ALLEN WOODHOUSE AGAINST LOCHS AND GLENS (TRANSPORT) LTD [2019] ScotCS CSOH

BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £5, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	Scottish Court of Session Decisions
You are here: BAILII >> Databases >> Scottish Court of Session Decisions >> ALLEN WOODHOUSE AGAINST LOCHS AND GLENS (TRANSPORT) LTD [2019] ScotCS CSOH_105 (18 December 2019) URL: http://www.bailii.org/scot/cases/ScotCS/2019/2019_CSOH_105.html Cite as: 2020 Rep LR 2, [2019] ScotCS CSOH_105, 2020 SLT 208, 2020 GWD 3-52, [2019] CSOH 105

[1] The pursuer was a passenger on a Volvo coach owned and/or operated by the

defenders. He was one of 51 passengers. On 26 March 2015, at around 2pm, while the coach

was being driven in a northerly direction on the A83 between Arrochar and Cairndow, at or

near a stretch of road known as “Rest and Be Thankful” and some 700 metres or so north of

the junction between the A83 and the B828, it left the road to the left, went down an

embankment towards Loch Restil, rolling over once in the process, and came to halt in an

upright position just short of the loch. A number of passengers were injured, including the

pursuer. The driver of the coach, who was an employee of the defenders, was also injured.

[2] The pursuer sues for damages, alleging that the accident was caused by the

negligence of the defenders. Quantum has been agreed in the sum of £15,000, inclusive of

interest up to the first day of the proof. Though the sum sued for in this action is relatively

modest, the action has been vigorously contested, no doubt because a decision on liability in

this case is likely to assist in the resolution of claims by others who were injured in the

accident. Because of its wider significance in this respect, this action, though initially

brought in the Sheriff Court, was transferred from that court to the Court of Session.

[3] Factual evidence about the accident and the period leading up to it was led from the

coach driver (Elizabeth Gallon), three of the passengers on the coach (Ronald Radmore,

Michael Long and Dawn Godwin), the driver of another car (Douglas McArthur) and two

police officers involved in road policing (PC Sam Lawrie and Inspector Adam McKenzie,

now retired) who attended the scene within about 30 minutes after the accident. I was

satisfied that all were doing their best to describe what happened, as they saw it, but, as I

explain when going through the facts, the accuracy of their evidence as to the details of what

[4] I also heard from a number of expert witnesses. Each party called an accident

investigator, Graham Greatrix for the pursuer and Christina Holland for the defenders.

They approached the matter in somewhat different ways, Mr Greatrix bringing more

practical experience to bear while Ms Holland tended to be more theoretical. As their

evidence developed, there was little between them on many of the relevant issues. Both

were well-qualified to give their evidence; their qualifications were not challenged. I formed

the view that both gave their evidence honestly. I initially had some doubts about whether

Ms Holland’s laconic replies to questions asked in cross-examination were an indication that

she was being deliberately obstructive; but ultimately I concluded that this reticence was in

part attributable to her natural manner and in part a perfectly legitimate mechanism for

giving herself time to think through the implications of the questions and to be careful in her

answers. Evidence from a forensic meteorologist, Dr Richard Wild, given in two Reports

prepared by him, was admitted in evidence by agreement between the parties without the

need for him to attend in person. I also heard from Professor Andrew Rae, Professor of

Experimental and Applied Aerodynamics at the University of the Highlands and Islands.

He spoke to the wind conditions at about the time of the accident, the topography of the

locus and its likely effect on wind patterns there, and the likely impact of gusts and

turbulence upon the coach at that time. He was an impressive witness and I had no

difficulty in accepting his evidence as honest and reliable. I shall consider aspects of all this

evidence when discussing the particular issues to which it relates.

[5] The basic narrative of the accident and the events leading up to it can be taken from

the account given by the driver, as supplemented by the evidence of other witnesses. Some

precise times for when the coach started and stopped were available from a reading from

the tachograph on the coach, which was spoken to in evidence by the driver. The accuracy

of such times depends on the setting of the clock on the machine and it was not suggested

that they would necessarily be accurate to the minute; but they were useful in particular for

fixing a fairly precise time for the accident and, therefore, in tying in the evidence of weather

[6] The passengers on the coach were enjoying a “7 day Spring Ceilidh Break” in the

Highlands. They were based at the Loch Awe Hotel, embarking on day trips each day. On

the day in question they had stopped for lunch at the Ardgarten Hotel, a few miles south of

Rest and Be Thankful. After lunch they set off from the hotel at about 1.45 pm, heading

generally northwards. The coach pulled in at the Rest and Be Thankful viewpoint, adjacent

to where the B828 Glen Mhor Road joins the A83 from the left. The idea was to let

passengers out to enjoy the view and take photographs, but (according to the evidence of

passengers on the coach) the weather had turned for the worse, very few passengers got off

the coach and, in consequence, the stop at the viewpoint was fairly brief.

[7] After setting off from the Rest and Be Thankful viewpoint, the driver noticed that the

front passenger door did not seem to be fully shut, so she stopped for a couple of minutes at

a lay-by on the left hand side of the road, a few hundred metres further on, to shut it

properly. She pulled out of the lay-by at about 2.00 pm, or just before, and resumed her

[8] The accident occurred less than a minute later, some 150 metres further up the A83.

