OUTER HOUSE, COURT OF SESSION [2013] CSOH 95
PD2550/11	OPINION OF LORD TYRE in the cause DANNY THOMAS RYDER Pursuer; against THE HIGHLAND COUNCIL Defender: ________________

Pursuer: Hood; Drummond Miller LLP

Defender: C Murray, Ledingham Chalmers LLP

13 June 2013

Introduction

[1] In the early hours of 9 December 2008, the pursuer's mother, Anne Jayne Sutherland or Ryder ("the deceased"), sustained fatal injuries in a road traffic accident while travelling south on the A99 from John O'Groats to Wick. She was 36 years of age. The pursuer seeks reparation from the defenders in respect of the loss and damage that he has sustained as a consequence of his mother's death. The grounds of fault alleged are concerned, firstly, with an operational decision taken regarding precautionary gritting of roads during the previous evening and, secondly, with the defenders' policy with regard to the overnight gritting of roads in winter conditions.

[2] Evidence was led at the proof from the pursuer, from various officials and employees of the defenders, and from certain eye witnesses regarding road surface conditions. There was no challenge to the credibility or reliability of any of these witnesses and I accept their evidence. Both parties also led expert evidence. On behalf of the pursuer, evidence was led from Mr Michael Hopwood, a civil/road engineer who has since 2007 been an associate with Hawkins & Associates Ltd, Wilmslow, Cheshire, carrying out investigations of "all types of civil and highway engineering matters". Between 1994 and 2007 Mr Hopwood was employed by Cheshire County Council, latterly as Principal Engineer. His professional experience included a period of years acting as duty officer making decisions regarding the treatment of roads affected by snow or ice. He had undergone a 5-day training course in accident reconstruction. On behalf of the defenders, evidence was led from Mr Adrian Runacres, who carries on business at Alresford, Hampshire, as an independent consultant undertaking collision investigations and research. Prior to June 2005 he was Head of the Investigations and Risk Management Department at the UK Transport Research Laboratory. He is the Technical Advisor and Secretary of the National Winter Service Research Group and has over 20 years' experience of highway maintenance and winter maintenance activities. Mr Hopwood and Mr Runacres both provided written reports. Neither had visited the locus of the accident. Both were instructed primarily to provide opinion evidence concerning the defenders' winter maintenance service policy and the decisions made by the duty officer during the day prior to the accident. At the end of the proof counsel for both parties provided me with detailed written submissions on the evidence and the law, for which I am most grateful.

The locus of the accident
[3] The A99 runs in a north-south direction along the east coast of Caithness for 17 miles from John O'Groats to Wick and then continues further south until it joins the A9 at Latheron. Between John O'Groats and Wick the A99 is not a trunk road, and the defenders are responsible for its maintenance, including winter maintenance. For the purposes of the defenders' Winter Maintenance Policy, discussed below, it is a Priority 1 route. Approximately four miles south of John O'Groats, the road passes through the settlement of Freswick. At this point the road is close to the sea at Freswick Bay and is about 10 metres above sea level. As the A99 approaches Freswick from the north it runs downhill and takes a slight bend to the right, at or about its junction with an unclassified minor road that runs east to Skirza. Evidence was given at the proof by PC Iain Mathers, a collision investigator with the Northern Constabulary, that the bend was sufficiently shallow that it could be negotiated safely in dry conditions at a speed in excess of the national speed limit of 60 mph. It did not therefore require warning signs. To the south of the junction with the road to Skirza the A99 straightens out and crosses a bridge over the Freswick Burn.

Circumstances of the accident
[4] At the time of the accident, the deceased resided at Huna, near John O'Groats, having moved there from Wick only a few days previously. She was in a relationship with a Mr Anthony Cundall who was at Huna with her during the night before the accident. The deceased was employed by Tesco as a price integrity assistant at their supermarket in Wick. On the day of the accident she was due to commence work at 6am. She owned a silver Toyota Aygo motor car registered in 2007 which she used to travel to and from work.

[5] At the time of the proof, Anthony Cundall could not be traced and did not give evidence. In a statement given to the police shortly after the accident and referred to by PC Mathers in his collision investigation report, Mr Cundall stated that he had been aware of the deceased getting out of bed at around 4am. He did not know when she left the house but she did not appear to be rushing. In favourable weather conditions the journey to Wick would take about 20-25 minutes. Nothing more is known of the deceased's movements prior to the accident. Mr Barry Metcalf, who lived in a house beside the A99 close to the Skirza road junction and who gave evidence at the proof, was awake from about 5.10am. He heard no car passing before 5.35 and then four or five cars passing during the next half hour, all except one travelling in the direction of Wick.

[6] At about 8.15am, after it had become light, a driver travelling south from John O'Groats noticed the deceased's car lying on its roof in a field to the west of the road and just south of the bridge over the Freswick Burn. The deceased was in the car and the engine was cold to the touch. He contacted the emergency services and an ambulance arrived at 8.45am. The crew could not gain access to the deceased but found no pulse present. One of the crew noted that she was wearing a seat belt. She was pronounced dead by a doctor at 9.37am. The cause of death was stated in the death certificate as "head injury - vehicular collision". I return below to consider the evidence as to the cause of the accident.

[7] On the basis of the foregoing information it seems likely that the accident occurred at some time between about 5.35 and 6am, and I so find. This was not a matter of controversy at the proof.

National Winter Maintenance Standards
[8] In 2001the UK Roads Board (UKRB) was created and issued a Code of Good Practice entitled "Delivering Best Value in Highway Maintenance" developed in partnership with national and devolved governments and local authorities. It was supported and recommended by, among others, the Scottish Executive, COSLA and the Society of Chief Officers of Transportation in Scotland (SCOTS). In 2005, the 2001 Code was superseded by a document entitled "Well Maintained Highways - Code of Practice for Highway Maintenance Management" published by the UK Roads Liaison Group, a specialist sub-group of the UKRB. (The 2005 Code has now itself been superseded by a version published in 2012.) Each Code contains the statement that "The suggested recommendations of this Code are explicitly not mandatory on authorities". However, the expert witnesses were in agreement that the Codes afforded a reasonable benchmark against which to measure the policies and practices adopted by a particular local authority. Both the 2001 and 2005 Codes contained sections dealing with Winter Service, i.e. winter maintenance, including an appendix dealing inter alia with decision-making procedure. Within the appendix there was a table entitled "Decision Matrix Guide" providing guidance as to decision-making with various permutations of road surface temperature, precipitation, and predicted road conditions. The matrix included the observation "All decisions require continuous monitoring and review."

[9] Mr Hopwood also referred to a publication by the Institution of Civil Engineers entitled "Highway Winter Maintenance, a practical guide" (2000). It was less widely available than the Codes and would be used more for training purposes than for day to day use.

The defenders' Winter Maintenance Policy
[10] The defenders' Winter Maintenance Policy at the time of the accident was set out in a document dated October 2003. The document notes that the defenders as roads authority have a duty under section 34 of the Roads (Scotland) Act 1984 to take such steps as they consider reasonable to prevent snow and ice endangering the safe passage of pedestrians and vehicles over public roads. With regard to treatment of roads (i.e. salting, gritting and snow ploughing), the policy is stated inter alia as follows:

"3.0 Treatment of Roads

Salting and gritting operations for ice and light snow conditions on adopted roads will be carried out as detailed below. Each gritting route will take a significant length of time to complete. The length of time taken will vary from day to day depending on actual weather conditions.

3.1 Network Hierarchy

It is not possible to provide a winter maintenance service on every road at the same time. In general priority routes will be treated before those of a lower priority. The following prioritised hierarchy has been adopted.

PRIORITY 1 - Main Routes

PRIORITY 2 - Other Strategic Routes including Bus Routes*

PRIORITY 3 - Main Urban Streets and School Bus Routes†*

PRIORITY 4 - All other routes as resources permit.

* Note route gritting will not necessarily be completed before buses start their journeys.

†* Contracted school buses or mini buses

The treatment routes, based on the priorities set out above, will be agreed by Area Committees and leaflets with maps showing the Priority 1 and 2 routes together with target treatment times will be distributed to the public at the start of each winter period.

3.2 Treatment Times - Monday to Saturday

The service will be provided between 6am and 9pm, although Priority 4 routes are unlikely to start before 8am and will not be treated after 6pm.

