookout when she approached, or was at, the give way sign on Eric Street, she would have seen the deceased's bicycle before she entered into the roundabout. Her failure to see the deceased was negligent, as is accepted by her by the admission of liability.[12]
By reason of the collision, the deceased suffered various injuries including a chest injury and a penetration wound to his heart which resulted in his death on 23 May 2013.[13]

Deceased's statements as to the circumstances of the accident

After the accident, the deceased was admitted to SCGH emergency department at approximately 3.30 pm on 12 May 2013.[14] The nursing triage notes relevantly read:[15]

Presenting problem:

Car v's cyclist

Assessment

63 yr old male BIBA [brought in by ambulance] post car vs cyclist - hit by car into windscreen. ? speed car approx. 20 km/hr. Unable to ambulant at scene. (emphasis added)

The following presently relevant emergency department continuation notes were made in relation to a cardiothoracic review by Dr Murphy at 5.20 pm on 12 May 2013:[16]

Cyclist vs car this afternoon 50km/hr

Was at roundabout → BMW failed to give way and hit pt on L) side → Landed on R) chest

Does not think he lost consciousness but is unsure

(emphasis added)

It is unclear whether these notes were made at the time the deceased was being spoken to by the doctors and being assessed, or whether the notes were compiled afterwards.[17]

It is unclear, by reference only to the notes quoted at [22] above, to what the 50 km/h speed relates. At first blush, the note appears to refer to the speed at which the car was [18]avelling.18 It was unlikely to be a reference to the speed limit in the area of 50 km/h. It is more likely a reference to the speed at which either the cyclist, or the car, was [19]avelling.19
It was likely that the information about the circumstances of the collision in the notes quoted at [22] above came from the deceased. By the time these notes were taken, the deceased had been in the emergency department for approximately 2 hours and had suffered significan[20]injuries.20 The deceased would have been distressed by and preoccupied with the nature of his injuries and his circumstances and would have been suffering significant pain. From a medical point of view he was alert and orientated and able to speak. However, that is not to conclude that any information he provided regarding the circumstances of the collision was accurate and reliable, or that the doctors accurately and reliably recorded any information provided. The doctors would have been recording information for the purposes of medical diagnosis and would only have been concerned to obtain the briefest information to allow them an understanding of the circumstances which caused the injuries to the deceased. They would not have been concerned to go into precise detail regarding the circumstances of th[21]accident.21
The following relevant notes were made on an 'Intensive Care Unit Flow Chart':[22]

ADMISSION DATE: 12.05.13

TODAY'S DATE: 12.05.13

DAYS IN ICU: 1

Diagnosis: cyclist vs car (50kph) collided [-] car (L) side and landed (R) chest. Nil LOC [loss of consciousness]. Flail chest # [fractured ribs] (L) 8 - 11/#??). #sternum clear cspine (L) elbow adema, abrasions

...

12/05/13

It is likely that this entry in the ICU flow chart was taken from earlier notes, most probably originating in Dr Murphy's notes quoted at [[22 Push bike v car accident.

Patient was cycling at 50 kmph, at round about, car failed to give way → Patient hit driver's side and landed on Rt side

Pt doesn't think he lost consciousness but not sure. SJA (St Johns Ambulance) called by bystanders. Pt not ambulant in the scene.

...

Impressions bike vs car accident

(emphasis added)

The primary judge was not satisfied that the information recorded in this note was taken directly from the deceased on 13 May 2013. His Honour was not persuaded that it was a contemporaneous record of something said by the deceased.[26] It was likely that the entry at [27] above was taken from the ICU flow chart quoted at [25] above. They reflect an interpretation by the au[27]or of earlier notes.27 It is unclear whether the notes were taken by a doctor or a nurse[28]nd for what purpose,28 but were more likely to have been[29]ecorded by a doctor.29 The time at which the notes were made and the observat[30]ns taken is unclear.30
At 8.58 am on 13 May 2013 a Registrar, Dr S Pradhan, made the following notes on a trauma services review:[31]

Bicyclist hit by car ≈ [approximately] 50 km/hr

That history is most likely to have been taken from earlier entries in the medical records and is not a fresh history taken from the deceased.[32]

An ICU review on 13 May 2013 relevantly noted:[33]

Days post trauma pushbike vs car (high speed) no loc [loss of consciousness]

It is unclear whether the reference to 'high speed' is a reference to the speed of the pushbike or the speed of the car. It is more likely to reflect a value judgement of the author from reading previous entries in the medical records than a note recording a conversation with the deceased.[34]

Admission notes for the deceased's admission into the HDU on 14 May 2013 stated:[35]

Admitted 2 days ago after car vs pushbike accident @ 50 kmph.

This entry had been carried through or taken from earlier medical records and does not record a further history taken from the deceased.[36]

The trial judge did not accept that the above notes were a reliable account of the speed at which the deceased was travelling, or believed he was travelling, when he entered the roundabout. While it was necessary for a proper understanding of how the injuries were caused to obtain details of the collision, it is likely those details would have been taken very quickly and the major focus would have been on assessing as quickly as possible the nature of the injuries suffered by the deceased, what medical investigations needed to be undertaken and what treatment should be provided.[37]
The trial judge also concluded that, even if the medical notes did accurately record that the deceased had told the doctors that he was riding his bike at 50 km/h, his Honour was not persuaded that, in the circumstances the deceased faced, his estimation of his speed was reliable.[38]
Grounds 3 and 4 challenge the conclusions noted at [32] and [33] above, so far as they relate to Dr Murphy's notes of 12 May 2013 which are quoted at [21] above.

Between 11.00 am and 12.00 pm on 15 May 2013 at the HDU, SC Bedwell took a statement from the deceased. A typed copy of the statement was emailed by SC Bedwell to the deceased at 3.47 pm on 16 May 2013. The typed statement described the accident in the following terms:[39]
1. I was riding towards the intersection with Eric Street, which is governed by a roundabout and I was riding to the left of a continuous white line that seems to designate a portion of the road for cyclists.
2. This is a route I have ridden on my pushbike for between ten or fifteen years, but no more than once a week. I would say that I am fairly familiar with this intersection.
3. As I approached this intersection, I had a good clear view from a long way back of traffic approaching the roundabout.
4. At this time I guess I was travelling around 25 - 30 kph.
5. I could see at least one, possibly even two, cars approaching the roundabout along Eric Street from my right, that is to say they were travelling in a westerly direction.
6. I think there may have been a car approaching the roundabout from the opposite direction to mine along Broome Street.
7. I also remember noticing a car going very slow, or even possibly stopped, at the intersection of the roundabout on Eric Street travelling in an easterly direction.
8. The vehicles travelling from my right had priority over myself when entering the roundabout, so I was watching them and trying to moderate my speed so that I didn't have to stop.
9. As I was watching the traffic to my right, I wasn't paying particular attention to the vehicle to my left, as I assumed that it would give way to me.
10. I entered the roundabout after the cars had passed in front of me.
11. I was probably going quite quickly for a pushbike, somewhere between 25 - 30kph.
12. When I was well into the roundabout, it became clear that the driver of the car to my left was pulling out.
13. I then had to brake very hard to try and slow down.
14. I turned my handlebars in an effort to steer around the car, but it was not effective.
15. I hit the side of the car and the next thing I remember is landing on the road.

At about 2.55 pm on 21 May 2013, the deceased emailed SC Bedwell in relation to the typed unsigned statement sent to him by SC Bedwell on 16 May 2013, part of which is set out at [35] above. In that email the dec[40]sed said:40

[T]hanks for your email - sorry it has taken me a few days, I've had problems getting my home computer set up and have only just seen it.

It's fine, actually pretty accurate.

The deceased went on to provide an update as to a presently irrelevant matter. By this email, the deceased accepted the contents of the unsigned statement as accurate and was prepared to sign it subject to updating the presently irrelevant matter. However, the deceased did not get to sign the statement before his death on 23 May 2013.[41]

The deceased's estimation of travelling at a speed of 25 ‑ 30 km/h, made on 15 May and confirmed on 21 May, was a more reliable estimation of his speed than any estimation he may have given to a doctor or nurse in the emergency department or ICU of SCGH, even though such an estimation was given closer in time to when the collision occurred.[42]

Mr Maddison was the only witness who was able to provide an estimation of speed. He estimated the deceased was travelling at 15 ‑ 20 km/h. He was not sure whether the deceased was peddling or free‑wheeling but thought he was coming at speed. He only saw the cyclist before he collided with the appellant's car for about 2 ‑ 3 s and over a distance of about 25 m.[43]
Mr Maddison was in the best position to make the most reliable estimation of the speed at which the deceased was riding his bike,[44] and his estimation was more reliable than the deceased's estimation of his own speed.[45] He was approaching the roundabout travelling in a westerly direction, that is, in the opposite direction to the appellant. The deceased was approaching the roundabout from Mr Maddison's left. Just after Mr Maddison entered the roundabout and before he reached the intersection with Eric Street, Mr Maddison saw the deceased when the deceased was about 5 m back from the give way sign facing him. The bicycle 'shot across in front or went across in front of' Mr Maddison's vehicle.[46]

Trial judge's approach Identification of issue raised by the pleadings

The trial judge began his substantive consideration by referring to the pleadings, which raised the issue of whether the collision and injuries to the deceased were caused or contributed to by the negligence of the deceased.[47] The essence of the appellant's contentions in those pleadings was that the deceased was travelling at excessive speed in the circumstances and failed to keep a proper look out.

The trial judge summarised the appellant's submissions on the issue of contributory negligence.[48] His Honour said the main basis on which the appellant pleaded the deceased was contributorily negligent was that he was riding at an excessive speed.[49] His Honour identified, in substance, three bases on which the appellant invited the court to find that the deceased was travelling at excessive speed:

(1) The court could find that the speed was excessive because the deceased was not able to react in time to avoid colliding with the appellant's vehicle.[50]

(2) The court could infer that the speed was excessive from the deceased's police statement that he 'was going fast for a bike'.[51]

(3) The court could make a finding, based on the deceased's statements recorded in the hospital notes, that he was travelling at 50 km/h, which was in all the circumstances excessive.[52]

Only the third basis required the trial judge to make a finding as to the speed, in km/h, at which the deceased was travelling. The trial judge also noted the appellant's counsel's closing submission that something less than 25 km/h would have been a safe speed in the circumstances of the accident.[53]

Rejection of appellant's case that the deceased was travelling at 50 km/h

After summarising the evidence,[54] the trial judge moved to consider what findings he should make as to the speed at which the deceased was riding his bicycle. His Honour began by considering what findings could be made based on the hospital notes.[55] In the course of doing so, the trial judge observed:[56]

The [appellant] bears the onus of satisfying me on the balance of probabilities that from the medical records, essentially the entry on 13 May in the ICU admission notes, and from other evidence, principally the eye witnesses to the accident, that [the deceased] was riding his bike at 50 km/h when he entered the roundabout.

The trial judge made the findings referred to above, including that:[57]

At the very least I am not persuaded that it is so clear that the ICU admission notes of 13 May 2013 record a fresh history taken from [the deceased]. Having regard to the onus of proof being on the [appellant] I am not satisfied that it is more probable than not that the ICU admission notes record a contemporaneous statement by [the deceased] that he was cycling at a speed of 50 km/h at the roundabout when the accident occurred.

[E]even if I accepted that the medical notes did accurately record that [the deceased] had told the doctors that he was riding his bike at 50 km/h, I am not persuaded, that in the circumstances [the deceased] faced, his estimation of his speed was reliable.

The trial judge then referred to the evidence of SC Patten as to what the deceased told him in the emergency department, accepting that the deceased had told the police officer that he was going fast for a pushbike.[59] The trial judge summarised his conclusion at this point in the following terms:[60]

In summary, the [appellant] has not satisfied me on the balance of probabilities that the medical records are a reliable record that [the deceased] said he was riding his bike at 50 km/h. Even if I was to accept the medical records do reliably record what a doctor or nurse was told by [the deceased] of what he believed was his speed at the time of the collision, I am not satisfied [the deceased's] estimation of his speed given in the emergency department or the ICU of SCGH was a reliable estimation of his speed. I am not prepared to find, based upon the entries in the medical records, that [the deceased] was riding his bicycle at 50 km/h. (emphasis added).

The trial judge then turned to consider the evidence of the account given by the deceased in his unsigned police statement. The trial judge explained why he concluded that the deceased's estimate in the statement of travelling at a speed of 25 - 30 km/h was more reliable than any estimation he may have given to a doctor or nurse in the emergency department or ICU of SCGH.[61] His Honour said that Mr Maddison was in the best position to make the most reliable estimation of the speed at which the deceased was riding his bike. He noted that Mr Maddison's estimation of 15 - 20 km/h was far closer to the deceased's estimation of 25 - 30 km/h. His Honour said that Mr Maddison's evidence:[62]

[S]upports my finding that if [the deceased] did give an estimation of speed of 50 km/h to medical staff at the hospital, it was an unreliable estimation. It is unlikely Mr Maddison, who was called by the [appellant], would have estimated the speed at 15 - 20 km/h if [the deceased's] speed was more than twice that, or three times greater than the lower estimation of speed.

The trial judge noted that neither the appellant nor Ms Staszewski (Mr Maddison's passenger) were able to give an estimation as to the speed at which the deceased was travelling.[63]

Finding as to the speed at which the deceased was travelling

Having rejected the proposition that the deceased was travelling at 50 km/h in the above passages, the trial judge turned to consider whether he could make any findings as to the speed at which the deceased was riding his bicycle.[64]
The trial judge first addressed the deceased's unsigned police statement.
His Honour did not accept the appellant's submission that the deceased was in the best position to make an estimation of his own speed. His Honour said:[65]

A person's estimation as to the speed at which they were travelling is likely to be more accurate and reliable if their bicycle was fitted with a speedometer. There is no evidence [the deceased's] bicycle was fitted with a speedometer.

