England and Wales Court of Appeal (Civil Division) Decisions

You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> Sutradhar v Natural Environment Research Council [2004] EWCA Civ 175 (20 February 2004)
URL: http://www.bailii.org/ew/cases/EWCA/Civ/2004/175.html
Cite as: [2004] EWCA Civ 175

[New search] [Printable RTF version] [Help]

JISCBAILII_CASE_TORT

		Neutral Citation Number: [2004] EWCA Civ 175
		Case No: A2/2003/1148

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM MR JUSTICE SIMON

Royal Courts of Justice
Strand, London, WC2A 2LL

20th February 2004

B e f o r e :

LORD JUSTICE KENNEDY
LORD JUSTICE CLARKE
and
LORD JUSTICE WALL
____________________

Between:

	Binod Sutradhar	Appellant
	- and -
	Natural Environment Research Council	Respondent

____________________

M. Beloff QC, C. Pugh and B. Cooper (instructed by Manches, Oxford) for the Appellant
Lord Brennan QC and R. Hermer (instructed by Leigh Day & Co, London) for the Respondent
Hearing dates: 13th and 14th January 2004
____________________

HTML VERSION OF JUDGMENT
____________________

Lord Justice Kennedy:

This is a defendant's appeal from a decision of Simon J who on 8^th May 2003 dismissed an application by the defendant to strike out the claim because, it is said, it discloses no good cause of action, or to give summary judgment in respect of it because the claimant has no real prospect of establishing a duty of care. For convenience I will continue to refer to the parties in this judgment as the claimant and the defendant.

The proper approach to such an application.

For present purposes no distinction need be drawn between the application to strike out and the application for summary judgment, and there is no dispute as to the approach which should have been adopted by the judge, and should now be adopted by this court. It is set out in paragraphs 4 to 13 of the judgment in the court below, and in the skeleton arguments prepared by both parties for this court. Without reference to authority it can be summarised as follows –

(1) The court must ask itself whether the claimant has a realistic, as opposed to a fanciful, prospect of success. For this purpose it must generally be assumed that the claimant can prove his allegations of fact, except where the defendant is able to show, without a mini-trial, that the claimant's prospects of proving necessary facts are minimal. The exception is not relevant in the present case.

(2) The claimant does not have to show that he will probably succeed, and where the case is complicated it may be difficult for the defendant to demonstrate that the claimant has no more than a fanciful prospect of success, but if that is demonstrated it is in everyone's interests that the litigation be brought to an end.

The facts as alleged.

The claimant is about 45 years of age and lives in Bangladesh. The defendant is a United Kingdom statutory body, and the British Geology Survey (BGS) is a department of the defendant. At the material time BGS was a world leader in the provision of hydrogeological and hydrochemical advice and analysis to government, non-government and private bodies.

Between 1984 and 1988 BGS, funded by the United Kingdom Overseas Development Agency, was involved in providing support and research services in connection with the Deep Tubewell II Irrigation Project (DTWII) in Bangladesh. When it became clear that BGS would not require all of the funds allocated to it for its contribution to DTWII BGS obtained from the ODA authority to use the remaining funding to survey the main aquifers in central and north eastern Bangladesh by taking and analysing samples of ground water from 150 tubewell sites. The work of taking samples was done by Mr Davies of BGS, accompanied for some of the time by his wife, between January and March 1992. Some analytical work was done in the field, the remainder was done in the United Kingdom because the necessary facilities were not available in Bangladesh, and then in 1992 a report was prepared by Mr Davies and Dr Exley which, it is said in paragraph 7 of the Particulars of Claim, "forms the basis of this claim".

The report had a limited circulation, but, it is said, it was foreseeable, and indeed it was intended, that it would be made available to the authorities in Bangladesh responsible for ensuring as far as they could that water from the sites sampled by BGS was safe to drink. In fact the water from those sites, on which many poor people depended was contaminated by arsenic. Those who prepared the report appear not to have tested for arsenic, although they tested for 31 other elements, and it is the claimant's case that, even though this study was only a short term pilot project, having regard to the way in which BGS knew that the report would be circulated to and relied upon by authorities who had no possibility of testing for themselves, BGS should in the circumstances either have tested for arsenic or at least have made it clear that they had not done so. As a result of their failure to take either course the authorities were not alerted, and indeed were lulled into the false belief that water from the sites tested was safe to drink. They therefore took no action to protect the claimant and he, like many others, continued to drink water from sites sampled by BGS. That resulted in he and many others suffering serious health problems attributable to ingestion of arsenic. We were told by Lord Brennan QC for the claimant that so far 699 potential claimants have been identified of whom 512 have applied for public funding.

The Particulars of Claim extend to 34 pages and contain extensive references to the 1992 report, which we have seen, but I am satisfied that for present purposes the summary which I have just given sufficiently encapsulates the claimant's case. That case can also be found to be succinctly summarised in paragraph 30 of the judgment of Simon J.

The defendant's response.

The action was commenced and the Particulars of Claim were served in August 2002, a decade after the publication of the report, and on 11^th October 2002 the defendant applied for the case to be struck out pursuant to Civil Procedure Rules Part 3.4, or for summary judgment pursuant to CPR Part 24. Mr Michael Beloff QC for the defendant submits that the claimant was not some one to whom the defendants when publishing the 1992 Report owed any duty of care. In the language of lawyers the relationship between the claimant and the defendant was not sufficiently proximate to give rise to such a duty. That is the point covered by the first 3 grounds of appeal. Mr Beloff further submits that in any event nothing in the 1992 Report should have misled those to whom it was addressed or to whom it was forseeably made available, and consequently even if the claimant was a person to whom the defendant owed a duty of care he has no realistic prospect of establishing any breach of that duty. That is the fifth and final ground of appeal. The fourth ground of appeal, which described the breach of duty relied upon as a pure omission, was not relied upon in this court.

The concessions.

For the purposes of the application giving rise to this appeal the defendant has made certain concessions, which are set out in paragraph 32 of the judgment of Simon J and in paragraph 2 of the skeleton argument prepared by the defendant for use in this court. It is conceded that it is arguable that –

(a) One of the purposes of the report was to provide information relevant to potability but not that it was intended to be a comprehensive assessment of potability;

(b) It was foreseeable that the report would be seen by those concerned and responsible for the provision of water for human consumption;

(c) It was, therefore, foreseeable, if the report was relied upon by those concerned with the provision of water for human consumption as representing that the water was safe to drink, so that they abstained from action, the claimant would suffer personal injury as a result .

The Construction issue.

Although most of the submissions made to us by Mr Beloff and by Lord Brennan were directed to the question of whether the defendant owed to the claimant a legally recognised duty of care, what I can describe for convenience as the Proximity Issue, it seems to me that in the light of Mr Beloff's concessions it is convenient to start with the construction issue because, as he submits, the claimant cannot succeed unless the 1992 Report could reasonably be read by those likely to see it who had responsibility for the safety of the claimant's drinking water as representing that the water was safe to drink. The report was not prepared in response to any request made by any of the authorities in Bangladesh, so, as it seems to me, nothing can turn on how it came to be authorised, or on what was done by those by whom it was prepared. All that matters is what they, on behalf of and with the authority of the BGS, said in the report, and how what they said could reasonably have been understood. As to that Mr Beloff makes a number of powerful points, which do not seem to have been considered in any detail in the court below, although it is clear from paragraph 36 of the judgment that the argument was advanced that "there was nothing in the 1992 Report which suggested that the water was being approved for human consumption". In paragraph 22 the judge had said –

"Both sides relied on the terms of the 1992 Report: the claimant to show that it addressed the question of toxicity of the ground water to humans; and the defendant to show the express limitations of the Report."

The judge then went on to say simply that both of those points were arguably correct, and that he did not have to construe the report for the purposes of the present hearing, so he need only set out extracts relied upon by the parties. I find that approach a little hard to understand so I turn to look at the way in which Mr Beloff puts his case in relation to the construction issue. In essence he submits that any one responsible for the safety of drinking water supplied to the claimant who read the report should have been aware of its limitations because they were apparent on the face of the document. In particular -

(1) On the title page the report described itself as -

"Short Term BGS Pilot Project to Assess the 'Hydrochemical Character of the Main Aquifer Units of Central and North-Eastern Bangaldesh and Possible Toxicity of Ground Water to Fish and Humans'"

(2) In the Executive Summary the Aims of the Study are set out. That section, so far as relevant, reads -

"This reconnaissance study was undertaken with the primary aim of producing a reliable body of data that could be used to describe the hydrochemistry of the main aquifer units of central and north-eastern Bangladesh. Such data as required to understand:

(1) The hydrochemical nature of the main aquifers

(3) The modes of occurrence of trace elements that may be toxic to biological systems.'

(3) Still in the Executive Summary, under the heading 'Scope for Additional Work' it is made clear that what was undertaken was a 'rapid reconnaissance survey' of a type which produces information of relevance 'not only to hydrogeologists but also to those who are concerned with availability and quality of ground water for domestic, agricultural, aquaculture and industrial usage.'

(4) In the body of the report paragraph 1.2 deals with 'Study Objectives'. It reads -

'This study aimed to acquire sufficient data to permit baseline hydrochemical characterisation of aquifer units recognised and recognition of processes controlling variations in ground water chemistry within these aquifer systems.

Such data should reflect and indicate ground water flow patterns, location of ground water recharge and discharge areas, and deliniation of ground water bodies. Detailed studies will be made of the occurrence and interaction of aluminium, manganese, silica, phosphorous and iron and their effects upon biological systems in the context of fish cultivation, crop irrigation and potable water supply' "

As Mr Beloff points out, potability is only to be considered in relation to specified elements.

(5) It is clear from a careful reading of the report which tests were performed, and there is no reference anywhere to arsenic.

(6) Paragraph 2.1 deals with data and sample collection in Bangladesh by Mr Davies, accompanied by his wife who is a medical anthropologist, nurse midwife and fluent Bangla speaker. She was able, it is said, to recognise health problems caused by poor quality water conditions, and a number of examples are given. Mr Beloff refers to that passage in the report as 'plainly anecdotal'.

(7) In paragraph 4 the results of the survey are discussed. The discussion as to water quality relates to irrigation and acquaculture not potability, and there is no reference back to the matters referred to in paragraph 2.1.

(8) If drinking water standards were being assessed there would inevitably, especially in Bangladesh, have been reference to microbiology, infectious diseases and standards such as those set by the World Health Organisation, and there is no reference to any of those matters.

On behalf of the claimant Lord Brennan submitted that the 1992 report is only one of several documents evidencing the defendant's aims in conducting the study. For reasons already given it seems to me that the defendant's aims in conducting the study are not relevant in relation to the construction issue. All that matters is what a person in the position of BGS could reasonably expect the report to convey to those responsible for the safety of water supply who might reasonably be expected to see it. As to that Lord Brennan points out that the report came into existence in the context of assistance to a Third World Country, which did not have the systems or the means to check the safety of water supply. That does seem to me to be potentially relevant, and in relation to the report itself Lord Brennan points to -

"(1) The reference in the title to possible toxicity of ground water to fish and humans (see above).

(2) The reference in the Executive Summary: Aims of Study (see above) to 'the modes of occurrence of trace elements that may be toxic to biological systems'. That, it is said and I would accept, would be understood by any competent person involved in water management as including humans.

(3) Still in the Executive Summary, in the section headed 'Results' there is a reference to the atlas prepared to show the distribution of elements within the main production aquifers of the study area, and it is said that -

'This atlas can be used to indicate the distribution of elements and ground water properties that can be of benefit or harmful to aspects of life in Bangladesh. Examples are:

goitre can develop where iodine is deficient in the water supply …'

(4) In the body of the report there are further references to the impact of water quality on human health in paragraphs 1.2 and 2.1 (which are set out above)

(5) In paragraph 4, Discussion of Results, in relation to one area discussed at 4.5, it is said that –

'there are reports of fish mortalities in the area due to the application of tubewell water to fishponds. There is a potential for the development of goitre and anaemia amongst the human population should the standard of nutrition decline'.

