Demetrios Karpasitis v Hertfordshire County Council [2025] EWCA Civ 788

[2025] EWCA Civ 788Case No CA-2023-002207
IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE (KING’S BENCH DIVISION)
VIKRAM SACHDEVA KC
Claim No. QB-2021-003247
Venue Royal Courts of Justice, Strand, London, WC2A 2LLDate 25/06/2025
Before
LORD JUSTICE BEANLORD JUSTICE COULSONLADY JUSTICE ANDREWS
Between
DEMETRIOS KARPASITISAppellantHERTFORDSHIRE COUNTY COUNCILRespondent
Mr M Porter KC and Ms A Karseras (instructed by Fieldfisher) for AppellantMr A Weitzman KC and Mr J Weston (instructed by DWF Law LLP) for RespondentHearing Hearing dates: 20-21 May 2025
Approved JudgmentThis judgment was handed down remotely at 10.00am on 25 June 2025 by circulation to the parties or their representatives by e-mail and by release to the National Archives..............................
[1]On 22 April 2020, a month into the first Covid-19 lockdown, Demetrios Karpasitis, then aged 42, decided to go on a ride on his mountain bike during the afternoon. He followed a familiar route in Hertfordshire which took him onto a path to the east of the A10 dual carriageway, from which it was separated by a grass verge. He travelled north up to the Paul Cully bridge which crosses over the A10 near Cheshunt. As far as the bridge the path was signposted as a shared footpath and cycle path. North of the bridge the path narrowed from 2.5 metres to approximately one metre in width. There was no sign denoting any changed use of the path.[2]Mr. Karpasitis continued north to a quiet road called Anchor Close where he turned around to go home. Heading south towards the Paul Cully bridge at around 6.30pm or 7pm he encountered a jogger also travelling south and wearing headphones. Travelling around 10mph, he took the decision to overtake the jogger. This required him to cycle on the grass verge to the right of the path, with the A10 on the right of the grass verge.[3]Unfortunately, there was a hole in the verge, just south of a road sign, which was sufficient to throw Mr Karpasitis from his bicycle. This caused a complex fracture of the second vertebra. His injuries were so serious that he gave up his job as a social worker.[4]The Respondent is the highway authority in respect of the grass verge (as well as the adjoining footway and carriageway) where Mr Karpasitis fell.[5]Directions were made by consent by Master Gidden for a split trial, with issues of breach of duty and factual and legal causation of some injury (ie liability), together with contributory negligence, being tried first.[6]The trial of these liability issues took place before Mr Vikram Sachdeva KC (sitting as a Deputy High Court Judge) on 14-16 and 23 March 2023. By a reserved judgment handed down nearly 7 months later, on 20 October 2023, the judge dismissed the claim. Mr Karpasitis appeals to this court pursuant to permission granted by Males LJ on 28 February 2024.

Relevant statutory provisions

[7]Section 41 Highways Act 1980 states as follows:
"41.— Duty to maintain highways maintainable at public expense. (1) The authority who are for the time being the highway authority for a highway maintainable at the public expense are under a duty, subject to subsections (2) and (4) below, to maintain the highway. (1A) In particular, a highway authority are under a duty to ensure, so far as is reasonably practicable, that safe passage along a highway is not endangered by snow or ice."
[8]Section 328 states as follows:
"328.— Meaning of "highway"
. In this Act, except where the context otherwise requires, "highway" means the whole or a part of a highway other than a ferry or waterway."[9]Section 329(1) states as follows:
"329.— Further provision as to interpretation. (1) In this Act, except where the context otherwise requires— …cycle track" means a way constituting or comprised in a highway, being a way over which the public have the following, but no other, rights of way, that is to say, a right of way on pedal cycles (other than pedal cycles which are motor vehicles within the meaning of the Road Traffic Act 1988) with or without a right of way on foot… "footway" means a way comprised in a highway which also comprises a carriageway, being a way over which the public have a right of way on foot only… "maintenance" includes repair, and "maintain" and "maintainable" are to be construed accordingly;" ... Section 58 states as follows: "58.— Special defence in action against a highway authority for damages for non-repair of highway. (1). In an action against a highway authority in respect of damage resulting from their failure to maintain a highway maintainable at the public expense it is a defence (without prejudice to any other defence or the application of the law relating to contributory negligence) to prove that the authority had taken such care as in all the circumstances was reasonably required to secure that the part of the highway to which the action relates was not dangerous for traffic. (2). For the purposes of a defence under subsection (1) above, the court shall in particular have regard to the following matters:- (a) the character of the highway, and the traffic which was reasonably to be expected to use it; (b) the standard of maintenance appropriate for a highway of that character and used by such traffic; (c) the state of repair in which a reasonable person would have expected to find the highway; (d) whether the highway authority knew, or could reasonably have been expected to know, that the condition of the part of the highway to which the action relates was likely to cause danger to users of the highway; (e) where the highway authority could not reasonably have been expected to repair that part of the highway before the cause of action arose, what warning notices of its condition had been displayed; but for the purposes of such a defence it is not relevant to prove that the highway authority had arranged for a competent person to carry out or supervise the maintenance of the part of the highway to which the action relates unless it is also proved that the authority had given him proper instructions with regard to the maintenance of the highway and that he had carried out the instructions…"

Issues at the trial

[10]The judge summarized the issues as follows- “36. First, statute. There are three issues relevant to liability… i) Was there a breach of s 41, ie was the highway in a condition which exposed to danger those using it in the ordinary way? ii) Was the accident caused by that breach? iii) Had the Highway Authority made out the statutory defence under s 58, ie of taking all reasonable care?37. Second, common law. i) Can misfeasance or an omission by the Defendant give rise to a common law duty of care? ii) Did the Defendant engage in any positive act or any omission which gave rise to a common law duty of care? iii) If so, what was the scope of that duty of care? iv) Was the Defendant in breach of any common law duty of care? v) Did any breach by the Defendant of a common law duty of care cause the Claimant's accident?38. Third, contributory negligence. If there is liability, did the Claimant's negligence contribute to the accident and if so, what is the appropriate apportionment?” i) Was there a breach of s 41, ie was the highway in a condition which exposed to danger those using it in the ordinary way? ii) Was the accident caused by that breach? iii) Had the Highway Authority made out the statutory defence under s 58, ie of taking all reasonable care? i) Can misfeasance or an omission by the Defendant give rise to a common law duty of care? ii) Did the Defendant engage in any positive act or any omission which gave rise to a common law duty of care? iii) If so, what was the scope of that duty of care? iv) Was the Defendant in breach of any common law duty of care? v) Did any breach by the Defendant of a common law duty of care cause the Claimant's accident?[11]The Council sought to establish their defence to the claim based on s 58 of the 1980 Act by giving evidence of a scheme known as their “Defect Management Approach” which required that footways categorized as “low pedestrian traffic-urban” routes should be inspected once every six months. Paragraphs 6.4-6.6 of the pleaded Defence read as follows:- “6.4 Prior to the date of the alleged accident, the last walked safety inspection of the relevant footway was performed by Jeff Cooke, a very experienced, diligent and conscientious highways inspector employed by Ringway. The walked inspection was performed on 13.02.2020, a little over two months prior to the occurrence of the alleged accident. During that inspection, although Mr Cooke did identify some defects which he characterised as "category 2 (medium)", no defect of any description was observed as being present at the location described in the Claimant's Particulars of Claim or referred to in the Claimant's photographs, also referred to at paragraph 2 of the Particulars of Claim. During the inspection, Mr Cooke recorded the following defect as being present on the footway:- "Failed wearing uneven footway raised root areas. Uneven frames, side out needed and damaged uneven kerbs. Also verge damage." 6.5 The reference to "verge damage" in Mr Cooke's inspection record is to an area where the carriageway meets the verge rather than to the actual verge itself. Furthermore, by selecting Category 2 for the defects he identified, this means the defects were nor regarded as hazardous and instead, would be placed into a pool of Category 2 defects to be monitored by Ringway and considered for repair when other planned works were due to take place. At the post-accident scheduled walked inspection performed by Mr Cooke in August 2020, at a point in time when on the Claimant's case the defect has been filled-in, the same Category 2 defects were noted to be present which indicates the defects were not present on the actual grass verge in February 2020. For the avoidance of any doubt, the Defendant avers that neither they, not their contractors, have carried out any remedial work to the defect identified by the Claimant. 6.6 The Defendant avers that at the date of the last walked inspection on 13.02.2020, no defect of any description was noted as being present at the location referred to by the Claimant in his Particulars of Claim.”[12]On 18 May 2022 Mr Simon Johnson, a partner in the Defendant’s solicitors’ firm, sent to Mr Lewis Ayre of the Claimant’s solicitors the following email:- “Since finalising the disclosure statement, I was yesterday furnished with a further document by Ringway, my client's highways maintenance contractors and so not a party to proceedings. This document was not in my client's control or possession when the disclosure statement was signed. I attach a copy by way of my client's continuing duty of disclosure. I was advised by Ringway as follows:- Please see Jeff’s vehicle data attached for that day. The data shows that Jeff was on site but it does not show where he stopped or for how long. Unfortunately the tracking location is not 100% accurate as it does rely on “pings” to/from satellites, similar to our phones. Although this will be covered in the statement from Jeff Cooke, the inspector who performed the last walked inspection before the accident, the vehicle tracking data relates to the Ringway vehicle he drove in to get to the walked inspection route, for the avoidance of doubt. Mr Cooke will also give evidence about the duration of the walked inspection, start and end points.”