By this time there was a very strong northerly wind, hitting the coach virtually head on. I

shall discuss the evidence about the strength and direction of the wind separately. The

driver’s evidence was that she pulled out of the lay-by and accelerated normally, not hard;

and had reached a speed of approximately 20-30 mph when she was hit by a strong gust of

wind from her left (nearside), which pushed her towards the middle of the road and over

the central white line. She immediately braked – she estimated that she managed to reduce

her speed to about 10-15 mph – and turned her steering wheel to the left, to regain her

(correct) side of the road. No sooner had she regained her proper position on the road –

maintaining the reduced speed of 10-15 mph – than the coach was hit by another severe gust

of wind, this time from her right (offside), which seemed to “lift up” the front of the coach

and force the coach to the left, off the road and onto the verge, despite the driver applying

[9] At this point in the road, the A83 runs alongside Loch Restil, which lies to the west of

the road (to the left if driving northwards). The road is high above the loch and there is a

steep bank down from the road to the edge of the loch. The slope down to the loch is not,

however, consistent. At the point where the coach first left the road, the grass verge was

reasonably flat and the coach ran along it, making gouge marks in the grass and mud, but

remaining upright. The effect of the driver applying the brakes and the build-up of mud in

front of the wheels meant that the coach was slowing down rapidly; and had it come to a

halt even some yards before it in fact came to a halt, it would have been on relatively level

ground and, despite the strength of the wind, would probably not have rolled over, despite

the continued force exerted by the wind. However, despite this braking effect brought

about by a combination of the brakes and the build-up of mud in front of the wheels, the

speed at which the coach was travelling took it further along the verge to a point where the

verge was not level but sloped down in a pronounced manner to the loch. It was when it

reached this point, on ground sloping steeply down towards the loch so that it was leaning

decidedly to the left while at the same time being subjected to the force of the wind coming

from its right, that the coach began to roll over. It completed one roll before settling in an

[10] The police arrived at the scene about 30 minutes after the accident. Other emergency

services arrived soon after. The police examined the scene of the accident. The next day,

27 March 2015, they took a statement from the driver. The decision was made not to bring

criminal charges against the driver or the defenders in respect of the accident. The police

took the view that the driver was not going too fast, and the accident was caused by the

wind. The coach was taken into custody of Police Scotland. For some unexplained reason

the police did not extract the data regarding the speed of the coach from the tachograph.

This would have had to be done within 24 hours of the accident, otherwise the data would

be wiped from the system. As a result there is no tachograph evidence regarding the speed

[11] The account given above is based primarily on the evidence given by the driver, with

assistance from the two expert accident investigators. There were a number of random

points which I should mention briefly; and a number of other more important points require

[12] There was no safety barrier at the edge of the road where the coach came off. There

is one now. At the time of the accident, the safety barrier at the loch side of the road stopped

at the lay-by where the driver stopped the coach briefly to fix the problem with the door.

[13] In her evidence the driver described how the second gust appeared to “lift up” the

front of the coach while forcing it to its left and, ultimately, off the road. I should make it

clear that it is not suggested that the coach was physically lifted up and dumped over the

edge of the road; nor that the front wheels were physically lifted up from the surface of the

road even for a short time. Nor was the accident caused predominantly – though there may

have been an element of this – by the sideways force of the wind overcoming the frictional

resistance between the tyres and the road surface, causing the coach to skid sideways on all

four wheels. What the driver was describing, as I understood her evidence, was the wind

hitting the coach on its right hand side, possibly at its front right quarter, taking some of the

weight off the front wheels of the coach, and forcing the front of the coach to yaw slightly to

the left, with the front edge of the coach turning towards the left despite the best efforts of

the driver to correct its alignment. The overall effect of this was to force the coach to the left

until the front left wheel went onto the verge, at which point the coach was doomed.

[14] The coach was examined on 30 March 2015 by Matthew Kelly, a Department of

Transport vehicle examiner. He was looking to see whether there were any pre-accident

defects which might have contributed to the accident. He examined the steering, the brakes,

the tyres, the suspension, the lights and the driving controls. He concluded that there were

no pre-existing or latent mechanical defects that would have been detrimental to the

vehicle’s handling or directional control at the time of the accident.

[15] The question was raised in the evidence, albeit briefly, about how the weight was

distributed within the coach. In terms of the number of passengers and where they were

seated, the coach was nearly full – there were 51 passengers in a 57 seater – so no real

difficulty arises there. But what about the luggage? On the basis that each passenger might

have 20 kg or so of luggage, how that luggage was distributed within the luggage

compartment on the coach might have been of some importance. However, I am satisfied

that this is a non-point. The passengers were staying at the same hotel throughout their

week long holiday and were returning that night. While they might have been carrying

small items of luggage for the day trip, their main luggage would have remained at the

[16] Other passengers on the coach did not support the driver’s evidence that she pulled

into a lay-by to fix a problem with the door soon after leaving the Rest and Be Thankful view

point. However, they did not pretend to have been concentrating on what the driver was

doing – they were talking to friends and/or looking at the scenery – and may well have run

the two stops into one in their own minds. I have no reason to think that the driver was

mistaken on this point of detail; she had no reason to make it up, since the point is probably

of no consequence; and her account of a brief stop after leaving the viewpoint and before the

accident is, on one view, supported by the only reading (the “Daily Driver Activity

Protocol”) recovered from the tachograph. I accept her evidence on this point.