3.3 Treatment Times - Sundays and Public Holidays (including Christmas Day and New Year's Day)

A service will be provided between 7am and 9pm on Priority 1 Routes, Service Bus Routes (if applicable) and Main Urban Streets only.

3.4 Precautionary Treatment

Precautionary treatment, carried out the previous evening, normally before 9pm, in advance of adverse weather, will in general be restricted to Priority 1 routes..."

The main point to be noted from the above is that the defenders' policy does not provide for the treatment of any roads for which they are responsible between 9pm and 6am. I should also mention that although the operation is generally referred to in the policy document as "gritting", and was often so described in the course of the proof, it was not a matter of dispute that what was in fact spread on the roads by the defenders at the material time was salt. The Winter Maintenance Policy is published in a leaflet and on the defenders' website.

[11] The defenders' policy is put into practice by means of annual Winter Maintenance Plans applicable to particular operational areas within the defenders' local authority area. Caithness constitutes one of those operational areas. At the time of the deceased's accident, the applicable Winter Maintenance Plan for Caithness was a plan produced in September 2008. This plan set out in detail the routes to be followed by gritting vehicles when carrying out morning treatment or evening pre-treatment. Morning Route M23, for example, required a gritting vehicle driver from the Wick depot to treat the A99 from Reiss to John O'Groats, the A836 to Canisbay, then Warth Hill, Everley, Skirza, Keiss, Howe, Quoybrae to Myrelandhorn, and Reiss - a mix of Priority 1, 2 and 3 routes with the Priority 1 routes being treated first. Where evening pre-treatment was carried out, it was carried out only on Priority 1 routes. So, for example, Evening Route E11 required a driver from the Thurso depot to treat the A836 from Thurso to Castletown, Gills Harbour Road, John O'Groats, and the A99 from John O'Groats to Reiss.

The defenders' decision-making practice
[12] Mr Iain Moncrieff, a civil engineer employed by the defenders as leader of their highways technical team at Wick, gave evidence of the defenders' procedure for taking decisions regarding the need for and timing of gritting operations. At the material time four senior engineers, including Mr Moncrieff, took it in turn to act as duty officer for the Caithness operational area for a week at a time. Each day, the duty officer would prepare a short report on road conditions, check weather forecasts for that night and the following day and, on the basis of the information available to him, decide at around lunch time what action, if any, was required that evening (bearing in mind that any evening treatment was normally restricted to Priority 1 routes) and/or the following morning. Sometimes the decision would be obvious, sometimes less so. Where a decision was taken not to pre-treat roads in the evening, it was possible, but unusual, for this instruction to be changed if further information was received during the afternoon which indicated that treatment was required. So far as the instruction for the following morning was concerned, the duty officer's instruction in cases of doubt might be to order the gritter drivers to be on standby at the depots at 6 am, at which time a decision would be taken by the foreman present at each depot as to whether to send them out to begin treatment immediately.

[13] In assessing the likely overnight and early morning weather conditions, the duty officer would have available to him information from two sources. First, there was a general synopsis provided at that time by the Met Office via email and covering a 24 hour period from noon on the day of issue until noon on the following day. The duty officer in Caithness would receive a forecast for "Highland Area A", an area which covered part of Sutherland in addition to Caithness. As well as providing a general 24 hour weather summary, the Met Office forecast predicted the likelihood of each of the following elements: ice, hoar frost, snow, fog, strong wind and rain. It also contained a prediction of minimum air temperature and minimum surface temperatures for urban roads, roads under 150 metres (i.e. above sea level) and roads over 150 metres. In each case there was an estimate of the period during which the temperature would fall below zero.

[14] The second source of information available to the duty officer in Caithness consisted of forecast data provided by two roadside "Icelert" ice sensors. One of these was situated at Drumholistan near Reay on the north coast. The other was at Old Hall on the A882 Wick - Halkirk road near Loch Watten. These two ice sensors produced 24-hour forecasts in graphic form of surface and dew temperature, surface state, rain state, cloud state and cloud amount which could be accessed remotely on screen by the duty officer at the time when he was making his decision. Mr Moncrieff explained that because of their respective locations, these two sensors tended to give worst case scenario forecasts. The Drumholistan sensor was located near the crest of a hill at one of the colder spots in the county and as high above sea level as any Priority 1 road was likely to reach in Caithness. The Old Hall sensor was located inland at a dip in the road where the road crossed a burn and where frost tended to settle.

[15] In addition to the two sensors with a forecasting function just described, the defenders had ice sensors at four other locations within the Caithness area. These consisted of a metal plate embedded in the surface of the road and they recorded inter alia road surface temperature, surface state (e.g. "wet", "ice" or "wtrtd" - the latter being an abbreviation of "wet - road treated"), and a somewhat crude estimate of the extent of presence or absence of salt on the road. There was no clear evidence as to how the sensor would categorise the surface as "wet" on the one hand or as "ice" on the other, and the expert witnesses were in agreement that a greater degree of reliance should be placed upon the surface temperature recorded than on the description of the road surface state. A further column in the record was headed "Alarm", and the word "alarm" would appear in the column whenever the temperature fell close to 0°C on a wet road on which the sensor did not detect the presence of salt. This information was recorded hourly, or at times in between if there was a significant change in conditions, and was available continuously online to the duty officer with a slight time lag of about half an hour. I return below to consider the significance of readings from one of these sensors on the night of the accident.

[16] The duty officer was on call at home during the night. He did not, as a matter of course, monitor weather conditions overnight or log in during the night to check temperatures at the various ice sensors around the county, although he could do so if he wished. There was in particular no procedure for monitoring sensor reports during the night to see if any were reporting "alarm" or indicating the presence of ice. Mr Moncrieff explained that he did not consider that there was anything to be gained by monitoring conditions overnight, as gritting could not in any event begin until 6am. Occasionally, if the Met Office forecast that conditions would deteriorate overnight, he would telephone the depot prior to 6 am to upgrade a decision from "standby" to "patrol", thereby saving around ten minutes in the commencement of gritting operations. Exceptionally, the duty officer could arrange for a gritter to be sent out during the night if this was necessary in connection with a police, fire or ambulance emergency. The duty officer would check the sensor reports when he came to work in the morning and again before issuing his instructions for the following day.

[17] The duty officer had no power to instruct a vehicle patrol during the night of roads within his operational area. Again Mr Moncrieff's view was that such a patrol would be pointless, partly because nothing could be done until 6am, and partly because information from a patrol would be little more than a snapshot of a particular location at a particular time and would add little to the data available from the sensors.

Weather forecast information for the night of the accident
[18] The Met Office general synopsis for Highland Area A for the period from noon on Monday 8 December 2008 to Tuesday 9 December 2008 included, so far as material to these proceedings, the following forecast:

"Ice: yes; high confidence; risk of ice overnight from wintry showers.

Hoar Frost: yes; low confidence; patchy hoar frost expected this evening.

Rain: yes; high confidence; see 24 hour weather summary below."

The minimum surface temperature forecast for roads less than 150 metres above sea level was -3°C, falling below zero between 1600 and 1000.

The 24 hour weather summary read as follows:

"Sunny spells and scattered rain showers becoming more frequent later this evening and during the overnight period turning increasingly wintry over 200m and falling as snow over 300m during the early hours. Further wintry showers expected tomorrow morning. Breezy."

The "readiness colour" of the forecast was red (on a scale of green, amber and red) indicating that road surface temperatures were expected to fall below zero with ice formation and/or snow accumulations. The forecast is recorded on the hard copy print used at the proof as having been received at 2.43pm on 8 December 2008. It appears, however, that it may also have been received earlier, because at 12.59pm Mr Moncrieff had issued his instructions (see below) in an email to Mr Stuart Bell and it is clear from the terms of that email that Mr Moncrieff had the Met Office forecast available to him at the time when it was sent.