I was advised by Ringway as follows:-

[13]The attached document with vehicle data tracked by GPS showed the movements of what is accepted to have been Mr Cooke’s vehicle between 09.53am and 10.26am on 13 February. Between 09.53 and 10.04 (or possibly 10.06) the vehicle calls at various addresses in the town of Cheshunt. It is then driven south from Cheshunt along the A10 past what was later to be the accident site, then turns round and goes northbound along the A10, again passing the future accident site, after which it turns off the A10 and returns to a Council depot at 10.26. During this journey the vehicle stops once for a period of three minutes. The total distance covered is about five miles.[14]The only other document compiled on 13 February 2020 disclosed in this case was an inspection report form completed by Mr Cooke just after 10.16. It states as follows:- "Location: Low (Rural) Priority Footway Description: Failed wearing uneven footway, raised foot areas, uneven frames, side out needed and damages uneven kerbs Also verge damage Work Location: Various points along section Defect Type: FSU6 Rough/uneven/crackd fwy-2M Priority: CAT2 Medium" This was in identical terms to the content of Mr Cooke’s previous inspection report for the same stretch of road filed in August 2019 (save for the date and some reference numbers). Work Location: Various points along section Defect Type: FSU6 Rough/uneven/crackd fwy-2M Priority: CAT2 Medium"

Defect Type: FSU6 Rough/uneven/crackd fwy-2M

[15]A witness statement of Mr Cooke dated 10 June 2022 was included in the witness statements exchanged between the parties. It included the following:- “3. I am familiar with the alleged accident location as it formed part of my inspection beat and in 2020, I carried out a walked scheduled inspection of both the footpath and the adjacent grass verge on 13.02.2020, approximately 2 months or so before the alleged accident occurred. There is now produced and shown to me at Exhibit JC1 a copy of the inspection record for 13.2.20 and my previous inspection on 16.8.2019. My usual approach to performing the inspections at any location was as follows. I would drive to a particular location from where I would park my vehicle, which I believe was a Vauxhall Corsa. This vehicle was issued to me by my employer and if I was scheduled to do a walked safety inspection on a particular day, I would drive to the start point from my home. On 13.2.20, I was not only inspecting the footway where the alleged accident occurred. That was but one section of my inspection schedule for that day. The whole of the Cheshunt area might take around 14 days to complete. 4. I would have parked my vehicle before getting out of it and starting my walked inspection. I was issued with a hand-held tablet to record any defects as I performed my inspection and I can see that the inspection record for this footway was entered while I was on site at just after 10.16am. I walked in the same direction along the footway as the Claimant says he was cycling, i.e. towards the Paul Cully footbridge, such that the grass verge would be to my right-hand side. 5. When I performed the walked inspection of the footpath (and I was never in any doubt that this was only a footpath as it was not wide enough to accommodate pedestrians and cyclists, and it had been designated by HCC as a footpath, not a shared cycling facility), I was aware that my inspection duties required me to look for defects in the fabric of the footpath together with defects or holes on the grass verge itself. There is now produced and shown to me at Exhibit JC2 a copy of the Claimant's photograph showing a very large hole just in front of the signpost heading in the direction of the Paul Cully footbridge. At the date of the last scheduled inspection of 13/02/2020, I had been working as a highways inspector for approximately 19 years and I was very experienced by that point. ... 8. Based on my experience, as I performed the inspection in February 2020, I anticipate that the defect, had it been present on the date of my last inspection, would have been visible, as the grass would not have grown a great deal over the winter period. Based on the Claimant's photograph, I would not have missed or overlooked such an obvious and glaring defect. 9. If I had seen a hole in the grass verge at this location, I would have investigated it. I would have arranged for a metal footway plate to be placed over the hole. After I had arranged to make the area safe with a footway plate, I would have reported the matter to the Ringway local network technician (for example, Robin Noades, whom I understand has provided a Statement for this matter) for further investigations. As an inspector, my primary role is to identify a defect and others at Ringway [would] implement the repair. Although I would not have expected any pedestrians, and especially cyclists, to be on the grass verge I might have been concerned that if a vehicle happened to park up on the verge at this location, it might present a danger as it would be big enough to swallow someone's tyre. That noted there is no real space to park a vehicle on the verge so close to the signposts and I cannot recall ever seeing vehicles parked on the verge during my inspections. There is no crossing at this point and the verge is adjacent to a very busy carriageway. For the avoidance of doubt, my 6-monthly inspection duties for this location did not extend to the adjacent carriageway. The carriageway was subject to monthly driven inspections by colleagues. 10. Had I seen a defect of the type shown in the Claimant's photograph during the scheduled inspection on 13.2.2020, I would have recorded on my hand-held tablet words to the effect of: - "Category 1, 24-hour – sunken grass verge please make safe with appropriate footplate." ... 14. In terms of my approach to the walked inspection, I would walk from the top of the footpath down towards the Paul Cully footbridge. It would take me around 10 minutes to walk that stretch of footpath. It is not particularly long. The walked inspection would have taken me past the exact location of the hole that can be seen on the Claimant's photograph. Had such a defect been present in the grass verge, I would not have overlooked it despite my thoughts that a pedestrian, or especially a cyclist, were unlikely to be anywhere near the grass verge at this location. I would have taken the steps outlined above to address it.”[16]On 24 February 2023, 18 days before the listed trial date, the Defendant’s solicitors served a notice pursuant to the Civil Evidence Act 1995. This stated:- “TAKE NOTICE that at the trial of this action the Defendant intends to rely upon a statement made by Mr Jeff Cooke dated 10th June 2022, served on 23rd June 2022. AND TAKE FURTHER NOTICE that the said Mr Jeff Cooke will not be called to give oral evidence at trial because he is retired from Ringway, the Defendant's highways maintenance contractors, and will not come to court to give evidence.”