The trial judge noted that a person's cycling experience and the types of speed at which they routinely travelled might give weight to their estimate of travelling speed. His Honour then referred to evidence suggesting that the deceased was an occasional, although regular, recreational cyclist who was not a serious or expert cyclist likely to be riding at a higher speed.[66]

The trial judge noted that there was no evidence as to how quickly the deceased's bike could be ridden. Nor was there evidence about the incline of the street towards the roundabout, or the impact incline could have on the speed at which the bicycle could be ridden.[67] His Honour said:[68]

I raise these matters, not to speculate about matters which are not in evidence, but to emphasise that it is the [appellant] who bears the onus of proof in satisfying me on the balance of probabilities that [the deceased] was contributorily negligent in causing the accident because he was riding his bicycle at an unsafe speed.

The trial judge referred to his rejection of the appellant's contentions as to what could be drawn from the hospital notes.[69]
Returning to the deceased's police statement, the trial judge said that the use of the words 'guess' and 'probably' indicated that the deceased was uncertain of what his speed was.[70] His Honour said that the deceased's statement that he was 'probably going quite quickly for a pushbike' was of little or limited weight in the absence of knowing the speed at which the deceased regularly rode his bicycle.[71] His Honour concluded that the deceased's reference to moderating his speed was to adjusting his speed as he prepared to merge with the traffic in the roundabout so he did not have to stop.[72]
The trial judge turned to address the deceased's statement that he had entered the roundabout after cars had passed in front of him. The trial judge said:[73]

The [appellant] submits that there is no evidence which confirms that [the deceased] entered the roundabout after cars to his right had passed in front of him and that he was wrong about that. I accept that submission. It reinforces my view that [the deceased's] memory of all of the circumstances leading up to the accident is not necessarily that reliable, which is understandable given the sudden and traumatic nature of the accident and the significant injuries suffered by him. It reinforces my view that his estimation of his own speed is not reliable.

The trial judge also said that the statement about cars passing in front of him supported the finding that, when he referred to trying to moderate speed, the deceased was:[74]

[A]djusting his speed by reference to the cars on his right, having regard to the need to give way to them and not wanting to stop at the roundabout. In those circumstances the reference to trying to moderate speed means to adjust his speed by reducing his speed.

The trial judge rejected a submission by the appellant that the deceased's statement was unconsciously influenced by the prospects of a compensation claim.[75]
The trial judge then turned to Mr Maddison's evidence, summarising its effect and concluding that Mr Maddison was in the best position to assess the speed at which the deceased's bicycle was travelling.[76] His Honour acknowledged and had regard to the difficulty in estimating the speed at which another object is travelling, particularly over a brief period of time.[77] He then expressed his conclusion as to the deceased's speed in the following terms:[78]

I am satisfied Mr Maddison's estimation of the speed at which [the deceased] was riding his bicycle is the most reliable estimation as to speed. It is a more reliable estimation than [the deceased's] own estimation as to his speed.

Further, his estimation of the upper range of 20 km/h is not dissimilar to [the deceased's] estimation at the lower speed of 25 km/h.

Without wanting to disparage any 63-year-old male recreational cyclists riding on their pushbike once a week, I think a speed of 15 km/h reflects a speed which could be realistically achieved by such a cyclist. A speed of 20 km/h would certainly be 'going quite quickly for a pushbike'.

I find that [the deceased] was probably travelling at a speed of approximately 15 km/h as he entered the roundabout and no more than 20 km/h.

As noted above, these findings are challenged by grounds 5 and 6.

Whether the deceased's manner of riding was negligent

After setting out the applicable legal principles,[79] the trial judge proceeded to determine whether at a speed of 15 - 20 km/h, having regard to all of the surrounding circumstances, the deceased's manner of riding his bike was negligent and contributed to the cause of the accident.[80]
The trial judge noted three aspects of the evidence which his Honour regarded as particularly important:

(1) As the appellant approached the roundabout, she slowed almost to a stop so that the deceased was entitled to assume she had seen him and was giving way as required by the give way sign.[81]

(2) The appellant drove her vehicle into the roundabout after the deceased entered the roundabout.[82]

(3) There was no suggestion by Mr Maddison or his passenger Ms Staszewski that the deceased's actions caused Mr Maddison to take any sudden evasive action, or that the deceased entered the roundabout in an unsafe manner or at an unsafe point in time.[83]

The trial judge found that the appellant had an unobstructed view of the intersection, and said that:[84]

Although [the deceased] was riding quickly for his pushbike he was entitled to ride in such a way as to expect [the appellant] would remain stationary at the give way sign. He rode in such a way that he did not pose or create a risk of a collision with Mr Maddison's vehicle. Mr Maddison was able to pass through the intersection behind [the deceased] without slowing down, stopping or taking any evasive action. Had [the appellant] not moved forward into the roundabout [the deceased] would have ridden safely through the roundabout.

I am not persuaded that his manner of riding his bicycle was in all of those circumstances negligent.

The trial judge then considered and rejected the appellant's contention that it could be inferred that the deceased was travelling at an unsafe speed from the fact that he was unable to avoid the collision.[85] The trial judge concluded:[86]

For the reasons I have already given, I am satisfied [the deceased's] manner of riding, in circumstances where he saw and believed [the appellant] was coming to a stop, or had stopped, at the give way sign to allow him to cross through the roundabout, was safe. It was reasonable that he believed it was safe for him to enter into the intersection providing he could safely do so having regard to traffic entering the roundabout from his right. [The appellant] drove into the intersection unexpectedly. Taking reasonable care for his own safety did not require [the deceased] to ride at such a speed, and in such a way, to be able to avoid a collision with [the appellant's] vehicle if she unexpectedly pulled out into the intersection in front of him. I have found [the deceased] was probably riding his bike at a speed of approximately 15 km/h and no more than 20 km/h. In my view that was in all of the circumstances a safe speed to be riding the bicycle when [the deceased] approached and entered the roundabout. A cyclist, taking reasonable care for his own safety, seeing a vehicle stationary at a give way sign, is entitled to enter the intersection in the expectation the driver will comply with the give way sign and not unexpectedly fail to give way and enter into the intersection after the cyclist has entered the intersection.

The trial judge then turned to address the issue of whether any contributory negligence by the deceased caused his injuries and death. The trial judge held that, at whatever speed the deceased was riding his bicycle, his speed was not causative of the collision.[87]
The trial judge found that the only cause of the collision was the appellant's failure to see the deceased and her entering into the roundabout after the deceased had already entered the roundabout. She entered the intersection without seeing the deceased. She therefore entered the intersection unexpectedly. In circumstances where the appellant had driven into the intersection immediately in front of the deceased's bicycle, at whatever speed he was riding, it is likely the deceased would have been unable to avoid the collision.[88]
The deceased rode his bike in a safe manner and was entitled to assume it was safe for him to enter the roundabout in the manner in which he did. It was reasonable for him to expect that the appellant had seen him and would not enter into the roundabout until he travelled through it. Taking proper care for his own safety, the deceased was not required to expect that the appellant would enter into the roundabout as she did and that he would need to take evasive action to avoid a collision. The speed at which he was riding his bicycle, whatever it was, was not causally relevant to the collision. The sole cause of the collision was the appellant's manner of driving, involving her failure to observe the deceased and to drive into the roundabout after the deceased had already entered the roundabout.[89]

The trial judge therefore found that the deceased was not contributorily negligent in his manner of riding his bicycle. His Honour ordered that the appellant pay the respondent 100% of the damages to be assessed and her costs of the preliminary issue.[90]

Ground 1: Failure to consider the evidence as a whole

The learned trial judge erred in law in that he failed to consider and evaluate the accumulation of the evidence as a whole and to consider its united force, and instead considered evidence in a piecemeal way in isolation from other evidence and wrongly applied the ultimate standard of proof to particular pieces of evidence considered in isolation from other evidence.

In my view, there is no merit to this ground. In the manner described above, the trial judge had regard to the whole of the evidence - the hospital notes, statements made by the deceased to police and eyewitness accounts - in reaching his conclusion as to the speed at which the deceased was travelling.
The appellant's submissions on this ground focus on the passage quoted at [42] above, reproduced again for convenience below:

The [appellant] bears the onus of satisfying me on the balance of probabilities that from the medical records, essentially the entry on 13 May in the ICU admission notes, and from other evidence, principally the eye witnesses to the accident, that [the deceased] was riding his bike at 50 km/h when he entered the roundabout.

Appellant's onus to prove the deceased was travelling at 50 km/h

The appellant's first criticism of this passage is that she was not, she submits, required to prove that the deceased was travelling at 50 km/h. She submits that she was only required to prove that the deceased was riding at excessive speed at the final stage of the reasoning process, having regard to the accumulation of the evidence as a whole and its united force. The appellant submits that the evidence that the deceased said that he was travelling at 50 km/h was only one piece of evidence relevant to that issue.
In my view, this submission ignores the structure of the trial judge's reasons, which was consistent with, and reflected, the manner in which the appellant had put her case. His Honour had identified one basis on which he understood the appellant to contend that the deceased was travelling at excessive speed as being that he should find, based on the medical notes, that the deceased was travelling at 50 km/h.[91] The trial judge, in summarising his understanding of that submission, referred to the following comments made by the appellant's counsel in opening:[92]

[T]his is a case where there is, as it happens, a strong evidential basis for - well, I can't say it's a strong - there is an evidentiary basis, in the words of [the deceased] himself, that he was travelling at 50 kilometres an hour.

Four days later he put it at 25 to 30 kilometres an hour, but in terms that it was still quick for a bike. It's - our friends say themselves in their written submissions that the speed limit on this road was 50 kilometres an hour. So at 50 kilometres an hour, he is traveling at the nominated speed limit for the road as he enters an intersection at which vehicles are present, at which he knows vehicles are present ...

[T]he point is he was aware that there was traffic at the intersection and at 50 kilometres an hour, was traveling at the nominated speed limit through the intersection on a bicycle.

And anyone who - in my submission, anyone who is familiar with the ordinary day-to-day incidents of traveling through an intersection is capable - is able to form the view as a matter of ordinary experience and common sense that that is just too [fast].

Further, in written opening submissions at trial, the appellant submitted that the deceased's 'original admission that he was travelling at 50 kph is more likely to be correct'.[93] The position was maintained in closing, where the appellant's counsel invited the trial judge to find that the deceased was travelling at 50 km/h.[94] In doing so he made extensive oral submissions as to the hospital notes, contending that they were reliable evidence of what the deceased had said about his speed and that the deceased's estimate was reliable.[95]
The trial judge was, therefore, correct to apprehend that one of the bases on which the appellant advanced her case of contributory negligence was that the deceased was travelling at 50 km/h, which was an excessive speed in the circumstances. To make out a case on that basis, the appellant needed to prove, on the balance of probabilities, that the deceased was travelling at 50 km/h. The only evidence of that fact to which the appellant was able to point was the hospital notes. Before he could be satisfied that the evidence established that the deceased was travelling at 50 km/h, the trial judge would have to be satisfied that the notes accurately recorded what the deceased had told hospital staff and that his estimation of his own speed was accurate.
That was what the trial judge was doing in the part of his reasons which are the focus of the appellant's complaint. He was addressing the submission that he should find the deceased contributorily negligent on the basis that he was travelling at 50 km/h, which was an excessive speed in the circumstances. His Honour correctly recognised that the appellant bore the onus of establishing that the deceased was travelling at 50 km/h before he could find the deceased to have been contributorily negligent on that basis. The trial judge rejected the submission because the essential facts on which it was based were not established on the balance of probabilities.
It may also be noted that, in rejecting that part of the appellant's case based on the proposition that the deceased was travelling at 50 km/h, the trial judge did have regard to the deceased's statements to police,[96] as well as the evidence of Mr Maddison.[97] Even in dealing with that aspect of the appellant's case based on the contention that the deceased was travelling at 50 km/h, his Honour had regard to the whole of the evidence and did not take a piecemeal approach to his task.
On appeal, senior counsel for the appellant denied that the references to the deceased's and Mr Maddison's evidence noted in the previous paragraph addressed the question of whether the deceased was travelling at 50 km/h.[98] I do not accept that submission. The trial judge's discussion of their evidence at this point of his reasons was expressly related to the proposition that the deceased was travelling at 50 km/h. For example, the trial judge said that Mr Maddison's estimate of speed 'supports my finding that if [the deceased] did give an estimation of speed of 50 km per hour to medical staff at the hospital, it was an unreliable estimation.'[99]
Rejection of the proposition that the deceased was travelling at 50 km/h resolved one of the bases on which the appellant's contributory negligence argument was advanced. That conclusion did not resolve the whole of the contributory negligence argument, as the trial judge recognised. Having rejected that aspect of the appellant's case, his Honour went on to make findings as to the speed at which the deceased was travelling.[100] In doing so, he did not suggest that the appellant bore the onus of proving that the deceased was travelling at any particular speed.

Reference to the notes of 13, rather than 12, May 2013

The appellant's second criticism of the trial judge's reasons quoted at [66] above is that the focus on the notes of 13 May 2013 'was (with respect) inexplicable and deflected from the question'.
In my view, the focus on the notes of 13 May 2013 was readily explicable. The notes made by Dr Murphy on 12 May 2013, either alone or with other evidence, were incapable of supporting a finding that the deceased told Dr Murphy that he was travelling at 50 km/h. The trial judge correctly made a finding, noted at [23] above, to the effect that it was not clear whether the reference to 50 km/h in the 12 May 2013 notes was to the speed of the cyclist or the car. I will explain my reasons for reaching that conclusion in dealing with ground 3, at [85] - [93] below.
The earliest reference in the hospital notes to the deceased travelling at 50 km/h was in the ICU admission notes of 13 May 2013. Later medical notes were either equally ambiguous as Dr Murphy's notes of 12 May 2013, or were likely derived from earlier notes. Therefore, success of that part of the appellant's case based on the contention that the deceased was travelling at 50 km/h required her to establish that the note of 13 May 2013 accurately reflected what the deceased had said and that his estimate of speed was reliable. Having made findings as to the lack of clarity in the 12 May 2013 note, the trial judge was correct to focus on the ICU admission notes of 13 May 2013.

