THE SUPREME COURT OF APPEAL

REPUBLIC OF SOUTH AFRICA

JUDGMENT

Case number: 653/07

In the matter between:

FOURWAY HAULAGE SA (PTY) LTD APPELLANT

and

SA NATIONAL ROADS AGENCY LTD RESPONDENT

Neutral citation: Fourway Haulage v SA National Roads Agency (653/07) [2008] ZASCA 134 (26 November 2008)

CORAM: SCOTT, FARLAM, BRAND, LEWIS ET JAFTA JJA

HEARD: 5 November 2008

DELIVERED: 26 November 2008

CORRECTED:

SUMMARY : Delict – pure economic loss – meaning of – policy considerations relevant in determining wrongfulness – remoteness of damage – application of flexible test

ORDER

On appeal from : High Court, Pretoria (Rabie J sitting as court of first instance)

The appeal is dismissed with costs

JUDGMENT

BRAND JA (Scott, Farlam, Lewis et Jafta JJA concurring)

[1] The appellant ('Fourway') is a long distance haulier. The respondent ('the Agency') owes its existence to the South African National Roads Agency Limited and National Roads Act 7 of 1998 ('the Act'). The dispute between them originates from an accident which occurred in the early evening of 26 September 2003 on the N1 national road between Polokwane and Mokopane in the Limpopo province. The two vehicles involved were an articulated truck and a light delivery van. The articulated truck was driven at the time by an employee of Fourway who was acting in the course and scope of his employment.

[2] The articulated truck was on its way from an asbestos mine in Zimbabwe to Durban harbour carrying about 34 tonnes of chrysolite asbestos, destined for export. As a result of the collision, the truck overturned and spilled its cargo onto practically the entire surface of a portion of the national road and its surroundings. Because of the hazardous nature of asbestos powder, the spillage required an extensive cleaning-up and decontamination operation.

[3] To facilitate the cleaning-up and decontamination process, the traffic authorities closed the section of the national road involved and diverted the traffic in both directions onto an alternative road. This lasted for about 24 hours. The section of the national road which was closed forms part of a toll road. The alternative route was not subject to toll. As a result of the closure, two toll plazas – as defined in the Act – could not collect toll fees. Based on these facts, the agency as the entity authorised by s 27 of the Act to levy and collect toll fees on toll roads, instituted an action in delict against Fourway for the damages it allegedly suffered in the form of loss of toll revenue in an amount of R105 996.67.

[4] At the commencement of the trial, the parties asked the court a quo (Rabie J) to order a separation of issues. In terms of the separation order, the issues relating to the liability of Fourway were to be decided first, while the quantum of the Agency's alleged damages stood over for later determination. The preliminary issues were decided in favour of the agency. Hence the court declared Fourway liable for such damages as the Agency may prove in respect of the lost revenue it would have collected at the two toll plazas involved, but for the closure of the road. It also ordered Fourway to pay the costs of the preliminary proceedings. Fourway's appeal against that judgment is with the leave of the court a quo.

[5] Part of the controversy on appeal was brought about by a shift in the focus of the defence advanced by Fourway and the resulting mutation of the issues involved. A convenient starting point for an account of the mutation is the opening address by counsel for the Agency, as plaintiff, at the beginning of the trial. With reference to the pleadings, counsel at that stage defined the issues between the parties as follows:

(a) Whether or not the respondent had the necessary authority to collect toll fees on that portion of the toll road which was closed as a result of the collision.

(b) Whether the collision occurred as a result of the negligence of the driver employed by Fourway.

(c) Whether the occurrence of the collision necessitated the decontamination operation and the closure of the road.

[6] Counsel for Fourway did not react to this definition of the issues. During the trial, Fourway formally conceded the issue referred to in (a) and the evidence led by the parties therefore dealt exclusively with the issues in (b) and (c). But in argument at the end of the trial, Fourway's counsel, for the first time, raised two further contentions. First he submitted that the Agency's claim was for the recovery of pure economic loss which required the existence of a legal duty on the part of Fourway and that the Agency had failed to plead or establish the existence of such a legal duty. Secondly he submitted that the Agency had failed to establish the requirement of legal causation with reference to the loss which formed the basis of its claim.