Oral evidence at trial

[17]The Claimant gave oral evidence as to the circumstances of the accident on 22 April 2020, as summarised at the outset of this judgment. The judge accepted that he was a sincere and honest witness.[18]The Claimant's father, Soderis Karpasitis, went to examine the accident site with his son on 6 May 2020 and again on 13 June 2020. On both occasions he took photographs and a video. On 6 May 2020 he described the hole as being "very wide and deep", with the grass quite overgrown. He took a walk along the verge and noticed around 10 holes of varying depth. On 13 June 2020 he found the verge and hole to have been covered in topsoil and the foliage had been trimmed. The edge of the verge where it met the pavement had been squared off.[19]David Crook, a friend of the Claimant’s father, gave evidence that the path was very well known to him, having lived nearby since 1977, and having cycled down the path 3 – 4 times per week for 40 years as part of his regular exercise route. He had always assumed that the path was a cycleway, and stated that the path was frequently used by cyclists.[20]Georgia Jessopp gave evidence that she had been driving southbound on the A10 between 6.30pm and 7pm on 22 April 2020. She noticed a man lying down on the grassy verge to the left of the carriageway, with another man providing assistance to him, so she pulled over. She walked up to Mr. Karpasitis and could see that his helmet had split in half but remained on his head. He had a deep cut on his forehead.[21]She immediately noticed what she described as a "very large" hole on the grass verge, a few metres past a road sign. The hole was, according to her, "very deep", and she said that "if I were to stand in it, it would be up to my knees". She could not be precise but she estimated the width of the hole as "at least a few feet wide". She commented that the grass surrounding the hole was quite overgrown, and despite being so large, the hole itself was quite inconspicuous. She stated that "[y]ou certainly would not be able to see this from the path or even when immediately in front of it on the grassy verge". Ms. Jessopp performed an initial first aid check on Mr Karpasitis. The judge observed that the significance of her testimony was that it was the best evidence of the size of the hole on the date of the accident.[22]Chris Martin was employed by Ringway Infrastructure Service Limited ("Ringway") as the contract manager for grass-cutting services. Since 2012 the Defendant had contracted with Ringway to carry out its responsibilities for highway maintenance. Mr. Martin gave evidence that records showed that the grass was cut in the relevant area on 6 and 7 April 2020 (barely two weeks before the accident). None of the operatives reported the presence of a hole at the accident location. If a hazard had been reported, he would have escalated it to a Category 1 defect call out.[23]Kyle Darcy was an experienced grass-cutting operative who was authorised to use ride-on lawnmowers and would have done so at the location of the accident in April 2020. If he had seen a hole of that size and description, he would have coned it off.

The evidence of Mr Allen-Smith

[24]Chris Allen-Smith was employed by the Defendant. He confirmed that the Council had adopted this Code of Practice at the date of the last walked inspection of the accident location on 13 February 2020. He stated that the Defendant's Defect Management Approach ("DMA") formed part of the contract with Ringway. The DMA distinguished between Emergency, Category 1, and Category 2 defects. Emergency and Category 1 defects were dealt with under the timescales in the DMA. As verges are specifically referred to in the DMA, Mr. Allen-Smith stated that holes on grass verges are a potential issue that he would expect Ringway inspectors to identify and deal with.[25]The contentious part of Mr Allen-Smith’s evidence was that he expressed the view that Mr. Cooke would have flagged the defect as Category 2 medium: indeed, he argued that the DMA did not permit verge defects to fall within Category 1.[26]Mr Vine, an employee of Ridgway, also expressed the view that Mr Cooke would have classified the hole as a Category 2 defect.

The treatment of Mr Cooke’s statement

[27]When Mr Allen-Smith was called to give evidence the witness statement of Mr Cooke had not been adduced by the Defendant: indeed, we are told it had been omitted from the judge’s bundle. It transpired that this was because the Defendant had been keeping its options open. During a short adjournment Mr Weitzman KC (for the Council) informed Mr Porter KC (for the Claimant) that his clients were withdrawing their application to rely on Mr Cooke’s evidence under the Civil Evidence Act 1995.[28]Mr Porter was thus put in the difficult position that Mr Allen-Smith had told the court that even if a hole such as the one which caused the accident had been present on 13 February 2020, Mr Cooke would only have classified it as a category 2 defect. Yet the witness statement of Mr Cooke, which Mr Allen-Smith accepted he had read, said the exact opposite.[29]Mr Porter applied to put the relevant passage from Mr Cooke’s witness statement to Mr Allen-Smith in cross-examination. Mr Weitzman submitted that this was not permitted unless the Claimant adduced Mr Cooke’s witness statement in its entirety. A transcript shows that Mr Weitzman said:- “I am making a decision that I’m not going to rely on any of it. What my learned friend is seeking to do is rely on parts of it and not the rest. He’s put to his election under the CPR. He can put it in and cross-examine Mr Allen-Smith. He can leave it out and not cross-examine him. What he can’t do is have his cake and eat it. It is cakeism.”[30]There were then some inconclusive exchanges. The judge, no doubt wisely, did not give an immediate ruling but invited Mr Porter to consider the position overnight. Mr Porter, after an adjournment, stated:
“I do have instructions to put the statement of Mr Jeff Cooke in evidence with all the caveats I expressed before, but obviously I am not accepting the content of the statement.”
[31]It is convenient to summarize the important evidence on this central issue. Mr Allen-Smith, supported by Mr Vine, said that even if a hole had been present on 13 February 2020 Mr Cooke would have classified it as category 2 (that is to say not a defect requiring urgent action). Mr Cooke’s witness statement stated in paragraph 3 that he had performed a walking inspection on 13 February 2020 and that if there had been a hole there in the dimensions alleged, he would have classified it as category 1 requiring urgent attention. However, the GPS data document indicates that Mr Cooke’s car only stopped for three minutes somewhere on the A10, which was nothing like enough time in which to have carried out a walking inspection.[32]The judge also had expert evidence before him. Dr Robert Davis, a cycling expert called by the Claimant, said that in his view Mr. Karpasitis was behaving as a proficient and law-abiding cyclist in not using the A10 carriageway; in cycling on what he had every reason to suppose was a shared use pedestrian and cycle path, there being no signs advising that the shared path had ended; and in overtaking the jogger in the way he did. The Defendant instructed a cycling expert, Mr Franklin, who attended the trial, but Mr Weitzman did not call him, did not adduce his report in evidence, and objected to Dr Davis being asked to comment on it. Indeed, he argued before us that expert evidence on the subject of cycling was inadmissible.[33]More substantial expert evidence was called from highway engineers: Mr Paul Ketteridge on behalf of the Claimant and Mr Michael Hopwood, the Defendant’s Highway Engineer, on his employer’s behalf. Insofar as the two experts differed the judge preferred the evidence of Mr Hopwood, as he was entitled to do, and I need only mention some of Mr Hopwood’s extensive findings:- “i) The closeness of the sign to the hole and the position of the hole relative to the footway suggest that Mr. Karpasitis must have made quite a sharp right turn and overcome a relatively high and steep difference in level, if he did not cycle between the two posts (as he claims). ii) There was quite a considerable height difference between the footway and the verge – around 300mm. iii) The photographs on 19 May 2020 show a sizeable hole/depression. The 6 May 2020 photographs perhaps do not show a sizeable hole/depression, although they do show something. iv) The likely cause of the hole, arising out of the investigation of 14 June 2022, was established and agreed on site to be caused by burrowing animals and/or the rotting underground remains of a tree stump………………….. x) Despite the legalities of the situation, it is a matter of common experience that cyclists use footways. This is nothing more than a typical situation whereby some cyclists have chosen to use a footway………………… xiii) The analysis of pedestrians and cyclists performed on 24 November 2021, which yielded an average of around 3 users per hour, suggests extremely low usage. There were about the same number of pedestrians as cyclists. xiv) There is no suggestion or evidence that usage by cyclists was an issue known to the Defendant prior to the accident in question, although there is witness evidence that the footway had been used by cyclists for many years. xv) For a defect to be dangerous, there has to be a real and identifiable risk of users encountering it. xvi) The defect photographed on 19 May 2020 was not dangerous, but was a Category 2 rut (ie not immediately dangerous, as with Category 1 defects, but which might become dangerous) to be added to a future programme of work……… xvii) Within the DMA it would correctly be identified as a rut greater than 100mm, and therefore "high" rather than "low" risk. All verge defects are categorised as either Category 2(M) or 2(L). Category 1 defects are those considered to be immediately dangerous and need to be repaired soon. Category 2 defects are those not considered to be immediately dangerous, but which might develop further and become dangerous, or which might benefit from being addressed in the long term to prevent future deterioration. They are typically added to a programme of work to be attended to at a later date, perhaps even months in the future. xviii) If the hole was of a comparable size to that seen on 19 May 2020, it should have been categorised as Category 2(H) at the highest…………… xx) His experience was that defects in grass verges were generally not considered by UK highway authorities to be Category 1 dangerous defects, because the verge is rarely intended to be used to travel along. In locations where there is no adjacent footway, verge defects may be dangerous if there is good reason to believe that the verge is ordinarily being travelled along, eg where it is used by horse riders or there is a nearby public right of way and walkers and/or horse riders are encouraged to use the verge to access that right of way. …………………... xxiii) If the alleged defect was present at the inspection on 13 February 2020 and in the same condition seen in the photographs take on 19 May 2020 he would have expected the inspector to have identified it as a defect. Mr. Cooke himself says he would not have missed or overlooked such an obvious and glaring defect. xxiv) He recorded the joint excavation findings on 14 June 2022 as follows: a) At a depth of about 300mm – 400mm a piece of rotting wood (possibly a remnant of a tree stump) was found. b) At a depth of about 500mm – 600mm what seemed to probably be a rodents' nest was uncovered. c) Up to a depth of about 600mm – 700mm, the soil was relatively loose. d) Below that the material changed to a firm clay. Numerous circular holes were found in the clay, varying from about 30mm to 100mm in diameter. Both he and Mr. Ketteridge considered that the holes were animal burrows. xxv) He considers it very likely that the hole/depression was not present on 13 February 2020, and developed shortly before the accident on 22 April 2020, for the following reasons: a) Mr. Cooke was a very experienced safety inspector, and he would have had a clear and unobstructed view of the verge. The grass on the verge would be likely not to have been particularly long as it would have been very early in the growing season, and he would have been no more than 1.5m from the hole (assuming he walked in the middle of the footway). There was no reason why Mr. Cooke would have missed such a relatively large hole in the verge. b) The grass verge was cut on 6 and 7 April 2020 and none of the grass cutting operatives encountered it. c) The cause of the hole has been established as being burrowing animals and/or an underground rotting tree stump. d) Whichever cause it seems very likely that the ground underneath the verge could have simply collapsed very suddenly. e) Alternatively the presence of animal burrows underground plus nesting material indicate that rodent activity caused a void into which the ground collapsed, rather than rodents using an existing hole to nest.” a) At a depth of about 300mm – 400mm a piece of rotting wood (possibly a remnant of a tree stump) was found. b) At a depth of about 500mm – 600mm what seemed to probably be a rodents' nest was uncovered. c) Up to a depth of about 600mm – 700mm, the soil was relatively loose. d) Below that the material changed to a firm clay. Numerous circular holes were found in the clay, varying from about 30mm to 100mm in diameter. Both he and Mr. Ketteridge considered that the holes were animal burrows. a) Mr. Cooke was a very experienced safety inspector, and he would have had a clear and unobstructed view of the verge. The grass on the verge would be likely not to have been particularly long as it would have been very early in the growing season, and he would have been no more than 1.5m from the hole (assuming he walked in the middle of the footway). There was no reason why Mr. Cooke would have missed such a relatively large hole in the verge. b) The grass verge was cut on 6 and 7 April 2020 and none of the grass cutting operatives encountered it. c) The cause of the hole has been established as being burrowing animals and/or an underground rotting tree stump. d) Whichever cause it seems very likely that the ground underneath the verge could have simply collapsed very suddenly. e) Alternatively the presence of animal burrows underground plus nesting material indicate that rodent activity caused a void into which the ground collapsed, rather than rodents using an existing hole to nest.”