[17] The general consensus in the evidence from the passengers was that the coach came

to a rest the right way up having rolled over once. That was consistent with the evidence

given in court by the driver, though she had initially told the police in her statement that she

was sure it had rolled over twice. One passenger, Ms Godwin, thought that the coach had

rolled over twice. She described undoing her seat belt and removing her sweater after the

coach rolled once, and then holding onto the overhead grab rail when it rolled over a second

time. I am satisfied that she must have been mistaken about this. The coach rolled once and

[18] It was an important part of the driver’s evidence in court that the coach had been hit

by a strong gust from the left before being hit by a gust from the right. This first gust

knocked her off balance and left her unprepared for the unexpected gust from the other side.

Although, in giving her account to the police, the driver referred to there having been two

gusts, she did not tell them that the gust from the right which forced the coach off the road

had been preceded by one from the left. Further, none of the passengers who gave evidence

spoke about a gust from the left preceding the one from the right; they described two gusts,

both from the right. The driver described the wind as windy, but quite normal for the

location. She was an experienced driver and had driven that road countless times; and she

was used to wind of that strength and the strong gusts that came with it. She described in

her evidence occasions when she would steer slightly into the wind to resist the force of the

wind pushing the coach in the opposite direction. The unusual thing, as far as she was

concerned, was the fact that on this occasion the gusts came from different directions. She

had not come across that before. In considering her evidence, I wondered whether what she

was describing was in fact not a gust from the left before the critical gust from the right, but

rather a sudden lull in the force of the wind from the right, perhaps because of being

blanketed by a feature of the adjacent land form, which might momentarily have eased the

force of the wind from the right and caused her to veer towards the centre line, making it

seem to her that there had been a gust from the left, the sort of phenomenon similar to that

routinely experienced by car drivers when temporarily blanketed from the wind by a high

sided vehicle. This explanation, if it is one, was not explored in evidence and I therefore do

not consider it further. I was, in any event, re-assured by the evidence of Professor Rae as to

how it might happen that within a very short space of time there could be strong gusts from

different directions. On balance I accept the driver’s evidence that she was hit by a strong

gust from her left before being hit by a strong gust from her right.

[19] Loch Restil lies in a narrow valley bordered by mountains to the east and the west.

At the point where the accident happened, the road runs parallel to the loch, though high

above it. In certain weather conditions the topography of the valley and surrounding

mountains can make the valley subject to its own unique weather patterns with gusts and

turbulent wind flows. Professor Rae described the topography in great detail under

reference to particular mountains and other features. His analysis was not seriously

[20] The driver described the conditions at the time of the accident as “very windy”.

Passengers on the coach spoke it similar terms about the conditions: “windy” but “not

disastrous”, “starting to get windy”. So did Professor Rae: “a nasty blustery day … with

swirling wind from any direction”. But such winds were not exceptional. The driver

confirmed that she was used to driving in those conditions. The area in the valley around

Loch Restil was known for winds of this strength, and conditions were often blustery.

[21] Precise weather information for the exact location at the time of the accident is hard

to come by. Weather information for the time was available from weather stations at

Bishopton (Glasgow), Glen Ogle and Dunstaffnage, each about 20 miles away from the

location, respectively to the south, the north east and the north west of the accident location.

Information from these sources indicates that from about early to mid-morning on 26 March

2015 the wind shifted to West-North-West. However, information from the Rest and Be

Thankful weather station, down the valley from the scene of the crash, provided every

10 minutes rather than every hour, suggests that from then on the wind there was more

northerly, blowing approximately north-south down the valley. That weather station

records lower prevailing wind speeds than the other weather stations but stronger gusts.

Professor Rae’s conclusions as to the wind conditions at and around the accident site are set

out in section 3 of his revised report, and include the following observations:

“(i) At the time of the accident the prevailing wind in the region was generally

from the west-north-west with a mean wind speed over 20 mph and gusting

(ii) The wind direction at Loch Restil at the same time (as measured at the Rest

and Be Thankful, at the southern end of the Loch) was northerly, possibly

due to the funnelling effect of the north-south aligned glen …;

(iii) With a prevailing northerly wind, the ridge comprising [certain particular

mountains, one with a sharp southern face], and the ridge extending

northeast from [another summit], will introduce further turbulence and

(iv) The location of the weather stations away from Loch Restil means that we

cannot know precisely the wind characteristics prevailing at the precise time

and location of the accident. However, while not directly useful, the weather-

station data do provide an indication of the general nature of the weather

conditions; all of the stations were consistent in recording blustery conditions

such that turbulent eddies, vorticity and longitudinal gusts cannot be

(v) The mean wind speed recorded at the Rest and Be Thankful over the 10-

minute interval covering the time of the accident (i.e. from 1400 to 1410) was

18.1 mph with a maximum recorded gust of 38.3 mph. The highest recorded

gust occurred in the following 10-minute interval (i.e. from 1410 to 1420) with

In terms of the Beaufort scale, a scale which is familiar to many, the wind speed at the locus

at about the time of the accident as noted in sub-paragraph (v) above equates to a prevailing

wind of Force 4-5 (moderate to fresh breeze), gusting to Force 7-8 (near gale to gale force).