[7] As we know from the result, the court a quo dismissed all defences relied on by Fourway, including those originally raised under what I categorised as (b) and (c), as well as the two new ones advanced for the first time in argument at the end of the trial. As to (b) and (c) the court found on the evidence presented that the negligence of Fourway's employee was the cause of the collision which necessitated both the decontamination process and the closure of the road. With regard to the defence based on the concept of pure economic loss, the court essentially held that the damage suffered by the agency did not amount to pure economic loss and that the question regarding the existence of a legal duty therefore did not arise. Finally the court held that the damages claimed could not be classified as too remote and that the requirement of legal causation had thus been satisfied.

[8] On appeal, it was conceded on behalf of Fourway that the court a quo was correct in deciding the issues under (b) and (c) against it. In consequence, the only issues on appeal turned on the contentions that were raised for the first time in argument at the end of the trial. They can be summarised thus:

(a) Whether the court a quo correctly came to the conclusion that the Agency's claim is not a claim for pure economic loss.

(b) If not, how the issue of wrongfulness should have been dealt with in the light of the fact that it was not pertinently raised in the pleadings.

(c) Whether the court a quo correctly came to the conclusion that the damages claim by the Agency cannot be regarded as too remote.

[9] The court a quo's finding that the damages claimed did not result from pure economic loss clearly emanated from its understanding of that concept. That understanding appears from the following statements in the judgment:

'The economic loss in this sense comprises patrimonial loss that does not result from a direct invasion of a subjective right of the person who suffered the loss.'

And that

'the aforesaid rights of the plaintiff. . . [ie the Agency's statutory rights to operate a toll road and to collect toll fees] were clearly subjective rights worthy of protection and which the plaintiff could enforce against other people.'

And that

'[c]onsequently, the loss suffered by the plaintiff is not a so-called pure economic loss, but the direct result of a direct infringement of subjective rights which was as such unlawful.'

[10] I do not share the court a quo's understanding of what is meant by 'pure economic loss' in the present context. I believe its meaning to be far less metaphysical. As explained by Harms JA in Telematrix (Pty) Ltd v Advertising Standards Authority SA 2006 (1) SA 461 (SCA) para 1, it means simply this:

'"Pure economic loss" in this context connotes loss that does not arise directly from damage to the plaintiff's person or property but rather in consequence of the negligent act itself, such as loss of profit, being put to extra expenses or the diminution in the value of property.'

(See also Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd 1985 (1) SA 475 (A) 497I-498H; Trustees, Two Oceans Aquarium Trust v Kantey & Templer (Pty) Ltd 2006 (3) SA 138 (SCA) para 14; Wille's Principles of South African Law 9 ed, (General editor: Francois du Bois) sv 'Delict' by Daniel Visser, 1105; Neethling, Potgieter & Visser, Law of Delict, 5 ed 268 et seq).

[11] Thus understood, the Agency's claim, in my view, falls squarely within the ambit of pure economic loss. As formulated, its claim was for loss of revenue in the form of toll fees resulting from the closure of the road. The Agency did not allege, nor did it set out to prove in evidence, that it was the owner of the road; that the road was physically damaged by the collision; or that the closure of the road resulted from any physical damage to the road. The Agency's argument on appeal, that in terms of s 7 of the Act it was in fact the owner of the road on which the collision occurred, is of no consequence and misses the point. For present purposes the question is not whether the Agency is in fact the owner of the road. The point is that it did not rely on such ownership to support its claim.

[12] Recognition that we are dealing with a claim for pure economic loss brings in its wake a different approach to the element of wrongfulness. This results from the principles which have been formulated by this court so many times in the recent past that I believe they can by now be regarded as trite. These principles proceed from the premise that negligent conduct which manifests itself in the form of a positive act causing physical damage to the property or person of another is prima facie wrongful. By contrast, negligent causation of pure economic loss is not regarded as prima facie wrongful. Its wrongfulness depends on the existence of a legal duty. The imposition of this legal duty is a matter for judicial determination involving criteria of public or legal policy consistent with constitutional norms. In the result, conduct causing pure economic loss will only be regarded as wrongful and therefore actionable if public or legal policy considerations require that such conduct, if negligent, should attract legal liability for the resulting damages (see eg Minister of Safety and Security v Van Duivenboden 2002 (6) SA 431 (SCA) paras 12 and 22; Gouda Boerdery BK v Transnet 2005 (5) SA 490 (SCA) para 12; Telematrix (supra) paras 13-14; Trustees, Two Oceans Aquarium Trust (supra) paras 10-12).