Questions from the judge after the trial

[34]On 21 July 2023, some four months after the conclusion of the trial, an email was sent on behalf of the judge to the parties’ solicitors asking the parties to provide submissions on two questions by 28 July 2023:- “1. What factual findings are sought in relation to the size of the hole on 22 April 2020? 2. Any further observations on the document at TB972, noting that the Claimant has already commented at paras 37 - 42 of his Closing Submissions.” [The reference was to the GPS tracking document][35]Both parties put in further submissions in response to this request.[36]Nearly three months later the judgment was handed down. The judge’s findings of fact were as follows:- “93. Two key questions are what size the hole was on the following dates: i) the index accident 22 April 2020, and ii) when it was last inspected, 13 February 2020. 94. Since no measurements were (understandably) taken on the day of the index accident, and it was not detected during the inspection on 13 February 2020, there is an issue as to how large the hole was on both dates, if indeed it was present at all. 95. The Claimant pleads that it was 0.8m x 0.7m in width, and 0.55m in depth on the date of the accident. He submits that the hole was not materially different in size to that found by his Highways expert Mr. Ketteridge on 19 May 2020. 96. The area was inspected by Mr. Jeff Cooke, an inspector with Ringway, on 16 August 2019 and 13 February 2020. On neither occasion was any hole in the verge in the vicinity of the sign noted. On 1 April 2020 there was a driven inspection of the carriageway. A still from the video indicates the area near the sign, but no hole in the verge is apparent. On 6 – 7 April 2020 the grass was cut by five men, three employed by Ringway and two by another company, using variously a tractor with side-arm, a ride-on mower, and a strimmer. None of them recorded that a hole was present. 97. Direct factual evidence of the size of the hole on 22 April 2020 was given by Ms. Jessopp, an independent witness who saw the hole on the day of the index accident from the footway and said that it was "very deep and I would say that if I were to stand in it, it would be up to my knees. I cannot be precise but I would say it was at least a few feet wide". She did not indicate how high her knees were from the ground, but she appeared to be of normal height for a woman, which approximates to around 45 or 50cm. 98. The Claimant's father, Mr. Karpasitis senior, said following a visit with the Claimant on 6 May 2020, "The pothole itself was very wide and deep". He took photographs and made a video recording. It is possible to see something in the photographs but it is not possible to clearly make out a substantial hole. However the video did show a substantial hole in the verge when the camera neared the hole. On 19 May 2020 Mr. John Ketteridge visited the site. He took photographs, which show a substantial hole, and measured the hole at being 0.8 x 0.7 x 0.55m. 99. On 13 June 2020 Mr. Karpasitis again visited the site with his father. However the hole had been filled in. The Defendant denies having filled it in; it is not clear from the evidence who did. The Defendant denies instructing that the hole be filled in, and has no record of ordering that the work be done. 100. On 25 June 2020 Robin Noades visited the site. He was unable to see the hole. On 6 October 2020 Mr. Ketteridge visited again and found the hole present, in a filled-in state, although he did not measure it. From the photographs he took it looks far smaller than it had appeared on 19 May 2020. 101. On 24 November 2021 Mr. Hopwood visited for the first time. He was unable to find the hole. On 29 March 2022 Mr. Allen-Smith, a senior manager for the Defendant, visited and viewed the hole. He measured it at 0.9m – 1m in diameter and 0.3m deep. 102. On 14 June 2022 there was a joint visit by both sides during which the hole was excavated and the contents examined. It is common ground that the hole was composed of decaying roots from a tree which had been removed, together with holes made by rodents and a rodents' nest. There is no suggestion that the hole resulted from inadequate drainage. 103. The Defendant points out that no measurement was made on the day of the index accident. It submitted that(1) it was for the Claimant to identify the dimensions of any alleged defect at the time of the accident,(2) there is no reliable evidence of the depth of the hole prior to 19 May 2020 (when Mr. Ketteridge visited), and(3) the dimensions of the hole changed over time, and(4) there is no reliable evidence of the depth of the hole and cannot prove that it was dangerous, either on the date of the index accident, or on the date of the inspection. 104. The Defendant's expert, Mr. Hopwood, considers that it is (very) likely that the hole developed just before the accident on 22 April 2020, from the collapse of earth as a result of rodents burrowing. There were many tunnels found in the excavated hole, and nesting materials, and the hole varied in size over time. Mr. Hopwood also considers it relevant that a very experienced highways inspector, Mr. Jeff Cooke, did not detect a hole on 13 February 2020, and the grass cutting operatives did not report any such hole on 6 – 7 April 2020, even after the grass had been cut. 105. The opinion of the Claimant's expert, Mr. Ketteridge, was that the substantial size of the hole meant that it had probably developed over a long period of time, and that it was probably there at the time of the index accident, and at the time of Mr. Cooke's inspection on 13 February 2020, and was of a similar size. The fact that the hole formed due to a combination of rotting tree roots and rodents creating holes does not change that. 106. The best evidence of the size of the hole on 22 April 2020 is the evidence of Georgian Jessopp, who came to help Mr. Karpasitis after the accident, and who gave persuasive and independent evidence of the size of the hole. She said that the hole was "very large" and "very deep", and she said that "if I were to stand in it, it would be up to my knees". She could not be precise but she estimated the width of the hole as "at least a few feet wide". 107. That description is not materially different from the size of the hole as measured on 22 April 2020 as on 19 May 2020, ie 0.8m x 0.7m x 0.55m. On the balance of probabilities I find that the hole was a similar size on 22 April 2020. 108. As to the size of the hole on 13 February 2020, I find that there was no significant hole in the grass verge on that date, for the following reasons. i) First, for reasons that I have stated already, where there is a conflict between them, I prefer the expert evidence of Mr. Hopwood to that of Mr. Ketteridge, although I note that neither expert claimed particular expertise in ground conditions and the formation of holes in grass verges. On this point Mr. Hopwood considered that it was very likely that the hole was created by the collapse into a void created by burrowing creatures between 13 February and 22 April. ii) Second, Mr. Ketteridge's sole reason for considering that the hole could not have developed in the time after the inspection on 13 February was its size. That is not a reason based on the potential mechanisms by which holes develop, but merely an assertion based on a measurement. Mr. Ketteridge provides no reason linked to the mechanisms by which holes form why relatively large holes in grass verges may not develop suddenly. iii) Third, the hole varied in size over short periods of time, having not been detected on 6 October 2020 but again being present on 30 October 2020. It was also present and measured at 0.9 – 1m in diameter and 0.3m deep on 29 March 2022. This variation is more consistent with the hypothesis that the hole could vary in size over short periods of time. 109. Given that a central question is whether Mr. Cooke missed a significant hole on 13 February 2020, it would not be logical to use as a reason why such a hole did not exist on that date the fact that Mr. Cooke would not have missed such a hole. 110. The grass cutting operatives did not find it, but their primary job was to cut the grass, not find holes in the verge, and not all of the means of cutting the grass would necessarily have detected the hole – the tractor with side arm, and the ride-on mower in particular. In determining what was present on 13 February I therefore place little weight on their failure to detect a hole.” i) First, for reasons that I have stated already, where there is a conflict between them, I prefer the expert evidence of Mr. Hopwood to that of Mr. Ketteridge, although I note that neither expert claimed particular expertise in ground conditions and the formation of holes in grass verges. On this point Mr. Hopwood considered that it was very likely that the hole was created by the collapse into a void created by burrowing creatures between 13 February and 22 April. ii) Second, Mr. Ketteridge's sole reason for considering that the hole could not have developed in the time after the inspection on 13 February was its size. That is not a reason based on the potential mechanisms by which holes develop, but merely an assertion based on a measurement. Mr. Ketteridge provides no reason linked to the mechanisms by which holes form why relatively large holes in grass verges may not develop suddenly. iii) Third, the hole varied in size over short periods of time, having not been detected on 6 October 2020 but again being present on 30 October 2020. It was also present and measured at 0.9 – 1m in diameter and 0.3m deep on 29 March 2022. This variation is more consistent with the hypothesis that the hole could vary in size over short periods of time.[37]Dealing first with the claim based on the Highways Act 1980, the judge found that the verge formed part of the highway; that the hole in the verge put the highway into a state of disrepair for the purposes of s 41 of the 1980 Act; and that it was dangerous. He said:- “113. Whether a hole is dangerous is primarily a question for a highway inspector's judgment. They are assisted by the DMA, but it merely provides guidance, and does not release the highway inspector from his duty of making an individual assessment.” 114. The DMA defines Category 1 defects as follows:
