Negligence and Foreseeability

NEGLIGENCE & FORESEEABILITY: Doctrine of Law or Public Policy

(Was there more than a snail in Ms Donaghue’s bottle of ginger beer?) [1]
Speech by the Honourable Justice Peter Underwood to the Australian Insurance law Association National Conference, Hobart 4-6 August 19996 August 1999

(Now published in (1999) 8 Australian Insurance Law Bulletin 73 and 85)


This paper attempts to do no more than step back and take a “big picture” view of the development of the tort of negligence in the writer’s lifetime by looking at a few key decisions during that time to see if there emerges any single discernible enduring legal principle that has stood the test of time. No attempt is made to examine the minutiae of this development for such an examination only obscures its true rationale. It has been said that this development has been no more than “the product of pragmatism and judicial intuition” and that the basic purpose of the tort “[is] to shift or distribute the losses [reflecting] a realisation that potential defendants today are often well able to foresee and guard against economic effects of their carelessness by building up internal reserves as a form of self-insurance, or by group insurance schemes which spread individual losses in the widest fashion.”[2] If this is right does it mean that in the last half century or so, learned judicial pronouncements on the tort of negligence have amounted to no more than pragmatic loss adjustment by the Courts for the communities they serve? Or does there exist some lofty objective principle that, like the Emperor’s clothes, is clearly visible to all except this uninformed writer? Is there a commonality of reasoning which makes a builder liable for the diminution in value of a building in the hands of a subsequent purchaser, a driver liable for injuries sustained by his or her passenger, and an auditor liable for a loss sustained by a non-privy investor? If there is no objective enduring principle to guide the development of the law of negligence, and if words of legal principle used by the appellate courts are no more than a cloak for individual judges’ views of the appropriate distribution of losses, are the courts and the doctrine of precedent the best way of managing the spread of loss in a rapidly changing society?


The tort of negligence is a relative newcomer to the law. As students of legal history are well aware, in the case of direct and immediate injury to the person and damage to property, liability was originally strict and the cause of action was known as trespass. Neither intention nor fault arose. In the case of indirect injury to person or property, liability for personal injury and damage to property depended upon proof of some element of fault in lieu of trespass. This form of action was known as action on the case.[3]With the advent of industrialisation, it soon became apparent that strict liability for direct trespass could not survive the modern conditions, and liability for personal injury and injury to property came to depend upon intention and fault, instead of causation.[4]In its transitory state from the old form of action on the case, negligence was described as carelessness in a matter where carefulness is obligatory. According to one early source, “negligence and wrongful intent are the alternative forms of mens rea, one or the other of which is commonly required by law as a condition of liability”.[5] By 1883, the foundation stone of the modern law of negligence was laid. I venture to suggest that no one in their wildest dreams, would have then contemplated its development and expansion over the next 100 years.

In Heaven v Pender,[6] a painter employed by a painting contractor was injured whilst using some staging erected by a dry dock company. The painting contractor was engaged by the ship owner and there was no contractual relationship between the dry dock company that erected the staging and the painting contractor. His men were simply using the staging erected by the defendant. In Brett MR’s judgment were sown the seeds of the modern tort of negligence. He said:

“Whenever one person is by circumstances placed in such a position with regard to another that every one of ordinary sense who did think, would at once recognise that if he did not use ordinary care and skill in his own conduct with regard to those circumstances he would cause danger of injury to the person or property of the other, a duty arises to use ordinary care and skill to avoid such danger.”[7]

The Master of the Rolls was in the minority when he expressed that principle in such wide terms. The other two members of the Court founded liability on the basis that the injured plaintiff was an invitee of the dry dock company and therefore an obligation arose by virtue of that relationship to ensure that appliances supplied for immediate use were not dangerous. The majority expressly eschewed Brett MR’s generalisation saying that such liability was confined to goods that are dangerous per se or dangerous to the knowledge of the defendant.

Early days

From the beginning, negligence meant a failure to so something which a reasonable man would have done in the circumstances or the doing of something that such a person would not have done in the circumstances.[8] It was early recognised that negligence would afford no remedy in law in the absence of a duty of care owed, not to the world generally, but to the particular plaintiff.[9] Principally, the development of the law of negligence has been built on this foundation namely, identification of circumstances which give rise to a duty of care. At first, the existence of a duty was determined by reference to the nature of the activity being undertaken eg, keeping dangerous goods, driving on the highway, being in a shop upon an implied invitation, and so on. A hundred years ago not only were the situations which gave rise to a duty of care carefully categorised, but also, the damage recoverable was confined to injury to person and property. Acknowledging that there was an obligation not to deceive another, the author of an early edition of Salmond on Torts[10] said “but I am commonly under no obligation to take care that the statements I make to him are true.”[11] How things change!

Fifty years later, things moved on. Although not quite a twinkle in my father’s eye in 1932, I was certainly then in the back of his mind when the House of Lords handed down its decision in the famous Donoghue v Stevenson case[12]. Until that time the conventional legal wisdom had been that the manufacturer of an article, other than one that was either dangerous to the knowledge of the manufacturer or dangerous per se, owed no duty of care outside that imposed by the contract under which the article was sold. In support of the hapless Ms Donoghue, it was contended that a new category of negligent liability should be opened up, namely, in cases in which a manufacturer puts onto the market a product intended for consumption, packaged in such a way that inspection prior to consumption is impossible. I interpolate to note that in those days drinking ginger beer appeared to be a hazard, for a Mr Bates was also the victim of a dubious bottle of ginger beer.[13] This one exploded, but, unlike Ms Donoghue twenty years later, Mr Bates recovered nothing on the basis that there was no privity of contract between him and the manufacturer and the ginger beer was neither dangerous per se nor dangerous to the knowledge of the defendant, even though with the exercise of reasonable care, the danger of explosion would have become apparent to the company.