The appellant's third criticism of the trial judge's reasons quoted at [66] above is that it identifies evidence as 'principal' or 'essential'. Again, in my view, that criticism is entirely unwarranted. The note of 13 May 2013 was the only statement in the hospital notes, which was not clearly derived from earlier notes, that the deceased was travelling at 50 km/h. Senior counsel for the appellant acknowledged that the only direct reference to a 50 km/h speed was in the hosp[101]l notes.101 That note was the essential basis on which this part of the appellant's case was based, and the principal other evidence which went to the issue of speed was the eyewitness accounts In any event, description of evidence as 'principal' or 'essential' does not mean that it was the only evidence considered.

Ground 2 contends that the trial judge erred in law in failing to give adequate reasons for his decision.
The fundamental requirement for reasons is to adequately disclose the actual intellectual process which has resulted in a particular determination.[102] Where there is a right of appeal, the reasons must be sufficient to give effect to that right. The basis for the decision must be apparent, as otherwise the losing party cannot know whether there has been a mistake of law or of fact. Just what that will involve depends upon the nature of the case.[103] As was noted in G v O,[104] whether the content of reasons is adequate will depend on the circumstances of the case and the matters that arose for the judge's consideration. The court will look at the reasons as a whole, if necessary in the context of the evidence, to determine if they achieve their required function and purpose. The fundamental elements of a statement of reasons are:

(1) a reference to the relevant evidence (which need not be detailed);

(2) a statement of material findings of fact and any ultimate conclusions;

(3) a statement of the reasons for making those findings and conclusions; and

(4) an explanation of how the law was applied to the facts as found.

In doing these things, the reasons must demonstrate an engagement with the losing party's case.

In my view, the trial judge's reasons comfortably satisfy all of the above requirements. They explain the intellectual process which his Honour adopted to reach the conclusions at which he arrived, in a manner which allows this court to identify any error. The appellant's complaints are really either just disagreement with the reasoning process, or complaint that error is revealed by that process, rather than a complaint that the reasoning process is unascertainable.
For example, the appellant says that the trial judge did not deal with the inconsistency between Mr Maddison's speed estimate of 15 - 20 km/h and his estimate that the deceased travelled 25 m over 2 - 3 s (indicating a speed of 30 - 45 km/h).[105] While the appellant made submissions to the trial judge relying on this inconsistency, the judge was not required to deal expressly with every submission made by a party. This was not a matter of such significance that a failure to refer to it amounted to a failure to engage with the losing party's case. The appellant's real argument is that the failure to refer to the inconsistency reveals that the judge erred by overlooking it. I deal with, and reject, that argument at [124].
In my view, ground 2 is entirely without merit.

Ground 3: Wrong findings as to Dr Murphy's notes

Ground 3 contends that the trial judge erred in his findings with respect to Dr Murphy's notes of 12 May 2013.
The note relevantly said:

Cyclist vs car this afternoon 50km/hr

Was at roundabout → BMW failed to give way and hit pt on L) side → Landed on R) chest

The appellant first contends that the trial judge erred in failing to find that these notes were an accurate record of the deceased's own contemporaneous estimate of the speed at which he was travelling immediately prior to the collision with the appellant's vehicle.
There is nothing in this note to indicate what the '50 km/hr' refers to. It is ambiguous as to whether the appellant's car or the deceased's bicycle was said to be travelling at 50 km/h, or whether the reference is to something else such as the speed limit in the area. The note is also materially inaccurate, as it refers to the appellant's car hitting the deceased on his left side, rather than the deceased running into the appellant's car as she pulled out in front of him. The note does not identify the person who provided the information to Dr Murphy, in particular whether the noted information was obtained directly from the deceased or via a third person such as a nurse or ambulance officer. Dr Murphy did not give evidence at trial. No person gave evidence of assistance in understanding the note, or the circumstances in which it was taken.
The appellant submits that, as her car was clearly not travelling anywhere near 50 km/h at the time of the accident, the reference to '50 km/hr' must be to the speed of the cyclist. She submits that, in construing the note, regard should be had to the other evidence. The appellant particularly refers to evidence to the effect that the deceased was travelling fast, and Mr Maddison's estimate that the deceased covered about 25 m in 2 - 3 s (equivalent to up to 45 km/h). The appellant says that, when Dr Murphy's note is read in the context of this evidence, it can be inferred that the reference to 50 km/h must be to the speed of the cyclist.
I do not accept the submission summarised in the previous paragraph. In my view, the evidence other than the hospital notes makes it highly unlikely that the deceased was, or thought that he might have been, travelling at anywhere near 50 km/h. In particular:

(1) The deceased's later statement to police indicates that neither he nor the appellant were travelling at 50 km/h, and estimates that he was travelling at approximately 25 - 30 km/h.

(2) Mr Maddison's evidence also indicates that neither the bicycle nor the car were travelling anywhere near 50 km/h. As the trial judge correctly noted, Mr Maddison is highly unlikely to have estimated the deceased's speed to have been 15 - 20 km/h if he was actually travelling at or near 50 km/h. Further, for the reasons explained at [110] - [124] below, Mr Maddison's estimates of the distance travelled by the deceased over time do not provide a reliable basis for determining the speed at which the deceased was travelling.

(3) It is inherently unlikely that a 63-year-old occasional recreational cyclist, who was not riding a racing bike, would have been capable, or would have considered himself capable, of reaching 50 km/h.

The weight of the other evidence indicates that neither the appellant's car nor the deceased could have been travelling anywhere near 50 km/h. In that context, there is no reason to attribute the reference to '50 km/hr' in Dr Murphy's note of 12 May 2013 as being to the speed of the cyclist rather than the car. If the other evidence established that neither the bicycle nor the car were actually travelling at or anywhere near 50 km/h, then there would be no reason to conclude that the reference to speed was to one or the other of the car or bicycle.
The notes of 12 May 2013 were ambiguous as to what the deceased had said. In my view, the other evidence taken as a whole does not resolve the ambiguity in those notes. The trial judge properly concluded that the note of 12 May 2013 did not make it clear whether the reference to speed relates to the car or the cyclist.[106] They are not a record of the deceased making a statement estimating his speed, as opposed to the speed of the car, to have been 50 km/h. In my view, the evidence as a whole does not support, and is incapable of supporting, an inference that the deceased told Dr Murphy that he was travelling at 50 km/h at the time of the accident.
The first record indicating that the deceased was travelling at 50 km/h are the ICU admission notes of 13 May 2013. However, the trial judge was not persuaded the notes of 13 May 2013 were a contemporaneous record of something said by the deceased. His Honour found that they were more likely to have been a doctor's interpretation of earlier notes. There is no ground of appeal challenging that finding about the ICU admission notes of 13 May 2013. As noted above, later medical notes were either equally ambiguous as Dr Murphy's notes of 12 May 2013, or were likely derived from earlier notes.
Therefore, in my view the evidence does not support, and is incapable of supporting, the inference that the deceased ever said that he was travelling at 50 km/h at the time of the accident. There is no other basis for concluding the deceased was travelling at or near that speed at the time of the accident.
In the alternative, the appellant contends that the trial judge erred in finding (if he did so find) that Dr Murphy's notes were not an accurate record of that matter in that it was against the weight of the evidence, was not a permissible inference and was unreasonable. In my view, the trial judge did not find that Dr Murphy's notes were an inaccurate record of what the deceased told Dr Murphy (although his Honour also did not find that they were accurate and pointed to a number of factors which could affect their accuracy). The problem with Dr Murphy's notes, from the appellant's perspective, is not that they are an inaccurate record of what the deceased said. Rather, the problem is that the notes do not purport to record what the deceased said about the speed he was travelling at the time of the accident.
In my view, ground 3 is not established.

Ground 4: Findings as to the reliability of the deceased's estimate

Ground 4 contends that the trial judge erred in finding that, insofar as Dr Murphy's notes of 12 May 2013 were an accurate record of the deceased's estimate of his speed, that estimate was unreliable. The appellant contends that there was no evidence for that finding or, alternatively, it was against the weight of the evidence, was not a permissible inference and was unreasonable.
The failure of ground 3 makes it strictly unnecessary to deal with this ground. Unless it is found that the deceased told Dr Murphy that he was travelling at 50 km/h, the question of the accuracy of the deceased's estimate that he was travelling at that speed does not arise. The trial judge was not satisfied that the deceased told Dr Murphy that he was travelling at 50 km/h at the time of the accident and this finding has not been successfully challenged on appeal.
In any event, having regard to the matters noted at [89] above, the ground is without merit. Common experience indicates that it would be highly unlikely for a cyclist such as the deceased to be able to sustain a speed of 50 km/h on a pushbike. The estimates of both the deceased in his statements to police and of Mr Maddison were that the deceased was travelling much slower that 50 km/h. In my view, the evidence, considered as a whole, did not leave it open to reasonably find that the deceased was actually travelling at 50 km/h, even if that hypothetically was what he said in the emergency department in an injured and traumatised state.

Grounds 5 and 6: Findings as to the deceased's actual speed

It is convenient to deal with grounds 5 and 6 together, as they both challenge the trial judge's finding that the deceased was travelling at a speed of approximately 15 km/h as he entered the roundabout and no more than 20 km/h.

Reference to 15 km/h as a 'speed which could reasonably be achieved'

Ground 5 contends that the trial judge erred in finding that 15 km/h 'reflects a speed which could reasonably be achieved by such a cyclist'.[107] The reference to 'such a cyclist' was to a '63-year-old recreational cyclist riding on their pushbike once a week'.[108]
The submissions advanced in support of ground 5 take this observation to be a finding that a 63-year-old male recreational cyclist cannot realistically cycle at more than 15 km/h. Senior counsel for the appellant confirmed that this was the basis on which the ground was advanced.[109] If that had been the trial judge's finding then, in my view, it would have been in error and unsupported by the evidence or common experience.
However, in my view, that was not what the trial judge was saying. The trial judge was saying that 15 km/h was a speed that could be realistically achieved by a cyclist such as the deceased, but was not saying that it was the highest speed that such a cyclist could realistically achieve. His Honour was making that observation in the context of a trial where the appellant had suggested a speed of 50 km/h. In context, the statement that 15 km/h reflected a speed that could be realistically achieved indicated the trial judge's view (which I share) that the alternative proposed speed of 50 km/h was unrealistic, at least for a cyclist such as the deceased. There was no error in that observation.
That is, the trial judge was not saying that the deceased was, or a cyclist such as the deceased would be, incapable of travelling any faster than 15 km/h. Rather, his Honour was assessing whether the speed which he found the appellant to be travelling at was realistic for a cyclist such as the deceased. In my view, the submissions advanced in support of ground 5 are based on a misinterpretation of the trial judge's observation. The appellant correctly accepted that ground 5 would not be made out if the impugned passage of the trial judge's reasons was construed in the manner indicated above.[110]

Ground 6 contends that the learned trial judge erred in fact in finding that the deceased was travelling at a speed of 15 km/h and no more than 20 km/h. The appellant contends that the finding was against the weight of the evidence, was not a permissible inference and was unreasonable.
In making this factual challenge, the appellant accepts that the role for this court is that which was explained by the High Court in Lee v Lee:[111]

A court of appeal is bound to conduct a 'real review' of the evidence given at first instance and of the judge's reasons for judgment to determine whether the trial judge has erred in fact or law. Appellate restraint with respect to interference with a trial judge's findings unless they are 'glaringly improbable' or 'contrary to compelling inferences' is as to factual findings which are likely to have been affected by impressions about the credibility and reliability of witnesses formed by the trial judge as a result of seeing and hearing them give their evidence. It includes findings of secondary facts which are based on a combination of these impressions and other inferences from primary facts. Thereafter, 'in general an appellate court is in as good a position as the trial judge to decide on the proper inference to be drawn from facts which are undisputed or which, having been disputed, are established by the findings of the trial judge'. (citations omitted, emphasis added)

As to the appellate restraint with respect to factual findings, the High Court observed in Robinson Helicopter Co Inc v McDermott:[112]

A court of appeal conducting an appeal by way of rehearing is bound to conduct a 'real review' of the evidence given at first instance and of the judge's reasons for judgment to determine whether the judge has erred in fact or law. If the court of appeal concludes that the judge has erred in fact, it is required to make its own findings of fact and to formulate its own reasoning based on those findings. But a court of appeal should not interfere with a judge's findings of fact unless they are demonstrated to be wrong by 'incontrovertible facts or uncontested testimony', or they are 'glaringly improbable' or 'contrary to compelling inferences'. (citations omitted)

Even where the appeal concerns the inferences to be drawn from established primary facts, it remains necessary for the appellant to demonstrate error. As Beaumont and Lee JJ observed in Minister for Immigration, Local Government and Ethnic Affairs v Hamsher:[113]

[T]he court is not obliged to proceed to make new findings of fact on all relevant issues and discharge the judgment appealed from if those findings differ from those of the trial judge and do not support the judgment. The court must be satisfied that the judgment of the trial judge is erroneous and it may be so satisfied if it reaches the conclusion that the trial judge failed to draw inferences that should have been drawn from the facts established by the evidence. The court is unlikely to be satisfied if all that is shown is that the trial judge made a choice between competing inferences, being a choice the court may not have been inclined to make but not a choice the trial judge should not have made. Where the majority judgment in Warren v Coombes (at 552 - 553) states that an appellate court must not shrink from giving effect to its own conclusion, it is speaking of a conclusion that the decision of the trial judge is wrong and that it should be corrected.

That observation has been adopted by this and other intermediate appellate courts.[114]

The choice as to the appropriate appellate standard of review is not binary. As Beech and Vaughan JJA observe at [206] - [213] below, the nature and extent of the trial judge's advantage informs what is required in order to reach, and the extent of appellate restraint in reaching, a conclusion of error. The nature of the finding, and the reasoning by which it was made, affects what is required in order to demonstrate appellable error. I agree with their Honours' observations in those paragraphs.