In relation to another area it is said at 4.6 that iodine content is low.

(6) In paragraph 5 of the report, headed Conclusions, it is said that -

'The hydrochemistry of the sedimentary aquifers of central and north-eastern Bangladesh reflects and is dependant upon a number of factors each of which is imperfectly understood at present.'

Those imperfectly understood factors are then listed, and the paragraph ends –

'The above factors appear to control the distribution of minor elements some of which can be toxic to aspects of agricultural, health and aquacultural fields of interest. Such reconnaissance hydrochemical surveys will give timely indication of the possible presence and distribution of such toxic substances in groundwater systems in Bangladesh'

(7) Lord Brennan then points out that many of the thirty one elements tested for, including iodine are recognised to be toxic, or at least undesirable if present in excess and he asserts that –

'With the exception of arsenic, every constituent which poses a threat to human health present in concentrations of greater than 0.001 mg/l in the sea water used by the defendant as a control was tested for.'"

To my mind there is much to be said in favour of the defendant in relation to the Construction Issue, but I recognise that this action is at a very early stage, and, as it seems to me, it may be possible for the claimant to show by evidence that in essence there was a risk that the report would be misread, and that the defendant, with its experience in the field, should have been alive to and guarded against that risk. That possibility derives some support from two statements which have been produced on behalf of the claimant, one from Sara Bennett, an environmental consultant specialising in water environment aspects of international development projects who has extensive knowledge of Bangladesh, and the other from Peter Ravenscroft, a hydrogeologost. The claimant may also be able to derive some support for this aspect of his case from what Mr Davies wrote about his report after its publication, although that seems to me to be much more questionable. However, the continued existence of the possibility to which I have referred causes me to conclude that at this stage it would be premature to decide in favour of the defendant on the construction issue.

The proximity issue.

I turn now to Mr Beloff's principal point, namely whether the relationship between BGS and the claimant was such as to give rise to a duty of care. At least since the decision in Donoghue v Stevenson (1932) AC 562 there have many attempts to set out the test to be applied. In that case Lord Atkin said at 580 that liability arises when the alleged tortfeasor can reasonably foresee that his acts would be likely to injure his neighbour, and that neighbours are "persons who are so closely and directly affected" by the defendant's act that they should be in his contemplation. We were referred to part of the speech of Lord Devlin in Hedley Byrne v Heller [1964] AC 465 at 516, and to Clay v Crump [1964] 1 QB 533, where an architect was held liable to an employee of building contractors for the collapse of a wall which, with the architect's approval, demolition contractors had left standing. In Dorset Yacht Company Limited v Home Office [1970] AC 1004, and again in Anns v Merton LBC [1978] AC 278, the House of Lords sought to assist courts which had to decide whether in a given situation the relationship between the parties was such as to give rise to a duty of care. Proximity as envisaged by Lord Atkin did for a time seem to be of lesser importance, but then in the High Court of Australia in Sutherland Shire Council v Heyman [1985] 50 ALR 1 Brennan J at 43 expressed the view that "the law should develop novel categories of negligence incrementally and by analogy with established categories. That approach was endorsed by Lord Bridge in Caparo Industries PLC v Dickman [1990] 2 AC 605 and by Lord Keith in Murphy v Brentwood DC [1991] AC 398 at 461, and it is the approach which Mr Beloff submits we should adopt in this case. In Caparo's case at 618 D Lord Bridge said that after the decision in Anns members of the House of Lords had noted "the inability of any single general principle to provide a practical test which can be applied to every situation to determine whether a duty of care is owed and if so what is its scope." At 617 H he said –

"What emerges is that, in addition to the foreseeability of damage, the necessary ingredients in any situation giving rise to a duty of care are that there should exist between the party owing the duty and the party to whom it is owed a relationship characterised by the law as one of 'proximity' or 'neighbourhood' and that the situation should be one in which the court considers it fair, just and reasonable that the law should impose a duty of a given scope upon the one party for the benefit of the other. But it is implicit in the passages referred to that the concepts of proximity and fairness embodied in these additional ingredients are not susceptible to any such precise definition as would be necessary to give them utility as practical tests, but amount in effect to little more than convenient labels to attach to the features of different specific situations which, on a detailed examination of all the circumstances, the law recognises pragmatically as giving rise to a duty of care of a given scope."

Lord Bridge then recognised the wisdom of Brennan J's incremental approach, and at 618 E he said -

"One of the most important distinctions always to be observed lies in the law's essentially different approach to the different kinds of damage which one party may have suffered in consequence of the acts or omissions of another. It is one thing to owe a duty of care to avoid causing injury to the person or property of others. It is quite another to avoid causing others to suffer purely economic loss."

That of course is a passage on which Lord Brennan relies.

In the Sutherland case Deane J had said at 55 -

"The requirement of proximity is directed to the relationship between the parties in so far as it is relevant to the allegedly negligent act or omission of the defendant and the loss or injury sustained by the plaintiff. It involves the notion of nearness or closeness and embraces physical proximity (in the sense of space and time) between the person or property of the plaintiff and the person or property of the defendant, circumstantial proximity such as an over riding relationship of employer and employee or of a professional man and his client and what may (perhaps loosely) be referred to as causal proximity in the sense of the closeness or directness of the causal connection or relationship between the particular act or course of conduct and the loss and injury sustained. It may reflect an assumption by one party of a responsibility to take care to avoid or prevent injury, loss or damage to the person or property of another or reliance by one party upon such care being taken by the other in circumstances where the other party knew or ought to have known of that reliance. Both the identity and the relative importance of the factors which are determinative of an issue of proximity are likely to vary in different categories of case. That does not mean that there is scope for decision by reference to idiosyncratic notions of justice or morality or that it is a proper approach to treat the requirement of proximity as a question of fact to be resolved merely by reference to the relationship between the plaintiff and the defendant in the particular circumstances. The requirement of a relationship of proximity serves as a touchstone and control of the categories of case in which the common law will adjudge that a duty of care is owed. Given the general circumstances of a case in a new or developing area of the law of negligence, the question what (if any) combination or combinations of factors will satisfy the requirement of proximity is a question of law to be resolved by the processes of legal reasoning, induction and deduction. On the other hand the identification of the content of that requirement in such an area should not be either ostensibly or actually divorced from notions of what is 'fair and reasonable'…. or from the considerations of public policy which underlie and enlighten the existence and content of the requirement."

Lord Brennan submitted that what was said by Deane J does not represent the English approach, but, as Mr Beloff pointed out in reply, Lord Nicholls in Stovin v Wise [1996] AC 923 at 932 E referred to what had been said by Deane J as a "valuable exposition".

In X v Bedfordshire County Council [1995] 2 AC 633 the House of Lords was concerned with the liability of local authorities to children when carrying out functions imposed on the local authorities by the state. The statutory setting was considered to be important, and in the Bedfordshire case the relationship was admitted to be sufficiently proximate (see 748 E), but in the Newham case, heard at the same time, where a doctor and social workers were criticised in relation to the investigation of alleged sexual abuse the necessary proximity was found not to exist. At 753 E Lord Browne-Wilkinson said -

"The social workers and the psychiatrist did not, by accepting the instructions of the local authority, assume any general professional duty of care to the plaintiff children. The professionals were employed or retained to advise the local authority in relation to the well-being of the plaintiffs but not to advise or treat the plaintiffs."

In D v East Berkshire NHS Trust [2003] 4 All E R 796 it was said in this court at 814d (paragraph 49 of the judgment) that the effect of later decisions has been to restrict the decision in Bedfordshire to the proposition that decisions by local authorities whether or not to take a child into care are not reviewable by way of a claim in negligence.

In Marc Rich and Co v Bishop Rock Marine Co Ltd [1996] 1 AC 211 a surveyor acting on behalf of the classification society had recommended that after repairs specified by him had been carried out a vessel should be allowed to proceed. It was lost at sea, but the House of Lords held that the cargo owners could not recover from the classification society. It was a case of indirect damage, and of economic loss, and the features which the House of Lords found to be persuasive are indicated by Lord Steyn at 238 A where he pointed that there was no contact between the cargo owners and the classification society. It was not even suggested that the cargo owners knew of the survey, they simply relied on the owners to keep the vessel sea worthy and to look after the cargo. Mr Beloff submits that a similar approach is indicative of insufficient proximity in the present case.

Naturally courts are reluctant to strike out a claim at an early stage in a developing area of law if when all the facts are know the claim might succeed (see, for example Coulthard v Neville Russell [1998] PNLR 276 and Siddell v Smith Cooper [1999] PNLR 511) but that cannot be a recipe for inaction if there is no realistic prospect of success.

Perrett v Collins [1998] 2 LL.R. 255 is a decision of this court on which Lord Brennan places considerable reliance. The plaintiff was a passenger in an aircraft which crashed, and there was a preliminary issue as to the liability to him of those who certified that the aircraft was fit to fly. At 257 Hobhouse LJ said -

"What the second and third defendants seek to achieve in this case is to extend the decisions upon 'economic' loss to cases of personal injuries. It represents a fundamental attack upon the principle of tortious liability for negligent conduct which has caused foreseeable personal injury to others. That such a point should be considered to be even arguable shows how far some of the fundamental principles of the law of negligence have come to be eroded."

At 259 he said -

"The denial of a duty of care owed by such a person in relation to the safety of the aircraft towards those who may suffer personal injuries, whether as passengers in the aircraft or upon the ground, would leave a gap in the law of tort notwithstanding that a plaintiff has suffered foreseeable personal injury as a result of the unsafety of the aircraft and the unreasonable careless conduct of the defendant. It would be remarkable if that were the law."

Hobhouse LJ then examined the history of this area of law and at 261 he said -

"In cases of personal injury it suffices that the activity of the defendant has given rise to the situation which has caused the injury to the plaintiff. Where the defendant is involved in an activity which, if he is not careful, will create a foreseeable risk of personal injury to others, the defendant owes a duty of care to those others to act reasonably having regard to the existence of that risk. The limiting factors are concepts of forseeability and reasonableness."

At 262 Hobhouse LJ said -

"Where the plaintiff belongs to a class which either is or ought to be within the contemplation of the defendant and the defendant by reason of his involvement in an activity which gives him a measure of control over and responsibility for a situation which, if dangerous will be liable to injure the plaintiff, the defendant is liable if, as a result of his unreasonable lack of care, he causes a situation to exist which does in fact cause the plaintiff injury. ….. Once proximity is established by reference to the test which I have identified, none of the more sophisticated criteria which have to be used in relation to allegations of liability for mere economic loss need be applied in relation to personal injury, nor have they been in the decided cases."

At 264 it was said that "Marc Rich should not be regarded as an authority which has relevance to cases of personal injuries."

Swinton Thomas LJ reached the same conclusion but at 270 he pointed to a factor which is not relevant in the present case saying –

"The first and second defendants have undertaken to discharge the statutory duty for the protection of the public, and in my judgment no injustice is done by imposing such a duty on them in respect of a negligent act."

The same point was emphasised by Buxton LJ and Mr Beloff submits that the present case is different. In preparing the 1992 Report BGS was not seeking to discharge any statutory or other obligation to protect the claimant, or any other person using the water from the 150 sites.

In Farah and others v British Airways and the Home Office (6^th December 1999 unreported) the issue considered on appeal was whether the Home Office can be liable to immigrants as a result of an Immigration Liaison Officer negligently and wrongly advising an airline that the immigrants they did not have the requisite documentation to obtain access to the United Kingdom. On an application to strike out the claim a circuit judge had found no sufficient proximity, but Lord Woolf MR and Chadwick LJ considered that decision to be premature.