[19] The Drumholistan ice sensor forecast a road surface temperature between about 0.5° and 1.5°C from about 5pm on 8 December until about 7am on 9 December. The forecast surface temperature does not fall to or below zero at any time. A total of 13 rain showers were forecast during the period, nine of which were of short duration. No graph showing the forecast from the Old Hall ice sensor was available at proof, but a printout was lodged showing the data that would have been available to the duty officer on 8 December 2008. This showed the road surface temperature forecast to fall to 0.3°C between about 12.40am and 2am on 9 December, rising to 1.1°C at 4.20am before falling back to 0.5°C between 5.40am and 7.40am. A number of periods of precipitation were forecast throughout the night and morning, including a continuous period between 2.20am and 4.40am

Decision taken by the duty officer
[20] Mr Moncrieff explained that in reaching his decision he would have taken account of both the Met Office forecast and the forecasts from the roadside sensors, giving neither preference over the other. He emphasised that the area covered by the Met Office forecast included part of Sutherland where roads were further from the sea and reached a greater height above sea level. It included locations such as Altnaharra and Forsinard where very low temperatures have often been recorded. Mr Moncrieff interpreted the Met Office report as indicating with a high level of confidence that ice would be present on roads within Area A but not throughout Area A. Priority 1 routes in Caithness were less than 200 metres above sea level. The area is flat and low-lying with the sea on two sides. The minimum surface temperature forecast of -3°C would not necessarily apply to Caithness. Below 200 metres the showers would fall as rain and would not cause ice to form. He then looked at the sensor data and noted, firstly, that temperatures at Drumholistan and Old Hall were not forecast to fall below zero and, secondly, that a number of rain showers were forecast at both locations. He concluded that an evening pre-treatment of roads would be a waste of time because the salt would be washed away by the rain. In relation to the decision not to order an evening pre-treatment, the emailed instruction sent by Mr Moncrieff to the Wick depot records the "cause" as being "Frost F'cast PM Low Conf - NA" - No Planned Action. A comment is added: "OH & Drum Graphs Ok". As regards the morning, Mr Moncrieff acknowledged that with a forecast of rain and temperatures within the area below freezing, the decision was not an easy one to make. He decided to hedge his bets by putting the gritters on standby, ready to proceed to treat the roads at 6am if necessary. Mr Moncrieff was aware of the existence of the Decision Treatment Matrix to which I have already referred but did not use it in making his decision. If he had used it his decision would not have been different.

Road surface conditions at the time and place of the accident
[21] There is no direct evidence of the road surface condition at the locus of the accident at around 5.45am on the morning of 9 December, and it is necessary to examine the material presented to the court in order to decide what finding, if any, may be made in this regard.

[22] Sensor information. I have already mentioned that the defenders had six ice sensors, including the two with a forecasting facility, at locations within the Caithness area. Parties were agreed that the most relevant of these for the purposes of this action was a sensor (without a forecasting facility) located at Ackergill near Wick Airport on the east coast of Caithness, approximately 30 metres above sea level and about nine miles south of the locus of the accident. The following table contains information extracted from the Ackergill sensor report for times between 7pm on 8 December 2008 and 8am on 9 December 2008:

Time	Surface Temp °C	Surface State	Alarm	Icelert Salt %
19.52	0.4	wtrtd	none	90
21.02	-0.6	wtrtd	none	60
21.52	-0.6	wtrtd	none	60
23.00	0.2	wet	alarm
23.52	0.2	wet	alarm
00.52	-0.4	wet	alarm
01.52	0.2	wet	alarm
02.52	-0.4	wet	alarm
04.00	-0.6	ice	alarm
04.52	-0.8	ice	alarm
05.26	-0.2	wet	alarm
05.52	-0.4	wet	alarm
06.33	-0.6	wet	alarm
06.52	-1.4	wet	alarm
07.26	-1.2	wtrtd	none	90
07.52	-0.6	wtrtd	none	90

It will be noted that the surface temperature had fallen below zero by 9pm. It then rose above zero by 11pm and remained above freezing during most of the next three or four hours until falling below zero again before 2.52am where it remained for the rest of the night. The sensor noted the presence of salt at 21.02 and 21.52pm, this being the product of a treatment carried out between 8am and 10.30am on 8 December. From 11pm no salt was detected and the sensor reported "alarm" throughout the night from the time when it ceased to detect the presence of salt on the road in which it was embedded until 7.26am when it once again detected its presence. The sensor asserted the presence of ice at 4am and at 4.52am.

[23] I note for the sake of completeness that road surface temperatures recorded at Drumholistan and Old Hall during the night of 8-9 December were not materially different from those forecast by the sensors. At Drumholistan the temperature did not fall below zero at any time although a temperature of 0°C was recorded at about 7pm before it rose again and remained between 1° and 2°C for the rest of the night. At Old Hall the surface temperature readings were consistently between 0.8° and 1.2°C from 12.53 until 5.27am, before falling to 0.4°C from 5.53 until 6.53am. The Old Hall sensor reported "alarm" at 10.53pm and between 2.53 and 5.53am. It is worthy of comment that on two occasions during the night the description of the road surface at Old Hall changed from "wet" to "wtrtd" although no new treatment had been carried out.

[24] Met Office Weather Radar data. In the preparation of his report, Mr Runacres, the defenders' expert witness, obtained data, not directly available as a production in the case, from the Met Office Weather Radar system for a 5 kilometre square area including the locus of the accident (though not that of the Ackergill sensor). This indicated that the area experienced a period of persistent rainfall between 9pm on 8 December and 1am on 9 December. No precipitation then appears to have occurred until at least 4am; no information is available for the period between 4am and 5am, and another shower affected the area between 5am and 6am. Recorded accumulation rates were not high, with a maximum hourly accumulation rate of 0.7 mm per hour.

[25] Eye witness evidence. Mr Barry Metcalf, who as I have already mentioned lived close to the locus of the accident, had gone outside at about 11pm to let his dog out and did not recall it raining at that time. He recalled the conditions at 6.30am as being "slightly frosty". Mr Metcalf's stepson, Graham Duffy, who lived at Freswick, was also up early that morning to drive his mother and sister to Wick so that they could catch a bus at 7.15am to Inverness. He recalled that the windows of his car had frost on them because he had had to leave the engine running for a period before driving off. He noted that the gravel on which his car was parked had ice on it. His headlights glistened on the road surface. He required to drive carefully as he felt the car steering to be light, as if there was ice on the road, but he did not actually skid. Because of the conditions, it took him longer than usual to complete the drive to Wick. Neither Mr Metcalf nor Mr Duffy had any clear recollection of whether they had been aware of rain during the night.

[26] Mr Colin Smith was the driver of the gritting vehicle allocated to Route M23 that morning. When he arrived on duty at 6am he was instructed by Mr Jim Hindmarch, the Community Works Officer, to begin gritting immediately. He described the route north of Reiss as icy with patches of snow on it, about 8 or 9 out of 10 on a scale of treacherousness. When he passed Freswick at about 6.40 - 6.50am the conditions were icy. Having reached Canisbay, he decided, in exercise of his discretion, to return along the A99 south and treat it again. He passed Freswick again at about 7.20am and conditions at the Skirza junction were still quite icy. There was little change because the salt takes some time to have effect. He reloaded his vehicle from a salt quarry south of Wick with the intention of carrying out a third run on the A99 but by then the road was closed because the accident had been reported. The ambulance crew who were called to the scene of the accident told the reporting police officer that they had attended with caution due to the icy road conditions. The Watch Manager of Highland and Islands Fire and Rescue Service, who arrived at the scene at 8.40am to assist in gaining access to the deceased's car noted that there were signs of ice on the road near the locus of the accident.

[27] Conclusions. On the basis of the foregoing evidence, I am satisfied that it is possible to make findings in fact, on balance of probabilities, regarding the weather and road surface condition at the locus of the accident. I consider that the Ackergill sensor data provide a reasonably reliable indication of the likely temperature and road surface state nine miles further north at Freswick, although I accept that these would not necessarily be the same at both locations at the same time. I also bear in mind Mr Runacres' observation that care is required in interpreting the Weather Radar data because recorded times and rates of precipitation do not necessarily represent those experienced at any particular location on the ground. Bearing these limitations in mind, I find that it is likely, on balance of probabilities, that one or more bands of rain passed across the locus of the accident between about 10pm and 1am. These were recorded on the Met Office Weather Radar system and had the effect by 11pm of washing the salt from the morning's treatment off the road at the Ackergill sensor. It is likely that a further shower passed over the area, including the locus, between 5am and 6am, causing the small temporary rise in road surface temperature noted by the Ackergill sensor at 5.26am. On the basis of the Ackergill data, I find it likely that by 3am at the latest, the road surface temperature at Freswick had fallen below zero and remained below zero for all or most of the time until after the accident occurred. Having regard to the evidence of the recent passage of rain showers, and the eye witness accounts of the presence of frost and ice on the road at or near the locus during a period beginning about one hour after the accident occurred, I find it likely, on balance of probabilities, that between 5.35 and 6am there were icy patches on the road in the vicinity of Freswick. It is not possible on the evidence to be more specific as to where these patches are likely to have been. It may be, as PC Mathers suggested, that it is particularly likely that ice would form on the bridge over the Freswick Burn, but as the deceased must have lost control of her car prior to arrival at the bridge this is not of significance in the case.