The difficulty facing the appellant in this case is that the trial judge's finding as to the speed at which the deceased was probably travelling was based on the evidence of Mr Maddison. Mr Maddison was a witness called by the appellant, and was the only living eye witness to the accident who was able to provide an estimate of the deceased's speed. His estimate that the deceased was travelling at 15 - 20 km/h was not challenged in cross-examination. His estimate of the deceased's speed was neither inherently improbable nor contradicted by other objective evidence. Mr Maddison had a clear view of the deceased as the deceased rode across his path. His unchallenged estimate which had been adduced in evidence by the appellant supported the trial judge's finding as to the deceased's speed.
The appellant relies on Mr Maddison's estimate that he saw the deceased cover a distance of 25 m over 2 - 3 s. The appellant says that this distance over time is equivalent to 8.33 - 12.5 m/s, which is 30 - 45 km/h. The appellant says that the trial judge failed to reconcile this aspect of Mr Maddison's evidence with his estimate that the deceased was travelling at 15 - 20 km/h. The appellant says that there is an inconsistency between Mr Maddison's estimate of the deceased's speed of 15 - 20 km/h and his estimation that the deceased travelled 25 m over 2 - 3 s.[115]
One difficulty with using estimates of distance and time in this manner is that small changes in the estimates can produce large changes in the resulting speed. Suppose that Mr Maddison's estimate as to distance travelled may have been out by up to 5 m and his estimate of the time taken to travel that distance may have been out by up to a second. If the deceased covered 20 m in 4 s, he would have been travelling at 5 m/s, or about 18 km/h. If the deceased covered 30 m in one second he would have been travelling at 30 m/s, or about 108 km/h.
Mr Maddison did not base his estimate of speed on a calculation of distance over time, and his estimate of distance and time does not, in my view, provide a proper basis for rejecting his estimation of the deceased's speed. The estimates of distance and time were given in qualified terms, as indicated by the following part of Mr Maddison's evidence-in-chief:[116]

Are you able to estimate how long - how long it was between the time that you first saw the cyclist until he hit the car?---Two or three seconds, maybe.

And are you able to say in metres how far it was between the point when you first saw the cyclist and the point at which he hit the car?---I saw him about 5 metres back from the give way line and I would say probably another 20 metres maybe from there to cross to the - maybe a little bit more. I'm not sure.

But are you saying 20 metres from where?---I would say from that give way line, from the roundabout side of the give way line.

So you saw him? So it's 20 plus 5 from the point - - - ?---Yeah, 25 metres before the give way line. It'll be 20 metres on the outside before he hit the car. (emphasis added)

In cross-examination, Mr Maddison gave the following evidence as to this issue:[117]

I think you said in your evidence, Mr Maddison that between when you first saw the cyclist, he was about five metres back from the give way sign?---That's right, yes.

Between that time and the collision was about two or three seconds?---Yep, that's what it appeared to be to me, yes.

(Inaudible)?---Well, yeah, but I mean, so I could see what was going to happen.

Then the distance further on from the give way sign to the point of collision, about 20 metres again?---A rough guess, yes. Twenty, 25.

And the distance that the car travelled from her give way sign to the point of collision, I think you said that was about - - -?---Five, yeah, roughly, yeah.

All these are rough figures?---Yeah. Yeah. I could see what was going to happen, what was going to happen, ...

(emphasis added)

These admittedly rough estimates of distance and time in my view provide an insecure foundation for reaching any conclusion about the speed being travelled by the deceased.
Further, contrary to the appellant's submissions,[118] in my view, estimates of the approximate speed travelled by a vehicle are likely to be more accurate than a speed calculated from estimates of both distance and time, especially where the time period is brief. While the capacity to estimate these matters will no doubt vary between individuals, ordinarily a regular road user would be expected to be more readily and reliably able to estimate speed than to estimate distance travelled and time elapsed over a period of the order of a few seconds. In Mr Maddison's case, he estimated the deceased's speed of 15 - 20 km/h in unqualified language while his estimates of distance and time were expressed in a more tentative way.
The same observations about the difficulties of drawing a conclusion of the deceased's speed from a witness' estimation of distance travelled over time apply to Ms Staszewski's evidence. Her estimate, to the effect that the deceased covered about 29 m over 3 s, appears to have been given in even more hesitant and qualified terms than Mr Maddison's estimate.[119]

Senior counsel for the appellant submits that the uncertainties as to the distance travelled by the deceased can be resolved by reference to scale plans of the accident scene. Counsel submits that the scale plan shows that the distance from the give way sign at the southern entry to the roundabout from Broome Street to the point at which the appellant's car came to a rest is 20 m. It is submitted that the estimate that the deceased was 5 m from the entry when Mr Maddison first saw him can be combined with the scale measurement. Counsel submits that this provides a reliable distance of 25 m travelled in the 2 - 3 s over which Mr Maddison observed the deceased before the collision.
For the following reasons, I do not accept the submissions summarised in the previous paragraph.
The plans to which counsel referred are trial exhibits 1.8, 1.9 and 1.10. They are noted as being prepared by 'Verifact', and purport to show the roundabout, the resting position of the appellant's car and a line of 12.0 m from the rear of the car to the western entry to the roundabout from Eric Street. The appellant, Mr Maddison and Ms Staszewski have respectively drawn a cross to indicate the point of impact between the deceased's bike and the car. Those witnesses were cross-examined by senior counsel for the respondent as to the location of the points of impact they had indicated in the plan. The marked copies of the plan were tendered by counsel for the respondent at the conclusion of the cross-examination of those witnesses.[120]
After the tender of those plans, Gerrie John Anderson, an employee of Verifact, was called by the appellant. He gave the following evidence-in-chief about the plan which became exhibits 1.8, 1.9 and 1.10:[121]

Leaving aside the handwriting on the document and leaving aside the handwritten X on the roundabout - - -?---Yes.

- - - do you know how this document came in to existence?---Yes. From the sketch I prepared.

So when you went to the site you prepared a sketch?---Yes.

And what did you do with that sketch once you had prepared it?---It - it gets emailed to a - a draftsperson in - in Brisbane and they prepare a scale plan.

Mr Anderson gave evidence to the effect that he had measured the 12 m distance from the give way line to the rear of the car. Senior counsel for the appellant then asked him:[122]

Okay. Do you recall taking any other measurements at the scene?---Well - no. That's the only one that I can see on the - on the - on the - on the diagrams.

In the course of cross-examination, Mr Anderson confirmed that he had not measured the width of the lane.[123]
The effect of Mr Anderson's evidence was that a draftsperson in Brisbane had drawn the plan from a sketch Mr Anderson had prepared. However, the 12 m measurement was the only measurement which Mr Anderson took. It follows that the draftsperson must either have drawn the location of give way lines, traffic islands and the road configuration from a sketch drawn by Mr Anderson without measuring, or from some other source the reliability of which is not the subject of any evidence. Given this state of the evidence, in my view, no reliable conclusions can be drawn as to the accuracy of the scale plan beyond the measurement taken by Mr Anderson. In my view, it would not be appropriate to use other elements of the drawing and its scale to attempt to make a finding as to other distances. Even if the scale plan is admissible for that purpose, it is not evidence of other distances which deserves any significant weight.

Absence of reference to estimates of distance and time in primary decision

The appellant says that the trial judge did not deal with the inconsistency between Mr Maddison's speed estimate of 15 - 20 km/h and his estimate that the deceased travelled 25 m over 2 - 3 s (indicating a speed of 30 - 45 km/h).[124] The trial judge did not ignore the evidence, referring to it in his summary of Mr Maddison's evidence.[125] Any failure by the trial judge to consider how these parts of Mr Maddison's evidence were to be reconciled did not result in any miscarriage of justice, as he would properly have preferred the estimate of 15 - 20 km/h for the reasons summarised above.

The deceased's own estimate of his speed, given in the police statement, was 25 - 30 km/h. However, he couched that estimate in qualified terms- 'I guess I was travelling around 25 - 30 kph', and 'I was probably going quite quickly for a pushbike, somewhere between 25 - 30 km/h'. This is what he regarded as 'quite fast for a pushbike'. The difference between the bottom of the range estimated by the deceased estimate (25 km/h) and the top of the range estimated by Mr Maddison (20 km/h) is only 5 km/h. Further, unlike Mr Maddison, the deceased was recounting a traumatic event in which he has received serious injuries, which were likely to impact on the reliability of his estimate.

Trial judge's finding not 'glaringly improbable' or 'contrary to compelling inferences'

I am far from satisfied that the trial judge's finding as to the deceased's speed was 'glaringly improbable' or 'contrary to compelling inferences'. In my view, it was well open to the trial judge to accept Mr Maddison's estimate as to the speed at which the deceased was travelling.
The appellant submits that the trial judge's finding as to the deceased's speed was not a credibility based finding and that this court should draw its own inferences and conclusions. I do not accept that submission. In my view, this is a case where the trial judge's factual findings 'are likely to have been affected by impressions about the credibility and reliability of witnesses formed by the trial judge as a result of seeing and hearing them give their evidence'.
In particular, the trial judge's finding is likely to have been affected by his impression of Mr Maddison's reliability gained from seeing and hearing him give evidence. Mr Maddison's tone and tempo (the degree or lack of hesitation in his answers), facial expression and body language would be capable of influencing the trial judge's assessment of his answers about speed, distance and time. This is not just a question of looking to the witnesses 'demeanour' to assess whether the witness is telling the truth (a process that is fraught with difficulty[126]). Ordinary spoken human communication proceeds from a combination of tone, tempo, gesture and facial expression. The judge in seeing and hearing the witness give oral evidence is able to perceive these aspects of the communication which the transcript cannot reveal. As Isaccs J observed in Dearman v Dearman:[127]

The mere words used by the witnesses when they appear in cold type may have a very different meaning and effect from that which they have when spoken in the witness box. A look, a gesture, a tone or emphasis, a hesitation or an undue or unusual alacrity in giving evidence, will often lead a Judge to find a signification in words actually used by a witness that cannot be attributed to them as they appear in the mere reproduction in type. And therefore some of the material, and it may be, according to the nature of the particular case, some of the most important material, unrecorded material but yet most valuable in helping the Judge very materially in coming to his decision, is utterly beyond the reach of the Court of Appeal.

More recently, this court recognised that there may be subtleties in the way questions were asked (or avoided) that are apparent in the heat of battle but which are not quite as clear in a more clinical examination of a transcript.[128]
For these reasons, a judge seeing and hearing the witness give evidence is in a much better position than an appellate court to assess the meaning and effect of the evidence, and the degree of confidence with which the witness expressed those views. That is so even if the judge was not fully conscious of the matters just adverted to, which are not ordinarily necessary to spell out in the court's reasons.
Further, in the present case it is relevant that the court's audio recording equipment was malfunctioning at the time Mr Maddison's evidence was taken. Although Mr Maddison's answers were picked up by the microphones, some of the respondent's counsel's questions were unable to be transcribed. For example, the transcript does not enable this court to ascertain the proposition put by counsel, with which Mr Maddison agreed, in the third paragraph of the passage quoted at [113] above.
The trial judge's conclusion as to the deceased's speed was not merely an inference drawn from facts which were uncontroversial or which had been found by the trial judge. If that were the case then this court would have been in as good a position as the trial judge to decide on the proper inference to be drawn.[129] Rather, the trial judge's finding was based principally on his acceptance of the reliability of the observations and estimation of an eye witness to the accident. The trial judge had a significant advantage over this court as a result of having seen and heard Mr Maddison give evidence, which placed his Honour in a better position than this court to assess the reliability of his estimate of the deceased's speed. I also agree with the observations of Beech and Vaughan JJA at [262] below. In these circumstances, the standard of appellate review identified in Robinson Helicopters applies.

Even if I am incorrect in concluding that the standard of appellate review identified in Robinson Helicopters applied in this case, that error would not affect my disposition of the appeal. If I were to form my own view as to the deceased's speed from the papers, I would give greater weight to the deceased's own estimate of his speed - both as to the probable range of 25 - 30 km/h and that he was travelling quite fast for a pushbike - than the trial judge. Having regard to all of the evidence, including that of Mr Maddison, I would draw the inference that the deceased was probably travelling somewhere between 20 - 30 km/h, most likely in the lower to middle part of that range.
However, the conclusion just expressed falls short of establishing error by the trial judge. It would, in my view, establish no more than that the trial judge made a choice between competing inferences which I may not have been inclined to make but which was open to his Honour. In my view, the trial judge's conclusion as to the deceased's speed did not fall outside the 'range of permissible inferences'[130], and he has not been shown to have erred in reaching that conclusion.
Further, as I explain at [180] - [192] below, even if I were to apply my own view that the deceased was probably travelling somewhere between 20 - 30 km/h, most likely in the lower to middle part of that range, I would not be satisfied that contributory negligence was established. Therefore, even if I were to give effect to my own view as to the deceased's probable speed, it would not result in the appeal being allowed.

The appellant also contends that the trial judge failed to accord procedural fairness to the appellant by departing from the respondent's own case in its written opening submissions that the deceased entered the roundabout at about 25 - 30 km/h. I am not satisfied that there was any denial of procedural fairness. The appellant called a witness who gave an estimate of speed which was not challenged in cross-examination. The appellant ought reasonably to have apprehended that the trial judge might accept the unchallenged evidence of one of its witnesses. The trial judge was not required to give notice to the appellant before doing so. In any event, the closing submissions of the respondent, after Mr Maddison's evidence was given, contended that his estimate should be accepted.[131] The appellant's counsel, who closed after the respondent, had the opportunity to respond to that submission.

For the above reasons, in my view, neither ground 5 nor ground 6 are established.

Ground 6A: Deceased's location when the appellant entered intersection

Ground 6A contends that the trial judge erred in fact in finding that the appellant entered the intersection after the deceased. As explained in particulars and oral submissions, this ground only arises if the court overturns the trial judge's finding that the deceased was probably travelling at 15 km/h and no more than 20 km/h. In that event, the appellant submits that the trial judge's finding that the appellant drove her vehicle into the roundabout after the deceased entered the intersection must be set aside. That is on the basis that the judge's finding was premised upon his finding as to the deceased's speed.[132]
As ground 6 has not been upheld, ground 6A does not arise. In any event, in my view there is no merit in the ground even on the assumption that the trial judge erred in finding that the deceased was probably travelling at 15km/h and no more than 20 km/h.
The finding that the appellant drove into the roundabout after the deceased had entered the roundabout was supported by witnesses' observations of the relative positions of the deceased and the appellant when the appellant started to enter the intersection. The finding was supported by the accounts given by the deceased and Mr Maddison, and was consistent with the appellant's evidence. Ms Staszewski's evidence did not touch on that issue.
The deceased's account in his police statement was that '[w]hen I was well into the roundabout, it became clear that the driver of the car to my left was pulling out'. Mr Maddison said:[133]

[A]s he got to the intersection, I saw him look across at my vehicle and I continued on going. He shot across in front or went across in front of my vehicle. By the time I got level with the intersection of Eric Street, I looked across to my right-hand side to see where he'd gone and I noticed he'd - he'd cut across to the kerb side of the roundabout. And just about that time, I noticed a dark green car on Eric Street travelling east that had just started to enter the - enter the intersection. (emphasis added)

The appellant's evidence was that:[134]

He [the deceased] was to my right, halfway through the roundabout when I first saw him, on the inside of the roundabout.