The line of authorities with which we are concerned in the present case was reviewed by the present Master of the Rolls in Watson v British Boxing Board of Control [2001] QB 1134, where it was submitted by an injured boxer that the Board should have required the provision of more skilled and effective medical assistance at the ringside. That submission found favour both at first instance and in the Court of Appeal, but, as was pointed out at 1149, there were special features of the case, in that the Board assumed responsibility for determining what medical care should be available, and participants relied upon the Board properly to discharge that responsibility. At 1151 Lord Phillips said -

"It seems to me that the authorities support a principle that, where A places himself in a relationship to B in which B's physical safety becomes dependant upon the acts and omissions of A, A's conduct can suffice to impose on A a duty to exercise reasonable care for B's safety."

Mr Beloff accepts that proposition which, after considering the authorities, the Master of the Rolls re-iterated at 1159 in these terms –

"These cases establish that where A advises B as to action to be taken which will directly and foreseeably affect the safety or well-being of C, a situation of sufficient proximity exists to found a duty of care on the part of A towards C. Whether in fact such a duty arises will depend upon the facts of the individual case and, in particular, upon whether such a duty of care would cut across any statutory scheme pursuant to which the advice was given."

He then turned to look specifically at proximity, and at 1161 he said –

"Had the board simply given advice to all involved in professional boxing as to appropriate medical precautions, it would be strongly arguable that there was insufficient proximity between the board and individual boxers to give rise to a duty of care. The board, however, went far beyond this. It made provision in its rules for the medical precautions to be employed and made compliance with these rules mandatory."

As Mr Beloff points out, BGS was in no position to dictate what precautions should or should not be taken in relation to the water which the claimant and others were to use, nor did it attempt to do so. Perret v Collins was extensively considered in Watson and the importance of the later case, so far as the present defendant is concerned, is that proximity is again shown to be a matter of significance even when the claim is for personal injury.

In Parkinson v St James's NHS Trust [2001] EWCA Civ 530 Hale LJ began her judgment by pointing out that the right to bodily integrity is the first and most important of the interests protected by the law of tort, but otherwise the decision seems to me to be of relatively little assistance in this case. Although, as the claimant points out, Brooke LJ does, at paragraph 18, refer to other factors which have emerged to assist in some cases when deciding whether a duty of care was owed, namely for what purpose a service was rendered and the principles of distributive as opposed to corrective justice, neither side before us sought to rely to any significant extent on those factors in this case.

In the present 18^th edition of Clerk and Lindsell on Torts at paragraph 7 -14 it is said that proximity may consist of various forms of closeness – physical, circumstantial, causal or assumed -

"It involves considering the relationship from the perspective of both the defendant and the claimant. At root, it will reflect 'a balancing of the plaintiff's moral claim to compensation for avoidable harm, and the defendant's moral claim to be protected from an undue burden of legal responsibility'. As such it will inevitably overlap with considerations of justice between the parties."

The words cited are from Richardson J in South Pacific Manufacturing Co Ltd v New Zealand Security Consultants and Investigations Ltd [1992] 2 NZLR 282 at 306. The problem which we face is whether it is possible to say at an early stage in litigation that the balance can be seen to be tipped so far in the defendant's favour that the action should not be allowed to proceed. Lord Brennan submitted that because this is a balancing exercise we should not interfere with the decision of the judge unless satisfied that he was plainly wrong. There is no authority for that proposition, and in my judgment it is incorrect. Of course we must have regard to the careful reserved judgment delivered by Simon J after a two day hearing, but in dealing with the material he had no advantage denied to us. We must therefore, in my judgment, decide for ourselves whether it is arguable that when the balancing exercise is carried out after all the facts have been ascertained, the claimant could succeed.

During the course of his submissions Mr Beloff produced to us a helpful table which compared five factors in five of the principle authorities with the same factors in the present case. He also submitted to us in written form the "daisy-chain of reliance and causation" connecting the 1992 Report relied upon to the claimant's health. He did not raise any argument in relation to public policy to safeguard the position of BGS, but contended that to find that there was a relationship of proximity between the claimant and BGS would go far beyond any decided case. It would not be an incremental step, and would be a step too far.

Conclusion on Proximity.

The features of the case to which Mr Beloff draws attention are not matters of contention, and cannot be challenged. They can be summarised thus -

(1) BGS did not create the hazard, namely the arsenic in the water. It was already there.

(2) BGS had no duty to provide the claimant or his fellow citizens with potable water. They had no power to do so, and they could not even warn him of any dangers.

(3) When BGS undertook the 1992 survey they did so with money provided by ODA, and it was to that organisation that the 1992 Report was rendered.

(4) Plainly it was not the primary purpose of the report to express a view about potability of water. Lord Brennan submits that it may have been a primary purpose, but the document speaks for itself.

(5) There was therefore no assumption of responsibility by BGS and if on publication of the report it had been said to BGS that the report was giving a legally enforceable assurance to a large cohort of the Bangladeshi population as to the safety of their water that would surely have been regarded as absurd.

(6) After publication of the report BGS, through Mr Davies, did to some extent promote the report, but not apparently in any way that altered the effect of the text, and although the document was plainly designed to be helpful BGS had ultimately no control over who saw the report, or how it was used.

I accept, as Lord Brennan emphasised, that there is a close link between foreseeability and proximity as tests for the existence of a legally enforceable duty, and that it is important to recognise that the 1992 Report was published in Bangladesh where facilities did not exist to test the conclusion which were expressed, but even so it seems to me that to say that this plainly restricted short term pilot project can be relied upon in the way for which Lord Brennan contends would not be an incremental step. It would be a mighty leap which would render the concept of proximity almost meaningless. I say by reference to the features of the case above, and without reference to factors such as the scale of the obligation undertaken by BGS, the uninsurability of the risk, and the desirability of not inhibiting activities in the Third World such as those which led to this claim, because those are largely matters of public policy. Simon J seems to have decided as he did largely because it is on the authorities so difficult to spell out what is meant by proximity (see especially paragraphs 43 and 60(iii) of his judgment) but if it is clear that at the end of the day that necessary pre-condition of liability, however defined, cannot be proved, then the court must act. The duty to do so seems to me to be even greater when the litigation is likely to be lengthy and expensive, because otherwise the unfortunate claimant will only be left with unrealistic expectations, and the defendant and the United Kingdom taxpayer will be left with an enormous bill.

I would therefore allow the appeal and strike out the action.

Lord Justice Clarke:

I regret that I have reached a different conclusion from that reached by Kennedy and Wall LJJ. I have reached the same conclusion as the judge, namely that the allegation that there was a relationship of proximity between the defendant and the plaintiff is not so fanciful that the action should be struck out or brought to an end under CPR Part 24 at this stage. I would therefore dismiss the appeal.

It is important to note that this is not the trial of a preliminary issue. I therefore agree with Kennedy LJ that, if the appeal is to be allowed, the defendant must demonstrate that the claimant has no more than a fanciful prospect of success. The question is whether it has done so. Like the judge, I remind myself that caution should be exercised before so holding in a case in which the facts have not been found. See e.g. X (Minors) v Bedfordshire CC [1995] 2 AC 633 per Lord Browne-Wilkinson at pages 740-1.

In this regard the judge referred by way of example to the decision of this court in Farah v. British Airways and the Home Office, unreported, 6 December 1999, where the issue was whether statements made by the Home Office to British Airways that immigrants did not have proper documentation could give rise to liability to the immigrants where the statements were made negligently. In Farah the judge struck out the claim on the basis that there was insufficient proximity between the claimants and the Home Office, since there was no contact between the Home Office and the Claimants. This court allowed the claimants' appeal on the basis that there were crucial factual issues which had to be determined at trial.

Agreeing with Lord Woolf MR, at paragraph 42 Chadwick LJ warned of the dangers implicit in the summary determination of the issue as to whether a duty of care exists in these terms:

"As Lord Browne-Wilkinson observed in Barrett v LB Islington [1999] 3 WLR 83, unless it is possible to give a certain and affirmative answer to the question whether the claims would be bound to fail, the case is not one in which it was appropriate to strike out the claim in advance of trial. Lord Browne-Wilkinson went on to point out that in an area of the law which was uncertain and developing, it could not normally be appropriate to strike out. He emphasised the importance of the principle that the development of the law should be on the basis of actual facts found at trial and not on the basis of hypothetical facts assumed (possibly wrongly) to be true on the hearing of the application to strike out. There are observations to the like effect in Lord Browne-Wilkinson's speech in X (Minors) v Bedfordshire County Council [1995] 2 AC 633 at 741; and in the judgment of Sir Thomas Bingham MR in E (A Minor) v Dorset County Council p 694 of the same report."

The judge correctly observed that in Farah it was the imprecision implicit in the concept of proximity that was at the heart of this court's reasoning as to why the matter should proceed to trial rather than be determined summarily under the RSC. Lord Woolf said at paragraph 32:

"I remind myself of the passages from the speeches in Caparo, to which I have referred, as to the requirement of proximity; in particular that it is no more than a convenient label which embraces not a precisely definable concept but merely a description of circumstances. The precise circumstances which are in issue here have yet to be ascertained and could only be ascertained at trial."

The relevant circumstances which might need investigation were set out by Lord Woolf at paragraph 35 with this comment:

"This is an area of developing jurisprudence. Where that is so, the question of whether or not an analogous situation should be recognised as giving rise to a duty of care, should be determined when the facts have been established."

The judge was in my opinion right to approach the issues in this case with those principles in mind. This is particularly important in a case where the only question raised by the defendant is whether it is sufficiently arguable that there is a sufficient relationship of proximity between the parties. That is because of the imprecision of the test.

The judge convincingly demonstrated that imprecision by extracts from two well-known passages in the speeches of Lord Bridge and Lord Oliver in Caparo Industries plc v Dickman [1990] 2 AC 605. Lord Bridge said at pages 617H-618B:

"What emerges is that, in addition to the foreseeability of damage, necessary ingredients in any situation giving rise to a duty of care are that there should exist between the party owing the duty and the party to whom it is owed a relationship characterised by the law as one of "proximity" or "neighbourhood" and that the situation should be one in which the Court considers it fair just and reasonable that the law should impose a duty of a given scope upon one party for the benefit of the other. But it is implicit in the passages referred to that the concepts of proximity and fairness embodied in these additional ingredients are not susceptible of any precise definition as would be necessary to give them utility as practical tests, but amount in effect to little more than convenient labels to attach to the features of different specific situations which, on detailed examination of all the circumstances, the law recognises pragmatically as giving rise to a duty of care of a given scope."

Lord Oliver said at page 633A-D:

"Thus the postulate of a simple duty to avoid harm that is, with hindsight, reasonably capable of being foreseen becomes untenable without the imposition of some intelligible limits to keep the law of negligence within the bounds of common sense and practicality. Those limits have been found by the requirement of what has been called a "relationship of proximity" between the plaintiff and the defendant and by the imposition of the further requirement that the attachment of liability for harm which has occurred be "just and reasonable." But although the cases in which the courts have imposed or withheld liability are capable of approximate categorisation, one looks in vain for some common denominator by which the existence of the essential relationship can be tested. Indeed it is difficult to resist the conclusion that what have been treated as three separate requirements are, at least in most cases, in fact merely facets of the same thing, for in some cases the degree of foreseeability is such that it is from that alone that the requisite proximity can be deduced, whilst in others the absence of that essential relationship can most rationally be attributed simply to the Court's view that it would not be fair and reasonable to hold the defendant responsible. "Proximity" is, no doubt, a convenient expression so long as it is realised that it is no more than a label which embraces not a definable concept but merely a description of circumstances from which, pragmatically, the courts conclude that a duty of care exists."