[28] It was submitted on behalf of the defenders that the evidence was insufficient to permit any reliable inference to be drawn as to the road conditions at Freswick prior to 6am. Signs of ice at the locus an hour later did not give a reliable indication: during that period the recorded temperature at Ackergill fell by more than 1°C. At 5.26am the Ackergill sensor had ceased to report the presence of ice. I regard the fact that the surface temperature was recorded as being below zero since 2.52am as more significant than the further fall that occurred after 6am and, in accordance with the view of others, I place greater weight on the recorded temperature than on the description of the road surface. I therefore reject the contention that no finding can be made.

The cause of the accident
[29] An investigation of the accident was carried out by PC Iain Mathers, a collision investigator with 15 years' police service and 18 months' experience as an investigator of fatal and other accidents in the Northern Constabulary area. He arrived at the locus at 10.50am on 9 December and recorded the road surface as being damp and in good repair. He examined the scene, took photographs, and made measurements. He examined the deceased's car, which had by then been rolled on to its wheels by the Fire and Rescue Service officers, and found no pre-accident mechanical defects. His findings with regard to the movements of the car at the locus were stated in his collision investigation report as follows:

"The Toyota Aygo motor car has negotiated a right hand bend, rotated in a clockwise direction and crossed the carriageway. The front of the motor car has then entered the west verge at an angle of approximately 10° and travelled along the verge for a distance of 17.83 metres. The motor car has then travelled down the banking and struck the top of the concrete retaining wall [i.e. of the Freswick Burn, just to the west of the bridge]. This is confirmed by the furrow marks made by rolling wheels found in the verge and the grind marks found on top of the retaining wall which were made by the underside of the motor car.

The rear offside wheel of the Toyota Aygo has then struck the concrete retaining wall causing the rear axle to be pulled from its offside mounting. This is confirmed by the damage to the rear offside wheel and the rear axle mounting point.

The Toyota Aygo has then travelled over the Freswick Burn and the leading edge of the bonnet collided with the south bank. This is confirmed by the area of disturbed earth, the debris and the damage to the front of the motor car.

The Toyota Aygo has then continued to pitch forward onto its roof, flip over the fencing above the bank and come to rest on its roof within a field. This is confirmed by the debris from the motor car, the damage to the roof, the gouge mark in the field and the motor car's resultant position."

PC Mathers' conclusion was that the driver of the car had entered the right hand bend on the approach to the Freswick Bridge and lost control of it "...as a result of inappropriate speed or harsh steering/braking for the conditions while negotiating a bend".

[30] PC Mathers found no indication that any other vehicle was involved in the accident. He measured the friction coefficient of the road surface and concluded that it would not have been a contributory cause of the accident if it was in the condition at the time when he carried out his measurement. He found no tyre marks on the road. His conclusion that the car had rotated through 360° was based on the fact that the marks on the verge broke the surface rather than merely rolling across it. He was able to calculate that at the time when the car struck the south bank of the burn it was travelling at 40 mph but had insufficient information to estimate the car's speed at the time when the driver lost control, beyond saying that it must have been at least 40 mph. He explained that the reference in his report to "the conditions" was a reference to the possibility that the road surface was affected by ice. If ice is present on the surface of a road, a car is more likely to rotate at a point where the driver adjusts his or her steering, for example to negotiate a bend in the road.

[31] An affidavit by Carolyn Ryder, currently resident in New Zealand, the mother of the deceased's former husband, was lodged. Mrs Ryder expressed the opinion that the deceased was a careful driver with whom she felt safe in a car.

[32] In a supplementary report, Mr Hopwood, the pursuer's expert witness, addressed the question whether it was possible to calculate the maximum speed at which the deceased's car was travelling when control was lost. He referred to the Traffic Accident Investigation Manual published by Northwestern University, which states that the coefficient of friction for an icy road which has not been rained on can be about 0.07 to 0.2 at speeds over 30 mph. Using PC Mathers' measurements, Mr Hopwood calculated that a driver would lose control of his or her vehicle at this bend at any speed in excess of 31 mph with a coefficient of 0.07 and 52 mph with a coefficient of 0.2. On the basis of available information regarding weather conditions at the locus, he expressed the view that the lower figure might be more appropriate. Mr Runacres regarded Mr Hopwood's range of possible coefficients as too narrow: thin ice capable of being broken by tyres locking could have a coefficient of 0.4 or 0.5. The calculation depended upon speculation regarding not only the road condition but also the line taken by the driver through the bend and the point in relation to the bend where control was lost. I did not find Mr Hopwood's evidence on this matter particularly helpful. As he freely acknowledged in cross-examination, he was not giving an opinion as to the speed of the vehicle prior to loss of control but rather an opinion as to the speed at which a loss of control might occur. It seems to me that, in absence of clear evidence as to the state of the road surface at the place where control was lost, such an opinion is speculative. I accept the evidence of PC Mathers that the car must have been travelling at a minimum of 40 mph, and I do not consider that Mr Hopwood's calculation entitles me to find that it was travelling at any particular speed higher or lower than that figure. In agreement with Mr Runacres, I consider that the calculation requires too many unsubstantiated assumptions to produce a result upon which any reliance can be placed.

[33] On behalf of the pursuer it was submitted that, given the presence of ice on the road at the locus and the absence of tyre marks, the court could, in the absence of any other plausible explanation, conclude that on the balance of probabilities the cause of the accident was that the deceased's car made contact with ice on the road causing the vehicle to lose its grip on the road surface and ultimately to leave the road altogether. On behalf of the defenders it was contended that even if as a matter of fact ice was present at or around the locus, there was no evidential basis from which to infer that it was a causative or contributory factor of the accident. There were other possible causes: another vehicle which did not collide with the deceased's car; an animal on the road; the driver distracted or driving too fast. The pursuer had failed to prove on balance of probabilities that ice was the cause of the accident.

[34] There is in my opinion no evidence to support any positive finding that the accident was caused by anything other than loss of control of the car by the deceased on encountering a patch of ice on the road. In particular, there is no evidence which would entitle me to find that the accident would not have happened but for the deceased driving too fast for the road conditions. Nor is there any evidence of involvement of another vehicle or an animal. No pre-accident defects were found in the car. I have already found that it is likely that there were icy patches on the road at or around the locus of the accident. That being so, I do not find it difficult to infer that, on balance of probabilities, the deceased's car left the road because she lost control due to ice on the road while steering round the shallow bend at the Skirza road junction. That would be consistent with the car rotating through 360° before leaving the carriageway to the north of the bridge, as PC Mathers considered had occurred. It would also be consistent with an absence of tyre marks on the carriageway itself. I accordingly so find.

Grounds of fault
[35] On behalf of the pursuer it was contended that the defenders were in breach of duties incumbent upon them at common law. It was submitted that the defenders were in breach of their duty as roads authority to take reasonable care for the safety of the pursuer in that

(i) they failed to treat inter alia the A99 at the accident site by way of salting or gritting on the evening of 8 December 2008;

(ii) they failed to monitor conditions overnight on 8-9 December 2008 by instigating a patrol or, at the very least, by monitoring the data being recorded by their ice sensors; and

(iii) on the basis that such monitoring ought to have been undertaken, and that the status of "alarm" recorded by the Ackergill sensor would therefore have been observed, they failed to take appropriate action overnight by ordering treatment of inter alia the A99 at the accident site by salting or gritting.

I shall address ground (i) first before addressing grounds (ii) and (iii) together.