Those are strong winds, in the gusts at least. But it was not suggested, indeed it could not

have been seriously suggested, that winds of that strength are so exceptional as to be

unforeseeable in that part of the Highlands. I should, however, note Professor Rae’s

evidence that the speed of the wind could be greater at the accident location than at the Rest

and Be Thankful weather station down the glen – which was at a lower altitude and

downwind of the accident site – and the strength of the gusts and the turbulence could be

affected by the topography. Further, he was able to explain how the driver could come to

experience two gusts from different directions in such a short space of time. A turbulent

eddy might operate as a vortex, rotating clockwise or anti-clockwise as it moved down the

[22] The best evidence of the speed at which the coach was travelling in the lead up to the

accident would have been the data recorded in the equipment linked to or forming part of

the tachograph. In the regrettable absence of such data being recovered from the

tachograph, evidence about the speed of the coach at the time of and leading up to the

accident depends on the driver, the passengers and the experts. That is unfortunate. The

driver’s evidence was that she was going at about 20-30 mph before being hit by the first

gust (from the left) and, after braking and returning to the correct side of the road, was

doing about 10-15 mph when hit by the second gust (from the right). This is almost certainly

an underestimate of her speed. The passengers in the coach did not think that she was

driving too fast, but this adds nothing of value without any information as to their

experience and what they would regard as too fast. There was evidence of a general nature,

which I accept, that a driver will generally underestimate the speed at which he or she is

driving. But more pertinently for present purposes, the experts carried out calculations (by

reference to the likely deceleration rate of the tyres on the road surface and/or muddy verge

and the length of the gouge marks left by the wheels of the coach on the verge from the

point of entry until the moment at which the coach tipped over while nearly stationary)

showing, and this was a matter of broad agreement between them, that at the moment when

the front left wheel of the coach first left the road and went onto the verge, the coach must

have been travelling at a speed of or in excess of 30 mph, possibly as fast as 40 mph.

Ms Holland thought that the speed of the coach at this point would probably have been

somewhere at the lower end of that range, because the deceleration rate for a tyre on a

slippery mud surface would probably be lower than that quoted in the research paper to

which she had referred; but a countervailing feature is that, as both experts accepted, the

deceleration rate would have increased as the tyre of the coach dug into the ground and

mud built up in front of it. My assessment, taking all the evidence into account is that the

coach was doing about 35 mph at the moment of the front left tyre first going on to the

verge. Assuming that the driver braked after the first gust and braked again after the

second gust, that suggests that she had been travelling at a rate considerably in excess of

35 mph when the coach was first hit by the gust from the left. On balance of probabilities I

conclude that the coach was travelling at about 40-45 mph when it was first hit by the gust

from the left and at about 35-40 mph when hit by the second gust from the right.

Ms Holland did a calculation to show that if the driver had accelerated at the typical

acceleration rate for a bus (0.8-1.2 m/s²), then, over the distance from the lay-by to the place

where the coach left the road, the “maximum likely speed” of the coach at that point, before

the first gust, would have been about 34-42 mph. That all depends on assumptions as to

rates of acceleration taken from the evidence of the driver, whose evidence about the speed

she was going is, as noted above, clearly wrong. Further, the bottom end of that range is

difficult to square with the calculation about the speed of the coach at the moment the wheel

went onto the verge. Mr Greatrix thought that the speed reached before the brakes were

applied might well have been as low as 34 mph or even lower, but this simply proceeds on

the hypothesis that the highest rate of acceleration is normally achieved when accelerating

from rest; and it again seems difficult to reconcile with the evidence about speed of entry

onto the verge. Standing the evidence, which seems probable and which I accept, that the

driver braked after both the first and the second gusts, I see no reason to depart from my

assessment that the coach probably reached a speed of about 40-45 mph before it was hit by

the first gust, dropping to about 35-40 mph before being hit by the second gust. There is no

significant difference between that speed and the top of the range selected by Ms Holland.

[23] Both experts produced calculations for reaction times, i.e. the time available for the

driver to react in such a way as to avert the accident, measured from the time and place

when the coach was hit by the second gust until the front left wheel came into sufficient

contact with the muddy verge so as to make the accident almost inevitable. The differences

between them were many and varied. They included differences about the extent to which

the tyre would have to be over the edge of the road before control would be lost; disputed

measurements of the width of the tyres, the width of the road and the width of the white

lines on the road; disputed assessments of the precise position of the coach on the road when

it was struck by the second gust; the effect of wind speed in calculating the time for the

coach to move from its pre-gust position to its potentially doomed position at the edge of the

carriageway; the different reaction times for applying the brakes as compared with

applying corrective steering (which would be quicker); consideration of treatises and

practical research about reaction times among different age cohorts and in different

situations; and challenges to the relevance of such reports to the present situation where,

unlike in the research, the driver did not know in advance that she was going to be called

upon to react to anything. On top of this there had to be fed into the calculations an

assessment of the time for the steering on the coach to respond to the movement of the

steering wheel (probably about half a second). The experts carried out calculations for

reaction times at the low speeds spoken to by the driver, but as noted above I do not find

those estimates of her speed to be reliable. Calculations of reaction times at higher speeds

are potentially more relevant. Thus, assuming a coach speed of 40 mph, a gust strength

equivalent to a wind of 46 mph hitting the side of the coach at 90°, and the front left wheel of

the coach being initially positioned 0.5 metres from the edge of the road, Mr Greatrix

calculated that the driver would have about 0.84 seconds to react. At a coach speed of