"'Category 1 Defects' (2 hours, 24 hours, 5 working days & 20 working days) – are Defects that require prompt attention because they represent an immediate or imminent risk of one of the following: injury to any party using or repairing the highway network…"
'Category 2 Defects' are all Defects that are not categorised as Category 1 Defects. Category 2 Defects will be sub-divided into: Category 2(H) – High Priority Category 2(M) – Medium Priority Category 2(L) – Low Priority" 115. There is a suggestion from the Defendant that Category 1 defects are dangerous and Category 2 defects are not. I do not accept that the question of dangerousness can be ascertained using this categorisation, for some Category 2 defects are in fact remedied, and must therefore be dangerous. If they were not dangerous, there would be no need to remedy them. 116. The question in law is not whether the Defendant has taken reasonable care to secure that the highway was not dangerous for traffic, but whether the duty to repair and keep in repair of the particular part of the highway under s41 has been satisfied. It requires consideration whether the Defendant has put the highway in such good repair as renders it reasonably passable for the ordinary traffic of the neighbourhood at all seasons of the year without danger caused by its physical condition; whether it has maintained and repaired the highway so that it is free of danger to all users who us that highway in the way normally to be expected of them. Foreseeability of harm alone is insufficient to establish dangerousness; the danger created must be the sort of danger which an authority may reasonably be expected to guard against, which can include consideration of the reasonable expectations of the public as to the standard of maintenance of the highway surface. A different standard will normally apply to verges as opposed to the carriageway. 117. The Claimant argues that the defect was dangerous because it would be a danger to a pedestrian crossing the verge or moving onto that verge; and the danger to a cyclist was even greater. 118. The Defendant concedes that if the hole had existed in the carriageway or footway it would have constituted an actionable defect for the purposes of section 41; it argues that the situation is different here because it was in the verge. 119. The Defendant argues that the obligation under s41 only extends to those who are entitled to use the highway for passage and to maintain the highway to avoid danger to traffic being driven in the way normally expected on that highway. The Claimant, being a cyclist, should not even have been cycling on the footway, let alone on the grass verge adjacent to the footway. 120. It is a criminal offence under s72 Highways Act 1835 to "wilfully ride upon any footpath or causeway by the side of any road made or set apart for the use or accommodation of foot passengers", so an argument could be made that there was no s41 duty where a cyclist rides on a footway. The Defendant withdrew the argument that the statutory duty was not owed to those on the verge due to illegality but maintains a "non-normal user" argument in respect of the verge. 121. I find that riding on the grass verge is capable of constituting a normal use of the grass verge. Although not common, and not the primary purpose of the grass verge, it is foreseeable that pedestrians and cyclists may use grass verges for passage. Both Highway experts considered it uncontroversial that the highway authority was obliged to look for defects in verges, and repair them, according to their assessment of dangerousness. The Defendant's DMA refers to defects in the highway, which would not have been relevant had it been thought that verges were not suitable for pedestrians (or cyclists), or that they could never be dangerous. 122. The Defendant also argues that the minimal use of the footway coupled with the unreasonableness of using the verge to overtake a pedestrian means that even a large hole in the verge could not result in a dangerous defect, or at least one that required rapid repair. That is reflected in the fact that the only defects in the DMA which relate to verges, those identifying ruts, can only ever be Category 2 defects, and (it is said) therefore not dangerous. 123. I find that the defect in this grass verge case was dangerous and called for repair, albeit it was not urgent, for the following reasons: i) There is persuasive evidence that footways are often used by cyclists, and specific evidence that this occurred at this location. The Defendant accepts that cyclists used this footway, and Mr. Allen-Smith admitted that it knew as such. Had he not admitted this, I would have inferred that the Defendant must have known that cyclists used the footway. ii) It is clearly foreseeable that pedestrians and/or cyclists may choose to go onto the grass verge, even though the spot involves riding up a small bank of 300mm, and that the particular spot was proximate to a traffic sign. iii) The verge was by the side of a footway, not just on the side of a carriageway, and I find that there was a real risk that a pedestrian might step into the hole or a cyclist may cycle into it from the footway, notwithstanding its proximity to the road sign. iv) Both Mr. Cooke (the highway inspector) and Mr. Martin (contract manager for grass-cutting) were both clear that if there was a substantial hole present on 22 April 2020, it would be Category 1, and had to be repaired quickly. v) It is the kind of damage that members of the public would reasonably expect would be remedied, albeit that due to the infrequent likely usage, such repair would not necessarily have to be urgent. 124. The DMA does not appear to have a category which is directly relevant to the defect in this case, as both Highway experts agreed. In my view a rut exceeding 100mm is not sufficiently similar to the defect in this case to be of great use in assessing dangerousness. 125. I find that, on the specific facts of this case, that stepping onto or cycling onto the grass verge in this case is a normal user of the highway, and that the public have a reasonable expectation that substantial holes in such a verge would be repaired within a reasonable period of their discovery. Such substantial holes constitute the sort of danger which an authority may reasonably be expected to guard against. I do not accept that this finding would place an unrealistic or disproportionate burden on highway authorities' limited budgets. 126. In all the circumstances I conclude that the defect was dangerous as of 22 April 2020. Causation 127. But for the breach of section 41, would the accident have occurred? And if there had been signposts requiring cyclists to dismount, would the Claimant have done so? 128. The question is whether, but for the defect in the highway, the accident would not have happened. It is clear that the defect in the highway caused the accident, by causing Mr. Karpasitis to fall from his bicycle. 129. As to the common law claim, if a "No cycling" sign had been erected just north of the Paul Cully bridge, I find that the Claimant would not have attempted to cycle on the footway north of the Paul Cully bridge, and the accident would not have occurred. 130. Accordingly, causation in respect of both the statutory and common law claims is made out.” "'Category 1 Defects' (2 hours, 24 hours, 5 working days & 20 working days) – are Defects that require prompt attention because they represent an immediate or imminent risk of one of the following: injury to any party using or repairing the highway network…" 'Category 2 Defects' are all Defects that are not categorised as Category 1 Defects.