[13] In this light, so Fourway contended on appeal, the Agency was obliged to allege in its pleadings not only that the negligent conduct relied upon was wrongful, but that it also had to allege and prove the facts relied upon to substantiate the considerations of policy giving rise to a legal duty on the part of Fourway's employee. As a result of the Agency's failure to adhere to these rules of litigation, so the argument went, neither the policy considerations relevant to the question of wrongfulness, nor the factual basis underlying such policy considerations, were identified and investigated during the trial. In consequence, so the argument concluded, it would be prejudiced if the issue of wrongfulness were to be summarily disposed of at the appeal. Fourway therefore suggested that, unless this court upholds its contention that the damages claimed are too remote – to which I shall presently return – the issue of wrongfulness should be postponed and decided with the rest of the issues concerning the quantum of the Agency's damages, which are standing over in any event.

[14] The proposition that a plaintiff claiming pure economic loss must allege wrongfulness, and plead the facts relied upon to support that essential allegation, is in principle well founded. In fact, the absence of such allegations may render the particulars of claim exipiable on the basis that no cause of action had been disclosed (see eg Trope v SA Reserve Bank 1992 (3) SA 208 (T) at 214A-G; Indac Electronics (Pty) Ltd v Volkskas Bank Ltd 1992 (1) SA 783 (A) 797E; Telematrix (supra) para 2). But, as we know, Fourway did not file an exception. The trial proceeded without any objection on its part. In the circumstances it would be futile to investigate whether an exception, if properly and timeously taken, would have been successful. As I see it, the question is rather whether, despite the lack of necessary allegations in the Agency's pleadings, Fourway had sufficient opportunity to produce the facts it would seek to rely on for the determination of the policy considerations pertaining to wrongfulness in its favour. Conversely stated, the question is whether Fourway has shown prejudice, in the sense that it would have conducted its case in a materially different way if the Agency's claim for pure economic loss had been properly pleaded. (See eg Shill v Milner 1937 AD 101 at 105; Robinson v Randfontein Estates GM Co Ltd 1925 AD 173 at 198; Collen v Rietfontein Engineering Works 1948 (1) SA 413 (A) at 433; Stead v Conradie 1995 (2) SA 111 (A) at 122A-H.)

[15] As I see it, the proposal by Fourway that the issue of wrongfulness be referred back for determination by the trial court therefore depends on the outcome of two discrete enquiries. First, can this court, on the basis of the facts available, decide that, as a matter of policy, Fourway should be held liable for the loss of revenue claimed by the Agency? If not, that would be the end of the matter. The Agency would have failed to make out a case. A decision on the other hand that the issue of wrongfulness should on the facts available be determined in favour of the Agency will lead to the next enquiry. The question is: can it be said that, if the issue of wrongfulness had been properly pleaded by the Agency, Fourway would have conducted its case any differently? If not, the Agency is entitled to succeed. It is therefore only a finding of potential prejudice on the part of Fourway that can justify a referral back to the trial court.