30 mph but with all other things remaining the same, he calculated that the reaction time

would be 1.28 seconds. If the coach was initially 0.575 metres from the road edge, and

everything else was the same, the reaction times at 40 mph and 30 mph would be

1.82 seconds and 2.60 seconds respectively. The figures differed according to a host of

variables, particularly the initial positioning of the coach and the assessment of where in

relation to the edge of the road the coach started to be out of control. Ms Holland, taking a

number of the same variables into account, but emphasising in particular the likely loss of

control as soon as any part of the tyre slipped onto the verge, put forward slightly lower

figures for reaction time. It is unnecessary to set them out. Her evidence was that in

circumstances where the reaction time (“perception response time” in her phraseology) was

less than 2 seconds, the driver would probably not have had sufficient time to apply

corrective steering. In cases where the reaction time was greater than 2 seconds, she

considered that it was not possible to say one way or another. This evidence from both

experts was in many ways illuminating; but it bore all the hallmarks of counting angels

dancing on the head of a pin. On any view the reaction time at the speed at which I have

found the coach to have been travelling at the time of the second gust, was likely to be very

short. The evidence left me satisfied that it was unlikely that the driver could be criticised

for failing to react in time by braking and/or steering when the coach was hit by the second

[24] For the pursuer, Mr Milligan QC argued that the mere fact that the coach came off

the road at that point gave rise to a prima facie inference of negligence on the part of the

driver. A well maintained coach, properly driven, should not come off the road. It was for

the defenders to rebut this inference by giving an explanation for the accident which was

(a) credible and reliable and (b) non-negligent. In support of this proposition he cited Mars

v Glasgow Corporation 1940 SC 202, 206, O’Hara v Central Scottish Motor Traction Co Ltd1941

SC 363, 379, Doonan v The Scottish Motor Traction Co Ltd 1950 SC 136, Sutherland v Glasgow

Corporation 1951 SC (HL) 1 per Lord Normand at 6-7, NG Chun Pui & Ors v Lee Chuen Tat and

Anr [1988] RTR 298, 300D-301E, Morton v West Lothian Council [2006] RepLR 7 at para [70]

and Kennedy v MacKenzie [2017] CSOH 118 at para [5]. The defenders appeared to rely on

the adverse weather conditions, but adverse weather alone is not sufficient to rebut the

inference, since drivers must drive to the conditions which they encounter: Weatherstone v T

Graham & Son (Builders) Ltd [2007] CSOH 94 at para [16], MacDonald v Aberdeenshire Council

2014 SC 144 at paras [64] and [70]. The weather was not unforeseeable, and the accident

could have been avoided by slowing down or steering away from the edge. The weather

was not a damnum fatale or Act of God: even the unprecedented rainfall in Caledonian Ry Co v

Greenock Corpn 2017 SC (HL) 56 was not a damnum fatale. It was simply a severe gust on a

[25] While insisting that it was for the defenders to show how the accident happened

without their negligence, Mr Milligan’s criticism focused specifically on the actions of the

driver. If she was going at the speed she thought she was going – 20-30 mph, reducing to

10-15 mph after the first gust and before the second – then she had plenty of time to react

and could have avoided the accident. In any event, at that speed, even if the coach had gone

onto the verge, it would have come to a halt before reaching that part of the verge which

sloped downhill and caused it to tip over. If, on the other hand, she was driving at a greater

speed – possibly in excess of 40 mph before being hit by the first gust – that, by her own

admission (and I shall come back to deal in more detail with the evidence on which

Mr Milligan relies as an admission), was too fast for the conditions. It left her with

insufficient time to react to the impact of the second gust; and it meant that when the coach

went onto the verge it did so at a speed which propelled it forwards onto the part of the

verge which sloped downhill and caused it to roll over. On either view, the defenders had

failed to discharge the onus on them of showing that the accident happened without their

[26] Mr Primrose QC, for the defenders, argued that this was not a case where the onus

shifted to the defenders to show that the accident happened without their fault. That rule

only applied where all the factors contributing to the accident were within the control of the

defenders. Under reference to O’Hara (supra), per Lord Normand at 378, he submitted that

even in such a case, the defenders were not required to prove a particular non-negligent

cause of the accident – the authorities established only that the defenders had to show that

there were circumstances which might have caused the accident without involving their

negligence. In such a case the burden reverted to the pursuer to prove negligence. The

cases of Moore v Fox [1956] 1 QB 596, 614-5 and Horgan v Alexander t/a the Inn at Ardgour 2019