In Donoghue v Stevenson, Lord Buckmaster was in the minority. What he had to say makes interesting reading a little less than four score and ten years later.

“There can be no special duty attaching to the manufacture of food apart from that implied by contract or imposed by statute. If such a duty exists, it seems to me it must cover the construction of every article, and I cannot see any reason why it should not apply to the construction of a house. Yet if a house be, as it sometimes is, negligently built, and in consequence of that negligence the ceiling falls and injures the occupier or any one else, no action against the builder exists according to English law, although I believe such a right did exist according to the laws of Babylon.”[14]

One can only imagine the apoplexy that his Lordship would suffer upon hearing of Bryan v Maloney.[15] Lord Atkin, whose famous dictum in Donoghue v Stevenson has rung down through the ages, eschewed the task of “elaborate classification of duties”[16] and searched the earlier cases for an expression of unifying principle to determine the existence of a duty of care. His Lordship immediately recognised that the dictum of Brett MR in Heaven v Pender (supra) was too wide, but was prepared to adopt it as a guiding principle, provided it was confined by the concept of proximity. As is well known, he described the proximity concept in these terms:[17]

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

So the modern concept of duty of care in the tort of negligence was born, but I doubt if even the farsighted Lord Atkin contemplated that his child might grow as it has, like Topsy, under the stress of changing policies, for Lord Atkin’s principle of the duty of care was limited to guarding against personal injury and property damage,[18] and then only in cases where such injury or damage was, or ought to have been, within the reasonable contemplation of the defendant.[19]

The first major expansion

The consequences of the alleged escape of Ms Donoghue’s snail from the bottle of ginger beer soon spread around the common law world.[20] For some thirty years after Donoghue v Stevenson, the tort of negligence jogged along under the perceived unifying principle of proximity which, in those days, meant reasonable foresight of injury to person or property. However, just about the time I was practising as a junior barrister in the Courts of Petty Sessions, the House of Lords handed down its decision in Hedley Byrne & Co Ltd v Heller & Partners Ltd.[21] Should there be a duty of care in the case of financial harm suffered in consequence of spoken words, or, as counsel for the appellant in the House of Lords so picturesquely put it:

“It would be strange if a person who handled his pen so carelessly as to put out X’s eye were liable to pay damages, but not if he handled it so carelessly in writing, that X was financially ruined.” [22]

Counsel for the respondent submitted that the event that attracted liability and created the requisite proximity in Donoghue v Stevenson was the inability to inspect the goods before consuming them. He contended[23] that the words of advice given by the Bank were not kept hidden and were free to be examined and therefore the necessary proximity was not there to create a liability. What did the House do in response to such able pleas? Absent a contract or fiduciary relationship betw[JU1] [JU2] een the person seeking the advice and the person giving it, was there any duty of care? In a breathtaking sweep, Lord Reid held that the neighbour principle in Donoghue v Stevenson had nothing to do with the case, observing that there is no comparison to be made between words and deeds[24]. His Lordship realised that the problem inherent in fixing the Bank with liability for their careless advice was to put a limit on the extent of any liability for the harm done. Overturning well established law,[25] Lord Reid seized upon some passing words in a rather obscure Scottish case[26] to pronounce that there was no reason to confine liability for careless advice to cases of contract or fiduciary duty. However, to limit liability for careless words once spoken, he declared that the law was that liability extended to all cases:

“… where it is plain that the party seeking the information or advice was trusting the other to exercise such a degree of care as the circumstances required, where it was reasonable for him to do that, and where the other gave the information or advice when he knew or ought to have known that the inquirer was relying on him.”[27]

With great respect to his Lordship, it seems to me that although his words have the clothing of legal reasoning, the extension of liability for negligent advice in 1964 had little to do with legal principle and a lot to do with policy namely, that where there has been the requisite reliance, it is in the community’s best interests that the loss be borne by the giver of the advice. If this is right, the question that immediately arises, is what was the basis for declaring that new policy and are the highest appellate courts of the land the best equipped vehicle for the identification and articulation of current policy for loss sharing? Consideration of this question must bear in mind that the courts in the common law world are designed to decide individual disputes between citizen and citizen, and citizen and State.

Lord Morris did at least apply Lord Atkin’s neighbour principle saying[28] that he could see no distinction in logic or principle between injury caused by reliance on the safety of the contents of a bottle of ginger beer and reliance upon words spoken. I pause to note that this application of legal principle, apparently crystal clear to Lord Morris, had until then, escaped the attention of many others with an interest in this field of law. My Professor of law who, only five years earlier, had taught me that there was a sharp distinction between injury caused by careless words and injury caused by careless deeds was one of them! In order to limit liability, Lord Morris said that the liability only arose when a person possessed of a special skill undertakes to apply that skill for the assistance of another person who relies upon that skill. This, he declared[29] was “settled” law!! Perhaps a more accurate description of his Lordship’s dictum would be “settled policy”, applied to provide a fetter on the extent of liability for loss caused by the negligently spoken word.[30]

Judicial activism was alive and well in Australia at that time, and it did not take long before the extension of liability created by Hedley Byrne was established in this country.[31] Relishing the task in front of the Court, Barwick CJ declared that “the matter so far as this Court is concerned is free of binding authority” and “the Court’s task is therefore to declare the common law in this respect for Australia.”[32] Like Lord Reid, the Chief Justice rejected as inappropriate, the Donoghue test to ascertain the existence of a duty of care and, although he denied it, it seems to me, with respect, that the Chief Justice resorted to categorisation of circumstances in which a duty of care would arise, just as was done in the case of personal injuries and injuries to property in the pre-Donoghue days.[33] When Evatt’s Case got to the Privy Council,[34] their Lordships thought it necessary to refine Hedley Byrne a little. Again, I respectfully contend, without resort to legal principle, the advice to Her Majesty was that before the duty of care arose in the case of words spoken, it was necessary that the party giving the advice “carry on the business or profession of giving advice or information of the kind sought”[35] or in some other way lets it be known that he has special skill in the field in question.[36]

Curiously, the House of Lords in Hedley Byrne paid scant attention to the proposition that the law had long denied recovery of pure economic loss arising out of a negligent act. One would have thought that if Hedley Byrne was all about the development and application of legal principle, there would have been considerable discussion about this long standing proposition as fundamental to the Donoghue principle was confinement of losses to those arising out of damage to property and injury to the person.

