Davies & Anor v Campfield & Anor

This is an appeal from the judgment of HHJ Jarman QC sitting in the County Court at Cardiff. Permission to appeal was refused by Singh J on paper but granted by Newey J at an oral hearing. The first and second claimants run a funeral business from their land, which includes garages, sheds and a paddock. The third claimant is a member of the same family as the other claimants and in one of the sheds he runs a stone masonry business making the headstones.

The claimants and the defendants are neighbours. The case is about water flowing from the defendants’ land to the claimants’ land in the period 2009-2015. The water has flooded on the claimant’s land and the claimants contend that the defendants are liable in nuisance and negligence for the flooding. After a 3 day trial the judge found in favour of the claimants. The defendants appeal.

The claimants’ case was that while water always has inevitably passed from the defendants’ land to the claimants’ land, before 2009 it was not a significant problem. In 2009 the defendants diverted a watercourse flowing through their land and this started to cause serious flooding on the claimants’ land. In 2015 the defendants effectively reversed the change which had been made in 2009 and the flooding stopped within 24 hours. Since then the position has been, in effect, the same as it was before 2009. So, the claimants contend, the flooding in 2009-2015 was caused by the defendants’ diversion works in early 2009.

The defendants’ case was that flooding on the claimants’ land had always happened, before 2009, during 2009-2015 and afterwards. Nothing very significant had changed in terms of the degree of flooding on the claimants’ land. As far as events on the defendants’ land are concerned, before 2009 the water flowing across the defendants’ land had been through an old 300mm clay pipe. A breakage had occurred in 2008 which led to flooding of a footpath on the defendants’ land. Mr Campfield had met representatives of the relevant agencies and agreed how to alter the water flow to stop the flow down the footpath. That was the diversion work in 2009. That alteration of the route the water took was done with the permission of the relevant agency (then Natural Resources Wales NRW). That permission itself gives the defendants a defence. In any case as a matter of causation neither the diversion works the defendants did in 2009 nor the so called reversal works in 2015 caused any significant changes to flooding on the claimants’ land. In particular the defendants said that the claimants’ site was subject to significant flooding for years before the diversion works in 2009 and also, after the reversal works in 2015, there was a serious flood on the claimants’ land in 2016.

The claimants called eight witnesses of fact and relied on a water engineering expert, Mr Haiste. The defendants called five witnesses of fact and also relied on a water engineering expert, Mr Jenkin. The judge had eleven bundles of evidence as well as a large number of photographs and DVDs with video clips.

The two experts met and had an agreed site visit. They produced a joint statement of matters on which they agreed and disagreed. In fact two statements were produced, dated 17^th February and 10^th March 2017. The parties invited me to read them and I have done so.

The water flowing into the land has two sources, a base flow component from an unknown source which may include some mine water and runoff from the urban catchment area upstream. The base flow will be reasonably constant while the runoff will vary with local rainfall.

Mr Haiste believed that the circumstantial evidence of the flooding that occurred during the period in 2009-2015 points to the diversion being the cause. In his opinion the flooding in the claimants’ land in that period was more severe than at the times before 2009 and after 2015. Mr Jenkin believed that the degree of flooding had not changed significantly over time and that the diversion work was not the cause of flooding.

The experts proposed three possible routes taken by the water on the defendants’ land. They can be called routes A, B and C although this labelling is a simplification of the evidence before the court below. Routes A and C were two possible routes of water flow in the pre-2009 period. Route B was the route caused by the diversion works in 2009. There was no dispute about route B but the experts and the parties did not agree about the route before 2009. It could be either A or C. Mr Jenkin’s opinion was that route A was the relevant route. That was along the 300mm clay pipe. Mr Haiste’s opinion was that the relevant route was route C.

During cross-examination it emerged that the experts disagreed about mechanism. They did not agree whether the important factor was the rate of water flow across the defendants’ land or the volume of water. Mr Jenkin’s view was that while flow rate could be important in some cases, in this case given the circumstances what mattered was the volume of water flowing and so the changes could not have made much difference. This supported his view that there was strong evidence that the claimants’ land flooded frequently and over a long period of time. Thus the fact that floods did occur in the 2009-2015 period is not proof that the diversion caused the flooding. There are other variables and the route of flow does not significantly affect the flood risk because the volume would be similar at all times. Mr Haiste’s view in cross-examination was that the flow rate mattered as well. The reason the diversion caused the flooding was because it must have increased the flow rate. In this focus on flow rate Mr Haiste’s opinion had shifted from that expressed in the agreed statement. However while that is a shift, what had not changed was Mr Haiste’s overall opinion about the cause of the flood and the pattern of flooding over time.