The appellant's evidence was of more limited significance as she saw the deceased 'within a split second' of when they collided. However, the deceased was, on her account, well within the intersection when she first saw him. The account of both the deceased and Mr Maddison was that the deceased was in the intersection when the appellant's vehicle entered the roundabout.
The trial judge began his consideration of the question of where the deceased was when the appellant entered the roundabout at [156] of the primary decision. He then referred to the above evidence at [157] - [160] before making the finding sought to be impugned by ground 6A at [160] of the primary decision. The reasons indicate that the trial judge's finding was not premised on his earlier finding as to the speed being travelled by the deceased. Rather, it was based on the observations of the position of the bicycle when the appellant's car started to enter the intersection. It is difficult to see how a contrary finding could have been made given the state of the evidence.
Therefore, even if ground 6 had succeeded, ground 6A would not be established.

[T]he learned trial judge erred in law in ruling that counsel for the appellant was not permitted to ask questions of the appellant's eye-witness, Mr Maddison, who had said that he observed the deceased 'coming at speed', as to what it was about the way the deceased was travelling that caused him to describe the deceased's riding in that way, or to adduce similar evidence from the appellant's other eye-witness Ms Staszewski.

To understand this ground, it is necessary to refer to the opening submissions, the course of Mr Maddison's evidence and the submissions made on the objection.
Unfortunately, the recording problems noted above meant that the trial judge's microphone did not appear to have been picking up his Honour's voice. A number of the trial judge's observations, including his ultimate ruling, are transcribed as 'indistinct'. However, the following aspects of the trial appear or can be inferred from the transcript.
When the appellant's counsel opened his case, he anticipated that Mr Maddison would give evidence that the deceased was travelling at a speed which appeared to him to be unsafe.[135] In an exchange between counsel and the trial judge, the trial judge expressed concern as to whether Mr Maddison could be asked to express an opinion in those terms. His Honour made it clear that he was not deciding the matter, and the appellant's counsel said he would take on board what the trial judge had said.[136] Counsel indicated that Mr Maddison would give that evidence, subject to the trial judge's ruling.[137] In relation to Ms Staszewski's evidence, counsel said:[138]

And at the risk of reopening the exchange that your Honour and I have had, will give evidence to the effect that the bicycle was travelling in a manner which appeared to be to - a speed which appeared to be too fast.

At the end of his opening submissions, the appellant's counsel asked the trial judge for some time 'to consider the point that your Honour raised, because I know there's a discussion about that topic in [Cross on Evidence]'.[139]
Mr Maddison's evidence-in-chief about the bicycle's speed began in the following manner:[140]

CUERDEN, MR: Can I ask you to - did you observe anything about the speed at which the cyclist was travelling?---Well, I thought it was too fast to go - - -

LAMPROPOULIS, MR: Well, I object.

THE WITNESS: - - - into the intersection.

LAMPROPOULIS, MR: Your Honour, can we have this - this was the issue that was raised earlier. Can we have this (inaudible)?

HERRON DCJ: (Inaudible) ask the question (inaudible).

LAMPROPOULIS, MR: Well, (inaudible).

CUERDEN, MR: I will, but there is an issue, but - but I might just come back to it.

Are you able, Mr Maddison, without putting it in terms of an opinion as to whether it was too fast, if I can just - can you describe what you saw in terms of the speed of the cyclist?---Fifteen to 20 Ks an hour.

That's your estimate of the speed?---Yes.

Can I ask you to give his - his Honour your description of what you observed as to the way in which the - he was riding the bicycle?---Well, I couldn't be sure if he was pedalling or freewheeling, but he was coming at speed.

(Inaudible) coming at speed?---Yes.

Now, what was it about the way he was travelling that causes you to describe it as travelling at speed?

(emphasis added)

The respondent's objection was then debated in the absence of the witness. The appellant's counsel submitted:[141]

In my submission, the witness is able, is entitled, to give evidence in terms of - he's not confined to giving evidence in terms of an estimate of speed. He's entitled to give evidence in terms of his overall impressions, and that includes whether, in his view, the [deceased] was travelling in a manner that he describes in the circumstances as being too fast or whatever language he chooses to use. (emphasis added)

The trial judge indicated his concern was whether the witness could go beyond estimating speed and say 'it was too fast in all the circumstances'.[142] Later, counsel for the appellant submitted:[143]

All I'm saying is that the witness is not confined to giving - in my submission is not confined to giving his estimate of the speed.

I am entitled to lead from him in my submission a more descriptive account of what he saw and if the manner in which he is best able to describe what he saw is the speed of the vehicle in a comparison to what he as a person at the intersection considered to be a safe speed, in my submission he's entitled to describe it in that way. (emphasis added)

After briefly hearing from counsel for the respondent, the trial judge made his ruling. Unfortunately the ruling is effectively transcribed as 'indistinct'. The appellant's counsel then said:[144]

CUERDEN, MR: And may I take it that that will apply equally to Ms Staszewski so I won't - can I take it as me having been ruled - well, your Honour is ruling against me leading evidence from her in those terms as well?

HERRON DCJ: Yes.

After Mr Maddison returned to the witness box, the appellant's counsel continued his examination-in-chief in the following terms:[145]

CUERDEN, MR: Now, Mr Maddison, I just want to ask you the question again to make sure that I haven't missed anything. I'm not sure that you've given the fullest explanation that you can. Can I ask you, I've asked you about your estimate of the speed and you've given your answer about that.

What I would like you to do, if possible, is not tell his Honour whether you think the cyclist was going too fast or something of that nature. And just do your best to describe in terms of what you saw - and take it nice and slow if you like. There's no rush. You can start from the first thing you saw and (indistinct).

Well, I just want to make sure that the witness can - doesn't - doesn't trespass.

What I want to ask you, Mr Maddison, is just to explain - to describe the - the nature of the - the way in which the cyclist was riding at the time you - from the time you first saw him?---From the time I first saw him, he shot - he shot a look at me as in my car and then cut across in front of me and went to the inside kerb lane. And - and about that time is when I saw the other car just started to enter the intersection. And when - and I saw - and I saw him actually hit the car just in front of the right-hand door on the fender.

On appeal, the appellant's counsel submits that the question:

Now, what was it about the way he was travelling that causes you to describe it as travelling at speed?

was an appropriate and permissible question which the trial judge should have allowed. The appellant's written submissions contend that the trial judge wrongly assumed the evidence would be inadmissible opinion.

In my view, the appellant's written submissions on appeal take the question and the trial judge's ruling out of context. The issue which was being debated in the objection was whether the witness could give evidence to the effect that he regarded the speed at which the deceased was travelling to be unsafe. The trial judge did not assume that the witness would give that evidence. Rather, the appellant's counsel indicated in opening that this was the evidence which the witness would give. The only ruling which, it may be inferred from the transcript, the trial judge made was that the witness could not express his own view as to the safety or otherwise of the speed being travelled by the deceased. That is confirmed by the questions which counsel asked Mr Maddison after the ruling was made on the objection. He asked Mr Maddison to describe what he saw without saying whether he thought the cyclist was travelling too fast.
In oral submissions, senior counsel for the appellant accepted that the substance of the trial judge's ruling was that he could not adduce evidence from Mr Maddison or Ms Staszewski that they regarded the deceased's speed to be 'too fast' or 'unsafe'.[146]
In my view, the trial judge was correct to rule that the witnesses could not give evidence as to their own subjective views as to whether the speed at which the deceased was travelling was safe. The trial judge's ruling otherwise did not prevent the witnesses from being asked to describe, in any terms they chose apart from by reference to safeness, what they saw or from making an estimate as to the speed at which the deceased was travelling. That evidence did not need to be given in terms of a number of km/h, as illustrated by Ms Staszewski being permitted to give evidence that the deceased was travelling 'quite swiftly' even though she could not estimate the speed.[147]
Counsel was not able to point to any authority in which evidence of a lay witness' subjective opinion as to the safety of a given speed was held to be admissible.[148]
The closest to which the appellant could come was the dissenting judgment of Basten JA in Ayre v Swan.[149] In that case, Ms Ayre sought to make a right-hand turn into a commercial driveway. There was oncoming traffic, relevantly a car driven by a Mr Smith and a motorcycle driven by Mr Swan. As Ms Ayre was undertaking her turn, Mr Swan increased his speed beyond the speed limit, passing Mr Smith on his inside. Mr Swan hit the back passenger side of Ms Ayre's vehicle and suffered injuries. Counsel relied on the following passage of Basten JA's judgment:[150]

The trial judge dealt with Mr Smith's evidence in the following passage:

Mr Smith gave an opinion that the defendant had sufficient time to complete her turn. I place little weight on that opinion having regard to the fact that it is a lay opinion not based on a qualified estimate as to speeds and based upon his own perspective at a time when he was unaware of the precise location of the plaintiff's motorcycle.

This analysis appears to have assumed that Mr Smith was expressing an opinion as to whether [Ms Ayre] could have safely turned in front of a speeding motorcycle; rather, it is tolerably clear that he was expressing an opinion that [Ms Ayre] had time to complete her turn in front of vehicles travelling, as he was, at the speed limit. There was no reason to reject his evidence in that respect: he was not giving expert evidence, but recounting an assessment he made as a driver involved in making decisions with respect to a turning vehicle at the critical time. (emphasis added)

This passage does not assist the appellant. The italicised portion indicates that the non-expert opinion evidence would have been inadmissible if it had merely made a value judgement as to the safety of Ms Ayre's conduct. The evidence which the appellant unsuccessfully sought to adduce in the present case concerns the making of such a value judgement as to the safety of the deceased's speed.
The first question is whether the evidence which the appellant sought to adduce in the present case was relevant. It will not be admissible unless it is relevant to a fact in issue at the trial.
As was noted in Smith v The Queen,[151] in determining relevance it is fundamentally important to identify what are the issues at trial, and whether the evidence, if it were accepted, could rationally affect the court's assessment of that issue.
One fact in issue in the present case was the speed at which the deceased was travelling. The evidence of Mr Maddison and Ms Staszewski as to their estimation of that speed, even if given in general terms such that he was travelling 'at speed' or 'quite swiftly', was rationally probative of that question. However, the fact that the witness considered the speed travelled by the deceased to be unsafe did not rationally affect the probability that he was travelling at any particular speed. One reason for this is that a witness' opinion as to the safety of a speed is based on unstated and unknown premises related to the witness' subjective conceptions of safety in relation to speed.
Another fact in issue in the present case was whether the deceased failed to take precautions that a reasonable person in his position would have taken against a risk of harm, including by travelling at a speed that was excessive in all the circumstances. That involved an objective question as to the extent to which a reasonable person in the deceased's position would have limited his or her speed in entering the intersection in the circumstances. Non-expert opinion as to whether that legal standard has been met, or whether a legally irrelevant standard such as 'safe' by the standards of the witness was met, is properly excluded on the ground that it is not relevant.
The evidence as to whether Mr Maddison and Ms Staszewski thought the deceased was travelling 'too fast', or at a speed that was 'unsafe', was properly excluded. The subjective value-judgement opinions of Mr Maddison and Ms Staszewski as to those matters could not assist the trial judge in identifying the precautions which a reasonable person in the deceased's position would have taken, or in determining the speed at which the deceased was travelling.
In my view, ground 7 is not established.

Ground 8: Finding as to contributory negligence

[T]he learned trial judge erred in law and in fact in finding that the deceased was not contributorily negligent, and he ought to have found that the deceased was contributorily negligent on the basis that he rode through the intersection at excessive speed and/or without having proper regard to the appellant's vehicle.

Section 5B of the Civil Liability Act 2002 (WA) (CLA) provides:

(1) A person is not liable for harm caused by that person's fault in failing to take precautions against a risk of harm unless -

(a) the risk was foreseeable (that is, it is a risk of which the person knew or ought to have known); and

(b) the risk was not insignificant; and

(c) in the circumstances, a reasonable person in the person's position would have taken those precautions.

(2) In determining whether a reasonable person would have taken precautions against a risk of harm, the court is to consider the following (amongst other relevant things) -

(a) the probability that the harm would occur if care were not taken;

(b) the likely seriousness of the harm;

(c) the burden of taking precautions to avoid the risk of harm;

(d) the social utility of the activity that creates the risk of harm.

Section 5K of the CLA provides:

(1) The principles that are applicable in determining whether a person is liable for harm caused by the fault of the person also apply in determining whether the person who suffered harm has been contributorily negligent in failing to take precautions against the risk of that harm.

(2) For that purpose -

(a) the standard of care required of the person who suffered harm is that of a reasonable person in the position of that person; and

(b) the matter is to be determined on the basis of what that person knew or ought to have known at the time.

That is, the standard of care is the same for contributory negligence as for negligence itself.[152]
In the present case, there is no issue that the risk of harm to the deceased in a collision with a motor vehicle was foreseeable and not insignificant, there was a high probability of death or very serious injury occurring if care were not taken. In that context, the critical question in assessing contributory negligence is what precautions a reasonable person in the position of the deceased would have taken, which the deceased failed to take.