Causation

(M) – Medium Priority Category 2(L) – Low Priority" Causation

[38]However, the judge went on to find that the s.58 defence had been established. His reasoning is at [131] – [139]. The critical element of that analysis was the finding that the hole was not present on 13 February [133]. That finding in turn seems to rely on Mr Cooke’s witness statement and the brief record of the inspection that day [138], and the judge’s decision not to place any weight on the GPS data because that would constitute “impermissible non-acceptance of part of a party’s own witness evidence” [136].[39]On the alternative claim at common law, the judge said:- “140. For liability in common law to arise, the Defendant must have committed a positive act which adversely affects the risk to users of the highway: Thompson (supra); Gorringe (supra); Yetkin (supra); Robinson v Chief Constable of West Yorkshire [2018] UKSC 4 [2018] AC 736. There is no authority suggesting that misfeasance or an omission by the Defendant gives rise to a common law duty of care. ……………………….145. There is no evidence that the Defendant has committed any positive act. Omitting to erect a sign indicating that cycling is not permitted is an omission. If that is incorrect, the positive act of constructing a shared facility south of the Paul Cully bridge without erecting a sign prohibiting cycling north of the bridge was not negligent, for given that the general presumption is that paths cannot be used for cycling absent expression permission; that the use of end signs is discretionary and indeed used as an exception rather than the rule; that the change in width of the path, together with its undulations and the lack of shared use signs were sufficient to indicate to a reasonable cyclist that the path was no longer one which could be used for cycling. It is not similar to failing to put up a "No Entry" sign at one end of a one way street (per Lord Brown's example in Gorringe): there are presumptive rules about not cycling on footpaths absent express authorisation which cyclists are taken to know.146. Nor can this case be considered to be comparable to Yetkin on the facts. In Yetkin the Defendant created the danger which caused the accident; whereas here the criticism is a failure to warn.147. I accept the Defendant's submissions that the use of end signs is not mandatory, indeed it is discouraged in the guidance; nothing in the guidance requires an end sign at a point like this; the reduction in width of the path implies a change in character and of use; and the absence of signs indicating shared use north of the Paul Cully bridge implies that the path is not shared use. There is therefore no liability in negligence in this case.”[40]The result was that the judge dismissed the claim. He stated that if he had found the Defendant liable he would have made a finding of one third contributory negligence against the Claimant.

The appeal to this court

[41]The Claimant’s grounds of appeal are brief but comprehensive:
“1. The learned deputy judge erred in finding that the Defendant Highway Authority had established the statutory defence afforded by section 58 of the Highways Act 1980. This ground raises an appeal on a point of law and against findings of fact. 2. The learned deputy judge erred in finding that there was no common law duty to sign the end of the cycle path. This ground raises an appeal on a point of law.”
[42]In granting permission to appeal Males LJ observed:- “1. The finding that the hole was not in existence on 13th February 2020 and only came into being thereafter appears to be critical. If that is correct, there was a basis for the judge's conclusion that the applicant had a defence to the section 41 claim; it would also appear to provide a defence to the common law claim for negligence in that the applicant did have and implemented a system for regular inspection of the verge, although given his conclusion on omission the judge did not need to get that far. 2. Although the judge expressly did not ground his finding on Mr Cooke's evidence (see at [109]), it is hard to think that this did not play a part. After all, if Mr Cooke carried out an inspection and did not find the hole, that is the best possible evidence that it was not there. Conversely, if he did not carry out the inspection, the position is (to say the least) much less clear. 3. It may be that the applicant is stuck with Mr Cooke’s evidence as a result of having decided to put his statement in evidence. However, Mr Cooke’s stated reason for declining to give oral evidence was feeble and the contemporary records (which are otherwise unexplained) on their face appear to show that his evidence was not true. 4. In these circumstances an appeal would appear to have a sufficient prospect of success to merit permission being given, particularly when account is taken of the delay in producing the judgment. That said, the delay does not appear to have affected the judge's reasoning, although the judgment does show some signs of having been prepared at different times. 5. Further, the case raises issues of the responsibility of a highway authority which could suitably be considered by the Court of Appeal.”[43]The Council’s solicitors served a Respondent’s Notice asking the court, if necessary, to uphold the order of the judge dismissing the claim for the following different and/or additional reasons:- “1. The Learned Judge was wrong at paragraph 91 of his judgment in finding that the opinion evidence of Dr Robert Davis as to what a proficient and law-abiding cyclist would or would not have done was admissible. This was not an opinion founded on a specialised body of knowledge but on common practice/usage [Liddell v Middleton [1996] PIQR 36].” 2. The Learned Judge was wrong to find that the relevant part of the grass verge was in disrepair for the purposes of s.41 Highways Act 1980. The Respondent relies, inter alia, on the following:a. The Learned Judge misdirected himself as to the law at paragraph 115 of his judgment that dangerousness was met because of a need to repair. The need for repair is not the test [Mills v Barnsley MBC [1992] PIQR 291].b. The Learned Judge’s finding that the grass verge was in need of a non-urgent repair [see paragraph 123] was inconsistent with his conclusion that the relevant area was dangerous for the purposes of s.41 Highways Act 1980.c. The Learned Judge at paragraphs 121 and 123 was wrong as a matter of law to rely upon the foreseeability of injury as a basis for a finding of dangerousness under s.41 Highways Act 1980.d. The Learned Judge was wrong to find at paragraph 125 that a cyclist using the grass verge is a normal user of the highway [Burnside v Emerson [1968] 1 WLR 1490]. 3. The court’s dismissal of the claim in common law negligence was correct for the following additional and/or alternate reasons: a. A common law duty of care was not owed because any alleged positive act did not create a danger. b. A common law duty of care was not owed because any relevant risk that arose was an obvious one against which the Defendant was not obliged to protect the Claimant at common law c. Even if a common law duty of care was owed to the Claimant, there was no liability in negligence because the scope of any duty did not extend to the risks arising from the Claimant electing to ride on the grass verge. 4. The Learned Judge’s apportionment of contributory negligence at 33% was wrong. The finding at paragraph 149 that it was reasonable for a cyclist to cycle on the grass verge, upon which the apportionment was predicated, was based on the inadmissible evidence of Dr Davis. The appropriate reduction should have been 50% as the Claimant was equally responsible for his injuries.” a. The Learned Judge misdirected himself as to the law at paragraph 115 of his judgment that dangerousness was met because of a need to repair. The need for repair is not the test [Mills v Barnsley MBC [1992] PIQR 291]. b. The Learned Judge’s finding that the grass verge was in need of a non-urgent repair [see paragraph 123] was inconsistent with his conclusion that the relevant area was dangerous for the purposes of s.41 Highways Act 1980. c. The Learned Judge at paragraphs 121 and 123 was wrong as a matter of law to rely upon the foreseeability of injury as a basis for a finding of dangerousness under s.41 Highways Act 1980. d. The Learned Judge was wrong to find at paragraph 125 that a cyclist using the grass verge is a normal user of the highway [Burnside v Emerson [1968] 1 WLR 1490]. a. A common law duty of care was not owed because any alleged positive act did not create a danger. b. A common law duty of care was not owed because any relevant risk that arose was an obvious one against which the Defendant was not obliged to protect the Claimant at common law c. Even if a common law duty of care was owed to the Claimant, there was no liability in negligence because the scope of any duty did not extend to the risks arising from the Claimant electing to ride on the grass verge.[44]This issue was raised in the Respondent’s Notice but it is logically antecedent to the Claimant’s main ground of appeal.