Decision not to treat roads on the evening of 8 December
[36] I have set out above the evidence of Mr Moncrieff as to his procedure and reasons for deciding not to order a pre-treatment of Priority 1 routes during the evening of 8 December. Evidence was also led on behalf of the defenders from Mr Stuart Bell, a civil engineer employed by the defenders as Community Works Manager at Wick. He had been employed by the defenders since 1985 and involved in winter maintenance since 1998. Mr Bell worked in the same department as Mr Moncrieff and also took turns as duty officer. He was not party to the decisions taken by Mr Moncrieff on 8 December, but remembered thinking that they were more cautious than his would have been. Having regard to the Met Office forecast and the sensor forecasts, he would not have ordered evening pre-treatment and would probably have regarded morning treatment as unnecessary.

[37] Mr Hopwood's view was that Mr Moncrieff's decision was the wrong one. Mr Hopwood would probably have instructed treatment because the danger of not doing so was too great due to the risk of ice overnight with a high confidence factor; risk of hoar frost during the evening, albeit with a low confidence factor, and a forecast of rain which would wash away any residual salt from the morning gritting. He considered the Met Office general synopsis to be much more important to decision-making than the sensor forecasts. He acknowledged that duty officers would have different perceptions of risk and that some might have chosen not to instruct gritting because hoarfrost was predicted with a low confidence factor and the forecast rain might wash away any salt that was put down. He could understand why they might instruct no treatment. The decision was finely balanced but in all of the circumstances, including the absence of any provision for overnight treatment, he considered that the duty officer has taken the wrong decision. He disagreed that putting salt down was a waste of time if rain was forecast; the spread rate could be increased and some of the salt might not be removed. It was not possible to be certain whether pre-treatment would have made any difference to the road surface at the time and place of the accident. The fall in salt percentage from 90 to 0 reported by the Ackergill sensor indicated that there had been sufficient rain at that location to wash the salt off.

[38] In his written report, Mr Runacres expressed the view (paragraph 8.7) that the predicted conditions warranted a decision to undertake evening treatment on those routes which included sites on Priority 1 roads that were likely to be affected by the hazards identified. In his written report he expressed a personal opinion that a prudent winter maintenance decision would have included treatment of Route E11 (including the A99 at the accident site), but also considered that it might have been appropriate not to undertake such a salting operation, depending on the information available to the decision-maker at the material time. In his evidence to the court Mr Runacres agreed that local knowledge was important, including knowledge of topography and experience of whether forecasts tended to overstate or understate temperatures at particular locations. He regarded road surface temperature as the key factor in the predicted conditions but acknowledged that he did not know which salting routes went into areas where the temperature was predicted to fall below zero. Having regard to the fact that salt concentrations recorded at Ackergill fell from 90 to 0 during a relatively short period of time, he was fairly confident that there was sufficient precipitation in that area to have washed off another application of salt.

[39] I found Mr Moncrieff to be an impressive witness. He gave his evidence in a measured way, explaining his thinking in detail and making concessions where appropriate. At the material time he had many years of experience of making decisions on the winter treatment of roads in the particular environment of Caithness, and my impression was that the decision not to instruct treatment on 8 December was made carefully and after having proper regard to both the Met Office synopsis and the sensor forecasts. It seems to me that there is force in Mr Runacres' acknowledgment that a duty officer familiar with the area has advantages over expert witnesses from further afield called to express an opinion on the reasonableness of a decision taken. Mr Moncrieff was aware from past experience that minimum temperatures forecast for the area covered by the Met Office synopsis for Highland Area A would not necessarily be reached by low-lying Priority 1 routes in Caithness. He was, in my opinion, entitled to place weight on the sensor forecasts which predicted that road surface temperatures at an exposed location on the north coast and at a known cold spot inland would both remain above freezing level. Mr Hopwood, Mr Runacres and Mr Moncrieff all agreed that a forecast of temperatures below zero together with overnight rain was an especially worrying one. On the one hand, the combination of the two made the formation of ice on road surfaces more likely; on the other hand there was at least a possibility that salt spread during an evening treatment would be washed away by the rain. The decision whether or not to pre-treat was a marginal one. Neither expert expressly considered that Mr Moncrieff's decision was outwith the range of decisions reasonably open to him. Both stated that they personally would have instructed treatment, although my understanding of Mr Runacres' view in this regard is that it proceeded upon an assumption that there were salting routes in Caithness which included Priority 1 roads likely to be affected by the hazards forecast, which was not the case. That being so, it seems to me that Mr Runacres' opinion, correctly interpreted, is at the very least not in conflict with Mr Moncrieff's decision. Mr Hopwood fairly acknowledged that some duty officers might have decided not to instruct treatment and that he would be happy to accept their risk assessment, albeit that he personally might have decided otherwise. I am satisfied that, in the whole circumstances, Mr Moncrieff's decision not to order pre-treatment of Priority 1 routes during the afternoon or evening of 8 December was not one that fell outwith the range of decisions open to him in the exercise of reasonable care as duty officer. I accordingly reject the contention that there was a failure by the defenders in this regard to exercise reasonable care for the safety of road users including the deceased.

[40] In reaching my decision on this issue I have had regard to the Opinion of the Court (delivered by Lord Macfadyen) in Taylor v Smith & Others 2003 SCLR 926. The issue in Taylor, by the time the case reached the Inner House, was whether a local authority which had taken a decision not to pre-treat for ice on the basis of information available at around midday was negligent in failing to review and alter that decision in the light of amended weather advice received at about 9.30pm, and was therefore somewhat different from the issue in the instant case. I do, however, note the following observation by the Court (at paragraph 31) in relation to use by the roads authority in their decision-making process of data from an ice sensor at a particular location:

"The decision which the defenders had to make was one for the Moray area as a whole. They cannot reasonably be expected to make individual decisions for individual lengths of road that are not known to present unusual hazards."

I regard that observation as supportive of the approach taken by Mr Moncrieff in attaching weight to the ice sensor forecasts as well as to the general synopsis.

[41] Even if I had held that there was a breach of duty by the defenders in relation to the decision not to instruct pre-treatment of Priority 1 routes, I would not have held that the pursuer has proved that such a breach caused the deceased's accident. Having regard to (i) the evidence, summarised above, regarding the passage of bands of rain across the area in which the accident happened; and (ii) the data from the Ackergill sensor indicating that the presence of salt on the road surface there diminished to zero in a relatively short space of time prior to 11pm, I consider that, on balance of probabilities, any salt spread on the A99 at the locus of the accident in the course of the afternoon or evening of 8 December would similarly have been washed off by rain by about the same time, and would not therefore have been present in the early hours of 9 December to prevent ice patches forming on the road surface. I note that Mr Hopwood suggested that allowance could be made for rain by spreading a greater amount of salt. I do not consider that there is evidence in the present case which would entitle me to hold, on balance of probabilities, that it would have been possible to achieve such a result with a pre-treatment on 8 December. For this reason too I consider that the pursuer's contention that the deceased's death was caused by the defenders' failure to pre-treat roads including the A99 at Freswick during the afternoon or evening of 8 December must be rejected.

Overnight monitoring and treatment
[42] It is convenient to address the pursuer's grounds (ii) and (iii) together. I agree with Mr Moncrieff's observation that there would be little point in carrying out patrols or logging into a sensor website to check for signs of the presence of ice if nothing could in any event be done about it until 6am, being the earliest time at which gritting operations could commence. The real issue, as it appears to me, is whether the defenders were in breach of their duty to exercise reasonable care by failing to treat the road surface at the locus of the accident during the night after the time when the Ackergill sensor reported conditions (wet road and surface temperature below zero) which rendered it likely that ice would form. As counsel for the pursuer acknowledged, this was in effect an attack on the defenders' policy of carrying out no treatment of any roads between 9pm and 6am on weekday nights. Before examining the law which may be applicable to this issue, it is convenient to summarise the evidence regarding the defenders' policy and the opinions of the experts with regard to it.

[43] Mr Richard Guest, a Chartered Engineer who was the defenders' Head of Roads and Community Works, had been employed by the defenders and by their predecessors, Highland Regional Council, since 1991. He explained in evidence to the court that neither the defenders nor their predecessors have ever operated a policy of treating roads 24 hours a day. (Treatment of trunk roads is the responsibility of the Scottish Government, which does operate a 24-hour maintenance policy). The defenders' winter maintenance policy had not changed significantly since before local government reorganisation in 1996. The 2003 policy document was prepared by the official then holding the office which Mr Guest now held. Mr Guest participated in those preparatory discussions. The draft policy document would have been considered by a team of officials including the area transport manager and then submitted to the defenders' Transport, Environmental and Community Services Committee, consisting of elected councillors, for approval. The Committee's decision would then be ratified by the full Council. A risk assessment (though not in formal written terms) was carried out, covering inter alia risk to the public and reputational risk to the defenders. The policy is not reviewed on a regular basis, but rather in response to any request from councillors if there was a perceived change of circumstances. At present a review was being undertaken to assess the resource implications of beginning gritting operations one hour earlier in the morning, i.e. at 5am instead of 6am. In order to comply with European driver hours regulations, this would require either an earlier finish in the evening or the employment of 43 additional drivers to cover all Priority 1 routes. The same number would be required for a 24-hour service, and would have to be employed between at least October and March even though during many nights no road treatment would be needed. The defenders' officials have not been asked by councillors to consider the introduction of a 24-hour service.