In his judgment (paragraph 28) the judge found in the claimants’ favour about the degree of flooding before 2009. In doing this he was rejecting a significant part of the defendants’ case on the facts. He found that the flooding in the period 2009-2015 was more severe. The judge’s finding was that before 2009 while a pond beside the paddock which was on the claimants’ land did increase in size during heavy rain, between 1980 and 2009 the water only reached into the sheds or garages on one or two occasions in the early 1980s and early 1990s and then only at a level of up to two inches. However in the period after 2009 the paddock looked like a lake on occasions and the water rose at times to the level of a couple of feet in the garages (paragraphs 29-32). The judge dealt with the pooling of water in the paddock in 2016 and held that was likely the result of exceptional rainfall in the preceding month (paragraph 37).

The judge also made another important finding of fact. He rejected the defendants’ case that there had been a 300mm clay pipe on the site before 2009. Therefore the water flow in that earlier period could not have been through that pipe (route A). The judge therefore was in effect accepting that the pre-2009 flow must have been via route C although he did not say so.

The judge made a third finding of fact. He rejected the defendants’ case that permission had been given by the relevant agency to divert the water flow in the way it was done.

Turning to the expert evidence, the judge preferred the view of Mr Haiste to the view of Mr Jenkin. He did so in a brief passage, stating (paragraph 45) that “in my judgment the findings of fact I have made fit better with Mr Haiste’s view and accordingly that is to be preferred on the facts of this case.” The findings of fact referred to are the findings of frequent severe flooding in the period 2009-2015, less severe flooding before 2009 and the point on 2016.

The defendants take five points on appeal. First they contend that the judge failed to find as a fact what the drainage position was before 1994. The significance of this point is that flow route C involves a small pond on the defendants’ land. It has a volume of about 50 cubic metres. There was a dispute between the experts about whether and to what extent that pond could attenuate the flow of water. Mr Jenkin for the defendants was sceptical that it could have a significant effect whereas Mr Haiste thought it could. Neither expert had done detailed calculations. The point is that the defendants say it is undisputed that the small pond only came into existence in 1994, so if its presence mattered that could not explain what happened before 1994. Ultimately the defendants’ real submission here is that the judge erred in rejecting the existence of the 300mm pipe. The judge should have found that there was a 300mm pipe and so route A was feasible and (since it is a pipe) water would have flowed quickly along it and so the historic flow rate was not slower than the rate in 2009-2015 and so in turn the diversion works in 2009 did not cause the severe flooding.

Second the defendants contend that the judge erred in rejecting Mr Jenkin’s evidence over Mr Haiste’s evidence. The judge’s approach did not engage with the detail of the expert’s opinions and he did not resolve the argument about whether the important factor was rate or volume. If he had approached it in that way he would have preferred Mr Jenkin’s evidence that volume was the more important factor because that evidence was more cogent and that in turn would have supported the defendants’ case overall. It would have led to the conclusion that the small pond on the defendant’s land cannot have attenuated the flow before 2009 and so that also would have supported the defendants’ case overall.

Putting the second point another way, the defendants’ say that the judge ought to have analysed the mechanism which was actually causing the flooding. That is what the experts disagreed about. If he had done so then he would have found for the defendants’.

Third the defendants contend that causation was not proved. In addition to the first and second points, the real point here is that the defendants submit the judge’s finding about 2016 was wrong. The fact that Wales overall experienced high rainfall in that year is not sufficient to prove that local rainfall in the Neath area in particular was sufficiently unusual to support the judge’s view. The judge had no evidence about local rainfall.

Fourth the defendants contend that the judge erred in his conclusion about the permission. The defendants contend there is no doubt a meeting took place on the site between Mr Campfield and a person from the relevant agency. They argue that the evidence is clear that the works Mr Campfield did were agreed and therefore covered by appropriate permission.

The fifth point is about the third claimant. The defendants do not challenge that he worked as a stone mason on the site but they submit that he is an independent sole trader and is not part of the business of the first and second claimants. The judge did not take that into account when he made the key finding that harm to the 3^rd claimant was reasonably foreseeable. If he had then he would have rejected that part of the claimants’ case.