SC Edin 9 (Sh. Weir) were both illustrations of the point that the reversal of the burden of

proof, and the application of the maxim res ipsa loquitur, only came into play in

circumstances where the pursuer did not know and could not know the cause of the

accident. In a case such as this there was no such disparity in knowledge; the pursuer had

access to the same information as was available to the defenders, and had instructed experts

to examine the cause of the accident. But even if the pursuer’s submission on shifting the

burden of proof were correct, the defenders had led enough evidence to establish that the

accident was not caused by their fault. The accident could be explained by the wind. The

coach was blown off the road by an exceptional gust, following on from a strong gust from

the opposite side. The pursuer had led no evidence of carelessness, lack of attention or

excessive speed. In so far as the driver said anything to suggest that, if she was going faster

than 20-30 mph, then she was driving too fast, her evidence had to be taken in context. If,

contrary to his submission, the driver was guilty of some failing in her response to the

sudden impact of the wind turbulence on the coach, then her actions and re-actions should

be assessed in light of having been taken in the “agony of the moment” and not judged by

too high a standard: NG Chun Pui at 302D, Easdon v A Clarke & Co (Smethwick) Ltd

[2008] CSOH 29, Dickson v Klinsman [2013] CSOH 111. Reliance was placed on the decision of the

[27] As an alternative line of defence, Mr Primrose relied upon damnum fatale (or force

majeure, vis major or Act of God, all of which amounted to the same thing). Under reference

to the cases of Tennent v Earl of Glasgow 1864 2 M (HL) 22, 27, and The Boucau [1909] P 163, he

submitted that it did not require to be proved that the operation of the external force was so

violent and unexpected that no human foresight or skill could possibly have prevented the

accident. It was enough that the force could not reasonably have been foreseen or that its

impact could not reasonably have been prevented. A gust of wind could amount to damnum

fatale, but it must be exceptional: Cushing v Peter Walker & Son (Warrington & Burton) Ltd

[1941] All ER 693, 694-5, Caledonian Ry Co v Greenock Corpn 1917 SC (HL) 56. Other cases

cited included Hogg v MacPherson 1928 JC 15, Burns v The Royal Hotel (St Andrews) Ltd 1958

SC 354 and General Construction Ltd v Chue Wing & Co Ltd [2013] UKPC 30. He submitted,

under reference to Burns in particular, that it was not always necessary positively to prove a

particular supervening accidental cause; the onus on the defenders could be discharged by

proving the absence of negligence, i.e. that the accident happened despite the exercise of all

reasonable care. In summary, the defenders’ case was that the combination of the two

severely turbulent gusts of wind from directly opposite directions, on an otherwise windy

day, were of such a magnitude as to render their occurrence not reasonably foreseeable as a

realistic possibility. The conjunction of such factors (the wind and the gusts) at that

particular moment and at that particular spot, with its particular topographical features,

coinciding with the coach passing that spot at that time, meant that the accident was not

caused by the fault of the driver but rather by the sort of freak conditions discussed in the

[28] Both counsel made detailed submissions on the evidence, including submissions as

to the credibility and reliability of the individual witnesses. I do not propose to rehearse

those submissions in any detail here, but I have taken them into account in reaching my

conclusions on the witnesses and on the salient facts all as set out above and below.

[29] In my judgment the pursuer is correct in his argument that in the circumstances of

this case the burden of proof switches to the defenders to prove that the accident occurred

without their negligence. I set out my interpretation of the relevant case law in para [70] of

my Opinion in Morton (supra) and I was not made aware that it has to date been the subject

of serious criticism either in the Inner House in that case (unreported, [2008] CSIH 18) or in

any subsequent case. Neither party before me in this case pointed to anything in that

analysis with which they particularly disagreed. My approach in this case is therefore, with

one qualification (or perhaps it is only clarification) in line with the following extract from

“[70] The usual situation in which the onus of proof shifts to a defender to explain

how the loss occurred without his negligence is in a case where all the circumstances

giving rise to the loss are within the control of the defender. This is the classic case

for the application of the maxim res ipsa loquitur. The cases cited to me by both

counsel bear this out. The justification for such a shift is obvious; not only are all the

circumstances under the control of the defenders but the defender will have the

means of knowing what occurred. The shift is necessary “to avoid the denial of

justice to those whose rights depend on facts incapable of proof by them, and often

exclusively within the knowledge and control of their opponent”: per Lord Justice

Clerk Cooper in Elliott v Young's Bus Service (supra) at p 456. … For the maxim to

apply, the circumstances of the accident, either ex facie or upon proof of certain facts,

must be “eloquent of negligence”, to adopt counsel for the defenders' expression. At

that point the onus shifts to the defender to prove, at the least, the existence of other

facts or circumstances which might have caused the accident without his negligence:

see O'Hara v Central SMT Co Ltd (supra) per Lord President Normand at p 378. I say

“at the least”, since some of the authorities appear to go further and suggest that the

defender must prove that the accident was not in fact attributable to his negligence:

see eg per Lord Evershed MR in Moore v R Fox and Sons (supra) at pp 614–615. For

reasons which will become clear, I do not need to resolve the issue of what the

defender need prove. Counsel for the defenders suggested, at one point in his

submissions, that the defender could shift the onus back to the pursuer simply by

offering a reasonable explanation of how the accident might have occurred consistent

with the absence of negligence. If the suggestion is that an explanation without full

legal proof was sufficient, I do not consider that it is supported by the authorities. It

appears to be based upon certain remarks of Lord Dunedin his dissenting speech in