All of that is now history. As those in the insurance industry know so well, what was not a breach of the civil law a little more than 30 years ago is now responsible for some of the biggest claims in history. Were he alive today, no doubt my Professor of law would be teaching today’s student that in the case of negligent misstatement, the relational test that emerged from Ms Donoghue’s bottle of ginger beer is not sufficient to establish a duty of care, and that more needs to be shown to create the requisite proximity. In this respect, he would also probably point out to his students that the broad brush of Lord Atkin in Hedley Byrne has been clipped somewhat for fear of imposing a crippling burden on the profession. He would be right to do so.[37] However, experience of appellate activism over the last two decades of my life would have warned my Professor against treating his lecture notes to that effect as “settled law” for there are just about as many different opinions in the High Court today (or rather yesterday) as there are justices, with respect to the true basis for limiting liability for the spoken word.[38]

Is there any “principle” underlying policy? Why should the manufacturer of a defective article have a duty of care to all those whom he ought to have foreseen might suffer injury, but an auditor’s liability is limited to those cases where there has been an assumption of liability or there was an intention to induce reliance?[39] According to Professor Flemming:[40]

“Unlike the [manufacturer], the auditor is unable to spread the cost amongst a multitude of clients. Insurance would be prohibitively expensive to the cost of the clients. While third party get the benefit without paying for it (‘free riders’). Unlike the consumer, it is argued, the investor is usually better placed to protect himself by making his own enquiries.”

That might or might not seem to be a very good reason for limiting the liability for negligent misstatement by an auditor, but its connection with “legal principle” seems to me, to be a little tenuous. Let me turn to another area of the law of negligence to see whether, during my life time, there has been development of legal principle or imposition of social policy.

Nervous Shock

The problem that became immediately apparent for appellate courts trying to express principle in “nervous shock” cases was, yet again, how to limit liability: a problem in the realm of policy rather than legal principle, it seems to me. How does the Donoghue neighbour test fit in here?

At first, damages for psychiatric injury unaccompanied by physical injury were not recoverable at all.[41] At the end of the last century, judges refused to recognise claims that were limited to mental illness alone. This was probably due to “ignorance of the medical subtleties of the subject, concerns of ‘opening the floodgates’ to limitless liability and its ramifications for the insurance industry”,[42] and so forth. However, this judicial timidity did not last long, although it persisted in Australia for a little longer than in other common law jurisdictions. As late as 1939, six months before World War II was declared, the High Court refused recovery to a mother who suffered psychiatric injury on witnessing the body of her drowned child being pulled from a ditch full of water. For the purposes of the appeal, it was assumed that the child drowned because of the carelessness of the local council.[43] Latham CJ was clearly concerned about the floodgates of claims being opened. He stated the Donoghue proposition that the mother must show that the Council owed her a duty of care and then said:[44]

“The duty which is suggested the defendant owed the plaintiff was a duty not to injure her child as to cause her nervous shock when she saw not the happening of the injury, but the result of the injury, namely the dead body of the child. It is rather difficult to state the limit of the alleged duty. If a duty of the alleged character exists at all, it is not really said it is confined to mothers of children who are injured. It must extend to some wider class – but to what class? There appears to be no reason why it should not extend to other relatives or to all other persons, whether they are relatives or not.”

The appeal was rejected upon the basis that a reasonable person would not foresee that a mother would suffer shock upon seeing the dead body of her child pulled out of a drain! Is that application of legal principle or simply social engineering to put a limit on claims? The law moved on after the War. When I was in Law School, the conventional teaching was that damages for psychiatric harm (nervous shock) unaccompanied by physical injury, was recoverable, but only if the injured plaintiff was in the area of potential harm from the tortious act[45]. The rationale was that the tortfeasor could only be expected to foresee that harm might be suffered by those who were so close to his act of carelessness (it was always “him” in those days) that they could be described as within the area of potential injury. So, in the space of about a century, the law moved from refusing to recognise claims for psychiatric illness unaccompanied by physical injury, to not only recognising them, but also permitting them to succeed, even in cases where the fear which caused the psychiatric illness was for the safety of someone other than the plaintiff.[46] The basis for this development has been ascribed to the proposition that “the distinction between mental shock and bodily injury was never a scientific one”.[47] The change in the law to permit claims for psychiatric harm has been simply described as “the true principle.”[48] One might be forgiven for thinking that the real basis for that development was the judicial notion that contemporary community standards were perceived to be that loss arising from psychiatric illness is better borne by the community than by the victim.

In the field of so called nervous shock, McLoughlin v O’Brian[49] is a significant decision. The plaintiff’s family was injured in a road accident. The plaintiff was two miles away from the scene of the accident at the time it happened, and learnt of it from what she was told by others and from what she saw when she went to the hospital where the members of her family had been taken. The defendant relied upon the fact that the plaintiff was not within the area of potential harm and therefore the injury to her was not foreseeable, as the Court had laid down in Bourhill v Young. That was a mistake! The House of Lords held that the prospect of the plaintiff, or a person in a class of persons of which the plaintiff was a member, suffering shock on learning of the involvement of a member of her or his family, was reasonably foreseeable and therefore the duty of care was owed, even though she did not come upon the scene of the accident and some of the shock was caused by what she was told by others. The decision is significant, not only because it extended the boundaries of nervous shock claims, but also because the House grasped the nettle and spoke of what was really involved in the decision. Lord Wilberforce cited the famous dictum in Donoghue, “persons who are so closely and directly affected by my act that I ought reasonably have them in my contemplation as being so affected …”, and said:[50]

“This is saying that foreseeability must be accompanied by and limited by the law’s judgment as to the persons who ought according to its standards of value and justice to have been in contemplation” [emphasis added]

Here in Australia, Gibbs CJ expressed approval of Lord Wilberforce’s dictum but went on to add: [51]

“In forming its judgment on such a matter the court is not at large, or free to indulge its own individual notions, but must be guided by existing legal principles and by analogies that may be drawn from decided cases.”