The judge also drew attention to the close relationship between the concept of proximity and that of fairness and reasonableness which is referred to by Lord Nicholls in his dissenting speech in Stovin v Wise [1996] AC 923, where he said at page 932B-C:

"The Caparo tripartite test elevates proximity to the dignity of a separate heading. This formulation tends to suggest that proximity is a separate ingredient, distinct from fairness and reasonableness, and capable of being identified by some other criteria. This is not so. Proximity is a slippery word. Proximity is not legal shorthand for a concept with its own objectively identifiable characteristics. Proximity is a convenient shorthand for a relationship between two parties which makes it fair and reasonable one should owe the other a duty of care. This is only another way of saying that when assessing the requirement of fairness and reasonableness regard must be had to the relationship between the parties."

Those passages show that proximity is an imprecise concept, where much depends upon the circumstances of the particular case, and that it is not possible to divorce the question whether the relationship between the parties is sufficiently proximate from the question whether it is fair just and reasonable to impose a duty of care upon the defendant on the facts of a particular case. On the contrary, as Lord Nicholls put it in the passage just quoted, proximity is a convenient shorthand for a relationship between two parties which makes it fair and reasonable that one should owe the other a duty of care.

This is of particular significance in the present case because Mr Beloff did not approach the issue of proximity in that broad way. He focused on the submission that in a case where the facts are different from the type of situation in which a duty of care has been held to exist in the past, the court should only hold a duty to exist if the facts were sufficiently similar to those of a previous case that find a duty of care would represent a small step. As Kennedy LJ observes in paragraph 23 above, Mr Beloff submitted that it would not (in Lord Bridge's words) be an incremental step but a step too far.

The submissions advanced by Mr Beloff to my mind provide strong support for the conclusion that it would not be fair, just and reasonable to expose the defendant to liability to a large number of Bangladeshi citizens. However, I am not persuaded that it would be right to hold that there is no real prospect of the claimant persuading the court at a trial that there was a sufficient relationship of proximity between the defendant and Bangladeshi citizens like him, essentially for the reasons given by Lord Brennan and accepted by the judge. I would briefly summarise my reasons as follows. In doing so I shall refer, not to the claimant in the singular, but to claimants in the plural because the decision in this case will in practice determine many others.

Mr Beloff described the chain of reliance and causation relied upon by Lord Brennan as a daisy chain. While that phrase has an attractive forensic ring to it, it is to my mind misleading. It was expressly conceded by Mr Beloff on behalf of the defendant that the claimants have an arguable case that, if they were owed a duty of care by the defendant, the defendant was arguably in breach of that duty because of a failure to exercise reasonable care and skill, that there was arguably sufficient reliance upon the report and that it is arguable that the claimants' physical condition as a result of drinking water with arsenic in it was caused by that breach of duty or negligence. Moreover, it was not submitted that, if there was sufficient proximity it would not, at least arguably, be unreasonable or unfair to impose a duty of care.

It is to my mind also important to have in mind these further concessions made by the defendant for the purposes of the application and therefore the appeal. As the judge observed in paragraph 32, it was accepted that it is arguable that (a) one of the purposes of the 1992 Report was to provide information relevant to the potability of the groundwater in the deepwells (although not that it was intended to be a comprehensive assessment of potability), (b) it was foreseeable that the 1992 Report would be seen by those concerned with and responsible for the provision of water for human consumption, and (c) it was therefore foreseeable that, if the 1992 Report was relied upon by those concerned with the provision of water for human consumption as safe to drink, the claimant was within a class of people that would suffer personal injury as a result.

The key features of the case relied upon by Mr Beloff are set out by Kennedy LJ in paragraph 24 above and were summarised by the judge thus in paragraph 34 of his judgement:

"(1) the Defendant was funded by the ODA, (2) the 1992 Report was prepared for the ODA and its subject-matter was agreed with the ODA, (3) the Defendant was not in contractual relations with the Bangladesh Government or any of its agencies, nor with the Claimant, (4) the Defendant did not give any undertaking to the Bangladesh Government or its agencies, or the Claimant, in advance as to the use to which the 1992 Report could be put, (5) the Claimant and the Defendant were never in direct contact with each other, nor were they aware of each other's existence, (6) the Claimant was not aware of the 1992 Report and did not act on the basis of it, (7) the Defendant did not intend the Report to certify that the water which had been tested was necessarily fit to drink, (8) the ODA and not the Defendant controlled the distribution of the Report, (9) the Defendant did not control the provision of drinking water to the Claimant, had no statutory or other responsibility for the provision of water to the Claimant or to certify the safety of the water, and had no obligation to advise those who had those responsibilities, (10) the Defendant had no obligation to provide the Report to those who had those responsibilities, (11) the Defendant was not responsible for the presence of arsenic in the water and had no responsibility for removing the arsenic."

Again as the judge observed, many of those facts were uncontroversial.

The judge set out the nature of the claimants' case in paragraph 30 of his judgment as follows:

"The primary source of drinking water in Bangladesh was groundwater from wells. The presence of arsenic in the water was a major health hazard. While in many countries local authorities would ensure the proper testing of drinking water, the Bangladesh Government was not in a position adequately to test the quality of the drinking water and, in particular, for the presence of potentially toxic trace elements, such as arsenic, The objects of the Defendant under its Charter included the provision of advice. The Defendant was paid by the ODA out of development aid funds designated for the benefit of the Bangladeshi people to conduct a hydrochemical baseline survey of the groundwater quality. The 1992 Report included an assessment of the water's toxicity to humans; and the Defendant therefore knew that it would be used for the benefit of consumers of ground water from wells. In 1992 the possibility of arsenic being present in the groundwater in the relevant area should have been known to reasonably competent hydrogeologists and should have been included as an element to be tested for in any properly conducted hydrochemical baseline survey. Readers of the 1992 Report could reasonably assume that, if the survey had not tested for particular trace elements, there was no need to do so. Alternatively, if in fact the Report was not intended to test for the potability of water for humans, then it was incumbent on the Defendant to have made it clear that it could not be relied on for that purposes. The Report on the survey was intended by the Defendant and the ODA (a) to be for the use and benefit of the Bangladeshi Government and agencies involved in the management of Bangladeshi water resources, and (b) to be widely distributed to such interested parties. As a consumer of drinking water in Bangladesh the Claimant was in a class which (a) would be bound to be affected by any action taken on their behalf by the Bangladeshi Government and agencies in relation to the management of water resources, and therefore (b) would be affected by the content of the 1992 Report."

I should first say a word about the 1992 Report. The survey which led to the report was promoted by the defendant and the ODA agreed to fund it. It was not just a desk study but involved testing and sampling a large number of sites. The aims of the study can be seen from the quotations from the report set out by Kennedy LJ in paragraph 9 above. I shall not repeat those quotations here save to note that its purpose was to assess the hydrochemical character of the main aquifer units of parts of Bangladesh and the possible toxicity of ground water to humans. Its aim was to produce a reliable body of data required to understand the modes of occurrence of trace elements that may be toxic to biological systems, which to my mind plainly included human biological systems.

One of the aims was to assess the potability of the water supply, although Mr Beloff submitted that the report shows that the potability was only considered in relation to specific elements. I agree that it is arguable that it is so limited but the question is how the report would be expected to be understood by those who were likely to read it. There is evidence which seems to me to be relevant to this question in a statement by Sara Bennett, to whom Kennedy LJ refers in paragraph 11. At the time she was the Environmental Specialist on phase 1 of the Northeast Regional Water Management Plan ("NERP"). She said in paragraphs 11 and 12 of her statement:

"11. As the Environment Specialist on NERP, I read the Davis & Exley 1992 report because I had overall responsibility for all environmental impacts and environmental health matters related to the Plan. In reading the report, I understood that:

(a) it had been prepared under the auspices of one of the most highly regarded and renowned geological organizations in the world;

(b) it proclaimed that it was considering the issue of toxicity to humans of trace elements in groundwater;

(c) it proclaimed that it was testing groundwater to determine the concentration of trace elements toxic to humans;

(d) its conclusions did not indicate that any trace element(s) in the groundwater posed a hazard to human health;

(e) there are no provisos attached to the report.

12. As an intelligent environmental specialist reading the report, I can simply say I took these statements at face value. I assumed I was reading a study by competent scientists from a reputable scientific organization, and thus that the investigators had (1) reasoned intelligently as to which trace elements needed to be tested, (2) tested for these elements, (3) articulated conclusions fully supported by the findings of their field studies, and (4) submitted their report to senior experts of their organization for formal technical review and clearance prior to external distribution."

It is plain that Ms Bennett was struck by the absence of any disclaimer or proviso in the report. She said in paragraph 17 that she understood from the report that the authors considered the study to be complete in itself as a reconnaissance of trace elements of possible relevance in the region to safe consumption by humans and that they did not consider it to be a partial study that omitted relevant trace elements from sampling and analysis due to resource, time, analytical constraints and other factors.

In my opinion that evidence supports the conclusion that it is arguable that the defendant should have tested for all trace elements including arsenic or made it clear in the report that it had not tested for arsenic and that it did not do so. Moreover, I agree with Kennedy LJ that the claimants have a sufficiently arguable case that, as he put it, there was a risk that the report would be misread or, as I would prefer to put it, that the report would be read as understood by Ms Bennett and thus that all relevant trace elements had been considered including arsenic.

Ms Bennett's statement also seems to me to contain evidence which supports the claimants' case on proximity. She said that the Government of Bangladesh relies heavily upon donor countries and multilateral donor organizations for resources to undertake data gathering which is funded by the governments of developed countries as a basic government function. She also said that in her experience once a survey is undertaken, its findings and data will be regularly used and relied on by other consultants in the field. It is at least arguable that those facts were known to the defendant.

Thus it is at least arguable that the defendant knew that the results of its survey and its report as to the potability of the water would be relied upon by the Government of Bangladesh in deciding what action to take and that any action (or indeed inaction) as a result of the report would be likely to affect the drinking water of the claimants and thus potentially their health. In these circumstances the claimants' case as set out above is arguable and potentially leads to the conclusion that there was a sufficient relationship of proximity between the defendants and the claimants. I accept that the factors relied upon by Mr Beloff point the other way but I do not think that the claimants' case on proximity is fanciful.

Nor do I think that the authorities lead to the conclusion that it is. This is a case in which the claimants complain, not of economic loss, but of personal injury. The courts have been less reluctant to hold that no duty of care exists in such a case. I recognise that there is no case quite like this in which a duty of care has been held to exist. However, the cases to which we were referred do not seem to me to lead to the conclusion that the claimants' case on proximity is fanciful. The question identified and answered by Lord Atkin in Donoghue v Stevenson [1932] AC 562 and applied many times since was this:

"Who, then in law is my neighbour? The answer seems to be – persons who are so closely and directly affected by my act that I ought reasonably have then in contemplation as being so affected when I am directing my mind to the acts or omissions which are called in question."

Lord Brennan relied upon the decision of this court in Perrett v Collins [1988] 2 Lloyd's Rep 254, where (at pages 260-2) Hobhouse LJ stressed the distinction between liability for pure economic loss and liability for causing physical injury to the human body and to goods. Hobhouse LJ recognised that problems may exist in cases of personal injury and that mere foreseeability may not be sufficient but there are four passages in his judgment which seem to me to be of particular assistance here.

The first is on page 261, where he said this of Donoghue v Stevenson and Grant v Australian Knitting Mills Ltd [1936] AC 85:

"Mere foreseeability did not suffice. In those cases the relevant consideration was whether or not the manufacturer was entitled to expect that there would be an intermediate inspection of the product before it was used by the consumer. This concerned the foresight of a chain of causation between the defendant's conduct and the plaintiff's injury."