Ballard v North British Railway Co (supra) at p 54; but such a proposition was plainly

rejected by the Inner House in O'Hara, in which one of the issues was whether the

defenders had established their explanation by corroborated evidence, a point which

would not have been relevant had they only been required to provide an explanation

falling short of proof. That proposition was also rejected by the Court of Appeal in

Moore. I do not understand Lord Guest's speech in Devine v Colvilles Ltd (supra) at p

100, to which counsel for the defenders referred, to indicate that a “reasonable” or

“plausible” explanation, unsupported by evidence to the appropriate standard of

The one point of qualification or clarification is this. I spoke of the onus shifting to the

defender to prove, at the least, the existence of other facts or circumstances which might

have caused the accident without his negligence, and of the possibility that the defender

might have to go further and prove that the accident was not in fact caused by his

negligence. I was not intending to exclude the possibility that the defender might avoid

liability if he could prove that the accident was not in fact caused by his negligence, even

[30] In the present case the considerations mentioned in the passage quoted from Morton

all point to the onus shifting to the defenders to prove a non-negligent cause of the accident

or at least to disprove negligence on their part. A coach travelling along an A road in the

Highlands should not ordinarily come off the road. The fact of it having done so gives rise

to a prima facie inference of negligence. It is for the defenders to rebut that prima facie

inference, if they can. Everything relevant to the accident involves matters within their

knowledge and control. They are responsible for selecting a suitable vehicle for the roads

and weather conditions reasonably to be expected; for maintaining the coach in a suitable

and roadworthy condition; for selecting a competent driver; for giving or ensuring

appropriate training for the driver to enable her to cope with road and weather conditions

which she might reasonably expect to encounter; and, vicariously, for the actions of the

driver in the course of the service. Although the weather was obviously a significant factor,

the defenders were prima facie in a position to take the steps needed to keep their passengers

safe from it. And it is the defenders who are in a position to know what has been done in

these respects. It is, of course, theoretically possible for the pursuer to find out what the

defenders have done and not done in relevant respects, but in reality the rules of pleading

and recovery of documents mean that there are likely to be gaps in what the pursuer can

explore with a view to pinpointing where the fault lies. In those circumstances I have no

difficulty in holding that in circumstances such as the present the burden shifts onto the

defenders to show that the accident occurred without their negligence.

[31] I am alive to the oft-repeated remark that onus seldom matters once the evidence is

out: see eg Gibson v BICC & Co 1973 SC (HL) 15, per Lord Reid at 22, Salt International v

Scottish Ministers 2016 SLT 82 at para [45], SSE Generation v Hochtief Solutions AG

[2018] CSIH 26 per the Lord President (Carloway) at para [273]. In a case such as this, at the initial

stage at least, the onus does, or may well, matter because it will, or may, determine what

matters are explored and what evidence is led. But once that evidence is led, it will be

subject to the usual process of cross-examination, to clarify matters or to expose weaknesses

in the defenders’ position. At the close of the evidence, when all the evidence is out, the

court will inevitably focus its attention on the areas of the defenders’ evidence which were

subject to challenge or criticism, to see whether those various challenges or criticisms are

made good. It will decide the relevant issues on the balance of probabilities. That is

important. The mere fact that the onus is transferred to the defenders to prove that the

accident occurred without any negligence on their part does not impose on the defenders

any higher standard of proof than the ordinary civil standard of the balance of probabilities.

The defenders do not have to exclude negligence on their part to any higher standard than

[32] I am persuaded on the evidence that the defenders have discharged the burden on

them of proving that the accident happened without their negligence. The evidence that the

coach was well maintained and did not suffer from any relevant pre-existing defect was not

challenged; indeed it was a matter of agreement in the Joint Minute lodged in process by the

parties. The only challenge, the only suggestion of fault advanced by the pursuer, was in

relation to the actions of the driver. What was said, in short, was this: if she was driving at

the speed at which she said she was driving (20-30 mph before the first gust, 10-15 mph

before the second) then she had plenty of time to enable her to react to the gust and should

have been able to prevent the coach leaving the road; whereas if she was driving faster than

that (40 mph, plus or minus) then she was driving too fast, cutting her reaction time so that

she could not react in time to prevent the accident. I have already found that the coach was

being driven at a speed of about 40-45 mph before the first gust hit it, and at a reduced speed

of 35-40 mph when the second gust struck. Accordingly the pursuer’s first alternative

criticism can be excluded. That leaves only the second alternative criticism for

[33] The pursuer’s case that if the coach was being driven at a speed of 40-45 mph then it

was being driven too fast was not supported by any expert or independent evidence to that

effect. It was not said, by anyone, that at that speed the coach would be less easy to handle

in the prevailing conditions, or less responsive to directional control, or in some way less

stable, more vulnerable to the impact of the wind, more prone to being deflected off course.