[44] In 2010 the defenders' total roads budget was £21.7 million of which about £4.7 million was allocated to winter maintenance. (The budget has since been reduced.) For the Caithness area the total budget was £3.5 million of which just over £800,000 was for winter maintenance. The Highland local government area has 4,213 miles of adopted roads, of which 1,270 miles (about 30%) are categorised as Priority 1 for winter maintenance purposes. The average population density of the Highland local government area is 9 persons per square kilometre, which in European terminology is classified as super sparsity. Mr Guest estimated the cost of a 24 hour operation in the Highland local government area at around £2.2 million on top of the existing roads budget. In deciding not to operate a 24 hour maintenance policy the defenders took account of the length of the road network within the local authority area, the length of individual main roads, and the very small number of vehicles using roads at night. Gritting roads carrying a very low volume of traffic was of questionable value: it takes the action of traffic on salt to remove ice from the road.

[45] The defenders are not alone in not operating a 24-hour maintenance policy, although numerically among Scottish councils they are in the minority. Most of the northern Scottish local authorities operated in 2008, and continue to operate, a similar though not necessarily identical policy. Comhairle nan Eilean Siar do not presently undertake gritting between 6pm and 6am except in response to "blue light" emergencies or snow clearing/gritting on specific high risk roads if necessary. Orkney and Shetland Islands Councils similarly do not treat roads after 6pm. The policy of Perth & Kinross Council in 2008 was to grit until 9pm and to aim to have all roads treated on weekday mornings by 7.30am. The policy provided however for the declaration of a "Blue Alert" by the local area roads engineer for potentially hazardous conditions such as rain followed by freezing conditions. This gave discretion to treat until midnight, but it was noted that the EU Directive on drivers' hours had affected the use of Blue Alerts, other than in cases of emergency, as the availability of drivers the following day had to be checked. Moray Council did not continue beyond 9.30pm because of resource constraints "except in exceptional circumstances". Mr Guest interpreted this as a reference to blizzard conditions; the defenders likewise might extend treatment beyond 9pm for continuous snow ploughing. Aberdeenshire Council aimed to finish its patrols by 9pm and re-commence at 5.30am but operated a standby policy for overnight emergencies including, in addition to "blue light" emergencies and attendance at road traffic accidents, dealing with localised incidents where there was added risk to the road user, such as an icy patch caused by surface water running across the carriageway, particularly at a bend. In contrast, other Scottish local authorities with extensive rural areas, including Argyll and Bute Council and the Scottish Borders Council, operated a 24-hour treatment policy for Priority 1 routes.

[46] Mr Hopwood criticised the defenders' policy not to treat roads between 9pm and 6am as "somewhat reckless". He considered that 6am was too late a start to have roads treated in time for rush hour, although he recognised that this was not a causative factor in the present case. In his 20 years of "hands on" winter maintenance and investigations this was the first time he had encountered a policy that incorporated an overnight break. He considered that the defenders' risk assessment "might have been flawed". He also offered an opinion that the reference in the 2005 UKRLG Code (see paragraph 8 above) to "continuous monitoring and review" should be interpreted as an indication that the Code did not assume overnight breaks. Given that the conditions on the night of 8-9 December 2008 were so uncertain, he would have expected the defenders to instigate an overnight patrol or, at the very least, to have monitored the road sensor information overnight. If they had done so they would have been aware of the icy conditions by 1 am and, had their policy allowed for overnight gritting, they could have taken remedial action before the time of the accident.

[47] Mr Runacres stated that the authors of the 2005 Code were fully aware that some Scottish local authorities operated a gritting policy which incorporated overnight breaks. The Code was not intended to restrict the entitlement of local authorities to set their own policies in this regard. Despite the fact that the Code was used as a benchmark by local authorities throughout the United Kingdom, there were considerable variations from council to council of policy on gritting periods and parts of the network treated. He interpreted the words "continuous monitoring and review" as an instruction to respond to the conditions reported by, for example, the ice sensors. If there was an overnight change, for example, it might be necessary to put the gritters out earlier than had been planned.

[48] The defenders conceded before me that they were subject to duties to the pursuer at common law, though not any statutory duty, on the basis of the duty imposed on them as roads authority by section 34 of the 1984 Act. In Rainford v Aberdeenshire Council 2007 RepLR 126, which proceeded on the same concession, the Lord Ordinary (Lord Reed) reserved his opinion as to whether it had been correctly made. He did not regard the question whether a duty of care existed as foreclosed by the authorities cited to him. He drew attention inter alia to the observation of the Lord Ordinary (Lord Glennie) in Morton v West Lothian Council 2006 RepLR 7 (at paragraph 64) that Scots law and English law appeared to differ on this point. In Goodes v East Sussex County Council [2000] 1 WLR 1356 (HL) and in Sandhar v Department of Transport [2005] 1 WLR 1632, it was held that no general duty of care was owed by a roads authority to road users to treat roads so as to prevent the formation of ice. Lord Clyde drew attention in Goodes to the difference between the two systems at page 1370:

"In Scottish practice a roads authority may be open to liability for personal injury caused by the presence of snow or ice making the passage of pedestrians or vehicles over pavements or roads unsafe. An example, where in the circumstances the claim failed, can be found in Grant v Lothian Regional Council 1988 S.L.T. 533. The claim there was brought both under section 34 and at common law.

It might be thought that there should be a liability upon a highway authority in England and Wales for damages in the event of injury occurring through a failure to take sufficient measures to preserve the safety of the highways under conditions of ice and snow. But there is no remedy there available at common law and if the statute is construed in the way I have preferred there is no remedy under the statute."

[49] The cases that have come before the Scottish courts have tended to consist either of challenges to the scheme of priority of treatment adopted by roads authorities, with the pursuer alleging that an untreated surface on which a vehicular or pedestrian accident has occurred ought to have been treated prior to the time of occurrence of the accident, or of allegations that the authority in question has failed properly to implement its scheme of treatment. Cases of these types include Gordon v Inverness Town Council 1957 SLT (Notes) 48, McGeough v Strathclyde Regional Council 1985 SLT 321, Grant v Lothian Regional Council 1988 SLT (Notes) 533, Syme v Scottish Borders Council 2003 SLT 601, Morton v West Lothian Council (above) and Rainford v Aberdeenshire Council (above). Of a similar kind is Cameron v Inverness-shire County Council 1935 SC 493, in which the pursuer contended unsuccessfully that the roads authority were liable, without any averment of negligence on their part, for losses caused by the blockage of roads by snow. All of these claims (except McGeough) failed on the basis that the pursuer had failed to prove (or in some cases aver) circumstances from which an inference of negligence on the part of the roads authority could be made. It was recognised explicitly in the more recent cases that roads authorities have a discretion to decide on priorities, and that the court should not interfere unless that discretion is exercised unreasonably. But the Scottish case law does not go so far as say that no duty of care is owed at all, and in McGeough the Inner House allowed a proof before answer where the court considered that the pursuer's complaint was properly to be regarded as being that the defenders' gritting policy had been negligently operated in practice. It seems to me that there is sufficient authority, including Inner House decisions in Cameron and McGeough and the obiter dictum of Lord Clyde in Goodes which I have quoted, for the existence in Scots law of a duty owed at common law by a roads authority to road users to take reasonable care to see that they do not suffer injury as a consequence of the presence of ice on a public road. Whether, in the circumstances of a particular case, a pursuer can satisfy the court that there has been a breach of that duty is of course another matter.