The second is also on page 261, where he referred to a number of cases which were concerned with the liability of some public institution or body for adverse consequences to members of the public and then said this:

"A minimum requirement of particularity and contemplation is required. But it has never been a requirement of the law of the tort of negligence that there be a particular antecedent relationship between the defendant and the plaintiff other then one that the plaintiff belongs to a class which the defendant contemplates or should contemplate would be affected by his conduct."

The third is on page 262. After referring to Clay v Crump [1964] 1 QB 262 and observing that the court rejected an argument that the opportunities for intermediate inspection sufficed to relieve the defendant architect of responsibility, he expressed his conclusion thus:

"Where the plaintiff belongs to a class which either is or ought to be within the contemplation of the defendant and the defendant by reason of his involvement in an activity which gives him a measure of control over and responsibility for a situation which, if dangerous, will be liable to injure the plaintiff, the defendant is liable if as a result of his unreasonable lack of care he causes a situation to exist which does in fact cause the plaintiff injury.

Once this proximity exists, it ceases to be material what form the unreasonable conduct takes"

The fourth is on page 264 where he was considering the decision of the House of Lords in Rich (Marc) & Co AG v Bishop Rock Marine Co Ltd [1994] 1 WLR 1071. He noted Lord Steyn's view that the law more readily attaches the consequences of actionable negligence to directly inflicted physical loss than to indirectly inflicted physical loss and said a few lines later:

"[Lord Steyn] expressly said that the fact that the carelessness of the surveyor did not involve the direct infliction of physical damage did not exclude the existence of a duty of care; indeed he could not have done so without overruling previous authority. …. where on general principle in the context of foreseeable risk of personal injury, a duty of care exists, lack of directness, unless it destroys the causative link, provides the defendant with no answer."

I have not found the decision in the Marc Rich case of any real assistance because it was not specifically concerned with the notion of proximity. Indeed, Lord Steyn said at page 241 that he was willing to assume without deciding that there was a sufficient degree of proximity.

In Watson v British Boxing Board of Control [2001] QB 1134 this court reviewed all or almost all the cases to which we were referred and Lord Phillips MR, giving the judgment of the court, said at paragraph 72:

"These cases establish that, where A advises B as to action to be taken which will directly and foreseeably affect the safety or well-being of C, a situation of sufficient proximity exists to found a duty of acre on the part of A towards C. Whether in fact such a duty arises will depend upon the facts of the individual case and, in particular, upon whether such a duty of care would cut across any statutory scheme pursuant to which the advice was given."

Although I recognise that the factors relied upon by Mr Beloff and set out above militate against there being a relationship of sufficient proximity between the defendant and the claimants, I agree with the judge that the contrary is arguable. It is conceded that it was reasonably foreseeable that if the defendant failed to exercise skill and care such that the report might be read as covering all relevant trace elements including arsenic it would or might cause physical injury to the claimants. It is at least arguable that the defendant would or ought to have appreciated that the Government of Bangladesh would rely upon the survey and the report without carrying out work or exercising an independent judgment of its own and that it would take decisions based upon them which would include decisions as to the water used for drinking by the claimants. In these circumstances it can be said that the defendant had a measure of control and responsibility over the testing of the potability of the water and that its acts or omissions in carrying out the survey and reporting on it would have a direct effect on the plaintiff.

In all the circumstances it seems to me to be reasonably arguable that the test of proximity identified by Lord Atkin as explained, for example, by Hobhouse LJ and Lord Phillips, is satisfied. Thus, on the facts set out above, it does not seem to me to be fanciful to say that the citizens of Bangladesh like the claimants were (at least potentially) so closely and directly affected by the negligent act or omission of the defendant in failing to test for arsenic and/or, having done so, in failing to make it clear to the reader of the 1992 Report that there might be trace elements (including arsenic) not tested for which might pose a hazard to human health, that it ought reasonably to have had them in contemplation when deciding what to test for and how to report the results. Put another way, it is not fanciful to say, in the words of Hobhouse LJ, that the claimants belong to a class which either was or ought to have been within the contemplation of the defendant by reason of its involvement in the activity of testing and reporting on the trace elements in the water which (as already indicated) gave it a measure of control over and responsibility for a situation which was liable to injure the claimants. In these circumstances it seems to me to be at least arguable that the case falls within the principle stated in paragraph 72 of the judgment in Watson quoted above.

I should finally mention two last points. The first is the reliance placed by Mr Beloff on the part of the decision of the House of Lords in X v Bedfordshire in which (despite the caution referred to by Lord Browne-Wilkinson at pages 740-1) it was held in the Newham case (as demonstrated in the extract from Lord Browne-Wilkinson's speech quoted by Kennedy LJ in paragraph 14 above) that the social workers and the psychiatrist did not, by accepting the instructions of the local authority, assume any general duty of care to the children on the basis that they were employed or retained to advise the local authority and not to treat the children.

For my part, I do not think that that decision, which was of course limited to the very different facts of that case helps to resolve the question whether there was arguably a sufficient relationship of proximity in this case, which does not depend upon the proposition that the defendant was engaged to advise or treat the claimants. Moreover, as Kennedy LJ observed, it was held in D v East Berkshire NHS Trust [2003] EWCA Civ 1151, [2003] 4 All ER 796, that the effect of later decisions, especially Barrett v Enfield London BC [2001] 2 AC 550, was to restrict the effect of Bedfordshire to the core proposition that decisions of local authorities whether or not to take a child into care are not reviewable by way of a claim in negligence.

The decisions in Barrett and D show how different the facts of those cases, and indeed Bedfordshire, are from this. They also stress the importance of the principle relied upon by the judge in this case that questions of this kind should not in general be determined without a proper examination of the facts. Moreover, it is I think noteworthy that, as explained in the later cases, the part of the decision in Bedfordshire relied upon by Mr Beloff was concerned not with the question whether there was a sufficient relationship of proximity but with the broader question whether it was arguable that it would be fair and reasonable to impose a duty of care on the facts of that case, whereas this case is not concerned with that broader question.

The second point is the correct approach of this court to an appeal of this kind. Kennedy LJ has expressed the view in paragraph 22 that we must decide for ourselves whether it is arguable that when the balancing exercise involved in resolving a question of this kind after all the facts have been ascertained the claimant could succeed. We were not referred to any authority on this question, except a dictum of Buxton LJ in Perrett at page 273, where he said this:

"However careful in assessing competing factors a judge may be, and however aware he may be of what those factors are, an Appellate Court is able to substitute its decision for his simply on the ground that it considers him to have been wrong: see eg Reeman v Department of Transport and Others [1997] 2 Lloyd's Rep 648."

I respectfully doubt whether, in the light of the principles discussed in the cases collected together in Assicurazioni Generali SpA v Arab Insurance Group [2002] EWCA Civ 1642, [2003] 1 WLR 577 at paragraphs 14 to 22, that represents the modern approach even after a trial. It is not, however, necessary to express a view on that question here because, whereas the appeal in Perrett was after a trial of preliminary issue, there has been no trial in this case. On an appeal on an interlocutory application of this kind, which involves a balancing of a number of disparate factors and a consideration of facts which have not been determined, it seems to me that the court should be reluctant to interfere with the decision of the judge unless he has reached a conclusion which was outside the range of reasonable responses to the question for decision. It is in this kind of case that particular respect should in my opinion be paid to the decision of the judge.

For these reasons I respectfully differ from Kennedy LJ as to the correct approach to an appeal of this kind but this difference of opinion does not affect the view I have formed as to the disposal of the appeal. The judge summarised his conclusions in paragraph 60 of his judgment as follows:

"i) This is a case that raises a novel point in a developing area of the law. Such cases should normally be decided on facts found at trial.

ii) On the basis of the facts set out in the Particulars of Claim it is not possible to give a certain answer to the question, whether the Claim is bound to fail.

iii) The issue of proximity is difficult to isolate from the issues of foreseeability and fairness in respect of which it is accepted that there can be no summary determination.

iv) There are various factual matters which are likely to be relevant to the issue of proximity which will have to be determined at trial.

I agree with those conclusions. Like the judge, I am of the view that questions of this kind should be determined at a trial, perhaps a trial of one or more discrete preliminary issues. I would accordingly dismiss the appeal, although I would add that, even if the claimants were able to proceed with their actions, they would face many great difficulties.

Lord Justice Wall:

I have had the advantage of reading in draft the judgments of both Kennedy and Clarke LJJ. Whilst, on a first reading of the papers I shared Clarke LJ's doubts about the appropriateness of the action being struck out without any judicial resolution of underlying factual issues, I have come to the clear conclusion, having heard the argument, that those doubts were misplaced. I am therefore in agreement with Kennedy LJ that the appeal should be allowed, and the action struck out. Moreover, for reasons which I will attempt to explain, I respectfully part company with Clarke LJ's analysis of the authorities on proximity.

For the Defendant, Mr. Michael Beloff QC both began and concluded his submissions with a nice advocate's flourish. He invited us to speculate on Lord Atkin's reaction were he to have been told that his classic statement of the neighbour principle in Donoghue v Stevenson [1932] AC 562, 580 would be used to formulate a duty of care in the instant case. Whilst such an invitation, expressed in that form, does not, of course, substantially advance the argument, it nonetheless serves to underline for me the proposition that the establishment of a duty of care in tort on the presumed facts of the instant case would undoubtedly be to break new ground. It has been said that the common law develops incrementally. The question for us, accordingly, is whether a judicial identification of the existence of a duty of care in the instant case is, as Lord Brennan QC for the Claimant argues, a logical and necessary development of the law of negligence, soundly based on existing authority, or whether, as Mr. Beloff argues, it is a step too far. I am, speaking for myself, in no doubt that it is the latter.

In what struck me as principled submissions, Mr Beloff made a number of important concessions in compliance with the rule that on an application to strike out a claim for negligence on the ground that the Defendant did not owe the Claimant a duty of care, the Claimant must be taken as being able to prove his pleaded case on the facts. Mr. Beloff accordingly accepted that both negligence and foreseeability of damage were arguable, although he made it clear that both would be hotly disputed if the matter came to trial.

The best way I have found into the question we have to decide is through the helpful analysis of previous authority on the existence or otherwise of a duty of care carried out by Lord Phillips of Worth Maltravers MR when giving the leading judgment in this court in Watson v British Boxing Board of Control [2001] QB 1134 (Watson v. BBBC) and the Master of the Rolls' application of the principles derived from those authorities to the facts of that case.

Simon J at paragraph 54 of the judgment under appeal cites the first sentence of the key passage from Watson v. BBBC ([2001] 1 QB 1134 at 1159, paragraph 72: set out at paragraph 76 below) but does not take up the Master of the Rolls' invitation to analyse the facts of the instant case (as the Appellant concedes them to be for the purposes of the application to strike out) in order to decide whether or not a duty of care is capable of arising on those assumed facts.

Instead, the judge reaches four conclusions which caused him to decline to strike out the claim. These were: -

(1) This is a case that raises a novel point in a developing area of law. Such cases should normally be decided on facts found at trial.

(2) On the basis of the facts set out in the Particulars of Claim it is not possible to give a certain answer to the question whether the Claim is bound to fail.

(3) The issue of proximity is difficult to isolate from the issues of foreseeability and fairness in respect of which it is accepted that there can be no summary determination.

(4) There are various factual matters which are likely to be relevant to the issue of proximity which will have to be determined at trial.

Unlike Clarke LJ I am unable to agree with the thinking behind these propositions. I agree with the first sentence of (1) and accept of course that the second sentence sets out the normal rule. I see the force of (3). However, I disagree with the judge on (2) and (4). I find it difficult to identify the "various factual matters" determination of which will be so crucial to the issue of proximity. Moreover, the fact that the task may be difficult is not, in my judgment, a reason for not attempting it, particularly as the exercise falls to be attempted on facts which are conceded for this purpose to be true. I also take the view, for the reasons given by Kennedy LJ in paragraph 24 of his judgment that if the court comes to the conclusion that no duty of care exists, it should act, both to prevent the expenditure of substantial further costs and to dispel unrealistic expectations on the part of the Claimants.