The judge set out some basic principles of the law on s 41 in terms which are not disputed:

“17. Section 41 creates a duty to put a highway in such good repair as renders it reasonably passable for the ordinary traffic of the neighbourhood at all seasons of the year without danger caused by its physical condition: Burnside v Emerson [1968] 1 WLR 1490, 1496 – 7 per Diplock LJ.” 18. The duty is reasonably to maintain and repair the highway so that it is free of danger to all users who use that highway in the way normally to be expected of them – taking account of the traffic reasonably to be expected on the particular highway……In every case it is a question of fact and degree whether any particular state of disrepair entails danger to traffic being driven in the way normally expected on that highway: Rider v Rider [1973] 1 QB 505, 514F – G. 19. Section 41 creates an absolute duty to maintain the highway, which includes the work of repair and the taking of measures which will obviate the need to repair, to forestall the development of a defect in the road which will, if allowed to develop, require remedial action. The standard of maintenance is measured by considerations of safety. The obligation is to maintain the road so that it is safe for the passage of those entitled to use it: Goodes v East Sussex County Council [2000] 1 WLR 1356 per Lord Clyde at p1368H – 1369A.”
[46]The Council submits that for a cyclist to ride on the verge cannot be “ordinary traffic”, the phrase used in Burnside v Emmerson. There are no appellate cases about liability for defects in verges. Mr Weitzman placed reliance on the decision of Scott Baker Jin R (Kind v Newcastle-Upon-Tyne Council [2001] EWHC 616 (Admin). Mr Kind had sought and obtained an order from the Newcastle Magistrates Court requiring the Council to “repair” the whole of the unmade verge of a country road to such a standard that it would be possible for vehicles and indeed HGVs to use it to pass cyclists and pedestrians. The Crown Court allowed the local authority’s appeal from this order and Scott Baker J dismissed the appeal. He said that:- “The finding that the verges are not suitable for all traffic does not amount to a finding that Prestwick Carr is out of repair. ... Prestwick Carr is in a reasonable state of repair to serve the ordinary traffic using it. ... The Crown Court had to look at the whole highway. This it did. It correctly found that the highway included the whole of the width between the enclosures, but that does not mean that the council had an obligation to level everything and make all parts of it like a motorway, flattening banks so that vehicles could pass over them. The question was whether the highway as a whole was reasonably passable for ordinary traffic.” This decision is a matter of obvious common sense, but it provides no support for the Council’s submission.[47]According to Mr Porter it was not argued before the judge that there can never be section 41 liability for a verge. Be that as it may, I am not impressed with the suggestion from the Respondents that a finding that a verge of this kind comes within section 41 would lead to an intolerable burden on highway authorities. A prima facie case of dangerousness under s 41 will not impose liability if the highway authority establishes the statutory defence under s 58. That section (cited above) requires the court to take into account the character of the highway and the traffic reasonably expected to use it; the standard of maintenance appropriate to such a highway; and the state of repair which a reasonable person would expect.[48]I accept Mr Porter’s submission that:- “Disrepair is fact sensitive. It is self-evident that a grass verge need not be maintained to exactly the same standard as a paved carriageway. Everything will depend upon the circumstances. The Judge was entitled, correct, and indeed bound, to find that:(1) The footpath was used as a cycle track, to the Respondent’s knowledge.(2) The Respondent had encouraged its use by cyclists by promulgating a cycling map depicting the footpath as a cycle route(3) The Respondent had chosen not to sign the end of the cycle track at the Port Culley Bridge which would have deterred law abiding cyclists, like the Appellant, from cycling on the footpath.(4) The footpath was too narrow to enable cyclists to pass each other or pedestrians in safety without going onto the grass verge.(5) In the circumstances it was reasonable to cycle not just on the footpath but on the grass verge.(6) Cycling on the verge accordingly constituted ordinary traffic and a large hole constituted a disrepair. Accordingly: (a) The ‘very large’ ‘very deep’ defect as described, particularly by the independent witness Georgia Jessop, was a dangerous defect. (b) The Judge was correct to find that the defect was dangerous and called for repair. He ought to have found that it called for urgent repair. Certainly, Mr Cooke would have dealt with it urgently. The distinction does not matter in this case as the accident occurred more than 2 months after the date of the last purported inspection and the Respondent cannot assert that a dangerous defect could have been left for that period. (c) The foreseeability of cyclists riding on the verge was clearly relevant to what constituted ordinary traffic. The Judge was correct to find that it was foreseeable that a defect of the dimensions described by Georgia Jessop would cause injury to a cyclist. It reinforces his finding that the defect was dangerous to ordinary traffic and represented disrepair of the highway. (d) In the circumstances a cyclist using the grass verge to pass another highway user was plainly foreseeable and represented ordinary use of the highway.” (1) The footpath was used as a cycle track, to the Respondent’s knowledge. (2) The Respondent had encouraged its use by cyclists by promulgating a cycling map depicting the footpath as a cycle route (3) The Respondent had chosen not to sign the end of the cycle track at the Port Culley Bridge which would have deterred law abiding cyclists, like the Appellant, from cycling on the footpath. (4) The footpath was too narrow to enable cyclists to pass each other or pedestrians in safety without going onto the grass verge. (5) In the circumstances it was reasonable to cycle not just on the footpath but on the grass verge. (6) Cycling on the verge accordingly constituted ordinary traffic and a large hole constituted a disrepair. (a) The ‘very large’ ‘very deep’ defect as described, particularly by the independent witness Georgia Jessop, was a dangerous defect. (b) The Judge was correct to find that the defect was dangerous and called for repair. He ought to have found that it called for urgent repair. Certainly, Mr Cooke would have dealt with it urgently. The distinction does not matter in this case as the accident occurred more than 2 months after the date of the last purported inspection and the Respondent cannot assert that a dangerous defect could have been left for that period. (c) The foreseeability of cyclists riding on the verge was clearly relevant to what constituted ordinary traffic. The Judge was correct to find that it was foreseeable that a defect of the dimensions described by Georgia Jessop would cause injury to a cyclist. It reinforces his finding that the defect was dangerous to ordinary traffic and represented disrepair of the highway. (d) In the circumstances a cyclist using the grass verge to pass another highway user was plainly foreseeable and represented ordinary use of the highway.”[49]I would add one further point. It was readily foreseeable that a pedestrian using the footway might go on to the verge (to avoid someone coming in the other direction), step into the hole and be injured. As it happens Mr Karpasitis was a cyclist rather than a pedestrian, but the dangerous nature of the defect applied to both.[50]It is convenient to mention at this point the first paragraph of the Respondent’s Notice. This argued that the expert evidence of Dr Davis to the effect that the Claimant was acting entirely properly in riding on the verge to pass the jogger was inadmissible as being on a topic within the court’s ordinary knowledge. It is extraordinary that such an argument should be run in the circumstances of this case. The Master made an order well before the trial that each party should be permitted to adduce the evidence of one expert on cycling. (Indeed, according to Mr Porter the suggestion had come first in correspondence from the Council’s solicitors.) There was no appeal against the Master’s order. It was too late at trial and is far too late on this appeal for the Respondents to submit that the evidence should never have been admitted in the first place.