The pursuer’s case that the coach was being driven too fast was periled on the evidence

Mr Milligan led – I use the word advisedly – from the driver, to the effect that she thought

that she was driving at 20-30 mph before the first gust and 15-20 mph before the second and

that it would not have been sensible to be going faster than that. In his written submissions

at the close of the proof Mr Milligan set out his record of the driver’s evidence in chief in this

regard. My own notes, though briefer, broadly support this summary of her evidence and I

did not understand Mr Primrose to suggest that it was not reasonably accurate. The

It [the first gust] pulled me over the white line. I put my foot on the

I don’t think I was going any more than 20 mph. I think if I braked I’d

Do you think that going more than 20 mph would be sensible in those

It would be dangerous to keep going at the same speed after a gust

If you were hit by a gust, to carry on at 34-42 would be dangerous?

Mr Primrose objected to this line of questioning on the basis that there was no record for it. I

allowed it to proceed under the usual reservations. In his closing submissions Mr Primrose

renewed his objection. I repel that objection. In a case where the burden is on the defenders

to disprove negligence, it is, in my view, open to the pursuer to test the evidence as it comes

out, despite the absence of notice on record. It is a matter of chance that in this case the

driver’s evidence was led by the pursuer. It might more naturally have been led by the

defenders, and cross-examined by the pursuer. The defenders needed to rely on her

evidence to negative any finding of fault in respect of how the coach was being driven. In

such circumstances it is open to the pursuer to cross-examine (or in this case examine) the

driver to seek to undermine her evidence. This is not a case where the pursuer was seeking

[34] However, although I allowed the evidence to be admitted, I do not accept that this

passage can bear the weight Mr Milligan sought to place on it. The exchanges between

counsel and the witness focus primarily on the wisdom or otherwise of maintaining her

original speed after the first gust had hit the coach. The driver’s evidence is clear: it would

not have been sensible to keep going at the same speed after the first gust had hit. Her

evidence was that she braked after the first gust and dropped her speed from an initial

speed of 20-30 mph to about 10-15 mph, maybe 20. She agreed that it would have been

dangerous to resume her earlier speed. But throughout the exchange she is focusing on the

relative speeds, before and after the first gust. She is not talking of absolute speeds. She

clearly considered that her initial speed, before the first gust, was sensible and safe. She

clearly thought that her reduced speed after the first gust and after braking was sensible and

safe. She put the wrong figures on her speed both before and after the first gust. She

thought, wrongly, that her initial speed was 20-30 mph and her reduced speed 10-15 mph,

maybe 20. But the fact that she was wrong in terms of assessing her actual speeds before

and after the first gust is neither here nor there. She is not to be understood as saying that,

now it has been pointed out to her that her initial speed was 40-45 mph and her reduced

speed 35-40 mph, then she accepts that she was going too fast. That would be to

misunderstand her evidence completely. Accordingly I reject the submission that the driver

accepted that if she was driving at 40-45 mph, reducing to 35-40 mph after the first gust,

[35] There is no other evidence either way on the question of speed. No reliance can be

placed by the defenders on the decision of the police not to charge the driver. That no doubt

reflects their own opinion on the question of speed and safety – no doubt informed amongst

other things by the applicable speed limit and their assessment of the weather conditions –

but the judgment in this case has to be mine, not theirs, it is not clear to me what evidence

they had before them when reaching their decision, and the views on this matter of the

police are therefore, for these purposes irrelevant. Just as an acquittal, had she been

prosecuted, would be irrelevant and inadmissible in a civil proof (see Hui Chi Ming v The

Queen [1992] 1 AC 34 at 42-43), as would a conviction have been until recent statutory

intervention, so too is the decision not to prosecute; the position is a fortiori. Nor is it directly

relevant that driving at a certain speed means that there is minimal time available for an

effective reaction in the event of the coach being blown off course by an unexpected and

violent gust of wind. Such considerations are too far removed. Absent any proved

connection between the coach’s speed and its vulnerability to being blown off course, absent

any evidence that the particular speed of the coach made an accident of this sort reasonably

foreseeable or in some way more likely, then the question of how to react in such

circumstances, and whether the speed of the coach allows sufficient time to react, does not

arise. Negligence depends on reasonable foreseeability – it does not require steps to be

taken in advance to meet a situation the occurrence of which is not reasonably foreseeable.

[36] In these circumstances I do not accept the pursuer’s case that the driver was at fault.

No other fault is raised as an issue. But I am left with this concern. My finding on the

evidence is that the weather conditions were unpleasant and the wind was strong – but

there was nothing exceptional about the conditions, winds of that strength were foreseeable,

and extreme turbulence, being a feature of the topography of that area, could also be

foreseen. For that reason I would have rejected the defence of damnum fatale, had it been

necessary to consider it. If, as I have also found, the driver was not at fault, how can the

accident be explained? It is possible that the combination of two gusts from opposite sides

of the bus was unique, though the fact that Professor Rae could readily explain the

phenomenon suggests otherwise. It may be that occurrences such as this, occurring without

fault on any side, though mercifully rare, are inevitable. It may be, on the other hand, that

the occurrence – a coach coming off the road in conditions which were by no means

unforeseeable, and there being no fault on the part of the driver – is an indication that more

could be done at a systemic level to guard against it happening again. I say no more about

this. I do not wish to speculate. When I raised the question of whether fault could be

attributed not to the driver but to the defenders as an organisation, I was told that this was

not in issue in the case, and the pursuer was not putting his case on that basis. I therefore

[37] For the above reasons I find that the pursuer’s case fails. I shall grant decree of

Scottish Court of Session Decisions