I approach the instant case through Watson v. BBBC because it is a case in which, as the Master of the Rolls himself remarks, he was "in no doubt that the judge's decision broke new ground in the law of negligence" (see [2001] QB 1134 at 1142 paragraph 7). We may therefore, I think, take it that it represents the latest, and more than incremental step (see [2001] QB 1134 at 1162 paragraph 86) in the development of the law of negligence and the furthest the courts have gone so far in extending the concept of the duty of care. It therefore provides a valuable guide against which to test the assumed facts of the instant case.

Immediately following the reference to the case breaking new ground, the Master of the Rolls in Watson v. BBBC approves the incremental approach to the development of the law of negligence preferred by Brennan J in the High Court of Australia in Sutherland Shire Council v Heyman (1985) 157 CLR 424, 481 itself subsequently approved by the House of Lords in Caparo Industries plc v Dickman [1990] 2 AC 605 (Caparo).

The Master of the Rolls begin his analysis of the authorities with the well known speech of Lord Bridge of Harwich in Caparo, the relevant extracts from which Kennedy LJ has set out in paragraph 12 of his judgment, and which I need not repeat. After a brief discussion of the concept of proximity in the context of Perrett v Collins [1998] 2 Lloyd's Rep 255, the Master of the Rolls identifies the principles alleged to give rise to a duty of care in Watson v. BBBC itself. They are those of "assumption of responsibility and reliance". The Master of the Rolls summarises Mr. Watson's case in the following way ([2001] QB 1134 at 1149 paragraph 43): -

(i) The Board assumed responsibility for the control of an activity the essence of which was that personal injuries should be sustained by those participating.

(ii) The Board assumed responsibility for determining the details of the medical care and facilities which would be provided by way of immediate treatment of those who received personal injuries while taking part in the activity.

(iii) Those taking part in the activity, and Mr Watson in particular, relied upon the board to ensure that all reasonable steps were taken to provide immediate and effective medical attention and treatment to those injured in the course of the activity.

The peculiar features of the duty of care alleged in Watson v. BBBC are identified by the Master of the Rolls as follows: (ibid paragraph 44)

(i) The duty alleged is not to take reasonable care to avoid causing personal injury. It is a duty to take reasonable care to ensure that personal injuries already sustained are properly treated.

(ii) The duty alleged is not directly, through the servants or agents of the board, to provide proper facilities and administer proper treatment to those injured. It is to make regulations imposing on others the duty to achieve these results.

Under the heading A duty to administer treatment ([2001] QB 1134 at 1150 to 1153 paragraphs 43 to 58) the Master of the Rolls then examines the cases dealing with the assumption of a responsibility to exercise reasonable care to safeguard a victim from the consequences of an existing personal injury or illness. Those cases, he says: -

……support the proposition that the act of undertaking to cater for the medical needs of a victim of illness or injury will generally carry with it the duty to exercise reasonable care in addressing those needs. While this may not be true of the volunteer who offers assistance at the scene of an accident, it will be true of a body whose purpose is or includes the provision of such assistance.

Lord Phillips MR then turns to cases which he identifies under the heading Indirect influence on the occurrence of injury ([2001] QB 1134 at 1153 to 1159 paragraphs 59 to 72). No case had been cited in which a duty of care had been established in relation to the drafting of rules and regulations governing the conduct of third parties towards a claimant. There were, however, cases involving negligent advice which had resulted in injuries to third parties. Thus in Clay v AJ Crump & Sons Ltd [1964] 1 QB 533 a building worker was injured when a wall collapsed on him. The wall had remained standing because the architect employed in supervising the building work had failed to advise that it was dangerous and should be demolished. This court held that the architect owed a duty of care under Lord Atkin's Donoghue v Stevenson neighbour principle.

However, in Marc Rich & Co AG v. Bishop Rock Marine Ltd [1996] AC 211, the House of Lords on a preliminary issue held that a classification surveyor who had surveyed a vessel laden with cargo and given it a clean bill of health did not owe a duty of care to the ship owners when the vessel sailed and sank a few days later with the loss of the cargo. Whilst the decision turned essentially on considerations of policy in relation to the role of a classification society in the context of the insurance of risks inherent in the carriage of goods by sea, the Master of the Rolls cites an extract from Lord Steyn's speech at [1996] AC 211 at 242: -

Given that the cargo owners were not even aware of NKK's examination of the ship, and that the cargo owners simply relied on the undertakings of the shipowners, it is in my view impossible to force the present set of facts into even the most expansive view of the doctrine of voluntary assumption of responsibility.

Following a discussion of X (Minors) v Bedfordshire County Council [1995] 2 AC 633 and Phelps v Hillingdon London Borough Council [2001] 2 AC 619 (to which I shall return), the Master of the Rolls then returns to Perrett v Collins. That case concerned the certification by an inspector from an organisation called the Popular Flying Association that a light aircraft assembled from a do-it-yourself kit was fit to fly. It was not, and the claimant, a passenger in it, was injured. The Court of Appeal rejected the defence that any negligence of the person certifying airworthiness was not the direct cause of the claimant's injuries, the argument being that the direct cause was not the certification but the faulty design of the aircraft. The claim against the certifying inspector was a conventional claim for negligence causing direct and foreseeable personal injury. The Master of the Rolls cites passages from the judgment of Hobhouse LJ in Perrett v Collins in which he says: -

A minimum requirement of particularity and contemplation is required. But it has never been a requirement of the law of the tort of negligence that there be a particular antecedent relationship between the defendant and the plaintiff other than one that the plaintiff belongs to a class which the defendant contemplates or should contemplate would be affected by his conduct. Nor has it been a requirement that the defendant should inflict the injury upon the plaintiff. Such a concept belongs to the law of trespass not to the law of negligence….. ([1998] 2 Lloyd's Rep 255, 261)

Where the plaintiff belongs to a class which either is or ought to be within the contemplation of the defendant and the defendant by reason of his involvement in an activity which gives him a measure of control over and responsibility for a situation which, if dangerous, will be liable to injure the plaintiff, the defendant is liable if as a result of his unreasonable lack of care he causes a situation to exist which does in fact cause the plaintiff injury. Once this proximity exists, it ceases to be material what form the unreasonable conduct takes. The distinction between negligent mis-statement and other forms of conduct ceases to be legally relevant, although it may have a factual relevance to foresight or causation ([1998] 2 Lloyd's Rep 255, 262).

The Master of the Rolls notes that in Perrett v Collins whilst both Swinton Thomas and Buxton LJJ agreed that the case fell into the category of negligence directly causing foreseeable personal injury, both also took the view that it was fair and reasonable to impose a duty of care in the circumstances of the case. The Master of the Rolls concludes this part of his judgment with the following proposition: -

72 These cases establish that, where A advises B as to action to be taken which will directly and foreseeably affect the safety or well-being of C, a situation of sufficient proximity exists to found a duty of care on the part of A towards C. Whether in fact such a duty arises will depend upon the facts of the individual case and, in particular, upon whether such a duty of care would cut across any statutory scheme pursuant to which the advice was given.

Lord Phillips MR then considers whether the principles derived from the established cases should lead to a finding that the BBBC owed Mr. Watson a duty of care. In coming to the conclusion that the BBBC did owe Mr. Watson a duty of care, the Master of the Rolls first examines the question of proximity: (see [2001] QB 1134 at 1159 to 1161, paragraphs 73 to 83). He rejects each of the five arguments advanced on the Board's behalf. I propose to set these out in summary form, followed immediately by a summary of the Master of the Rolls' response: -

(1) The Board exercised a public function which it had assumed for the public good.

The broad function of the Board is to support professional boxing. Its members are those involved in professional boxing. Caring for the needs of boxers, and in particular the physical safety of boxers is the primary object of the Board

(2) If the Board had made its rules pursuant to a statutory power it would be tolerably clear that it could not be held liable in negligence in relation to the matter in which it chose to exercise its discretion: - X (Minors) v Bedfordshire County Council: Stovin v Wise [1996] AC 923.

It was not profitable to speculate on what the position might be if the Board had a statutory function in relation to boxing. However, if the Board were given the statutory function of directing what medical assistance should be provided to boxers at the stadium, it would be at least arguable that it owed boxers a duty of care in exercising that function

(3) There was insufficient proximity between the Board and the objects of the duty (the boxers) because the duty was allegedly one owed for an indeterminate time to an indeterminate number of people.

The duty owed was to a determinate class – professional members who were members of the Board (about 550 professional boxers in 1993). The numbers of those to whom the duty is alleged to be owed in the present case are not incompatible with the requirements of proximity

(4) The Board did not create the danger of injury or the need for medical assistance: it therefore owed no greater duty of care than that owed by a rescuer, namely a duty to take reasonable care not to make the situation worse.

There is a clear distinction between the role of the Board and the role of a fire service or the police service. The latter have the role of protecting the public in general against risks, which they play no part in creating. There is a general reliance by the public on the fire service and the police to reduce those risks. In these circumstances there is no close proximity between the services and the general public. There are also reasons of public policy for not imposing a duty of care to individuals in relation to the performance of their functions

In contrast, the injuries which are sustained by professional boxers are the foreseeable, indeed inevitable, consequence of an activity which the Board sponsors, encourages and controls. Moreover, the board arrogated to itself the task of determining what medical facilities will be provided at a contest by (i) requiring the boxer and the promoter to contract on terms under which the board's rules will apply and (ii) making provision in those rules for the medical facilities and assistance to be provided to care for the boxer in the event of injury. These facts brought the board into close proximity with each individual boxer who contracts with a promoter to fight under the board's rules. The comparison between the board and a rescuer was not apt.

(5) The Board did not provide medical treatment or employ doctors. It acted as a regulatory rule-making body. It was accepted that if a boxer asked specifically for advice about precautions to be taken and the Board had given advice it would have been under a duty to exercise care in giving that advice. As to boxers generally, however, the Board had no duty of care.

Had the Board simply given advice to all involved in professional boxing as to appropriate medical precautions, it would be strongly arguable that there was insufficient proximity between the board and individual boxers to give rise to a duty of care. The board, however, went far beyond this. It made provision in its rules for the medical precautions to be employed and made compliance with these rules mandatory. As Mr Morris accepted, by reason of its control over boxing the board was in a position to determine, and did in fact determine, the measures that were taken in boxing to protect and promote the health and safety of boxers.

The Master of the Rolls concludes that the facts of Watson v. BBBC produced a relationship of close proximity between the Board and those of its members who were professional boxers.

The Master of the Rolls next examines reliance ([2001] QB 1134 at 1161 to 1162, paragraphs 84 and 85). He found the submission that Mr. Watson did not rely on the exercise of care by the Board unrealistic. The Board was a body which held itself out as treating the safety of boxers as of paramount importance. Boxer members, including Mr. Watson, could reasonably rely on the Board to have taken reasonable care in making provision for their safety. The Board was a body with special knowledge giving advice to a defined class of persons in the knowledge that it would rely upon that advice in the defined situation of boxing contests. The judge was entitled to conclude that there had been reliance by Mr. Watson on the exercise of care and skill by the Board in looking after his safety. This was a further factor which tended to establish the proximity necessary for a duty of care.

The Master of the Rolls then considered the question: was it fair, just and reasonable to impose a duty of care in Watson v. BBBC? Because the facts were so unusual, there was no category in which a duty of care had been established from which one could advance to the case by a small, incremental step. It was difficult if not impossible to avoid a degree of subjectivity, but the approach had to be to apply established principles and standards. The Master the Rolls was attracted to formulation of legal principle by Hobhouse LJ in Perrett v Collins, which I have set out at paragraph 15 of this judgment.