The appellant's written submissions contend that the finding that the deceased was not contributorily negligent was in error even on his Honour's finding that he was travelling at 15 - 20 km/h. However, in oral submissions, senior counsel for the appellant accepted that the success of ground 8 depended on the success of ground 6.[153] That is, the appellant ultimately accepted that the trial judge's finding as to contributory negligence cannot be impugned unless she succeeds in overturning the finding that the deceased was probably travelling at 15 km/h and no more than 20 km/h.
The appellant notes that the trial judge's conclusion that the deceased was not contributorily negligent was expressly premised upon his finding that the deceased was riding at 15 - 20 km/h. If that finding were to be set aside, then, she submits, the basis for the trial judge's conclusion as to contributory negligence would fall away.
The appellant also points to that aspect of her case which contended that the deceased failed to pay proper regard to the appellant's vehicle. The appellant contends that the deceased was paying attention to Mr Maddison's vehicle to his right to the exclusion of the appellant's car, and that the accident was a result of a combination of the deceased's failure to keep a proper lookout and his excessive speed.[154]
As to the question of 'excessive speed', senior counsel for the appellant was unable to identify a speed which the appellant contended would be safe.[155] That is, counsel could not identify a speed or range of speeds at which a reasonable person in the deceased's position would have entered the roundabout, taking care for his or her own safety.[156] However, counsel contended that contributory negligence could be established without it being necessary to identify a speed which would not have been excessive. In written submissions, the appellant contends:[157]

As said above, at whatever precise speed the deceased was travelling he was in fact unable to avoid a collision with a slow-moving vehicle which had moved 7- 12m into the roundabout to the point of collision. As the respondent's senior counsel acknowledged, he should have been able to do so.

In oral submissions, counsel for the appellant contended that a reasonable person in the deceased's position, in going through the roundabout, would pay attention to everything that was going on around him and stand ready for the car at the give way sign to come out at any time.[158] The appellant submitted that the deceased's contributory negligence was in the combination of speed and a failure to keep a proper lookout, and those two elements needed to be considered in light of each other.[159]

As the trial judge's finding as to speed stands, ground 8 cannot succeed, as the appellant now accepts.[160] That concession was, in my view, properly made. In my view, on the facts found by the trial judge, his Honour's conclusion that the deceased was not contributorily negligent was correct, for the reasons which his Honour gave. That is sufficient to dispose of ground 8.
Even if I were to apply my own view as to the speed at which the deceased was probably travelling, I would not conclude that the deceased was contributorily negligent.
The critical facts were that the appellant slowed her vehicle almost to a stop at the give way sign and entered the intersection only when the deceased was already in the roundabout.
It may readily be accepted that, as was observed by the High Court in Sibley v Kais:[161]

The failure to take reasonable care in given circumstances is not necessarily answered by reliance upon the expected performance by the driver of the give way vehicle of his obligations under the regulations; for there is no general rule that in all circumstances a driver can rely upon the performance by others of their duties, whether derived from statutory sources or from the common law. Whether or not in particular circumstances it is reasonable to act upon the assumption that another will act in some particular way, as for example by performing his duty under a regulation, must remain a question of fact to be judged in all the particular circumstances of the case.

I also take into account that, as the appellant emphasises, being on a bicycle means that, in the event of a collision with a car, more serious injury was reasonably foreseeable by a person in the deceased's position than by a driver of a car.[162]
However, in the particular circumstances facing the deceased, it was reasonable for him to assume that the appellant had stopped and was giving way to him, as she was required to do. To enter and travel through the roundabout at up to 30 km/h would not be regarded as unsafe by a reasonable person in those circumstances. A reasonable person in the deceased's position would not, in my view, find it necessary or appropriate to pass by a vehicle apparently obeying a give way sign so slowly as to allow sufficient time to stop if, unexpectedly, the vehicle were suddenly to contravene the give way sign. Having regard to the social utility of the road system depending upon travellers being able to proceed at a reasonable rate, and the burden involved in so reducing the rate of travel on road, a reasonable person would not find it necessary to travel at less than 30 km/h when entering and passing through an intersection at which the vehicle required to give way was apparently doing so.
Further, the inability of the appellant to identify the speed which a reasonable person in the deceased's position would have regarded as safe presents a fundamental difficulty for her case. To say that the party was negligent by travelling at excessive speed requires the court to be satisfied that:

(1) The party was travelling at a certain speed, or within a certain range of speeds; and

(2) The speed, or range of speeds, at which the party travelled was greater than that to which a person in the party's position, taking reasonable care for his or her own safety, would have limited himself or herself.

In some circumstances, speed may be so excessive as to be so clearly above that which would be travelled by any reasonable person that it is unnecessary to make any finding as to the particular speed to which a reasonable person would limit himself or herself. However, a speed in the range of 20 - 30 km/h in the circumstances of the present case is not in that category. The incapacity of the appellant to identify a speed at which a reasonable person in the deceased's position would have travelled reflects the insufficiency of the evidence to prove that the deceased failed to take a precaution which would have been taken by a reasonable person in his position.
Nor, in my view, does the evidence establish a failure by the deceased to pay proper regard to the appellant's vehicle. The deceased had, according to his police statement, seen the appellant's car going very slowly or stopped at the intersection. He reasonably assumed she was giving way to him. In those circumstances, it was reasonable for the deceased's attention to be focussed on Mr Maddison's vehicle to his right, to which the deceased was required to give way. There was no other evidence that the deceased failed to keep a proper lookout for the appellant's vehicle.
The evidence also does not support a finding that the deceased focussed his attention on Mr Maddison's vehicle to the exclusion of the appellant's car. The deceased's statement to police is that he was watching cars to his right and did not pay 'particular attention' to the deceased's vehicle. That is not to say that he was completely ignoring the appellant's car. Further, the deceased said that, when he was well into the roundabout, 'it became clear that [the appellant's car] was pulling out'. That is, the deceased saw the appellant pull out as he was well into the roundabout, but did not have the opportunity to avoid a collision.
It may be accepted that the trial judge found that the appellant came into the intersection slowly. However, it would take very little time for a car moving from the entrance to a roundabout, even if traveling slowly for a car, to block the way of a cyclist passing through the intersection, leaving the cyclist with insufficient time to react and take successful evasive action.
In these circumstances, it is not possible to infer negligence by excessive speed and a failure to keep a proper lookout from the fact of the collision. It is not possible in all the circumstances of the present case to find contributory negligence on the basis that the collision would not have happened if the deceased had not been negligent.
Moreover, breach of duty of care is to be determined prospectively, both at common law and under cases to which s 5B of the Civil Liability Act applies.[163] Under s 5K(1), extracted at [171] above, the principles applicable to determining liability for negligence also apply to contributory negligence. It follows that contributory negligence is similarly to be assessed prospectively and not with the benefit of[164]ndsight.164 Section 5K(2)(b) also expressly requires that contributory negligence is to be determined by what the person harmed 'knew or ought to have known at the time'. To reason from the fact of the accident, and the deceased's inability to prevent that accident, to a conclusion that the deceased failed to take reasonable care tends to involve impermissible hindsight reasoning.
In my view, ground 8 is not established, either on the primary facts found by the trial judge or on the assumption that the deceased was travelling at up to 30 km/h as he passed through the intersection in the circumstances otherwise found by the trial judge.

Ground 9 contends that the trial judge erred in finding that any contributory negligence of the deceased did not cause the collision. It is unnecessary for this court to determine that ground.
The question of causation would only arise if it were established that the deceased failed to take reasonable care for his own safety as he passed through the intersection. If negligence consisted of travelling at excessive speed then a question of factual causation would arise,[165] namely whether the harm would have occurred if the deceased was not travelling at a higher speed than a reasonable person in his position would have travelled. (It may be noted that the appellant's inability to nominate a speed at which a reasonable person in the deceased's position would have travelled would also present a difficulty for her case on causation). The question of causation is not usefully analysed in the absence of any finding as to conduct which constitutes contributory negligence.

For the above reasons, in my view none of the grounds of appeal are established, and the appeal should be dismissed.

Agreement with and adoption of part of Mitchell JA's reasons

We have the considerable benefit of having read a draft of Mitchell JA's reasons for decision.
Mitchell JA has explained the circumstances of the litigation ([1] - [6] above) and the facts as found by, and conclusions of, the primary judge ([7] - [63] above). There is no need for us to repeat those matters. We adopt what has been said by Mitchell JA. Save in one respect we also agree with, and adopt, Mitchell JA's reasons for dismissing grounds 1 to 5.
The exception is that, in rejecting grounds 3, 4 and 5, we do not consider it necessary to conclude that it is inherently unlikely that a 63‑year‑old occasional recreational cyclist such as the deceased would

have been capable of reaching a speed of 50 km per hour.[166] This is not to say that we consider this conclusion to be incorrect. At the hearing of the appeal senior counsel for the appellant did not cavil with the proposition that one might say 50 km per hour might be difficult to achieve[167] after having earlier seemingly accepted that 50 km per hour was an improbably high speed for a 63‑year‑old cyclist riding a pushbike.[168]

We simply consider it is not necessary to reach a conclusion on this point. We would dismiss grounds 3, 4 and 5 for the remainder of the reasons that Mitchell JA has given.
So too we would dismiss ground 7 for the reasons given by Mitchell JA.

Ground 6: Challenge to finding that travelling at approximately 15 km per hour

We wish, however, to develop our own reasons in relation to ground 6. Ground 6 provides:

[T]he learned trial judge erred in fact in finding that the deceased was travelling at a speed 15km/hr [sic] and no more than 20 km/hr, in that the finding was against the weight of the evidence, was not a permissible inference and was unreasonable.

The appellant's case for contributory negligence was not limited to speed. The allegation of contributory negligence was a combination of speed and whether the deceased had failed to keep a proper lookout. Those two things fell to be considered together. But ground 6 was directed solely to the primary judge's finding at [141] of his Honour's reasons that when entering the roundabout the deceased was travelling at a speed of approximately 15 km per hour and no more than 20 km per hour (which his Honour went on to find was, in all of the circumstances, a safe speed to be riding his bicycle when the deceased approached and entered the roundabout).[169]
The primary judge seemingly accepted that this speed (ie a speed of approximately 15 km per hour and not more than 20 km per hour) was 'riding quickly for [the deceased's] pushbike'.[170]

The nature of the appellant's challenge to the factual finding

In challenging the factual finding that, in entering the roundabout, the deceased was probably travelling at a speed of approximately 15 km per hour, and no more than 20 km per hour, the appellant went further than saying that the finding was against the weight of the evidence. The appellant submitted that the primary judge's conclusion was 'not open' and the reasoning for it was 'captious'.[171] The appellant argued that she had adduced a body of evidence to the effect that the deceased was travelling faster - at least 25 - 30 km per hour and up to 50 km per hour - and the primary judge was not entitled to ignore it.[172]
The appellant accepted, correctly, that she must demonstrate that the primary judge's factual finding was wrong, not merely that an alternative finding was open. It was said, however, that the primary judge's findings were not credibility based and that the principles of appellate restraint applicable to the review of such findings did not apply.[173] We accept that the factual finding about the deceased's speed of travel as is the subject of ground 6 was not credibility based. But, as Mitchell JA observes at [105] and [127] above, the principles of appellate restraint adverted to by the appellant do not only apply to credibility based findings. Appellate restraint with respect to interference with factual findings applies when a finding is likely to have been affected by impressions about the credibility and reliability of witnesses formed by the trial judge as a result of seeing and hearing the witnesse[174]ive their evidence.174
Lee v Lee has not changed anything in this respect. As Allsop CJ explained in Aldi Foods Pty Ltd v Moroccanoil Israel Ltd,[175] the High Court's references to Fox v Percy[176] in Robinson Helicopter Co Inc v McDermott[177] are 'plainly to findings of fact reached after assessing competing witnesses for their reliability and credibility'.[178]
In a case where assessment of witness reliability attracts the same sort of issues for, and approach to, fact finding as that of witness credibility, an appellate court will not interfere with the factual finding unless it is demonstrated to be wrong:[179]
- by reference to incontrovertible facts or uncontested testimony;
- because the finding is glaringly improbable or contrary to compelling inferences; or
- because the trial judge failed to use, or has palpably misused, his or her advantage as trial judge.
Not all factual findings are reviewed on appeal by reference to these standards. The nature and extent of the trial judge's advantage informs what is required in order to reach, and the extent of appellate restraint in reaching, a conclusion of error. It is necessary, for example, to distinguish between a finding based on a credibility determination and a finding based on inferences drawn from undisputed facts or facts as found. Thus the nature of the finding, and the reasoning by which it was made, affects what is required in order to demonstrate appellable error.[180]
The basis for appellate restraint is grounded in the 'natural limitations' of the appellate process[181] and the degree of comparative advantage possessed by a trial judge based on his or her position as trier of fact at first instance[182] - an advantage to which 'due regard' must be had.[183] Those advantages are many and varied.[184] The specific advantages can vary from case to case. For example, in Abalos v Australian Postal Commission McHugh J thought it to be of some significance that the trial judge in that case had enjoyed the advantage of some in-court demonstrations as well as seeing and hearing the witnesses.[185] There can also be an advantage in having had a view.[186] This court has referred to possible subtleties in the way questions were asked or avoided that may be apparent to the trial judge but not quite as clear in a more clinical examination of a transcript.[187]
The trial judge's comparative advantage - also referred to as his or her 'relative' advantage[188] or the extent of his or her 'perceived advantage'[189] - depends not only on the particular evidence received by the trial judge, and the circumstances in which it was received, but on the nature of the findings. Outside the area of credibility based findings it may be the case that:

The advantages of the trial judge may be more subtle and imprecise, yet real, not giving rise to a protection of the nature accorded credibility findings, but, nevertheless, being highly relevant to the assessment of the weight to be accorded the views of the trial judge.[190]

So, where a finding concerns a matter of impression or judgment, a trial judge's comparative advantage in coming to that evaluative determination should be given 'proper weight'.[191]
The finding the subject of ground 6 answered a question of objective fact: what speed was the deceased travelling at when he entered the roundabout? As will be seen, however, the primary judge's conclusion was informed by his assessment of the relative reliability of one eyewitness' estimate of speed as against the deceased's estimate. Moreover, absent some mechanical means of precise measurement (none being available in the present case), the assessment of speed involves a matter of impression or judgment. So far as this was the position for the witness it was no less so for the primary judge. These two matters must be taken into account in considering the primary judge's advantage over this court and the consequence that has for this court's examination of the evidence and review of the primary judge's reasons.
That said, the statutory requirement that this appeal proceed by way of rehearing necessitates a real review. The court cannot excuse itself from the task of weighing conflicting evidence and drawing its own inferences and conclusions.[192] In those circumstances, determining the proper weight to be afforded to the primary judge's advantage is best done after considering the evidence and the primary judge's reasons. We therefore turn to those matters.