The section 58 defence and the treatment of Mr Cooke’s statement

[51]At the heart of the argument before us was the question of whether, by putting passages from Mr Cooke’s witness statement to Mr Allen-Smith (and to a lesser extent to Mr Vine) in cross-examination, Mr Porter was bound to accept at face value the whole of Mr Cooke’s witness statement. The judge found in effect that he was; that no weight should be given to the GPS vehicle tracking data; and that Mr Cooke had, therefore, carried out a walking inspection on 13 February 2020. In doing so he accepted Mr Weitzman’s submission that this followed from the decision of this court in Property Alliance Groupv Royal Bank of Scotland plc [2018] EWCA Civ 355; [2018] 1 WLR 3529. Before considering what the Property Alliance case establishes I would observe that, with respect, the approach of the judge in paragraph 136 of his judgment overlooks a basic principle of fact-finding. The usual approach is to start by looking at contemporaneous documents, especially those from an impartial source, before considering what witnesses have to say about them. The Defendant appears to have persuaded the judge to do the opposite. Mr Weitzman’s submission that the GPS document “did not speak for itself”, and could only have been of any value if supported by expert evidence (for which permission would have had to be obtained in advance of the trial), involves an extraordinarily technical approach to the admissibility of hearsay evidence, which might have been valid in the criminal courts before the passing of the Criminal Justice Act 2003, but is inappropriate to an ordinary civil dispute such as the present one.[52]That would be so even if the GPS document had stood alone; but it did not. It had been disclosed to the Claimant’s solicitors, as noted above, under cover of an email from the Defendant’s solicitors saying that it “showed Jeff’s vehicle data” for the day in question and that Mr Cooke would deal with it in his witness statement. Taken together, the GPS document and the solicitor’s email demonstrate an overwhelming prima facie case that on 13 February 2020 Mr Cooke drove his vehicle along the A10 for a period of about ten minutes, during which the car only stopped once for three minutes.[53]Mr Cooke’s witness statement does not address the GPS document at all. Mr Weitzman told us on instructions that this was because, by oversight, it was not shown to him. The oversight was extraordinary. It seems to me an inevitable inference that no walked inspection of the relevant footway and grass verge took place on 13 February 2020. Small wonder then, that Mr Cooke did not wish to attend the trial, or that Mr Weitzman withdrew the application to rely on his witness statement.[54]It is, no doubt, the general rule that a party cannot put in part of a witness statement served by an opposing party since that, as this court noted in Property Alliance, would be to contradict the terms of CPR 32.5(5). Also, as Mr Weitzman correctly submits, it is a general evidential rule that a party cannot impugn the truthfulness of its own witness. While it is entitled to call other evidence that may provide a different or inconsistent picture, it generally cannot go further and seek to show that the witness cannot be believed on their oath (see Ewer v Ambrose (1825) 3 B. & C. 746, cited in McPhilemy v Times Newspapers Ltd [2000] 1 WLR 1732). But this proposition does not give an answer to the question of whether the only way that Mr Porter could challenge Mr Allen-Smith’s evidence as to what Mr Cooke would have done was by putting the whole of Mr Cooke’s witness statement in evidence. In Property Alliance the issue was whether a party could put in evidence part of a witness statement disclosed by the other side, not whether the party could put part of such a statement to another witness in cross-examination.[55]A more useful case is Anonima Petroli Italiana S.P.A. And Neste Oy -v- Marlucidez Armadora S.A.“The Filiatra Legacy” [1991] 2 Lloyd’s Rep 337. The plaintiffs sought to recover the value of oil which went missing during a shipment from Turkey to Italy. They put in the written statement of a Captain Bellucci, who had carried out tests when the oil was disembarked at a port in Italy. Over the course of the trial they sought to distance themselves ever further from his evidence, so that by closing they were submitting that he had not carried out the tests he said he had, a submission which the trial judge accepted when finding for the plaintiffs.[56]When considering the approach to be taken to Captain Bellucci’s evidence this court identified the general evidential rule: a party cannot impugn the evidence of its own witness unless that witness is deemed to be hostile, although by reference to other evidence in the case they might submit that he was mistaken. The position of the party was to be distinguished from that of the judge:
“It may perhaps be the case that there is a distinction between what is permissible for a party and what is within the powers of the judge. As to the latter, we recognise that in principle an adversarial system of civil procedure requires the judge to decide the issues placed before him, on the evidence given at the trial, so far as such a decision is necessary for a resolution of the dispute. It is not in general for the judge to raise issues which the parties themselves have chosen not to raise. Nevertheless, the object of the procedure is to achieve justice and we should be slow to hold that, if the judge were to conclude that the remainder of the evidence made no sense unless an item of unchallenged evidence was untrue, he would have no power to satisfy himself of the position, subject of course to every precaution necessary to avoid unfairness to the parties and to the witness himself. Whether such a power ought in a given situation to be exercised is a different matter altogether and, no doubt, the occasions when this would be proper would be rare indeed.”
[57]In the present case Mr Porter should have been allowed to put it to Mr Allen-Smith that the evidence he had given about what Mr Cooke would have done had the hole been present on 13 February 2020 (in particular, that he would have classified it as category 2) was misleading in the light of the signed statement of Mr Cooke – which Mr Allen-Smith admitted he had read – which stated the contrary. Mr Allen-Smith had created a misleading impression which Mr Porter should have been allowed to correct. Any other approach would conflict with the overriding objective of the Civil Procedure Rules of enabling the court to deal with cases justly.[58]On a proper application of The Filiatra Legacy Mr Porter was entitled to submit to the judge that Mr Cooke’s written evidence, to the extent that he claimed to have carried out a walked inspection in February 2020, was inconsistent with the contemporaneous evidence of the GPS tracking document; and, in the absence of any explanation from Mr Cooke, should be rejected. In fairness to Mr Cooke, I note that the greater part of his witness statement contains phrases such as “I would have done” X or Y, rather than “I remember that I did” X or Y. The witness statement, made more than two years after the accident, is not couched in terms of confident recollection of an actual event. It was unnecessary for Mr Porter to argue that the GPS tracking document showed that he was a liar, only that he was mistaken. There was more scope for a finding that Mr Allen-Smith had been seeking to mislead the court; but that was not a reason to prevent a challenge to that part of his evidence except on the basis of requiring the Claimant to treat Mr Cooke as his own witness.

Conclusion on the Section 58 defence

[59]As I have already indicated, Mr Porter should not have been required to “put in” the witness statement of Mr Cooke in order to challenge Mr Allen- Smith on his assertion that Mr Cooke would have categorised the hole as “non-urgent”. But the fact is that he did put it in. Once he had done so, neither the Property Alliance case nor any other authority required the judge to treat its entire contents at face value. That part of Mr Cooke’s witness statement which asserted that he had carried out a walked inspection of the relevant part of the footway and verge on 13 February 2020 should have been treated by the judge as manifestly incredible (see eg per Newey LJ in Kireeva v Bedzhamedov [2022] EWCA Civ 35 at [34]) and given no weight since: a) it was flatly contradicted by the GPS tracking document, which the Council’s solicitors had said showed Mr Cooke’s movements on the day in question; b) Mr Cooke’s brief inspection report was in identical terms to that compiled six months earlier; and c) Mr Cooke had expressed unwillingness to attend the trial for no better reason than the fact that he had retired.[60]Without this evidence from Mr Cooke, the whole basis of the Council’s defence under s 58 unravels. Mr Hopwood may well have been right to say that the hole was caused by burrowing rodents. But the cause of the hole seems to me of minimal importance if the Council had not discharged the burden of showing that there had been any inspection of the relevant area for nine months. Where Mr Cooke’s evidence plainly was capable of belief was his assertion that if, in the course of a walked inspection, he had seen a hole in the verge of anything like the dimensions of the hole into which the Appellant rode he would have categorised it as requiring urgent attention. I consider that the judge ought to have held that the Council had failed to establish their statutory defence to the claim based on s 41.[61]In those circumstances, if Coulson and Andrews LJJ agree, I do not regard it as necessary to extend an already long judgment by considering whether the judge was right to reject the alternative claim based on common law negligence.

Contributory negligence

[62]The final paragraph of the Council’s Respondent’s Notice criticises the judge’s finding that, if the Appellant had succeeded on primary liability, there would have been a deduction of 33% for contributory negligence: it is said that the apportionment should have been altered in the Council’s favour. There is a long line in authority, culminating in the Supreme Court decision in Jackson v Murray [2015] UKSC 5 that it is extremely difficult to upset a judge’s apportionment of primary liability and contributory negligence. This is not one of those exceptional cases where such an alteration would be justified.

Disposal

[63]I would allow the appeal and enter judgment for the Appellant for damages to be assessed, subject to a deduction of 33% for contributory negligence.[64]I agree.[65]I also agree.