The Master of the Rolls applied that formulation to the facts of Watson v. BBBC in the following way ([2001] QB 1134 at 1162, paragraph 87): -

Mr Watson belonged to a class which was within the contemplation of the board. The board was involved in an activity which gave it, not merely a measure of control, but complete control over and a responsibility for a situation which would be liable to result in injury to Mr Watson if reasonable care was not exercised by the board. Thus the criteria identified by Hobhouse LJ for the existence of a duty of care were present. In this case the following matters are particularly material.

(1) Mr Watson was one of a defined number of boxing members of the board

(2) A primary stated object of the board was to look after its boxing member's physical safety.

(3) The board encouraged and supported its boxing members in the pursuit of an activity which involved inevitable physical injury and the need for medical precautions against the consequences of such injury.

(4) The board controlled every aspect of that activity.

(5) In particular, the board controlled the medical assistance that would be provided.

(6) The board had, or had access to, specialist expertise in relation to appropriate standards of medical care.

(7) The board's assumption of responsibility in relation to medical care probably relieved the promoter of such responsibility. If Mr Watson has no remedy against the board, he has no remedy at all.

(8) Boxing members of the board, including Mr Watson, could reasonably rely upon the board to look after their safety.

Finally for present purposes, the Master of the Rolls addressed policy considerations ([2001] QB 1134 at 116, paragraphs 89 to 91). He rejected the argument that a duty of care should not be imposed on the Board because it was a non-profit making organisation and did not carry insurance. The Master of the Rolls likened this argument to that advanced in Perrett v Collins and it was to be rejected for the same reasons.

The Claimant's pleaded case.

As Kennedy LJ has pointed out, the Particulars of Claim run to some 34 pages. Rather than attempt my own summary, I am content to adopt that undertaken by the judge who, in paragraph 30 of his judgment summarises the Claimant's case in the following way: -

…. The primary source of drinking water in Bangladesh was groundwater from wells. The presence of arsenic in the water was a major health hazard. While in many countries local authorities would ensure the proper testing of drinking water, the Bangladesh Government was not in a position adequately to test the quality of the drinking water and, in particular, for the presence of potentially toxic trace elements, such as arsenic. The objects of the Defendant under its Charter included the provision of advice. The Defendant was paid by the Overseas Development Agency (ODA) out of development and funds designated for the benefit of the Bangladeshi people to conduct a hydro chemical baseline survey of the groundwater quality. The 1992 Report included an assessment of the water's toxicity to humans; and the Defendant therefore knew that it would be used for the benefit of consumers of ground water from wells. In 1992 the possibility of arsenic being present in the groundwater in the relevant area should have been known to reasonably competent hydrologists and should have been included as an element to be tested for in any properly conducted hydro chemical baseline survey. Readers of the 1992 Report could reasonably assume that, if the survey had not been tested for particular trace elements there was no need to do so. Alternatively, if in fact the Report was not intended to test for the potability of water for humans, then it was incumbent on the Defendant to have made it clear that it could not be relied on for that purpose. The Report on the survey was intended by the Defendant and the ODA (a) to be for the use and benefit of the Bangladeshi government and agencies involved in the management of Bangladeshi water resources, and (b) to be widely distributed to such interested parties. As a consumer of drinking water in Bangladesh the Claimant was in a class which (a) would be bound to be affected by any action taken on their behalf by the Bangladeshi government and agencies in relation to the management of water resources, and therefore (b) would be affected by the content of the 1992 Report.

Analysis.

How do the principles identified in the previous cases and applied by the Court of Appeal in Watson v. BBBC relate to the question in the instant case of whether or not the Defendant owed a duty of care to the Claimant on the basis of his pleaded case? In my judgment they demonstrate that the relationship is too remote for a duty of care to be imposed by the court on the defendant.

Although the Defendant was involved between 1984 and 1988 in testing the hydraulic efficiency of the Deep Tubewell (DTW) designs and undertaking a hydrological analysis; and although it is asserted that this was a "provision of support and research services for the benefit of the Bangladesh Government and people" it is not suggested that the Defendant had any role in the actual installation of the 4000 DTWs in Bangladesh between 1983 and 1992. The Defendant's role, between 1984 and 1988 was to provide "research services for the benefit of the Bangladesh Government and/or the Bangladeshi people in connection with the DTW II Project". The result was a report entitled "Pilot Study into the Optimum Deep Tubewell Design" about which no complaint is made. The Defendant did not therefore build the DTWs from which the water was extracted, nor, of course, did it have any control over or responsibility for the quality of the water which was extracted.

In the terms of Lord Phillips' analysis at paragraph 72 of his judgment Watson v BBBC ([2001] QB 1134 at 1159 set out at paragraph 16 above), the duty for which the Claimant contends has, I think, to be formulated in a different way. I hope I am not misrepresenting the Claimant's case if I formulate it in the following way. If A (the Defendant) being an expert with the requisite expert knowledge writes a report for B (the ODA), knowing that B will pass the information contained in the report to C (the Bangladesh Government); and if A fails in that report to alert B (and, through B, C) to the existence of a hazard which will directly and foreseeably affect the safety or wellbeing of D (one of a class of people comprising all those citizens of Bangladesh who are exposed to the hazard) with the consequence that C does not take the steps which it would otherwise have taken to protect D (and the class of which he is a member) from the hazard, does a sufficient proximity exist to found a duty of care on the part of A towards D?

I have a number of difficulties with identifying a duty of care between A and D this formulation. The first is Lord Atkin's neighbour principle. It is perhaps worth remembering that the famous statement of principle contained in that case and which Clarke LJ cited at paragraph 22 of his judgment is introduced with words which limit the extent of its application: - see [1932] AC 562, 580. Thus:

…. Acts or omissions which any moral code would censure cannot in a practical world be treated so as to give a right to every person injured by them to demand relief. In this way rules or law arise which limit the range of complainants and the extent of their remedy. The rule that you are to love your neighbour becomes in law; you must not injure your neighbour, and the lawyers' question: Who is my neighbour? receives a restricted reply. You must take reasonable care to avoid acts or omissions which you can reasonably foresee would be likely to injure your neighbour……

The famous identification of the lawyers' neighbour follows. My neighbour must be in a class of persons

so closely and directly affected by my act that I ought reasonably to have them in contemplation as being so affected when I am directing my mind to the acts or omissions which are called in question

Speaking for myself, I am unable, in the instant case to bring the Claimant and the Defendant within the neighbour principle on the conceded facts of this case. The most favourable of the modern authorities, from the Claimant's perspective, must be Perrett v Collins and Watson v. BBBC. I find it very difficult to fit the relationship between the Claimant and the Defendant into Hobhouse LJ's identification in Perrett v Collins, of the "measure of control" necessary for the establishment of a duty of care in tort in the passage which I have set out at paragraph 15 of this judgment. I have equal difficulty with the elements both of control and determinate class which the Master of the Rolls in Watson v BBBC identifies as necessary for the existence of a duty of care.

Even assuming for this purpose that the Claimant does belong to a class which was so closely and directly affected by its act that the Defendant ought reasonably to have had the Claimant in contemplation as being so affected when conducting the survey, I do not think it can be properly be said that the Defendant in the instant case had any measure of control over, or responsibility for the presence of the arsenic in the water, or that by its unreasonable lack of care it caused a situation to exist which did in fact cause the Claimant injury. The Defendant did not construct the wells. Self-evidently, it did not put the arsenic in the water. It was not responsible for the supply of potable water to the Claimant and the wide class to which he belonged. It had no control over the manner in which the wells were used, and no control over who drank ground water from them.

The body which plainly owed a duty of care to the Claimant to provide him and the class to which he belonged with a clean water supply was the Bangladesh Government and any subsidiary body which that government created and made responsible for the supply of uncontaminated drinking water. The fact that those bodies did not undertake or were incapable of undertaking the task does not, in my judgment, of itself render the Defendant liable in their place. The highest it can be put, I think, is that the Defendant owed a duty to the ODA, and through the ODA to the Bangladesh government competently to carry out its geophysical survey of the hydrochemistry of the aquifers. That does not, however, in my judgment, impose on it a duty of care to take reasonable steps to ensure either that the Claimant had a supply of potable water, or that the groundwater from the DTWs was safe to drink. These were not matters for which the Defendant had any responsibility or over which it had any measure of control

I therefore agree with Mr. Beloff that the authorities which Lord Phillips MR analyses in Watson v BBBC (and indeed Watson v BBBC itself) support the proposition that in circumstances where a duty of care has been found not to exist, or where it has been doubted, the lack of control over, or responsibility for the dangerous situation has been a central factor pointing to the absence of such a duty.

I do also have difficulty with the identification of a duty of care owed to the entire class to which the Claimant belongs. At its widest, that class must comprise anybody who drank groundwater contaminated with arsenic from any Tubewell from which the Defendant had taken a sample or samples. None of the authorities, it seems to me, embrace a class of potential claimants as wide as that envisaged in this case. Lord Brennan was at pains to assure us that of the current figure of 699 potential claimants, the 512 who had applied for public funding were being carefully vetted. But that, with respect, is not the point. The potential class of claimants, if not unlimited, must run to many thousands of people. If a person or body is to be held to have a duty of care towards, and to be made potentially liable to an extremely wide class of claimants, the nature and basis of the duty of care alleged needs, in my view, to be very clear: indeed, the wider the class of potential claimants, the greater the need, it seems to me, to identify the nature of the duty with clarity and precision.

I also agree with Mr Beloff that the authorities, notably in this respect, Watson v BBBC place reliance both on the particular extent to which there is some particular contract, transaction, event or situation which can be said to provide a particular nexus between the Claimant and the Defendant; and on the primary aim or purpose of the Defendant's function or activity. Neither factor is present in the instant case.

Mr. Beloff identified some 18 features of the relationship (or absence of it) between the Claimant and the Defendant which, he submitted, arose from the Claimant's pleading, and which could not be disputed. In addition to those to which I have already referred, Mr. Beloff relied on the fact that the Report was funded by and prepared for the ODA, which also agreed its subject matter. There was no contractual relationship between the Defendant and either the Claimant or the Bangladesh government. The Defendant gave no undertaking to either as to the use to which the Report would be put: indeed, the Claimant was unaware of the existence of the Defendant or the Report. (The same, of course, can be said for the entire class of potential claimants). The primary purpose of the Report was to produce a body of data that could be used to describe the hydrochemistry of the main aquifer units of central and of north-eastern Bangladesh. The Defendant did not control the distribution of the Report and was not under any statutory or other obligation to provide the report to those responsible for the provision of water for domestic use to the Claimant.

Whilst I would not regard the mutual lack of knowledge between the Claimant (and others within the class of potential claimants) and the Defendant as to each other's existence as being, of itself, necessarily fatal to the Claimant's claim, the points which Mr. Beloff makes seem to me to buttress the arguments that the class to which the Claimant belongs is too wide, and that the lack of responsibility and control which Mr. Beloff identifies are inimical to the creation of a duty of care between them.

X (Minors) v Bedfordshire County Council (X v Bedfordshire).

Insofar as Mr. Beloff's reasoning relies upon the decision of the House of Lords in X v Bedfordshire, Lord Brennan submits that its force has been substantially reduced by its re-assessment by this court in D v East Berkshire Community NHS Trust; Mak v Dewsbury Healthcare NHS Trust; RK v Oldham NHS Trust [2003] 2 FLR 1166 (D v East Berkshire). Whilst, speaking for myself, I do not perceive X v Bedfordshire as central to the argument, it was considered in detail by Lord Phillips MR in BBBC v Watson, and I propose, accordingly, to examine it in the current contact.