[16] The enquiry, whether as a matter of policy Fourway should be held liable for the pure economic loss suffered by the Agency, raises a question which is logically anterior: what are the considerations of policy that should be taken into account for purposes of the enquiry? In accordance with what criteria should the relevant considerations of policy be identified? Must we accept that policy considerations are by their very nature incapable of pre-determination and that the identification of the policy considerations that should find application in a particular case are to be left to the discretion of the individual judge? Does this mean that in the context of pure economic loss the imposition of liability will depend on what every individual judge regards as fair and reasonable? I believe the answer to the last two questions must be 'no'. Liability cannot depend on the idiosyncratic views of an individual judge. That would cloud the outcome of every case in uncertainty. In matters of contract, for example, this court has turned its face against the notion that judges can refuse to enforce a contractual provision purely on the basis that it offends their personal sense of fairness and equity. Because, so it was said, that notion will give rise to legal and commercial uncertainty (see eg Brisley v Drosky 2002 (4) SA 1 (SCA) paras 21-25; South African Forestry Co Ltd v York Timbers Ltd 2005 (3) SA 323 (SCA) para 27). I can see no reason why the same principle should not apply with equal force in matters of delict. A legal system in which the outcome of litigation cannot be predicted with some measure of certainty would fail in its purpose. As pointed out by Lord Scott of Foscote in Lagden v O'Conner [2004] 1 AC 1067 (HL) para 86:

'One of the main functions of the law of obligations, contractual or tortious, is to provide, or attempt to provide, a set of yardsticks for determining whether a legal injury has been inflicted on a person (the claimant) by another person (the defendant) and, if so, for determining the amount of the damages that the defendant must pay by way of reparation. If the two parties are unable to agree, an answer can be found by recourse to litigation. But the cost of litigation, often excessive both in absolute terms and in relation to the amount in dispute, and the inevitable delay, worry and anxiety that accompany court proceedings provide impelling reasons why the yardsticks by means of which legal liability is to be measured should be kept as simple and uncomplicated as practicable.'

[17] We therefore strive for certainty. The question is, how can that be achieved in an area directed by considerations of public or legal policy? I believe we must accept at the outset that absolute certainty is unattainable. The moment this court took the first tier policy decision – in Administrateur, Natal v Trust Bank van Afrika Bpk 1979 (3) SA 824 (A) – to abolish the absolute exclusion of liability for pure economic loss, it abandoned the bright line of absolute certainty. The second tier policy decision as to when liability should be imposed must of necessity be accompanied by some degree of uncertainty, at least at the early stages of development in this area of the law. That much was recognised and predicted by Rumpff CJ in Administrateur, Natal itself (see 831B). This measure of resulting uncertainty also seems to be an experience shared by those jurisdictions where the same first tier policy decision has been taken. Thus it was stated, for example, by Gaudron J in the Australian High Court, in Perre v Apand (Pty) Ltd 1999 198 CLR 180 (HC of A) para 25:

'The law as to liability for economic loss is a "comparatively new and developing area of the law of negligence". It has not yet developed to a stage where there has been enunciated a governing principle applicable in all cases. Perhaps it never will.'

And by McLachlin J in the Canadian Supreme Court in Canadian National Railway Co v Norsk Pacific Steamship Co Ltd (1992) 91 DLR (4^th) 289 at 366:

'Judges seem able to pick out deserving cases when they see them. The difficulty lies in formulating a rule which explains why judges allow recovery of economic loss in some cases and not in others.'

(Compare also K Zweigert & H Kötz An Introduction to Comparative Law 3 ed 625 et seq; B S Markesinis The German Law of Torts, A Comparative Introduction 3 ed 42 et seq; Daniel Visser & Niall Whitty The Structure of the Law of Delict in Kenneth Reid and Reinhard Zimmermann A History of Private Law in Scotland Vol II Obligations 461 et seq.)

[18] What is more, it seems that in those jurisdictions where attempts have been made to obtain certainty by formulating new bright line rules – in lieu of the old rule excluding all liability for pure economic loss – there was little success in achieving this goal. In England, for example, the first such attempt was made in Anns v Merton London Borough Council [1978] AC 728 [HL]. Under the Anns test the court will find wrongfulness – or in English legal parlance, the existence of a duty of care – if the harm was foreseeable and there is no policy reason for negating liability. In a number of subsequent judgments of the House of Lords, there was, however, a retreat from Anns because of its expansionist tendencies. Eventually Anns was expressly overruled in Murphy v Brentwood District Council [1991] 1 AC 438 (HL) at 457.