His Honour immediately proceeded to say that Chester v Waverley Corporation was no longer good law! I respectfully ask what is the legal principle that rejected Mrs Chester’s claim but accepted that of the plaintiff in Jaensch v Coffey? How is it possible to avoid the uneasy feeling that the law changes according to the individual notions of social justice of the incumbents of the appellate bench from time to time? Even Lord Wilberforce’s frankness was disguised in words such as “the law’s judgment” of standards, as if “the law” was some immutable objective beacon that existed other than in the statements made by members of an appellate court from time to time. I should add that in Jaensch v Coffey, Murphy J felt no need to speak so coyly of public policy.[52] His Honour delivered himself of a one page judgment that addressed the shortcomings of the then social security system and the inadequacies of ad hoc extension of tort remedies.

The principles expressed by Lord Wilberforce in McLoughlin came under scrutiny by the House of Lords less than a decade later in consequence of the appalling deaths of 92 persons at a soccer match and the injury of many more due to the admitted carelessness of the police.[53] Relatives of some of the dead and injured learned about the tragedy by watching it on television. Not all of them were spouses. The House of Lords retreated a little from McLoughlin and denied a remedy for those who learned of the tragedy from the television. Why? The reasoning of the Law Lords is not harmonious, but it may be noted that Lord Oliver of Aylmerton referred to the judgment of Deane J in Jaensch[54]and frankly admitted that the question of who should recover and who should not “must be based upon policy rather than upon logic”. Such frank admission immediately raises the question, whose policy, and upon what consideration is it based? The Alcock case has been applied in the UK on more than one occasion,[55] but has yet to be considered by an authoritative appellate court in Australia.[56]

Pure economic loss

I turn now for a moment to a different kind of loss; pure economic loss. At the time Ms Donoghue was drinking her ginger beer allegedly containing a decomposed snail (for it was never established that there was any snail – such are the idiosyncrasies of the common law), it was unthinkable that there was any duty to guard against purely economic losses.[57] How that has changed in the space of a life time. The problem about imposition of liability for pure economic loss is the risk that where such loss is unaccompanied by physical harm, there may be created “liability in an indeterminate amount for an indeterminate time to an indeterminate class”.[58] This is a much bigger risk than that which concerned the High Court when it rejected Mrs Chester’s claim. Today, that liability has been imposed and we are presently witnessing the difficulties that have to be faced to ensure that there will not be liability in an indeterminate amount for an indeterminate time to an indeterminate class.

The opposition to extending the Donoghue principle to cover pure economic claims is well described elsewhere.[59] That opposition has produced some absurd results. The mine owner whose mine is damaged as a result of the negligence of a subcontractor can recover, not only for the physical harm suffered, but also for the related, or consequential, economic loss. Not so the miners who are thrown out of work by the subcontractor’s negligent act and who suffer substantial loss of income. Whilst on this point it might be of interest to those with a special interest in insurance law to note that an insurance company which paid out pursuant to the terms of the policy of insurance as a result of the negligent act of a third party, failed to establish that the tortfeasor owed it any duty of care and was unable to recover that loss from the third party![60]

The long standing exclusory rule prohibiting claims for pure economic loss which my old Professor of law believed was not only settled law but enshrined in stone, came under attack in Australia in Caltex Oil (Australia) Pty Ltd v The Dredge “Willemstad.[61] In that case, Caltex delivered its oil by contract to another (AOR). This company refined the oil and sent it back to Caltex by pumping it through an underwater pipe that it owned. A dredge was operated carelessly and damaged the pipe which disrupted the oil flow. AOR recovered the cost of making good the pipe and consequential financial losses. What about Caltex, which suffered no damage to property, but did suffer economic loss, being the cost of making alternative arrangements for the transport of its oil while the pipe line was out of action? Under the exclusory rule, Caltex, having suffered no damage to its property, would be unable to recover. Gibbs CJ traced the history of the exclusory rule[62] and concluded that if it was applied to this case, Caltex would not recover. Having said that, the Chief Justice went on to say:[63]

“The fact that the loss was foreseeable is not enough to make the [economic loss] recoverable. However, there are exceptional cases in which the defendant has knowledge or means of knowledge that the plaintiff individually (Caltex), and not merely as a member of an unascertained class, will be likely to suffer economic loss as a consequence of his negligence and owes the plaintiff a duty to take care not to cause him such damage by his negligent act.”