In challenging the primary judge's finding the appellant acknowledged Mr Maddison's evidence - elicited by the appellant herself in evidence‑in‑chief - that he estimated the deceased was travelling at 15 to 20 km per hour.[193] To counter that the appellant referred to the following categories of evidence which were said to be against the finding:[194]
1. The medical records insofar as they recorded references to a speed of 50 km per hour.
2. Various statements on the part of or adopted by the deceased - which included the deceased's own estimate of the speed at which he was travelling.
3. Mr Maddison's other evidence (especially so far as it was said to be inconsistent with Mr Maddison's estimate of speed) as well as that of Ms Staszewski.
4. An alleged admission - in the form of the respondent's written opening submissions for trial[195] - that the deceased entered the roundabout at about 25 - 30 km per hour.
The appellant also relied on the fact that the deceased was unable to stop to avoid the collision with the appellant's car.[196] According to the appellant this was evidence that the deceased was travelling at an excessive speed.[197] In that regard the appellant relied on her evidence, as accepted by the primary judge, that she moved into the intersection slowly, at about 5 to 10 km per hour.[198]
The primary judge was not satisfied that the medical records were a reliable record that the deceased said he was riding at 50 km per hour or - if the deceased did state such a belief - that the 50 km per hour was a reliable estimation.[199] The challenge on appeal to those findings (express in grounds 3 and 4 and implicit in grounds 1 and 2) fails for the reasons given by Mitchell JA. Having regard to the way in which the appellant's case was presented, the medical record evidence also fell to be considered in coming to a conclusion as to whether the deceased was travelling at an excessive or unsafe speed based on the deceased's statements as to his speed. However, given the conclusions that have been reached as to the medical records' evidence, that material is an insufficient basis, alone or in combination with the other matters on which the appellant relies, to impugn the primary judge's finding as to speed.
More significantly, the deceased was found to have made or adopted a number of statements. These included:
1. The oral statement on 12 May 2013 to Senior Constable Patten at Sir Charles Gairdner Hospital (SCGH) that he 'was going fast for a bike' (a statement made a matter of hours after the collision).[200]
2. The statements in the unsigned police statement that:[201]
  1. At this time [ie approaching the intersection] I guess I was travelling around 25 - 30 kph.

The vehicles travelling from my right had priority over myself when entering the roundabout, so I was watching them and trying to moderate my speed so that I didn't have to stop.

I entered the roundabout after the cars had passed in front of me.
I was probably going quite quickly for a pushbike, somewhere between 25 - 30kph.
When I was well into the roundabout, it became clear that the driver of the car to my left was pulling out.
I then had to brake very hard to try and slow down.
I turned my handlebars in an effort to steer around the car, but it was not effective.
I hit the side of the car and the next thing I remember is landing on the road. (emphasis added)
The police statement was taken from the deceased on 15 May 2013 while he remained an inpatient at SCGH.[202] The statement was, in substance, adopted by the deceased in an email sent on 21 May 2013.[203] The primary judge found that the deceased's estimations made on 15 May 2013 and confirmed on 21 May 2013 were more reliable estimations of his speed than any estimation the deceased may have given at an earlier time in the emergency department or the intensive care unit.[204] Insofar as ground 4 challenges that finding, the challenge fails, for the reasons given by Mitchell JA.
The so-called 'admission' in the respondent's written opening submissions for trial was plainly based on the deceased's unsigned police statement. In terms of evidentiary value the 'admission' could rise no higher than the statement. Accordingly, this aspect of the evidence relied on by the appellant requires no additional consideration over and above consideration of the statements in the unsigned police statement.
Mr Maddison's evidence in estimating the deceased's speed was given in unqualified terms. His estimate of the deceased's speed was that the deceased was traveling at 15 to 20 km per hour.[205] (By contrast, when asked about the appellant's speed, Mr Maddison put it in terms of 'very slowly',[206] and when pressed to put some context to that answer said: 'I couldn't put a speed to her cos I don't know.)[207] Mr Maddison described the deceased as having 'shot across' in front of Mr Maddison's vehicle[208] and as 'coming at speed'.[209] Mr Maddison did not notice any change in the speed at which the deceased was travelling from the point at which he first saw the deceased.[210]
On appeal the appellant emphasised other evidence Mr Maddison gave as to time and distance.
Mr Maddison estimated that two to three seconds passed from the time he first saw the deceased to the collision.[211] However, this evidence was expressed in qualified terms. Mr Maddison first said it was two to three seconds 'maybe'.[212] In cross‑examination the time lapse estimate was put in terms of 'that's what it appeared to be to me.'[213] In terms of distance Mr Maddison said that he first saw the deceased 'probably five' metres back from the give way line before the intersection[214] and that he observed the deceased for 'probably another 20 metres' - expressed as being 'on the outside' - before the collision with the appellant.[215] However, in cross‑examination Mr Maddison expressed the latter figure as being '[t]wenty, 25'.[216] Mr Maddison also accepted that these were 'all rough figures'.[217]
Before the primary judge[218] - and then on appeal[219] - senior counsel for the appellant used Mr Maddison's estimates of time and distance to come to an inferred or calculated speed and sought to compare that inferred or calculated speed with Mr Maddison's estimate of speed. As a matter of mathematics the deceased's speed was 30 km per hour if he travelled 25 metres in three seconds or 45 km per hour if he travelled 25 metres in two seconds. (The appellant did not do the maths on 30 metres - the relevant distance under one aspect of the answer in cross‑examination - but then the speed became 36 km per hour to 54 km per hour.)
Ms Staszewski's evidence was more limited than Mr Maddison's. Ms Staszewski said that the deceased was travelling 'quite swiftly' but was unable to estimate his speed.[220] Ms Staszewski was initially unable to say whether, when she first observed the deceased, he had reached the roundabout.[221] When pressed, however, Ms Staszewski estimated that the deceased was before the roundabout by a distance of one and a half times the width of the courtroom (which the parties agreed meant a distance of nine metres).[222] Ms Staszewski said that the time between first seeing the deceased and the collision was three seconds.[223]
Senior counsel for the appellant extrapolated from the nine metres to come to a total distance travelled to the collision of 29 metres, over the three seconds, to come to an inferred speed based on Ms Staszewski's evidence of around 35 km per hour.[224]

The primary judge found that people's estimations of speed and time are notoriously unreliable.[225] No party challenges that finding on appeal. His Honour also accepted that Mr Maddison only had a brief period of time in which to make his assessment as to the deceased's speed.[226] That was undoubtedly correct. Mr Maddison's evidence was that he observed the deceased - from first sighting to the collision - over about two to three seconds.[227] And, although Mr Maddison was proceeding in a westerly direction (thus being in a position to observe the deceased moving from his, Mr Maddison's, left to right as the deceased moved in a northerly direction) Mr Maddison's evidence was that the deceased 'shot across' in front of Mr Maddison's vehicle.[228] Nevertheless, Mr Maddison said that he had a clear line of sight[229] and visibility was clear.[230]
The primary judge held that Mr Maddison was in the best position to make the most reliable estimation of the speed at which the deceased was riding.[231] While not the subject of a ground of appeal, the appellant's submissions, both written and oral, sought to challenge that finding.[232] Indeed, the appellant went so far as to say that there was 'no rational reason' to conclude that Mr Maddison was better placed than the deceased to best estimate the deceased's speed.[233]
There is no substance in this complaint. There were only two estimates before the primary judge. The appellant did not see the deceased until a 'split second' before the collision[234] and Ms Staszewski (the passenger in Mr Maddison's car) was unable to give an estimate of the deceased's speed.[235] The only estimates before the primary judge were those of the deceased and Mr Maddison.
While the primary judge accepted that the deceased's estimates in the unsigned police statement were a more reliable estimate than any estimate he had given in the immediate aftermath of the accident, it was still, in our view, well open for his Honour to conclude that Mr Maddison was in the best position to make the most reliable estimate of the deceased's speed. Mr Maddison had a clear view of the deceased as the deceased travelled in a direction perpendicular to Mr Maddison's line of view. He was an independent observer, called by the appellant, who was not personally involved in the collision. By contrast the deceased had been involved in a serious accident and suffered traumatic injuries. That alone provided ample basis for the primary judge's finding that the deceased's memory of the circumstances leading up to the accident was not necessarily that reliable.[236] Further, it was well open to the judge to take the view, as his Honour evidently did, that a person seeing a cyclist travelling left to right (or vice versa) across the observer's path is better placed to reliably estimate the cyclist's speed than the person on the moving bicycle.
Two other aspects of the evidence supported the conclusion that Mr Maddison was in the best position to make the most reliable estimate of the deceased's speed:
1. The deceased's estimates were qualified. He referred to travelling around 25 - 30 km per hour 'I guess' and said that he was 'probably' going quite fast for a pushbike.
2. The deceased's recollection - as set out at par 14 of the unsigned police statement - was that he entered the roundabout after cars had passed in front of him. As submitted by the appellant at trial,[237] the deceased's recollection was wrong. Mr Maddison's evidence,[238] as was accepted by the primary judge,[239] was to the contrary. There were no cars that entered from the right in front of the deceased.
The primary judge had additional reasons for concluding that Mr Maddison was in the best position to make the most reliable estimate of the deceased's speed.[240] It is not necessary to examine those further reasons. The matters already referred to demonstrate that, contrary to the appellant's submissions before this court, there were rational reasons to conclude that Mr Maddison was in the best position to assess the speed that the deceased was travelling and it was reasonably open to the primary judge to so conclude. There was no error in this finding.
Having found that Mr Maddison was in the best position to assess the speed at which the deceased was travelling, the primary judge accepted Mr Maddison's evidence.[241] In so doing his Honour observed that Mr Maddison's upper speed of 20 km per hour was not dissimilar to the deceased's lower estimation of speed of 25 km per hour.[242] The primary judge also made the observation, as is the subject of ground 5, which should be understood in the sense that a speed of 15 km per hour was one that could realistically be achieved by a 63‑year‑old recreational cyclist such as the deceased.[243]
Mr Maddison's estimate was that the deceased's speed was 15 - 20 km per hour.[244] However, the primary judge's finding was that the deceased was 'probably travelling at a speed of approximately 15 km per hour ... and no more than 20 km per hour'.[245] His Honour fixed on the lower speed rather than the range. No reasons were given for that apparent incongruity - one arising insofar as his Honour had earlier accepted Mr Maddison's evidence.[246]
There are four other aspects of the primary judge's reasoning that should be mentioned.
First, as formed one of the central themes of the appellant's submissions on appeal, the primary judge did not address the apparent inconsistency between Mr Maddison's evidence as to estimate of speed and his (and Ms Staszewski's) estimates of time and distance leading to a higher inferred speed. His Honour did not seek to reconcile that evidence. We will return to this point later in these reasons.
Second, the primary judge said there was no evidence of the incline of Broome Street towards the roundabout along which the deceased was riding.[247] That is not quite correct. Photographs were tendered. These showed that in travelling in a northerly direction along Broome Street towards the intersection the deceased had the benefit of a slight downward slope.[248] However, the photographs are insufficient to form a considered view as to the degree of the downward incline. Nor was there any evidence, expert or otherwise, as to the likely consequences that the downward incline as evidenced by the photographs would have on the speed at which a cyclist such as the deceased might be able to travel.
Third, the deceased's unsigned police statement also referred to the deceased as having a 'light wind' behind him.[249] The primary judge referred to this in the sense of repeating the contents of the statement.[250] No other mention was made of that particular aspect of the evidence.
Fourth, at [129] ‑ [131] of the primary reasons his Honour considered what was meant by the deceased's description, at par 12 of his unsigned statement, that the deceased was 'trying to moderate' his speed so that he did not have to stop. His Honour concluded that the reference to moderating speed meant to adjust speed 'by reducing his speed'.[251] There is no difficulty if that passage is understood simply as the primary judge's interpretation of the passage in the deceased's statement. But it would be otherwise if it were a finding that in actual fact the deceased had slowed down. The uncontradicted evidence was that the deceased did not change speed.[252] Moreover, there was no need for the deceased to do so to deal with traffic to the right as there was no car before Mr Maddison's car.
When the primary judge's reasons are read as a whole, the passage at [131] should be understood in the first sense rather than as suggesting that his Honour was making a finding that the deceased had slowed down. Nowhere else is there any suggestion that the deceased had in fact slowed down to enter the roundabout. Nevertheless, the apparent error in par 12 of the deceased's unsigned police statement is a further indication that there are questions as to the reliability of the deceased's recollection of the circumstances of the accident.
On appeal the appellant did not seek to make anything of the primary judge's incorrect observation that there was no evidence as to the incline of Broome Street. The appellant was correct not to do so. Given the lack of specific evidence before the primary judge as to the degree of incline, and the possible effect on the deceased's speed, this court would have been in error to overturn the primary judge's factual finding by mere extrapolation from the photographs.[253] So too nothing of substance can be drawn from the additional evidence, not mentioned by the primary judge in coming to his conclusion as to speed, that the deceased had a light wind behind him.