[50] In my opinion, however, the present case is not on all fours with the cases that I have mentioned in which the pursuer's attack was on the system of priority of treatment operated by the roads authority, or on an alleged failure properly to implement such a system. The pursuer in the present case goes much further and attacks not merely the priorities or the practical operation of the defenders' Winter Maintenance Policy, but rather a key element of the policy itself, namely the decision not to operate a 24-hour treatment service. It seems to me that this case raises, in a way in which previous cases have not, a question of whether the issue is justiciable by the court. The defenders' primary position was that it was not, although it is fair to say that the point was developed further in the pursuer's response to it than in the defenders' own argument.

[51] So far as I am aware, the issue of justiciability of policy decisions has not been addressed in any previous Scottish cases concerning winter treatment of roads. It has of course been addressed at length in a series of decisions of the House of Lords with other factual contexts. I take as my starting point the following dictum of Lord Browne-Wilkinson in X (Minors) v Bedfordshire County Council [1995] 2 AC 633 at 738:

"Where Parliament has conferred a statutory discretion on a public authority, it is for that authority, not for the courts, to exercise the discretion: nothing which the authority does within the ambit of the discretion can be actionable at common law. If the decision complained of falls outside the statutory discretion, it can (but not necessarily will) give rise to common law liability. However, if the factors relevant to the exercise of the discretion include matters of policy, the court cannot adjudicate on such policy matters and therefore cannot reach the conclusion that the decision was outside the ambit of the statutory discretion. Therefore a common law duty of care in relation to the taking of decisions involving policy matters cannot exist."

It has however, been acknowledged that a distinction between policy decisions on the one hand and operational decisions in implement of policy on the other is an inadequate tool with which to discover whether it is appropriate to impose a duty of care or not: see Stovin v Wise [1996] AC 923, Lord Hoffmann at 951. In Barrett v Enfield London Borough Council [2001] 2 AC 550, Lord Hutton at page 583 expressed the provisional view that

"...the fact that the decision which is challenged was made within the ambit of a statutory discretion and is capable of being described as a policy decision is not in itself a reason why it should be held that no claim for negligence can be brought in respect of it. As I read it this is what is said by the Privy Council in its judgment in Rowling v Takaro Properties Ltd [1988] AC 473, 501G. It is only where the decision involves the weighing of competing public interests or is dictated by considerations which the courts are not fitted to assess that the courts will hold that the issue is non-justiciable on the ground that the decision was made in the exercise of a statutory discretion."

Speeches delivered in Phelps v Hillingdon London Borough Council [2001] 2 AC 619 are to the same effect. At page 652-3, Lord Slynn of Hadley stated:

"It does not follow that the local authority can never be liable in common law negligence for damage resulting from acts done in the course of the performance of a statutory duty by the authority or by its servants or agents. This House decided in Barrett v Enfield London Borough Council that the fact that the acts which are claimed to be negligent are carried out within the ambit of a statutory discretion is not in itself a reason why it should be held that no claim for negligence can be brought in respect of them. It is only where what has been done has involved the weighing of competing public interests or has been dictated by considerations on which Parliament could not have intended that the courts would substitute their views for the views of ministers or officials that the courts will hold that the issue is non-justiciable on the ground that the decision was made in the exercise of a statutory discretion."

In the same case, Lord Clyde said at p.673-4:

"A distinction may be suggested between on the one hand matters of policy or discretion and on the other hand matters of an operational or administrative character. But this kind of classification does not appear to provide any absolute test for determining whether the case is one which allows or excludes a duty of care. The classification may provide some guide towards identifying some kinds of case where a duty of care may be thought to be inappropriate. Where a statutory authority has to make a choice between various courses of action, all of which are within its powers, and the choice involves a weighing of resources and the establishment of priorities, it will in general be inappropriate that someone injured through the particular decision which the authority has made should have a remedy in damages. But it was recognised by Lord Browne-Wilkinson in X (Minors) v Bedfordshire County Council that even in matters of a discretionary character the authority may be liable in damages if its decision falls without the ambit of the discretion, as where the action taken is so totally unreasonable as to amount to an abuse of the discretion."

Commenting on these decisions in Gorringe v Calderdale Metropolitan Borough Council [2004] 1 WLR 1057, Lord Steyn observed (page 1059):

"...Except on a very careful study of these decisions, there is a principled distinction which is not always in the forefront of discussions. It is this: in a case founded on breach of statutory duty the central question is whether from the provisions and structure of the statute an intention can be gathered to create a private law remedy? In contradistinction in a case framed in negligence, against the background of a statutory duty or power, a basic question is whether the statute excludes a private law remedy? An assimilation of the two inquiries will sometimes produce wrong results."

This line of case law has been cited and applied in various Scottish judgments not involving winter maintenance of roads including, for example, Aitken v Scottish Ambulance Service 2011 SLT 822 and Santander UK plc v Keeper of the Registers of Scotland 2013 SLT 362. I take from the dicta which I have quoted that as a general rule a statutory authority will be liable at common law for breaches of duty to take reasonable care. Where, however, the statutory duty imposed upon the authority requires the making of policy decisions involving the weighing of resources and of competing public interests, taking into account considerations which are not properly within the domain of the court, the issue of whether an individual has sustained injury as a consequence of a policy decision thus made will not be regarded as justiciable.

[52] Applying these principles to the circumstances of the present case, it seems to me that the choice before me lies between either (a) holding that the decision of the defenders not to operate a 24-hour winter maintenance service is not justiciable; or (b) holding that that decision is justiciable and then proceeding to assess whether it falls within or outwith the ambit of the discretion afforded to the defenders in the implementation of the statutory duty conferred upon them by section 34 of the 1984 Act. My conclusion is that whichever of these is the correct approach to adopt the pursuer's case must fail. I am inclined to prefer the first approach. It seems to me that the decision whether or not to allocate sufficient resources to permit the operation of a 24-hour winter maintenance service for some or all of the roads classified as Priority 1 routes within the Highland local government area falls within the category of decisions which this court is not fitted to determine. It is, as I have indicated, a decision ultimately taken by elected representatives, on the basis of information and advice from officials, with very significant financial consequences for the budgetary process of allocation of limited resources. It requires the balancing of competing public interests as the expense of a 24-hour service which according to Mr Guest's estimate would increase winter maintenance costs by almost 50%, could not be met out of the existing roads budget and would presumably require to be funded by cutting spending on some other council service. In my opinion this goes beyond the setting of operational priorities and is dictated by considerations in respect of which Parliament could not have intended, when enacting the 1984 Act, that the court would substitute its view for that of the body elected to make such decisions.

[53] If, however, I am wrong in categorising this issue as non-justiciable, I would hold that the defenders' decision not to operate a 24 hour winter maintenance service did not constitute a breach of the common law duty which they owed to the pursuer. On the basis of the case law to which I have referred, it would be a necessary, though not a sufficient, requirement for a finding of breach of common law duty of care by the defenders that I regarded the decision not to operate a 24-hour winter maintenance service as outwith the ambit of their discretion in implement of the statutory duty imposed upon them. I am not persuaded that this requirement is satisfied. I have noted in paragraph 44 above the factors which, according to Mr Guest who, in my opinion, is well placed to speak to this matter, were taken into account by the defenders in deciding in 2003, and in adopting no subsequent change of policy, that they would not allocate resources to the provision of a 24-hour treatment service. Putting the matter at its lowest, these factors do not seem to me to amount to an unreasonable basis for the policy adopted by the defenders. I have also noted that similar policies are applied by a number of Scottish local authorities with significant rural areas, including almost all of those geographically adjacent to the Highland local authority area.

[54] I have summarised the expert evidence led in relation to the defenders' policy but, as I think both Mr Hopwood and Mr Runacres recognised, this is primarily a matter for the court in respect of which expert opinion is of limited value. I accept, as a matter of fact, that no English local authorities operate with an overnight break even though many of them have extensive rural areas. I have already observed that the numerical majority of Scottish local authorities also operate without an overnight break. None of this persuades me that the defenders, as the local authority responsible for the administration of the Highland area with its exceptionally sparse population and long, lightly-used roads, acted outwith the scope of its discretion in deciding not to allocate resources to 24-hour winter treatment of its Priority 1 routes. I accept Mr Runacres' evidence that the 2005 Code was produced in full awareness of the fact that not all Scottish authorities operated an overnight service, and that its terms were not intended to interfere with the discretion of roads authorities in that regard.