It might well be accepted social policy that a careless defendant should pay for economic loss suffered by a plaintiff whom he knew individually, and whom it was reasonably foreseeable might suffer such loss by reason of the careless act, but upon what legal principle is such a proposition based when the law of negligence since long before Ms Donoghue drank her ginger beer, denied recovery for pure economic loss? Gibbs CJ went on to say that “it would be unwise to formulate a principle that would cover all cases in which such a duty is owed”! It might be wise for the judge to adopt that course, but what of the practitioner who has a duty to advise clients of their potential liability? Absent clear legal principle how can insurers, who in so many cases accept the relevant risks, gauge the magnitude of those risks in advance of a claim being made? Stephen J took the view that reasonable foresight (the Donoghue test) was, by itself, inadequate to determine whether a duty of care existed in any given situation. He said that in addition to reasonable foresight, there must be a sufficient proximity between the tortious act and the injury suffered.[64] Stephen J declined to identify what constituted a sufficient proximity, leaving that to emerge case by case. Mason J (as he then was) considered that recovery for economic loss was appropriate in a case where the defendant could reasonably foresee that a specific identifiable plaintiff might suffer consequential loss by his or her negligent act.[65] Jacobs J took the conservative approach and said that the tortious act of cutting the pipe line caused Caltex to suffer physical loss, viz, “immobilisation through the pipe line of the crude oil and the products thereof”[66] and the economic loss was recoverable as it was consequential upon the physical loss. The other member of the Court in Caltex, Murphy J, opened his single page judgment with the words:

“There is no satisfactory general principle governing recovery of economic loss caused by negligence. The difficulties in this branch of the law arise mainly from the doctrine of foreseeability but also from unresolved questions of public policy.”[67]

With great respect to those who have a different view, I think that his Honour was absolutely right.[68]

In this country, the search commenced for a unifying principle – presumably a legal one – by which the existence of a duty of care could be determined.[69] Although, as has been noted, Stephen J touched upon this concept in Caltex, the search began in earnest with the judgment of Deane J in Jaensch v Coffey[70]. Deane J explained in Jaensch[71] that :

“The requirement of ‘proximity’ … should be accepted as a continuing general limitation or control of the test of reasonable foreseeability as the determinant of a duty of care.”

In a later case,[72] his Honour explained that in cases of personal injury, reasonable foresight (Donoghue) was enough to establish the requisite proximity, but in other cases, such as negligent misstatement, more was required, such as reliance. In Stevens v Brodribb Sawmilling Co Pty Ltd[73] Deane J pushed his concept along a little further, and in San Sebastian Pty Ltd v The Minister,[74] it was adopted as part of the common law of Australia. Proximity became the limitation upon the test of reasonable foreseeability.[75] Its advantage is that it is flexible. Its disadvantage is that it is flexible! At the end of the day however, it seems to me that proximity is no more than a new word to cover what is in reality, appellate judgment of what in any given set of circumstances is appropriate social policy.

In this context, may I be permitted to refer to one final case, viz, Bryan v Maloney?[76] This case went to the High Court on appeal from the Full Court of the Supreme Court of Tasmania. The High Court upheld the decision of the trial judge and the Full Court which found a builder liable to a later purchaser for the reduction in value of a house caused by the construction of defective foundations. These defects did not render the house dangerous. They simply reduced its value. Prior to her purchase, the respondent had never seen or spoken to the appellant. The common quest in the High Court was to see if there was the requisite proximity between the owner and the builder to give rise to a duty of care owed by the latter to the former. In pursuit of this quest, the majority joint judgment[77] noted that there was no decision of the Court that directly determined the point and said:[78]

“Necessarily, as has been indicated, the resolution of that question requires the articulation of both the factual components of the relevant category of relationship and the identification of any applicable policy considerations. Ultimately, however it is a question of law which must be resolved by the ordinary processes of legal reasoning in the context of the existence or absence of the requisite element of proximity in comparable relationships or with respect to comparable acts and/or damage.”

The refreshingly frank reference to policy considerations is immediately undermined by the reference to legal reasoning. With respect to their Honours, it might sound to some as if the majority judgment says no more than liability will depend upon our view of the case and what we think should be the relevant policy considerations![79] The majority determined that the house itself was a “connecting link” in the proximity test. Their Honours said,[80] “It is a permanent structure to be used indefinitely and in this country, it is likely to represent one of the most significant, and possibly the most significant investment which the subsequent owner will make during his or her life time.” Add to that the undisputed fact that the loss was reasonably foreseeable and liability in a new class of cases becomes law.

The so called unifying concept of proximity has since been applied to a number of different fact situations in which pure economic loss has been suffered and is likely to prove a growth area in the law of negligence.[81]

Policy or legal principle?

The point of this necessarily brief and, therefore, incomplete, overview of the development of the tort of negligence during my life time is to see whether any legal principle is discernible, for the leading judgments are redolent with reference to the legal principle. After all, it cannot be denied that today, the tort of negligence is the dominant cause of action at common law for recovery of loss.[82] Has the expansion of the categories of loss recovery since Ms Donoghue sat down to her bottle of ginger beer occurred only through the application of the single “neighbour principle” enunciated by Lord Atkin? Deane J pointed out:[83]

“Lord Atkin did not seek to identify the precise content of the requirement of the relationship of ‘proximity’ which he identified as a limitation upon the test of reasonable foreseeability. It was left as a broad and flexible touchstone of the circumstances in which the common law would admit the existence of a relevant duty of care to avoid reasonably foreseeable injury to another.”

This is no doubt so, and clearly the path the High Court has chosen to follow, but it seems to me that this is no more than saying that recovery of reasonably foreseeable loss caused by careless conduct will be allowed unless there is good policy reason not to do so, for to impose liability to make good all reasonably foreseeable loss would, of course, impose “liability in an indeterminate amount for an indeterminate time to an indeterminate class”.[84] If this is right, the question immediately arises whether the appellate courts are the appropriate vehicle for the determination of this policy and, if they are, whether they are going about it in the right way.