In our opinion, a trial judge will generally have a significant advantage over an appellate court in evaluating the reliability of a witness' estimate of speed.
In this context the question of reliability is not merely one of the reliability of the witness. The issue is one of the reliability of the evidence itself - something that partially, but by no means exclusively, depends on the reliability of the witness. An estimate of speed involves an intuitive synthesis where the observer's reasons cannot always be fully communicated. Estimates of speed and distance are one of those established exceptions where a non-expert witness can give 'inferential evidence'.[254] So too estimates of time. Such evidence is admissible despite having the appearance of opinion evidence because it is evidence of direct observation which can only be described in the language of inference or conclusion. As explained in Sherrard v Jacob:

[T]he category in question includes instances in which the primary facts and the inferences to be drawn therefrom are so adherent or closely associated that it may be hard, if not impossible, to separate them. A witness may reach a conclusion on observation without being able to muster or recall, or it may be even to describe, all the facts which have united to produce that conclusion.[255]

In respect of such matters of common experience within the knowledge of ordinary people, the reception of such inferential evidence 'enables evidence to be conveniently summarised or distilled in common form which in practice could not reasonably be called for in all its multitudinous detail'.[256] Thus evidence of this nature is admissible, although involving inference and opinion, when it would be inconvenient, if not impossible, for the observer to separate his or her inference from the facts on which the inference is based.[257]
The primary judge had an advantage over this court in forming a general view as to the reliability of Mr Maddison as a witness in the usual way in which a trial judge has an advantage in evaluating the credibility and reliability of a witness who gives oral evidence. It is accepted that there is an advantage in seeing and hearing a witness. Mitchell JA has explained some of the reasons for this at [128] - [130] above. Seeing and hearing the witnesses also contributes to the trial judge's 'feel' of the case, which an appellate court cannot share from simply reading the transcript.
All the more so, in our opinion, a trial judge has a considerable advantage over an appellate court in evaluating the reliability of a witness' estimate of speed. The witness' evidence itself involves a matter of impression or judgment insofar as the witness estimates a particular speed. The trial judge's consideration of the reliability of the evidence involves him or her evaluating and forming a conclusion as to the soundness of the witness' inferential conclusion having regard to the other evidence (including the witness' observational position and any relevant experience in such estimation) and the witness' general reliability. Where, as here, there are competing estimates and various general statements as to speed, that too must be taken into account. For these reasons, so far as the estimation of speed is a matter of impression or judgment for the witnesses, the trial judge's finding is not simply one of drawing an inference from undisputed facts or primary facts as found.[258]
There were five categories of evidence that the appellant relied on in support of ground 6's complaint that the primary judge's finding as to speed was against the weight of the evidence, constituted an impermissible inference or was unreasonable (see [214] - [215] above).
We have considered the evidence as a whole and, in doing so, have evaluated the combined force of the matters on which the appellant relies. For ease of exposition, we will deal with them individually.
The medical records have already been considered. The more general statements of the deceased - as well of those of Mr Maddison and Ms Stazewski - are inconclusive. The primary judge's upper figure of 20 km per hour is 'going fast' or 'quite quickly' for a pushbike. So too that figure meets the general description of 'coming at speed' or 'quite swiftly'. The statement in the respondent's opening submissions for trial, as said to constitute an admission, goes no higher than what was said in the deceased's unsigned police statement. The deceased's inability to stop and avoid the collision does not provide any sound basis to impugn the primary judge's finding as to speed. There was no evidence, expert or otherwise, to the effect that a recreational cyclist of the approximate size, weight and age of the deceased should have been able to stop his bicycle within two to three seconds if he was travelling at 20 km per hour or slower. Having so addressed the bulk of the countervailing evidence relied on by the appellant, two aspects of the evidence remain for consideration: the deceased's own estimates (25 ‑ 30 km per hour) and the time and distance estimates of Mr Maddison and Ms Stazewski from which senior counsel for the appellant calculated inferred speeds of 30 ‑ 45 km per hour or 35 km per hour.
The primary judge preferred Mr Maddison's eyewitness estimation over that of the deceased, finding it to be more reliable. For the reasons already given, there was good reason to do so. The deceased's estimates were expressed in more equivocal terms and were potentially affected by his involvement in the collision. Also, in two aspects the deceased's recollection of the events of the collision, as set out in the unsigned police statement, were found to be incorrect.
In any event, as the primary judge noted,[259] the upper range of Mr Maddison's estimate (20 km per hour) broadly accorded with the lower range of the deceased's estimate ('around' a range commencing at 25 km per hour or 'somewhere' starting at 25 km per hour).
We turn to the part of the appellant's argument which relied on Mr Maddison's and Ms Stazewski's estimates of time and distance resulting in the substantially higher inferred speeds. The appellant contended that, as far as Mr Maddison was concerned, this resulted in an inconsistency which demonstrated that Mr Maddison's estimate of speed was unreliable.[260]
We accept that the primary judge did not seek to reconcile this evidence, and the inferred speed, with Mr Maddison's estimate of speed. We are not satisfied that this bespeaks error on the part of the primary judge. On a fair reading of the primary reasons as a whole the primary judge was acutely aware of the issue that arose as to the reliability of the evidence before him as to the deceased's speed. His Honour accepted that estimations of speed were notoriously unreliable. So appreciating the inherent unreliability of such evidence, and having carefully considered Mr Maddison's and the deceased's respective reliability, the primary judge attended closely to the issues of reliability presented by the evidence as a whole.
We are, in any case, unable to accept that the evidence of time and distance - leading by calculation to a higher inferred speed - is, alone or in combination with the other matters on which the appellant relies, sufficient reason to reject the primary judge's finding as to speed based on Mr Maddison's direct evidence (a conclusion we reach having regard to the evidence as a whole including the five categories of evidence relied on by the appellant on appeal).
Evidence as to time and distance will often be difficult to state or estimate by direct recollection. The latter may sometimes be dealt with by forensic reconstruction. That was not attempted in the present case. The appellant sought to place some store on the plans that were introduced into evidence.[261] While these contained a scale it was accepted that there was no positive evidence to substantiate that the plan was an accurate depiction.[262] As Mitchell JA's reasons show, the only measured part of the plan was the 12 metres between the commencement of the roundabout (in proceeding in an easterly direction along Eric Street) to the rear of the final stopping position of the appellant's car following the collision (see at [119] - [123] above). The accuracy of the remainder of the plans was not the subject of evidence. The only material evidence as to distance was that elicited from Mr Maddison and Ms Staszewski.
Senior counsel for the appellant made the submission that as a matter of ordinary experience it was easier for witnesses to recount observations as to bare data - such as time and distance - than to make a compound assessment of speed.[263] That submission assumes that an observer's assessment of speed of a vehicle or bicycle is compound in nature. We do not accept that premise. Further, to the extent that attention is to be directed to what might be expected of an ordinary road user, we agree with what Mitchell JA has said at [115] above. To our mind, however, of more significance for the present case than broad generalisations was Mr Maddison's ability to give accurate and reliable estimates of time, distance and speed (and, to a lesser extent, Ms Staszewski's ability to give accurate and reliable estimates of time and distance).
Whether either witness was more capable of making an accurate and reliable estimate of time and distance, as opposed to speed, was not directly explored in the evidence before the primary judge. In the case of Mr Maddison it can be inferred, from the terms in which he expressed his estimates, that, from his perspective, the contention advanced by senior counsel for the appellant does not hold good. Mr Maddison estimated the deceased's speed in unequivocal and unqualified terms.[264] By contrast Mr Maddison's evidence as to time and distance was hedged in terms of 'maybe',[265] 'maybe ... maybe a little bit more. I'm not sure'[266] and a 'rough guess',[267] 'roughly'[268] or 'rough figures'.[269]
It is also relevant that, while Mr Maddison was able to give an estimate as to the deceased's speed in unequivocal and unqualified terms,[270] he expressed himself as being unable to estimate the appellant's speed.[271] A feature of the evidence before the primary judge is that neither party's counsel explored the basis for Mr Maddison's unequivocal and unqualified estimate of speed. No doubt, for the respondent, this was a sound forensic decision. But it was open to the appellant to examine by questions, without infringing the prohibition on cross‑examining one's own witness, the basis for Mr Maddison's inferential conclusion, his degree of confidence in the estimate and what experience, if any, he had in estimating speed. In the absence of any such questioning the apparently discerning evidence of Mr Maddison takes on an added dimension of reliability.
We are not persuaded that the appellant's inferred speed calculations have sufficient weight to sustain appellate interference with the primary judge's finding as to speed based on his Honour's acceptance of Mr Maddison's estimate. The inferred speed calculations suffer from the imprecision in the witness' evidence as to time and distance. The calculations are highly dependent on accuracy as to time. For example, to take Mr Maddison's estimate, let it be assumed that the distance was 25 metres but that the time between first observation and collision was 3.5 seconds (only half a second more than Mr Maddison's two or three seconds 'maybe'). The inferred speed is then in the order of 26 km per hour. At four seconds the inferred speed is 22.5 km per hour. The inferred speed also falls if the distance has been overestimated. At 20 metres and 4 seconds the speed becomes 18 km per hour. The uncertainties in both distance and time are capable of having a compounding effect.
The analysis urged by the appellant involves an unjustified reliance on calculations which depend on variables which are imprecise and which were hedged with qualification by the witnesses. In other contexts the High Court has commented on the difficulties of such calculations.[272] In Pledge v Roads and Traffic Authority the High Court considered the Court of Appeal in New South Wales to be in error in overturning a finding at first instance in reliance on time, speed and distance calculations that involved so many imponderables as to make the calculations little more than speculation. In that case it was said to be not irrelevant that the calculations in question were not sought to be made during the trial where the assumptions on which they were based could have been fully tested and explored.[273] In some respects, the position in the present case is not dissimilar. Neither Mr Maddison nor Ms Staszewski were taken to an inferred speed calculation and asked to consider their evidence in light of the calculation. The calculations were not fully tested and explored in evidence before the primary judge.
Obviously, each case must depend on the evidence and facts before the court in that case. On the evidence in this case we are not satisfied that the calculated inferred speeds - and their inconsistency with Mr Maddison's estimate of 15 ‑ 20 km per hour - provides an adequate basis to uphold the challenge to the primary judge's finding as to the deceased's speed. In part that is due to the advantage that the primary judge enjoyed as trial judge. In part it is due to the apparent reliability, on our assessment of the evidence, of Mr Maddison's estimate of speed in comparison to the other evidence on which the appellant relied. In part it is due to the limited weight that can be given to the calculations given the imprecise foundation on which they are based.
Mitchell JA has concluded that, in the circumstances of this case, the challenge to the primary judge's finding as to the deceased's speed attracts the standard of appellate review identified in Robinson Helicopter Co Inc v McDermott.[274]
We consider that will not necessarily be the case for every finding based on evidence as to estimates of speed. In this case, however, the finding at first instance was grounded in the primary judge's conclusion that Mr Maddison's evidence was reliable after his Honour reminded himself that such evidence was notoriously unreliable and his Honour then proceeded to conduct a painstaking review of the reliability of the evidence touching on speed. The primary judge had the benefit of seeing and hearing Mr Maddison give evidence. His Honour was in a position to be able to compare Mr Maddison's reliability in giving his inferential evidence as to speed - thereby being better able to evaluate Mr Maddison's impression and judgment as to the deceased's speed as conveyed by his evidence - and juxtapose that with Mr Maddison's more qualified evidence as to time and distance and complete inability to make an estimate as to the speed of the appellant.
In these circumstances we too would apply Robinson Helicopter Co Inc v McDermott and employ the sort of standards discussed in [207] above in assessing ground 6. Even if, contrary to that conclusion, the finding ought not attract the same high degree of appellate restraint accorded to a credibility finding, the matters we have referred to demand that the proper weight to be accorded to the primary judge's advantage is considerable. In either case, for the reasons we have given, we are not satisfied that the appellant has demonstrated error in terms of ground 6.
Whether the primary judge's finding was in error depends on whether it should have been held that the deceased's speed in entering the roundabout exceeded 20 km per hour. The primary judge accepted that Mr Maddison's estimate of 15 ‑ 20 km per hour was reliable. For the reasons we have given above the finding as to the deceased's speed has not been shown to be wrong by reference to incontrovertible facts or uncontested testimony. Nor is it glaringly improbable or contrary to compelling inferences. There is nothing to suggest that the primary judge failed to use or misused his advantage as trial judge. Based on our review of the evidence and examination of the primary reasons we do not accept that the primary judge's finding was wrong having regard to the subject matter of the conclusion and the advantages enjoyed by the primary judge over this court.
There is one further minor matter that should be mentioned.
Having accepted Mr Maddison's evidence, the primary judge should have expressed his ultimate conclusion as to the speed at which the deceased entered the roundabout conformably with that evidence. The primary judge did not do so in stating that the deceased was probably travelling at a speed of 15 km per hour, when Mr Maddison expressed the deceased's speed by reference to a range starting at 15 km per hour. However, identifying error is concerned with substance not verbal semantics. In circumstances where, in closing submissions, the appellant accepted that a speed less than 25 km per hour was a safe speed that was not excessive in the circumstances of the accident,[275] his Honour's factual conclusion of substance was that the deceased was travelling at a speed of no more than 20 km per hour. That was the finding as expressed at [141] of the primary reasons - a conclusion that was in keeping with Mr Maddison's evidence as accepted by the primary judge.
The primary judge was not in error in finding that the deceased was travelling at no more than 20 km per hour as he entered the roundabout. Ground 6 should be dismissed.

Ground 6A challenged the finding that the appellant entered the intersection after the deceased. However, as expressed, ground 6A only did so if ground 6 was upheld. As we would dismiss ground 6, ground 6A also fails.
In any case, against the eventuality that ground 6 was upheld, the respondent argued that the impugned finding did not solely depend on the primary judge's finding as to the speed that the deceased was travelling at as there was stand-alone evidence to support that finding.[276] We accept that submission for the reasons that Mitchell JA gives at [139] - [145] above. Ground 6A should also be dismissed for those reasons.
Senior counsel for the appellant accepted that if the appellant did not succeed on ground 6 then she could not succeed on grounds 8 or 9.[277] Grounds 8 and 9 thus fall with ground 6.
Moreover, for the reasons given by Mitchell JA at [169] ‑ [177] and [180] ‑ [191] above, even had ground 6 succeeded to the extent that the proper finding was a speed of around 25 km per hour, we would not find contributory negligence on the part of the deceased. In further answer to senior counsel for the appellant's submission that a reasonable cyclist in the deceased's position, going through the roundabout, would stand ready for the appellant's car to come out at any time we would also refer to what was said by Newnes JA in Wensink v Marshall (although here we would substitute 'cyclist' for driver' with all necessary consequential changes):

[I]t is clear that the exercise of reasonable care does not require a driver to drive in such a way as to be in a position at all times to anticipate anything that a pedestrian or another motorist might do, or to be in a position to reduce the speed of their vehicle to a level which will avoid any risk of collision.[278] (emphasis added)

At a glance

Court

Decision date

Before

Source

Judgment (302 paragraphs)

Legislation Cited (2)

Cases Cited (22)