[19] Another attempt at a bright line rule is often referred to as 'the three-stage test' which is attributed to a passage in the speech of Lord Bridge of Harwich in Caparo Industries PLC v Dickman [1990] 2 AC 605 (HL) at 617-618. (See eg D v East Berkshire Community Health NHS Trust [2005] 2 AC 373 (HL) para 2 where reference is made to 'the familiar test laid down in Caparo'. See also Sutradhar v Natural Environment Research Council [2006] 4 All ER 490 (HL) para 32.) According to this test a plaintiff can establish wrongfulness (in the South African sense) only when it can prove three things: first, that the causing of damage was reasonably foreseeable; secondly, that a relationship of 'proximity' or 'neighbourhood' existed between the parties; thirdly, that in all the circumstances of the case, it is fair, just and reasonable to impose liability on the defendant. Somewhat ironically, however, Lord Bridge never claimed to create a bright line rule. He did not even profess to formulate a 'test'. That, I think, is apparent from the very passage in his speech usually relied upon in support of the 'three-stage test'. After Lord Bridge referred to the ingredients of foreseeability, proximity and the situation in which the court considers it fair, just and reasonable to impose liability, he continued (at 618A-B):

'[T]he concepts of proximity and fairness . . . 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 features of different specific situations which . . . the law recognises pragmatically as giving rise to a duty of care . . ..'

And in the same case Lord Oliver of Aylmerton said (at 633F):

'I think that it has to be recognised that to search for any single formula which will serve as a general test of liability is to pursue a will-o'-the-wisp.'

[20] In some decisions of the House of Lords it is explicitly recognised that the question whether the required relationship of proximity exists is dependent on policy factors (see eg in Barrett v Enfield Borough Council [2001] 2 AC 550 (HL) at 559). And in Cooper v Hobart (2002) 206 DLR (4^th) 193 para 37 the Supreme Court of Canada also recognised that, whatever the test formulated, the imposition of liability ultimately depends on 'residual policy considerations'. Proceeding from this premise, academic authors have engaged in the constructive exercise of identifying the relevant considerations of policy that can find application in determining whether, in a particular case, the negligent conduct of the defendant can sustain a claim for the plaintiff's pure economic loss (see eg John Hartshorne 'Confusion, Contradiction and Chaos within the House of Lords post Caparo v Dickman (2008) 16 Tort Law Review 8 et seq; Jonathan Burchell 'The Odyssey of Pure Economic Loss' in T J Scott & Daniel Visser Developing Delict – first published as Acta Juridica 2000 – 99 et seq).

[21] Does this mean we are back to the proposition that, in the field of pure economic loss, liability depends on the idiosyncratic views of the individual judge as to what is reasonable and fair? Fortunately, I think the answer must again be 'no'. In the first instance some degree of certainty is established by the identification of categories where liability will be imposed. In Telematrix (para 15) one such category was recognised, by way of example, with reference to the liability of collecting banks. Another example is to be found in Perre v Apand (paras 28-30) where liability for the failure to provide accurate information or advice – ie for negligent misstatements – was recognised as a category of liability for pure economic loss in the context of Australian law. (For the South African law on the topic of negligent misstatements, see eg Kern Trust (Edms) Bpk v Hurter 1981 (3) SA 607 (C); Bayer South Africa (Pty) Ltd v Frost 1991 (4) SA 559 (A) at 568B-D; OK Bazaars (1929) Ltd v Standard Bank of South Africa Ltd 2002 (3) SA 688 (SCA) 695G-I.) I believe it can be predicted with confidence that in time further categories will become discernible and so the law will develop in an incremental way.

[22] Further insurance against uncertainty and unpredictability derives from the principle which was formulated as follows by Nugent JA in Minister of Safety and Security v Van Duivenboden 2002 (6) SA 431 (SCA) para 21:

'When determining whether the law should recognise the existence of a legal duty in any particular circumstances what is called for is not an intuitive reaction to a collection of arbitrary factors but rather a balancing against one another of identifiable norms.'

(See also eg Telematrix paras 15-16). In a case like the present where the claim for pure economic loss falls outside the ambit of any recognised category of liability, the first step is therefore to identify the considerations of policy that are of relevance. As part of the identification process assistance can of course be gained from previous decisions, both at home and abroad, as well as from the helpful analysis by academic authors such as those to which I have already referred.