I tend to agree with the suggestion that the use of covert policy behind the label of a principle such as proximity makes the reasons for judgment sound a touch “hollow and jaundiced”.[85] Stronger criticism states:

“The proximity concept must not continue to be use to shroud what is in fact pure public policy. Judges should openly express the true premises of their decisions and, if recovery is considered to be undeserved, in the light of greater moral, social, economic, administrative or philosophical perceptions, then those reasons, and not others inherently uncertain in nature, should be expressed as the true foundation for denial of recovery.”[86]

It has been said that the proximity concept had been designed:

“… to obscure the fact that decisions in hard cases are based upon controversial value judgments by the courts, and to preserve the appearance of value free adjudication by reference to a fundamental pre-existing legal principle.”[87]

In his powerful dissent in Bryan v Maloney,Brennan J (as he then was) said that without definition of its content, the “notion of proximity would be a juristic black hole into which particular criteria and rules would collapse and from which no illumination of principle would emerge.”[88]

In a strong criticism of the adoption of the proximity principle,[89] McHugh J classified the policy factors applicable to a determination of whether a duty of care arose in any given situation as administrative, ethical or moral, economic, justice and public interest.[90]

An administrative factor identified by McHugh J is the capacity of the courts to handle the claims, either because of the likely volume of the cases or because of their nature. He referred to the early reluctance of the courts to entertain claims for psychiatric harm as a “policy administrative factor”, but the approach in fact taken by the courts was to deny that such an injury was reasonably foreseeable. McHugh J identified as an example of an ethical or moral policy factor whether the imposition of a duty would require the taking of affirmative action. The economic policy factor is, of course, wide and far reaching. Will the cost of finding a duty of care outweigh the benefits? The justice policy factor concerns the question of whether it is just and fair in contemporary society to impose a duty on a particular class of persons. Finally, the public interest policy factor requires a consideration of the impact of the result in a case between identifiable litigants upon the general public in the future. However expressed, McHugh J was of the opinion that these are in reality the factors that put a limit on the imposition of liability for all foreseeable loss.

The Future

Fundamental to the common law doctrine of precedent is the proposition that each case is decided upon its own facts by the application of principle from preceding cases in the relevant hierarchy of courts. The true principle in any case is often only discovered by careful identification of the facts to which the statements of law relate. But if each case is decided upon its own facts, does it not follow that the doctrine of precedent is an inherently unsuitable vehicle for the determination of policy? Evidence not relevant to the issues raised by the parties’ pleadings is inadmissible. Thus, evidence relevant to the policy factors clearly identified by McHugh J are not put to the tribunal that is required to make a policy decision. Decisions like Bryan v Maloney have far reaching economic and social consequences for the building, insurance, financing and other sectors of the community, but the common law system is designed in such a way that it prohibits the courts from receiving evidentiary material illuminating those consequences. It is accepted that the common law needs to be flexible in order to adapt to changing conditions and new developments in society, but shrouding reasons for judgments in bogus words of legal principle simply obscures the real task at hand.

I do not pretend to have the answers to this conundrum. However, I believe that until there is frank acknowledgement that considerations of public policy are central to the development of the law of negligence, no adequate consideration will be given to improving the system. As I presently see it, the courts are better situated than the parliaments to develop the law of negligence, but the manner in which this is done is unsatisfactory. It is unsatisfactory principally because decisions affecting a wide variety of community interests are made without those interests having any input into the decision making process. As I have said, a decision that a duty of care is owed by a builder to a subsequent owner of a house has an enormous impact on builders, manufacturers of building materials, insurers, house owners and the like. It also has, of course, an impact on the cost of housing. Although there is no doubt that those involved in those businesses have a legitimate interest in the outcome of such a decision, they have no right to be heard. The case is simply decided upon the facts then before the court. Whether imposition of a duty of care would “open the floodgates of litigation” or impose an unwarranted financial burden on some part or parts of the community is left to the intuitive notions of the individual members of the appellate court. Not uncommonly, there is a difference of opinion amongst the members of the court about such matters.

It cannot be beyond the wit of mankind to devise a more satisfactory scheme, one that will frankly acknowledge that the facts of a particular case raise major questions of policy and that the decision will affect the wider community; a scheme that will in such cases, permit the receipt of relevant material from those affected in the wider community. I appreciate that as the Constitution is presently framed, the High Court of Australia is prevented from receiving such material,[91] but State Appeal Courts are empowered to receive fresh evidence[92] and although the rules presently governing the exercise of that power would not permit the reception of evidence of the kind I postulate should be received, there is no reason why those rules cannot be changed to permit that to be done. Perhaps as a starting point, regard could be had to the provisions governing interpleader applications.[93] Although the purpose of such applications is to determine competing claims to property, if a legitimate interest of the kind I have just mentioned arises in an action for damages for negligence, why should not a representative of that interest be entitled to apply to be heard in the action on the issue of law? I appreciate that rules would need to be worked out to specify such matters as the status of the interest, the identification of the representative of that interest and the effect of any intervention upon the principal parties to the litigation. As I have said, I do not pretend to have the answers to the problem, but if McHugh J is right and policy factors of the type he identified are at work determining the boundaries of the law of negligence, is it not only fair that those with a legitimate interest in those factors should be heard?

One academic writing to the effect that policy considerations require judges to address community standards from time to time in the decision making process at appellate level[94] argued that there is a need for such judges to resort to social research and suggested as illustrations of available material, “reports of law reform bodies and of governments, material culled from academic and other authoritative sources and public opinion polls.”[95] He suggested that procedures be altered so that counsel can refer to such material in argument or that the Court be given authority to order a social research inquiry. There are obvious difficulties with this approach in our common law adversary system, but none-the-less all ideas of this kind are worth consideration and discussion.

I have attempted to do no more than:


  • show that over a short space of time there has been a vast expansion of the tort of negligence;
  • demonstrate that the vehicle for such expansion has been the concept of a duty of care;
  • argue that the basis of this expansion has been appellate court judges notions of contemporary public policy; and
  • suggest that it is time to consider whether there might better ways for judicial ascertainment of appropriate contemporary public policy at the cutting edge of the tort of negligence?Although there might be some doubt about whether there ever was a snail in Ms Donoghue’s bottle of ginger beer it seems to me that there is no doubt that it contained the genie of public policy that has ever after infected the law of negligence.peter underwood[96] 

    A foot note:


    A week after this paper was first delivered the High Court handed down its judgement in Perre & Ors v Apand Pty Ltd [1999] HCA 36 and already, my paper became outdated! It seems from the 167 page judgment that the members of the High Court are not ad idem about the test for ascertaining the existence of a duty of care in claims for economic loss unaccompanied by physical harm. I think it is not unfair to say that the state of law on this important area is in disarray. The proximity test, after a promising start has been banished in disgrace. Vulnerability was the test that found favour with the Chief Justice and Gaudron J. The Chief Justice thought that this was no more than an aspect of foresight. Gaudron J said that where a person knows that their acts may harm another, either as a known individual (as in Caltex Oil) or as a member of a class (a new development) there is a duty of care if that other “is in no position to protect his or her own interests.”