At this stage it is convenient to mention another ground of appeal raised in the appellants notice. It had been envisaged that the trial judge would conduct a site visit but in the end he decided not to. The defendants contended that was an unreasonable decision by the judge. Singh J rejected that ground on paper and before Newey J the defendants accepted that was not a free standing ground of appeal. Nevertheless they maintained that the fact the judge did not visit the site was a factor which contributed to the errors they contend the judge made relating to the 300mm pipe. I will deal with it on that basis.

I am not persuaded by either the fourth or fifth grounds and I can deal with them shortly first. On the fourth point the evidence before the judge was the testimony of Mr Campfield and the documentary evidence. The defendants are right that the documentary evidence – which consists of letters from various agencies including NRW, the local authority and the official concerned with the right of way, support Mr Campfield’s testimony that a meeting took place between him and a relevant person at the NRW and support his testimony that at that meeting the issue of the diversion was discussed. The problem for the defendants is about specificity. Mr Campfield was clearly asked to take steps to prevent the flooding on the footpath and he did so. That is not the issue. The question is a narrower one. Using the way I have described the water flow routes, his case was that he was given permission to divert the flow onto route B (or at least permission to use any route he chose). That mattered because it is flow along route B in particular which was said to be causing the problem.

However the documentary evidence does not support the view that any permission was in those terms. The relevant paragraph of the judgment is paragraph 39. The judge referred to a letter in 2008, which is the closest document in time to the relevant meeting. The judge rightly held that it does not amount to permission to use Route B. It may refer to using route C, which would be flatly against Mr Campfield but at best it is ambiguous about that and the judge held it was indeed ambiguous. The high point of the defendants’ case is a letter in 2013 from the local authority (NRW’s successor) which states expressly that the officer “agreed works with you”. The problem with that letter is again that it is unspecific. A further letter emerged recently and was placed in the appeal bundles by the defendants without objection from the claimants. However it is similarly unspecific.

What would be required is evidence that what was agreed was specific enough to allow Mr Campfield to use Route B. The judge held that no record of the permission has been found and that it was unlikely that such permission would have been given orally without a follow up confirmation in writing and he held that Mr Campfield’s recollection was inaccurate. The defendants submit that the judge did not find Mr Campfield to be untruthful. That is correct but it does not help. These findings, about the likelihood of a written follow up and about the accuracy of Mr Campfield’s recollection, are findings the judge was clearly entitled to make. On the state of the evidence it is fatal to the defendants’ case.

The defendants’ final point on this is to note that many of the letters relied on do not come from the NRW or its successor body and letters from those sources ought to be given more weight as opposed to letters from different organisations. This is a fair point but it does not plug the gap in the defendants’ case. The letters relied on by the defendants are unspecific as well. I reject this ground.

The fifth ground relates to the third claimant’s position as a sole trader. The relevant passage in the judgment is paragraph 48. Here the judge notes that Mr Campfield accepts that he knew the garages and shed were used in the past for the storage of funeral business vehicles and wood and that he knew that the third claimant’s shed was erected in 2008. The judge makes the point that Mr Campfield said he did not know until much later that this shed would be used for a stone mason business but the judge finds that it was reasonably foreseeable that this shed would be used for business purposes associated with the funeral business. The point (unspoken but obvious to the judge) was that a stone mason business making headstones was plainly associated with the funeral business, and so the third claimant’s case was proved.

In my judgment it makes no difference to this factual analysis that the stone masonry business was being carried on as an independent sole trader rather than as part of the other claimants’ business. What matters was that it was foreseeable the shed would be used for business purposes associated with a funeral business, and that is what was happening. The defendants’ counsel did not cite any authority to support the submission that the fact it was an independent business made any difference as long as the third claimant had brought a claim in his own right. I cannot imagine why it would matter. I reject this ground.

I turn to address the first, second and third points, which are the major issues on this appeal. Although they are distinct questions, they can be taken together. The claimants’ submission was the judge’s judgment was right for the reasons he gave. They submitted that these points are in truth attempts by the defendants to overturn findings of fact by a trial judge which are all findings which were open to the judge on the evidence. They say that is the key difficulty in the way of the defendants’ appeal.

A major part of the defendants’ case is the submission that the judge ought to have grappled with the experts’ rival mechanistic theories first, made findings about that, and then applied them to the rest of the case. This submission has the merit on appeal of being a point of principle about the manner in which the judge approached the task as a whole. If it is a good point then it would undermine the whole approach the judge took. If it is a good point then the judge’s short passage in which he prefers Mr Haiste’s opinions to those of Mr Jenkin would not be adequate.