[23] The policy consideration that immediately comes to mind is directly linked to the initial doubt as to whether pure economic loss should be actionable at all. That reason was referred to by Rumpff CJ in Administrateur, Natal v Trust Bank van Afrika Bpk (supra) – where this court eventually decided to cut the Gordian knot – (at 833A) as 'die vrees van die sogenaamde oewerlose aanspreeklikheid' (ie the fear of so-called boundless liability). In the light of this fear the relevant consideration is succinctly stated as follows by Gaudron J in Perre v Arpand (supra) para 32:

'The first policy consideration is the law's concern to avoid the imposition of liability in an indeterminate amount for an indeterminate time to an indeterminate class.'

(See also eg Canadian National Railway Co v Norsk Pacific Steamship Co (1992) 91 DLR (4^th) 289 at 359; M M Corbett 'The Role of Policy in the Evolution of our Common Law' 1987 SALJ 52 at 59.)

[24] From this consideration it follows, in my view, that liability will more readily be imposed for a single loss of a single identifiable plaintiff occurring but once and which is unlikely to bring in its train a multiplicity of actions. That is the reason why liability was imposed, for example, in Coronation Brick (Pty) Ltd v Strachan Construction Co (Pty) Ltd 1982 (4) SA 371 (D) 386D-H and not in Shell and BP South African Petroleum Refineries (Pty) Ltd v Osborne Panama SA 1980 (3) SA 653 (D) at 659G-H. The present case self-evidently falls in the same class as Coronation Brick and not in the class of Shell and BP. The loss claimed was suffered by a single plaintiff and is finite in its extent. To illustrate the point: the position could very well be different if the plaintiff was a businessman who claimed for the loss he suffered because of a missed flight to London, being the loss of a lucrative business opportunity, owing to the closure of the road.

[25] But the absence of indeterminate liability itself will not automatically give rise to the imposition of liability. That much was expressly held in Trustees, Two Oceans Aquarium Trust para 20. The reason why this court refused to come to the aid of the plaintiff in that case, despite the absence of indeterminate liability, was that the plaintiff was in a position to avoid the risk of the loss claimed by contractual means (see para 24). Conversely, the plaintiff's inability to protect itself by contract was one of the policy reasons why this court decided in Indac Electronics (Pty) Ltd v Volkskas Bank Ltd 1992 (1) SA 783 (A) at 799H-J to impose liability on a collecting bank. Support for the same consideration is to be found in Australian cases where delictual liability was extended to plaintiffs who were said to be 'vulnerable to risk' because they were unable to protect themselves against the risk of the particular loss by other means (see eg Woolcock Street Investments (Pty) Ltd v CDG (Pty) Ltd (formerly Cardno & Davies Australia (Pty) Ltd) [2004] HCA 16 (para 80); Perre v Apand (supra) para 11 (Gleeson CJ) and para 50 (McHugh J). In the present case the Agency can, in my view, be said to be 'vulnerable' to the risk of the loss that eventuated because it could not readily protect itself against that risk by concluding a contract with every user of the toll road.

[26] Another policy reason why the extension of delictual liability is sometimes refused is that it would impose an additional burden on the defendant which would be unwarranted or which would constitute an unjustified limitation of the defendant's activities (see eg Minister of Law and Order v Kadir 1995 (1) SA 303 (A) at 321C-J; Steenkamp NO v Provincial Tender Board, Eastern Cape 2006 (3) SA 151 (SCA) paras 37-40; Road Accident Fund v Shabangu 2005 (1) SA 265 (SCA) para 18; X (Minors) v Bedfordshire County Council; [1995] 2 AC 633 (HL) at 750). The converse of this consideration appears from the statement by McHugh J in Perre v Apand (supra) para 50 that the imposition of liability would not 'unreasonably interfere with Apand's commercial freedom because it was already under a duty to [a third party] to take reasonable care'. That, I believe, is also the position in this case. Fourway's driver was already under an obligation towards other users of the road to drive with reasonable care. Imposing liability on him – and his employer – for economic loss resulting from his negligent driving would thus not foist any additional burden upon him at all.