Mr. Beloff relied on X v Bedfordshire as a case in which the court struck out claims on the basis that, as Lord Browne-Wilkinson said at [1995] 2 AC 633, 704-1 there were cases "where it is evident that, whatever the facts, no common law duty of care can exist". He also relied upon it for the proposition that the lack of any actual control over or responsibility for the dangerous situation (as opposed to the giving of advice that may only influence the decision of others who have such control or responsibility) will be a factor in determining that there is no duty of care (the subsidiary role point). This point, it seems to me, is not dependent upon X v Bedfordshire, as my discussion of Perrett v Collins and Watson v BBBC demonstrates.

The essence of the decision in X v Bedfordshire seems to me one of policy based on the nature of the statutory duties imposed on local authorities under legislation dealing with child protection and education. In both the child abuse and the education / special needs cases, the House of Lords declined to impose on local authorities a common law duty of care owed to individual parents and children in the performance of their statutory functions. Equally, local authorities were not to be held vicariously liable for the actions of social workers and other professionals engaged to advise them on the proper exercise of those functions. The only exception related to psychological advice given in the context educational assessments and in the determination by an educational authority of a child's educational needs.

X v Bedfordshire was revisited several times by the House of Lords itself before the decision of this court in D v East Berkshire. The critical feature of that latter case, so far as the children involved were concerned, was the conclusion that so far as the duty of care owed to children was concerned, X v Bedfordshire could not survive the advent of the Human Rights Act 1998, and that it was no longer legitimate to rule that, as a matter of law, no common law duty was owed to a child in relation to the investigation of suspected child abuse or the initiation and pursuit of care proceedings. However, the position of the adults was different. Since the parents' and the children's Convention rights might well be in conflict, there were cogent reasons of public policy why a duty of care towards parents should not be imposed on local authorities taking child care decisions in pursuit of their statutory responsibilities.

Accordingly, in D v East Berkshire this court decided that the effect of X v Bedfordshire should be limited to the core proposition identified by Lord Slynn of Hadley's statement in Barrett v Enfield Borough Council [2001] 2 AC 550, 569 namely that X v Bedfordshire established that decisions by local authorities whether or not to take a child into care were not reviewable by way of a claim in negligence. That seems to me, with respect, to be an accurate summary of the state of the authorities.

The decision of this court in D v East Berkshire does not, however, in my judgment affect the outcome of the instant case in any way. In terms both of proximity and the rights of children under Article 8 of the Convention (particularly where the statute which creates the duty exercisable towards them requires the court to treat their welfare as paramount and the local authority to safeguard and promote it) the identification of a duty of care owed by local authorities to children in their care seems to me a logical and proportionately incremental step. It is, I think, another matter to impose such a duty in relation to a child's parents.

All this is, in my judgment, a long way away from the instant case, in which Convention Rights are not engaged, and in which no question of policy deriving from a breach of Statutory duty arises. In my judgment, accordingly, Lord Phillips' analysis in Watson v BBBC of the remoteness issue based on X v Bedfordshire does not affect the outcome of the instant case.

The 'Daisy Chain' of Reliance and Causation

It is, I think, difficult to fit the concept of reliance into the mould created for it by Lord Phillips MR in Watson v BBBC: - see paragraph 79 above. It is accepted that both the individual Claimant in this case and the class he represents have no knowledge of the existence either of the Defendant or the 1992 report. In paragraph 57 of his judgment, under the heading The Claimants' relationship with the Defendant, Simon J identified four stages in what he described as "the chain in relation to reliance". These were: -

(1) the Defendant produced the 1992 Report for the ODA as part of the DTWII project;

(2) copies were sent to Mott McDonald Ltd (which was the contractor for whom the Defendant worked on the DTWII project) and various agencies associated with the Bangladesh Government (to which quarterly reports on the DTWII project were sent) included the BADC (Bangladesh Agricultural Development Corporation) (which was responsible for the DTWII project) and the North Eastern Regional Project Flood Action plan, the Department of Geology at the University of Dhaka;

(3) from there copies of the Report were passed on to other interested parties, including those concerned with water resource management and supply, and were relied on by them for water quality data;

(4) the Claimant, relying on such organisations, carried on drinking the water.

Mr. Beloff described this as a "daisy chain" demonstrating both the lack of any primary responsibility on the part of the Defendant for the provision of safe drinking water to the Claimant and the indirect nature of the alleged damage. He also submitted that, assuming for this purpose the Defendant had detected the presence of arsenic in the water, there were on the pleadings and on the evidence filed on behalf of the Claimant at least two further necessary links which had to be inserted into the chain between stages (3) and (4) before any action would have been taken to protect the Claimant and the class to which he belongs. These were (1) a baseline survey to understand the magnitude of the problem and to produce an accurate map of its extent (the evidence of Mr. Ravenscroft and Ms Bennett); and (2) the implementation of an arsenic mitigation programme, together with educational programmes and alternative water supplies. It would only be at that point, he submitted, that the Claimant and others would stop drinking the contaminated water, subject to their level of education / understanding and the availability of an alternative supply of water.

Clarke LJ, as I read his judgment, takes the view that the "daisy chain" argument goes essentially to foreseeability of damage, and is accordingly covered by Mr. Beloff's concession that if there is a duty of care, negligence and foreseeability of damage are arguable. Whilst I see the force of that position, I agree with Mr. Beloff that the chain of causation argument also goes to the concepts of primary responsibility and control and can legitimately be prayed in aid in support of the argument that no duty of care is capable of existing on the facts of the case.

Fair just and reasonable

In so far as the imposition of a duty of care requires it to be fair just and reasonable for such a duty to be imposed, I would like to associate myself expressly with Kennedy LJ's observations in paragraph 24 of his judgment. It does not seem to me either fair, just or reasonable, and would be a mighty leap to make the Defendant liable to a substantial portion of the Bangladeshi population for damages for personal injuries arising from drinking polluted water on the assumed facts of this case. Whilst Kennedy LJ reaches that conclusion, as do I, without reference to wider policy decisions, it does not seem to me those the considerations which he identifies can be wholly ignored, and any imposition of a duty of care which would undoubtedly have the effect of discouraging aid projects aimed at benefiting the populations of under developed countries would, in my judgment, require either a very clear factual nexus or a clear voluntary assumption of responsibility for those consequences.

The propriety of striking out.

Clarke LJ, in his judgment, places considerable emphasis on the unreported decision of this court in Farah and others v British Airways and the Home Office (6 December 1999) (Farah). The question in that case, as identified by Lord Woolf MR (as he then was) was whether the Home Office could be liable for the loss caused to immigrants as a result of an immigration liaison officer negligently and wrongly advising an airline that the immigrants did not have the required documentation to obtain access to the United Kingdom, if, as a result of this, the airline did not fly the immigrants to the United Kingdom. The judge had struck out the claim on the ground that there was insufficient proximity between the claimants and the Home Office to create a duty of care. This court reversed that decision.

Lord Woolf was properly concerned with the precise nature of the relationship in these circumstances between the airline, the Home Office and the immigrants. Comparing the situation in Farah, and the situation in which errors by immigration officers in the course of exercising their normal statutory duties did not give rise to a duty of care, Lord Woolf said (at paragraph 34 of his judgment): -

It seems to me there is a real distinction between the two situations. The activities of a liaison officer when giving advice to a carrier such as British Airways as to the documentation of would-be immigrants to this country, is not part of the ordinary functions of an immigration officer. It is a responsibility which is exercised by the liaison officer to assist the airline. It is also to assist the immigrants because if they did not have the valid documentation, it is not in their interest to travel the long distances to this country only to be refused entry. It is also in the interests of the Home Office who have the difficult task of controlling immigration into the country.

35. If those who do not have the appropriate documentation can be prevented or deterred from flying to this country, that makes the administration of the Immigration Act 1971 easier. In my judgment, the fact that there is this interest between the Home Office, the carrier and the would-be immigrant, coupled with the fact that any advice given would obviously have a direct affect upon the immigrants concerned, is part of the reasons why the facts in this case should be investigated before a conclusion is reached as to whether a duty of care exists or not. This is an area of developing jurisprudence. Where that is so, the question of whether or not an analogous situation should be recognised as giving rise to a duty of care, should be determined when the facts have been established.

In the situation identified in Farah, the conclusion reached by this court is plainly correct. As the citation from Lord Woolf's judgment makes clear, the question, in essence, was who owed what duty and to whom? The Home Office could not rely on the absence of any duty of care arising from the exercise of immigration officers' statutory discretions per se and it was arguable (dependent on the facts found) that the Home Office owed a duty of care both to the airline and the immigrants. Identification of the correct duty or duties of care in these circumstances would plainly depend on the facts found at trial.

I find it impossible, however, to extract from Farah or the other authorities the proposition that it can never be possible to strike out a claim on the basis that, whatever the agreed facts, the case needs nonetheless to go to trial for formal findings of fact to be made. The test, as Lord Browne-Wilkinson says in Barrett v London Borough of Islington [2001] 2 AC 550, 557 is whether or not is possible to give a affirmative answer to the question whether the claim would be bound to fail. Plainly, as the authorities demonstrate, in many if not most cases it will normally not be appropriate to strike the claim out without an investigation into, and judicial determination of, the facts. But if a court reaches the clear conclusion that the claim is bound to fail because, on the way the claimant advances his case, no duty of care can properly be said to exist, it is, as I have already stated, important both that substantial expenditure in costs should cease and expectations should not be raised.

In the instant case the relationships between the parties (or lack of them) are, in my judgment, perfectly clear, and an assessment based on the most favourable factual analysis leads to the conclusion, on the authorities, that no duty of care arises.

Other matters.

Like Kennedy LJ I was impressed by Mr. Beloff's arguments on what Kennedy LJ has described as the "construction" issue. However, as Kennedy LJ points out, Lord Brennan is able to point to a number of references in the Report to toxicity in the context of human health, most notably, of course, the reference to the development of goitre where iodine is deficient in the water supply. Accordingly, I agree with Kennedy LJ that it would be premature to decide in favour of the Defendant on this issue at this stage.

Clarke LJ places emphasis on the statement of Sara Bennett. In my judgment, however, that statement goes primarily to the issues of negligence and reliance, not to the existence of a duty of care, although for the reasons identified in paragraph 106 her evidence, and that of Mr. Ravenscroft can be prayed in aid in support of the "daisy chain / primary responsibility / control argument. Equally, it seems to me the Claimant obtains only very limited assistance from the statement of Mr. Ravenscroft, to which Clarke LJ also refers. Mr. Ravenscroft makes the point that he did not think that Mr. Davies was in a position to decide which elements would be tested, and that those decisions were made in the United Kingdom by the laboratory. For example, he asked Mr. Davies to include bromide in the suite of chemicals to be tested, and reports Mr. Davies' reply that whilst he would ask for it to be included, he could not guarantee that it would be because it was the chemists in the United Kingdom who decided which elements would be tested. It is also perhaps significant that Mr. Ravenscroft did not suggest that the water be tested for arsenic, and did not notice that arsenic was not included when he received and read the report in 1992.

However, in my judgment it is not profitable to enter into these areas of the case save in so far as they tend to confirm the impression that the claim would be likely to fail in any event. Mr. Beloff placed his case fair and square on the proximity issue: in my judgment and for the reasons I have attempted to give, that argument succeeds.

Conclusion.

Had the judge conducted the exercise based on Watson v BBBC which I have attempted to undertake, I venture to think that he would have been driven, as I have been, to the conclusion that the Defendant, even on the facts posited by the Claimant did not owe a duty of care to the Claimant and the class which the Claimant represents, and that it would not be fair just and reasonable to impose such a duty. It must equally follow that the case should be halted at this stage, before substantial further costs are incurred.

For these reasons, and in full agreement with those given by Kennedy LJ, I too would allow this appeal.