    Repeating the dangers of creating an indeterminate liability for an indeterminate class of persons, McHugh J cautiously embraced the vulnerability theory but added that if the plaintiff could have taken steps to protect itself then the duty of care will not exist. McHugh J said that vulnerability often includes reliance and an assumption of responsibility and that knowledge of vulnerability was necessary before the duty of care would be found to exist. Gummow J said that there was no general formula for the determination of the existence of a duty of care and that it will depend on the “salient features” of each case. He said in effect, that there will be an emergence of “a coherent body of precedents.” Hopefully that will be the case although so far as I can see, that has not happened in the last half century and I cannot see any basis for supposing it will in the next.

    Kirby J said that he favoured a three stage approach to determine if there existed a duty of care namely, was harm foreseeable, was there a proximity between the defendant and the plaintiff and “is it just and fair that the law should impose a duty of care?” His Honour acknowledged that his approach imposed “upon lawyers advising clients, trial judges and even appellate courts unreasonable burdens, and upon clients intolerable uncertainties and costs” but said that they exist anyway, – adding “the policy evaluation stubbornly remains”.

    For his part, Hayne J said that nothing was to be gained by rejecting the concept of proximity and substituting a test of “fair just and reasonable”. He concluded that there was great similarity between this case and Caltex Oil in that the class of plaintiffs likely to be harmed by the defendant’s conduct was determinate. So far as I can see Callinan J simply took the view that “it must be accepted that this is an area of the law in which the courts should move incrementally and very cautiously indeed.” He said, “it is not yet possible to identify a bright line of demarcation between those cases of pure economic loss in which damages are recoverable and those in which they are not.”

    All of this leads me to wonder where on earth Ms Donoghue’s snail is off to now!


    [1] This paper was first given at the Australian Insurance law Association National Conference, Hobart 4-6 August 1999.

    [2] Takaro Properties Ltd v Rowling [1978] 2 NZLR 314 at 322-323.

    [3] Fleming, The Law of Torts, 9th ed at 24.

    [4] Goodhart and Winfield, Trespass and Negligence (1933) 49 LQR 359. Early landmark decisions which recognised this development include Rylands v Fletcher (1866) LR 1 Ex 265; River Wear Commissioners v Adamson (1877) 2 App Cas 743.

    [5] Salmond, Law of Torts, 7th ed

    [6] (1883) 11 QBD 503

    [7] Ibid at 509.

    [8] Blyth v Birmingham Waterworks Co (1856) 11 Ex at 784.

    [9] Thomas v Quartermaine (1887) 18 QBD 694; Butler v Fife Coal Co Ltd [1912] AC 159.

    [10] WT Stallybrass 7th ed (1928).

    [11] Ibid at 25.

    [12] [1932] AC 562.

    [13] Bates v Batey & Co Ltd [1913] 3 KB 351.

    [14] Supra at 577-578.

    [15] (1995) 182 CLR 609.

    [16] Supra at579.

    [17] Supra at 580

    [18] Charlesworth, The Law of Negligence, 2nd ed (1947) at 14 states that “An expression of opinion given without due care and skill by one person to another knowing that it is likely to be acted upon by a third party gives no right of action to the third party if he acts upon it to his detriment.”

    [19] Cf Woods v Duncan [1946] AC 401; Bourhill v Young [1943] AC 92.

    [20] For example, Grant v Australian Knitting Mills [1936] AC 85, an Australian case in which the neighbour principle was applied to aid the unfortunate plaintiff who contracted dermatitis from sulphites negligently left in the material of which his underpants were made.

    [21] [1964] AC 465.

    [22] Ibid at 470.

    [23] Ibid at 474.

    [24] Ibid at 483.

    [25] Derry v Peek supra; Candler v Crane Christmas & Co [1951] 2 KB 164; Le Lievre v Gould [1893] 1 QB 491.

    [26] Nocton v Lord Ashburton [1914] AC 932.

    [27] Supra at 486.

    [28] Supra at 496.

    [29] Supra at 502.

    [30] The decision attracted considerable academic discussion and criticism e.g., D M Gordon, Hedley Byrne v Heller in the House of Lords (1964) 38 ALJ 39 in which the author described the decision as a “drastic break with tradition” and one to be regarded with “misgiving”.

    [31] Mutual Life and Citizens Assurance Co Ltd v Evatt (1968) 122 CLR 556.

    [32] Ibid at 563.

    [33] Ibid at 569.

    [34] (1970) 122 CLR 628.

    [35] Ibid at 630.

    [36] Such restriction has since found little favour in Australia. See e.g., Shaddock v Parramatta CC (1981) 150 CLR 225.

    [37] Smith v Bush [1990] 1 AC 831; Caparo Industries v Dickman [1990] 2 AC 605; Esanda Finance Corporation Ltd v Peat Marwick Hungerfords (1997) 188 CLR 241.

    [38] Esanda Finance Corporation Ltd v Peat Marwick Hungerfords, (supra)

    [39] R Lowe Lippman Figdor & Franck v AGC (Advances) Ltd [1992] 2 VR 671; San Sebastian Pty Ltd v The Minister (1986) 162 CLR 340.

    [40] Law of Torts, 9th ed at 712.