[27] The only policy consideration relied upon by Fourway as to why it should not be saddled with the Agency's loss was that it would be more appropriate to spread the burden by increasing toll fees in order to accommodate losses of this kind. I do not agree with this argument. On the contrary, I can see no reason why innocent users of the toll road should effectively be held responsible for the negligent conduct of Fourway's employee. The fact that the loss would be spread more widely may alleviate the burden imposed upon the individual innocent motorist, but it does not detract from the principle that it would be a choice to the prejudice of the innocent in favour of the negligent driver.

[28] During argument the issue was raised as to the relevance of the consideration that Fourway's employee was transporting chrysolite asbestos which can, by all accounts, be described as dangerous cargo. Is this another policy consideration for imposing liability on Fourway for the Agency's loss, or not? My view is that it is not. As I see it, it would make no difference in principle in the determination of wrongfulness whether the cargo consisted of asbestos or of an innocuous substance like sand. I have given the policy reasons why Fourway should, in my view, be held liable. I cannot see that it would have any effect on any of them if the cargo was not asbestos, but sand. If the Agency lost revenue because the toll road had to be closed in consequence of the negligence of Fourway's employee, Fourway should, in my view, in principle be liable, whether the closure was necessitated by the spillage of asbestos, sand or cement. Where the dangerous nature of the cargo could have a bearing is on the issue of foreseeability of damage. I think it is more readily foreseeable that the closure of the toll road will be necessitated by a spillage of asbestos than a spillage of sand. But by the same token I believe that the issue of foreseeability should more appropriately be considered under the rubric of legal causation and not as part of determining wrongfulness.

[29] Now that I have decided the issue of wrongfulness in favour of the Agency, the further question arises as to whether it can be said that Fourway had been prejudiced by the Agency's failure to specifically raise the issue of wrongfulness in its particulars of claim. I think not. The policy considerations that should, in my view, be taken into account all appear from the allegations in the pleadings supported by the evidence which was led at the trial. In argument, counsel for Fourway could not think of a single policy consideration in favour of their client that could have been supported by evidence led on its behalf if the Agency had specifically referred to the issue of wrongfulness in its pleadings. In short, despite an express invitation to that effect, counsel for Fourway were unable to submit that their client's case would have been presented any differently if the Agency's pleadings were in perfect order.

[30] That brings me to causation. In this regard it has by now become well-settled that, in the law of delict, causation involves two distinct enquiries. First, there is the enquiry into factual causation which is generally conducted by applying the 'but-for' test, as described by Corbett CJ in International Shipping Co (Pty) Ltd v Bentley 1990 (1) SA 680 (A) at 700E-G. The facts that are common cause on appeal leave no doubt that, but for the collision caused by the negligence of Fourway's driver, the Agency would not have suffered the loss. Factual causation is therefore not in issue. The dispute turns on the second enquiry, under the rubric of causation, namely whether the negligent conduct of Fourway's driver is linked sufficiently closely or directly to the loss suffered by the Agency for legal liability to ensue, or whether the loss is too remote. This issue is referred to by some as remoteness of damage and by others as legal causation.

[31] In the final analysis, the issue of remoteness is again determined by considerations of policy. Broadly speaking, wrongfulness – in the case of omissions and pure economic loss – on the one hand, and remoteness on the other, perform the same function. They are both measures of control. They both serve as a 'longstop' where most right-minded people, including judges, will regard the imposition of liability in a particular case as untenable, despite the presence of all other elements of delictual liability.

[32] Since wrongfulness – in the context of omissions and pure economic loss – and remoteness are both determined by considerations of policy, a certain degree of overlapping is inevitable. However, wrongfulness and remoteness are not the same. They involve two different enquiries in respect of two different elements of delict, each with its own characteristics and content (see eg LAWSA, 2ed Vol 8 (1) sv 'Delict' by J R Midgley and JC van der Walt, para 132). Even where negligent conduct resulting in pure economic loss is for reasons of policy found to be wrongful, the loss may therefore, for other reasons of policy, be found to be too remote and therefore not recoverable. An example of a case where this happened is to be found in a decision of this court in International Shipping Co (Pty) Ltd v Bentley (supra).