In my judgment the defendants’ submission is wrong in principle. The judge approached the matter by deciding what the primary facts actually were and then considering how the experts’ opinions stood against those findings of primary fact. He found that there had been less significant flooding up to 2009, serious flooding after the diversion and then the work in 2015 had fixed the problem. Against those primary facts he could test the experts’ rival theories. Mr Haiste’s opinions predicted that those would be the facts whereas Mr Jenkin’s opinions predicted different facts. His volume theory predicted that there would be no significant change over time. That is why the judge was able to prefer Mr Haiste’s evidence in a brief paragraph 45. That is what the judge meant when he said “the findings of fact I have made better fit with Mr Haiste’s view”.

Not only was the judge entitled to take this approach, in my judgment it was a preferable approach to take as a matter of principle. When there is direct evidence about what the degree of flooding actually was at the material times, to start with the rival theories and then use a preference for one theory over another to resolve a dispute about the primary facts risked coming to the wrong conclusion. The theories should be fitted to the primary facts, not the primary facts fitted to the theory.

If one expert’s theory fits the facts and the other’s does not then the court is entitled to prefer the former over the latter. So I reject the defendants’ case that the judge ought to have decided that volume was the relevant mechanism rather than flow rate.

Turning to the judge’s findings of fact, the first aspect was about the position before 2009. There was clear evidence from a number of witnesses. The judge dealt with this thoroughly in paragraphs 14 to 28. He noted that there had appeared to be a stark conflict of evidence based on the witness statements of the defendants’ witnesses but a more nuanced picture emerged in cross-examination. In reply counsel for the defendants referred me to a letter written in 1987 which talked about “much flooding”. The judge does not refer to that letter but he had ample evidence to support the finding he reached. The judge found the flooding prior to 2009 was less significant that in the 2009-2015 period. There can be no realistic challenge to that finding.

The problems in 2009-2015 are not disputed but the defendants do dispute the judge’s findings that the events in 2016 can be explained as due to an extremely wet winter. The point taken is that the only records the judge referred to were rain records for the whole of Wales which cannot tell you whether the specific locality experienced unusually high rainfall.

The judge’s finding about 2016 is made in paragraph 37. The finding is in a context in which he had just held in the previous paragraph that the reversal works Mr Campfield carried out on 28^th February 2015 had an immediate effect. There had been flooding on site just before the works were carried out and the judge was given “before” and “after” photographs showing the disappearance of the water within 24 hrs of the works. There was also evidence from Mr Wright that the water level dropped overnight. So the judge here was finding that the reversal works had made a significant difference very quickly. Of course the point is that this is supportive of the claimants’ case on causation.

Then the judge dealt with the pooling of water in 2016 and held it was likely due to exceptional rainfall. He noted that the records were for the whole country and not for localised events but he also noted that the claimants had given evidence that that winter was extremely wet and they used pumps because they were concerned it might overtop. Finally in this context he made a finding that hardcore which had been laid down by the claimants near the garages was unlikely to explain why there was no further flooding.

In my judgment the attempt to undermine the finding about 2016 must fail. The judge had the limited nature of the records well in mind. National records are not strong evidence but he also had other corroborative evidence. Overall he was entitled to draw the inference he did about 2016.

I turn to deal with the point on the 300mm pipe. The judge drew his conclusions on that at paragraphs 40-41. There were a number of points. First there are photographs which Mr Campfield said was the outlet of the pipe. However the judge found they are not clear. The claimants submit that is supported by the cross-examination of Mr Campfield on the subject in which he seems to have accepted that the photographs were not conclusive. An appellate court is in no position to take a different view on a matter like that. The judge’s conclusion was clearly open to him.

Second Mr Jenkin had said he recalled seeing a rim of a pipe but although a camera was available he did not photograph it. That had occurred when Mr Jenkin visited the site without Mr Haiste. When the two experts visited together the rim could not be found nor any other evidence of the pipe. As the judge said, they even used a shovel to dig to find it.

Third the judge noted that Mr Campfield had tried to find it by digging slit trenches but has failed.