    [41] Victorian Railways Commissioners v Coultas (1888) 13 App Cas 222.

    [42] Per Mullany & Hanford, Tort Liability for Psychiatric Damage at 2.

    [43] Chester v The Council of the Municipality of Waverley Corporation (1939) 62 CLR 1.

    [44] Ibid at 7.

    [45] Bourhill v Young [1943] AC 92.

    [46] Hambrook v Stokes Bros [1925] 1 KB 141; King v Phillips [1953] 1 QB 429; Boardman v Sanderson [1964] 2 WLR 1317; Storm v Geeves [1965] Tas SR 252.

    [47] Bourhill v Young supra at 103.

    [48] King v Phillips supra at 440.

    [49] [1983] AC 410.

    [50] Ibid at 420.

    [51] Janesch v Coffey (1984) 155 CLR 549 at 554-555.

    [52] Ibid at 557-558.

    [53] Alcock v Chief Constable of South Yorkshire Police [1992] AC 310

    [54] Ibid at 418.

    [55] Page v Smith [1995] 2 All ER 736; White v Chief Constable of South Yorkshire Police [1999] 1 All ER 1.

    [56] Cf Coates v GIO of NSW (1994) 36 NSWLR 1 especially Kirby P (as he then was) at 9-11; Tri Minh Pham v Lawson unrep Full Court of the Supreme Court of SA 25 March 1997.

    [57] For example, Anglo-Algerian Steamship Co Ltd v The Houlder Line Ltd [1908] 1 KB 659; The Elliott Steam Tug Co Ltd v The Shipping Controller [1922] 1 KB 127.

    [58] Per Cardozo CJ in Ultramares Corp v Touche (1931) 174 N.E. 441.

    [59] Relational Loss, The Exclusory Rule and the High Court of Australia, Hogg, (1995) 3 TLR 26.

    [60] Simpson v Thomson (1877) 3 App Cas 279. Contra if the action had been brought in the name of the insured pursuant to the right of subrogation.

    [61] (1976) 136 CLR 529.

    [62] Ibid at 545 et seq. Referring to Cattle v Stockton Waterworks Co (875) LR 10 QB 453. See also Simpson v Thompson (1877) 3 App Cas 279.

    [63] Ibid at 555.

    [64] Ibid at 575.

    [65] Ibid at 593.

    [66] Ibid at 605.

    [67] Ibid at 605.

    [68] In the United Kingdom, the House of Lords determined in Anns v Merton London Borough Council [1978] AC 728 that the test for determining a duty of care was a two stage operation. The first question was to ask if there was the requisite proximity and that was determined by reasonable foresight. The second stage was to ask if there were any policy considerations that barred recovery notwithstanding reasonable foresight. Anns was not followed in Australia. See The Council of the Shire of Sutherland v Heyman(1985) 157 CLR 424. There has since been a retreat from Anns. See D & F Estates Ltd v Church Commissioners for England [1989] AC 177; Murphy v Brentwood DC [1991] AC 398.

    [69] For an interesting article on the effect of Caltex Oil (Australia) Pty Ltd v The Dredge “Willemstad” on the common law in Australia and elsewhere see Karen Hogg, Relational Loss, The Exclusory Rule (1995) 3 TLR 2.

    [70] Supra at 47.

    [71] Supra at 583-584.

    [72] Sutherland Shire Council v Heyman supra.

    [73] (1986) 160 CLR 16.

    [74] (1986) 162 CLR 340.

    [75] It has been applied in many cases since. See, eg, Cook v Cook (1986) 162 CLR 376; Hawkins v Clayton (1988) 164 CLR 539; Burnie Port Authority v General Jones Pty Ltd (1994) 179 CLR 520.

    [76] (1995) 182 CLR 609.

    [77] Mason CJ, Deane and Gaudron JJ.

    [78] Supra at 619.

    [79] I should make frank disclosure that I was a member of the Court whose judgments were upheld by the High Court Bryan v Maloney Tas unrep A77/1993!

    [80] Supra at 625.

    [81] For example, Hill v Van Erp (1997) 188 CLR 159, a case in which a solicitor was held to owe a duty of care to the intended beneficiary of part of the estate of a client for whom the solicitor drew the will but who did not inherit because of the solicitor’s negligence in getting the intended beneficiary’s husband to attest the execution by the client.

    [82] See Swanton, Another Conquest in the Imperial Expansion of the Law of NegligenceBurnie Port Authority v General Jones Pty Ltd (1994) 2 TLJ 101.

    [83] Jaensch v Coffey supra at 584.

    [84] Op cit 54.

    [85] Dr Norman Katter, Duty of Care in Australia 58.

    [86] NJ Mullany Proximity, Policy, and Procrastination (1992) 9 Aust Bar Rev 80 at 83.

    [87] J Smillie, The Foundation of the Duty of Care in Negligence (1989) 15 Mon LR 302.

    [88] Supra 72 at 655.

    [89] Essays on Torts Neighbourhood, Proximity and Reliance ed PD Finn.

    [90] Ibid at 40.

    [91] In Mickelberg v R (1989) 167 CLR 259 the High Court held by a majority, that an appeal from a decision of a State court, authorised by the Constitution, s73 did not confer a power to receive fresh evidence.

    [92] See e.g. Supreme Court Act 1970 (NSW), s70; Rules of Court (Qld) O70 r10; Supreme Court Civil Procedure Act 1932 (Tas) s48; Rules of Court (WA) O 63 r10.

    [93] See e.g., Rules of Court (NSW) Pt 56 rr 2 & 5; Rules of Court (Tas) O 69 r1; Rules of Court (WA) O17 r1.

    [94] Stanley Yeo Professor of law, Southern Cross University “Judicial Law-Making and Community Standards (1999) 8 JJA 203.

    [95] Ibid 211

    [96] Senior Pusine Judge, Supreme Court of Tasmania.