[33] The question therefore remains: is the loss claimed by the Agency too remote? With regard to this question it has been held that the test in our law for determining remoteness is a flexible one (see eg International Shipping Co (Pty) Ltd v Bentley (supra) 701A-F; Smit v Abrahams 1994 (4) SA 1 (A) at 15E-G; OK Bazaars (1929) Ltd v Standard Bank of South Africa Ltd (supra) para 23). According to the 'flexible' test, (also referred to as the 'supple' test), so Fourway submitted, remoteness is determined by considerations of reasonableness, fairness and justice. As support for this submission it sought to rely on the judgment of Botha JA in Smit v Abrahams (supra) at 14F-15G. I do not agree with this submission and I do not believe it derives support from what Botha JA said. Considerations of fairness and equity must inevitably depend on the view of the individual judge. In considering the appropriate approach to wrongfulness, I said that any yardstick which renders the outcome of a dispute dependent on the idiosyncratic view of individual judges is unacceptable. The same principle must, in my view, apply with reference to remoteness. That is why I believe we should resist the temptation of a response that remoteness depends on what the judge regards as fair, reasonable and just in all the circumstances of that particular case. Though it presents itself as a criterion of general validity, it is, in reality, no criterion at all. In essence I agree with the following statement by McHugh J in Perre v Arpand (supra) para 80:

'But attractive as concepts of fairness and justice may be in appellate courts, in law reform commissions, in the academy and among legislators, in many cases they are of little use, if they are of any use at all, to the practitioners and trial judges who must apply the law to concrete facts arising from real life activities.'

[34] As to the dicta of Botha JA in Smit v Abrahams (supra) it is apparent that they are founded largely on the judgment of Van Heerden JA in S v Mokgethi 1990 (1) SA 32 (A) at 40I-41D. What Van Heerden JA said in that case is not that the 'flexible' or 'supple' test supersedes all other tests such as foreseeability, proximity or direct consequences, which were suggested and applied in the past, but merely that none of these tests can be used exclusively and dogmatically as a measure of limitation in all types of factual situations. Stated somewhat differently: the existing criteria of foreseeability, directness, et cetera, should not be applied dogmatically, but in a flexible manner so as to avoid a result which is so unfair or unjust that it is regarded as untenable. If the foreseeability test, for example, leads to a result which will be acceptable to most right-minded people, that is the end of the matter (see eg LAWSA (supra) para 132).

[35] In this case it can, in my view, be accepted with confidence that any of the various criteria will lead to the conclusion that the loss suffered by the Agency is not too remote. If, for example, the direct consequences criterion is applied, it is clear that the loss followed directly from the wrongful and negligent conduct of Fourway's driver; there was no so-called novus actus interveniens that broke the chain of events. If, on the other hand, one applies the foreseeability test, it was in my view reasonably foreseeable that a collision could cause spillage and that, because of the dangerous nature of the cargo, spillage could result in the closure of the toll road which could lead to a revenue loss by the Agency. What is more, I do not find the conclusion that Fourway should be held liable for the loss in any way untenable. Consequently, considerations of fairness and equity do not arise. In any event, the only consideration of fairness advanced by Fourway was that it would be fairer to spread the loss amongst users of the toll road by way of an increase in toll fees. As I have indicated earlier under the heading of wrongfulness, I cannot see the fairness of this proposal at all.

[36] For these reasons the appeal is dismissed with costs, including the costs of two counsel.

……………………..

F D J BRAND

JUDGE OF APPEAL

APPEARANCES:

FOR APPELLANT: J H DREYER SC

J A DU PLESSIS

INSTRUCTED BY: MACGREGOR STANFORD KRUGER INC, PRETORIA

CORRESPONDENTS: E G COOPER ATTORNEYS, BLOEMFONTEIN

FOR RESPONDENT: A C FERREIRA SC

I ELLIS

INSTRUCTED BY: FRIEDLAND HART INC, PRETORIA

CORRESPONDENTS: ISRAEL SACKSTEIN MATSEPES INC, BLOEMFONTEIN