Fourth the judge dealt with an inspection chamber. Inside the chamber you can see that the flow changes direction and points generally towards the claimants’ sheds. That is supportive of the idea of the 300mm pipe. The pipe was said to run in that direction from the inspection chamber. Mr Haiste accepted that the direction was at least generally towards the shed. From the photographs I was shown that seems to be correct.

The judge recognised this was evidence in the defendants’ favour in paragraph 41. His conclusion was:

“Given the changes in configuration over the years to the defendants’ land, there may be an explanation for this configuration, although I agree it tends to support the existence of the 300mm pipe. Given the history of infill and of creating the bank, what Mr Campfield and Mr Jenkin saw which they thought was part of the rim of pipe is not strong evidence to corroborate the existence of this pipe. However in my judgment this evidence taken together is not sufficient to outweigh the absence of any further evidence of the existence of this pipe despite the efforts by Mr Campfield, the authority and the experts to find it. There is no evidence of the connection of such a pipe to the inspection chamber. In my judgment, it is unlikely that what was seen was in fact the rim of a 300mm pipe and unlikely such a pipe existed.”

[The reference to the authority making efforts to find the pipe is a reference to an unsuccessful attempt by the Neath Port Talbot local authority to find the pipe.]

The defendants say that in this decision the judge has not taken into account other evidence in favour of the existence of the 300mm pipe. Before addressing that evidence I will deal with the site visit. In my judgment the fact the judge decided a site visit was not necessary does not undermine his findings about the 300mm pipe in any way. That is because he had the benefit of the evidence of two independent experts who had visited the site themselves and looked for but failed to find it.

Turning to the other evidence relied on, the first topic is the pre-1994 point. The argument is in effect that the route C – which must be the flow route based on the judge’s rejection of the 300mm pipe (route A) – is said by the defendants to need the small pond in order to function in the manner contended for by the claimants with a relatively low flow rate which would explain the reduced flooding before 2009. However that pond was only built in 1994 so the argument goes it cannot explain what happened before that year. I do not accept this submission. It is similar to the defendants’ earlier point that the judge ought to have decided theory first then primary facts second. The judge approached the matter by deciding as a matter of primary fact what had taken place on the site over the years. That included the period in the early 1980s. He was entitled to take that approach. In any event the cross-examination on the flow rate theory and the effect of the small pond did not establish that there was an irreconcilable problem with the theory in explaining events prior to 1994.

A second aspect of this point is that Mr Jenkin’s view was that the ditch through which the water was flowing was in a state of “geomorphological flux”. In other words he viewed it as a relatively young structure, not consistent with that being the path along which water had flowed in the past before 1994. The defendants submit that the judge ought to have dealt with that and accepted Mr Jenkin’s view. However Mr Haiste did not agree. In his view the relatively young channel was running inside an older wider structure which suggested that water had historically flowed along the same line. Also the claimants had referred to a 1960s OS map which has the words “issues” and “sinks” marked in places which were consistent with the idea that water did indeed flow along that path in the past.

A third aspect of the point is that the defendants contend there was evidence of other witnesses of the existence of an outfall. The defendants placed particular emphasis on Mr Robertson and Mr Whitton because they had worked at the site for the previous owner. That previous owner had held all the relevant land and so the individuals could move freely about the entire plot. The judge observed that the statements of Mr Robertson, Mr Whitton and Mr Campfield contained paragraphs which were remarkably similar in wording to one another (paragraphs 16) but the defendants rightly point out that the judge did not find Mr Robertson or Mr Whitton to be untruthful.

The defendants are correct that the judge did not address the geomorphological flux point nor did he record in terms the point that other witnesses had supported the presence of an outfall. However I am not persuaded that any of this justifies interfering with the judge’s finding of fact concerning the 300mm pipe. The judge aware of the point that the 1960s OS map carries the words “issues” and “sinks” although he did not refer to it again in making the finding on the 300mm pipe. The judge heard all three of Mr Robertson, Mr Whitton and Mr Campfield give oral evidence. He noted the nature of their witness statements and also that Mr Robertson was friendly to the defendants (paragraph 21). The major problem the judge had was that the efforts to find the 300mm pipe had failed. That was strong evidence that the pipe did not exist. The judge weighed that against the best objective evidence that the pipe did exist, which was the change in direction, and came down against the defendants. He was entitled to take that approach. That is enough to dispose of this part of the appeal. However I will add that having seen all the evidence I have been shown, and without the benefit of having heard the witnesses which the judge had, I would reach the same conclusion on this point as the judge.

I will dismiss this appeal.