[55] In the light of my conclusion that the defenders were not in breach of their common law duty of care to the pursuer by not operating a 24 hour winter maintenance service, I do not consider that it is necessary for me to deal at any length with the pursuer's submission that the defenders were under a duty to monitor road conditions overnight in the Caithness area on 8-9 December 2008 by instigating a patrol of Priority 1 routes or, at the very least, by overnight monitoring of data from the ice sensors. I accept the defenders' submission that they were under no such duty because no treatment could have been instructed by the duty officer to commence prior to 6 am. Mr Moncrieff's instruction on 8 December had been that the gritters should be on standby at 6am to commence treatment immediately if necessary, as in fact occurred. In these circumstances there would have been no practical utility in conducting overnight monitoring by way of vehicular patrol or remote reading of the sensor data.

[56] Even if I had held that the issue of the defenders' overnight treatment policy was justiciable and that the defenders, in order to comply with their common law duty of care to the pursuer were bound to operate a winter maintenance service which allowed for overnight treatment in certain conditions, that would not in my opinion have been sufficient for the pursuer to succeed in this action. It would still have been necessary for the pursuer to prove, on balance of probabilities, that the proper execution of such an overnight service would or ought to have resulted in the A99 at the locus of the accident having been treated with salt prior to the time of the accident and that such treatment would have prevented the accident. When dealing with a similar issue in Morton v West Lothian Council (above), Lord Glennie observed at paragraph 67:

"In order to move from the general duty of care to a specific duty to take care to treat a particular road, and to do so by a particular time in the morning, it would need to be shown that such treatment was not just a requirement of the particular strategy adopted by the roads authority acting reasonably, but that it would necessarily have been a requirement of all strategies within the range of strategies that a roads authority, acting reasonably in response to the predicted and actual weather conditions, could properly have adopted."

[57] In his report, Mr Hopwood expressed the view (at paras 5.7.2 and 9.9) that if the defenders had been monitoring the Ackergill sensor they would have been aware of the presence of ice by 1am and able to take remedial action (i.e. gritting) within two or three hours. In his conclusions, however, he also observed (at para 9.7) that if the defenders had "...followed national advice and gritted roads as soon as the forecasted rain ceased", there was no reason to assume that the road would still have been icy at the time of the accident. On the hypothesis that overnight gritting would have been possible, Mr Moncrieff's evidence was that he would have sent the gritters out once the last band of rain had passed through. It will be recalled that the Met Office evidence indicated that bands of rain passed through or near the area in which the accident occurred prior to 1am and then again between 5am and 6am. That being so, I consider that one of the strategies which could properly have been adopted by the defenders, on the hypothesis that overnight gritting could be instructed, would have been to wait for the latter band of rain to pass through before beginning treatment, which, as Mr Moncrieff pointed out, was in effect what happened. Alternatively, if the defenders had carried out a treatment within 2-3 hours after 1am, it is possible that the salt would again have been washed off by the time of the accident. Having regard to all of these factors, I do not find it proved that, if the defenders had operated and properly executed a 24 hour winter maintenance service, it is more likely than not that the A99 at the locus of the accident would have been treated during the night of 8-9 December 2008 and that such treatment would have prevented the occurrence of the accident.

Contributory negligence
[58] Counsel for the defenders accepted that if I found the defenders liable to make reparation to the pursuer, there was no basis in the evidence for a finding of contributory negligence on the part of the deceased. I agree, and would have made no such finding.

Quantum of damages
[59] I require to express an opinion as to the damages that I would have awarded had I found the defenders liable to make reparation to the pursuer with regard to the deceased's death. The only claim made by the pursuer is made under section 4(3)(b)(ii) and (iii) of the Damages (Scotland) Act 2011, i.e. for grief and sorrow caused by the deceased's death and for the loss of such non-patrimonial benefit as the pursuer might have been expected to derive from the deceased's society and guidance if she had not died.

[60] The pursuer was aged 17 at the date of the accident and 21 at the time of the proof. According to his evidence (which was not challenged, and which I accept), he had been brought up by the deceased and did not recall any contact with his father. When the pursuer was about five or six years old, the deceased married Leon Ryder. Thereafter the pursuer referred to Mr Ryder as his dad and to Mr Ryder's parents as his grandparents. He assumed the surname Ryder. The deceased and Mr Ryder had three children together. From time to time the pursuer stayed with Mr Ryder's parents while a house for the family was being built next door. He considered his home life at that time to have been a normal family life. However, while he was in his teens he had violent arguments with Mr Ryder and was taken into care. This caused a rift with the deceased who did not at first believe an allegation made by the pursuer against Mr Ryder. He remained in care for two years and did not see the deceased during the first year. Thereafter she visited him and took him out most weekends. After his two years in care, the pursuer returned to the family home, although he moved in with his "grandparents" next door for a year to alleviate pressures of space. By this time he considered that his relationship with the deceased had returned to normal. The pursuer commenced a college course in Thurso. He spent time with the deceased and Mr Ryder in the evenings and at weekends. Approximately 11 months before her death, the deceased separated from Mr Ryder and moved to a house in Wick. The pursuer also obtained accommodation in Wick. During this 11 month period the pursuer and the deceased spent more time together and became closer. She was the person in whom he would confide and from whom he would seek advice. Following her death, he did not feel the same without her and felt he had lost motivation. He missed her greatly especially on occasions such as Christmas. He now had no family. He was in touch with Mr Ryder and his family, having previously fallen out with them for a time, but not as often as he would wish. The deceased had herself been adopted, and although she had made contact during her lifetime with her birth mother, the pursuer had had no contact with the latter since the deceased's funeral. The pursuer was now living and working in Banchory and trying to rebuild a life for himself.

[61] The pursuer's account of his relationship with the deceased was largely corroborated, with some differences of detail, by Carolyn Ryder in her affidavit. She considered in particular that, rather unusually, the pursuer and the deceased had grown into a closer relationship as adults than they had had when the pursuer was younger. This relationship was cut short by the deceased's untimely death.

[62] Parties were agreed that I should adopt as a starting point the guidance provided by a Court of Five Judges in Hamilton v Ferguson Transport (Spean Bridge) Ltd 2012 SC 486. In that case the court set aside as excessive jury awards of £120,000 to the 17 year old daughter of a 50 year old woman killed in a road traffic accident and £90,000 to the 57 year old father of a 26 year old man killed in an accident at work. The court discussed at length the difficulties which arose out of attempting to reconcile judicial awards with jury awards for relatives who had suffered similar non-financial losses. In the course of his judgment, Lord President Hamilton, with whom the other members of the court agreed, expressed the view that awards in two recent judicial decisions had "markedly undervalued" relatives' claims. These two decisions were Bellingham v Todd 2011 SLT 1124, in which two young children of a 40 year old man killed in a road accident were awarded £25,000 each and a 25 year old son from a previous relationship was awarded £15,000, and Wolff v John Moulds (Kilmarnock) Ltd 2012 SLT 231, in which three adult daughters of a 66 year old man who died of mesothelioma were awarded £15,000, £15,000 and £18,000 respectively. Subsequently, in McGee v RJK Building Services Ltd [2013] CSOH 10, the two daughters, aged 44 and 37 respectively, of a 71 year old man who died in an accident at home were awarded £35,000 each and a son who accepted that his relationship with the deceased had not been as close as that of his sisters was awarded £27,500.

[63] In the present case the history of the pursuer's relationship with the deceased is somewhat unusual; there have clearly been periods of comparative estrangement. However, those difficulties had come to an end at least a year before the date of the accident, and I accept the pursuer's evidence that at the time of his mother's death they had become close. I take into account that the deceased was only 36, and the pursuer only 17, at the date of the accident, and that the pursuer could and would have expected his mother to remain an important part of his life for many years to come. He is especially unfortunate in having no other close relatives to provide him with support. I accept the submission on behalf of the pursuer that in the circumstances of this case, the best guidance is afforded by Bellingham v Todd, uprated to take account of the observations of the court in Hamilton v Ferguson Transport (Spean Bridge) Ltd. Having regard to the ages of the pursuer and the deceased, I would have regarded £40,000 as an appropriate award under section 4(3)(b), with interest at 4% per annum on one half of that sum from the date of the deceased's death until the date of decree, and thereafter at 8% per annum until payment.

Disposal
[64] In the light of my decision on the merits of the cause, I shall grant decree of absolvitor.