by Jeroen Kortmann
- I. Introduction
- II. What is ‘nonfeasance’?
- III. Theoretical justifications for distinguishing between feasance and nonfeasance
- III:1. Introduction
- III:2. A positive duty to act would be too much of an invasion of an individual’s freedom
- III:3. ‘Why pick on me?’
- III:4. Economic arguments
- III:5. The law does not reward the performance of affirmative duties
- III:6. Nonfeasance does not ‘cause’ harm
- III:7. Nonfeasance is less culpable
- IV. Liability for nonfeasance in continental european private law
- V. Liability for nonfeasance in English private law
- V:1 Historical background
- V:2. The law in England at present
- V:3. The defendant has created the risk
- V:4. The defendant has undertaken to act and/or has induced the plaintiff to rely on him
- V:5. The defendant is in a special position of responsibility towards the plaintiff
- V:6. The defendant is in a special position of control over the source of the damage
- VI. Comparative analysis and conclusion
- Table of Cases
- Table of Legislation
This article was submitted for the degree of Master of Studies in Michaelmas term 1999. The article includes few references to material that appeared after that date. Some of the foreign material that is referred to may be rather outdated. The author would like to emphasise that most of this article is part of a larger project, provisionally titled ‘Altruism in Private Law’ (to be submitted for the degree of Doctor of Philosophy), and should be regarded as ‘work in progress’. Any comments or suggestions would therefore be very much appreciated.
I owe a debt of gratitude first and foremost to my supervisor, Mr Robert Stevens, for the enthusiasm with which he guided me through the past year of research. I would also like to thank Dr Gerhard Dannemann and Dr David Ibbetson for agreeing to read drafts of sections of this thesis which touched upon their respective areas of specialisation. Their comments and suggestions were very helpful.
Dr Ibbetson allowed me to read draft-chapters of his forthcoming book, A Historical Introduction to the Law of Obligations, which proved to be indispensable as will become apparent from the frequent references to it in the fifth chapter of this thesis. I also had the pleasure of reading a draft of an article titled ‘Barrett, Negligence and Discretionary Powers’, by Professor Paul Craig and Mr Duncan Fairgrieve. I am very thankful to these authors for allowing me a ‘sneak-preview’.
I very much appreciate the effort of Nick Smith and Shannon McBriar in proof-reading drafts of this thesis.
Finally, I would like to take this opportunity to thank the many institutions that have so generously supported me: the European Commission through the TMR (Training and Mobility in Reasearch) Programme, the Radboudstichting for granting me a ‘stimuleringssubsidie’, the Art and Humanities Research Board for awarding me a studentship and finally Merton College, for awarding me a Domus Graduate Scholarship. Without their continued financial assistance I would not have been (and still would not be) able to afford my stay in Oxford.
‘But he, willing to justify himself, said unto Jesus, And who is my neighbour?
And Jesus answering said, A certain man went down from Jerusalem to Jericho, and fell among thieves, which stripped him of his raiment, and wounded him, and departed, leaving him half dead.
And by chance there came down a certain priest that way: and when he saw him, he passed by on the other side.
And likewise a Levite, when he was at the place, came and looked on him, and passed by on the other side.
But a certain Samaritan, as he journeyed, came where he was: and when he saw him, he had compassion on him,
And went to him, and bound up his wounds, pouring in oil and wine, and set him on his own beast, and brought him to an inn, and took care of him.
And on the morrow when he departed, he took out two pence, and gave them to the host, and said unto him, Take care of him; and whatsoever thou spendest more, when I come again, I will repay thee.
Which now of these three, thinkest thou, was neighbour unto him that fell among thieves?
And he said, He that shewed mercy on him. Then said Jesus unto him, Go, and do thou likewise.’ (Luke, 10:29-37)1
From a private law point of view, the parable of the Good Samaritan raises two questions:
1) Do we have a legal duty to give aid to our fellow human beings? In particular: can we be held liable for damages if we fail to do so? This question falls mainly in the area of tort-law.
2) If we do come to the rescue, like the Samaritan did, will we have any claim for the expenses that we incurred in doing so? Such claims could arise in tort, in unjust enrichment or in “negotiorum gestio“.
This thesis will only deal with the first question: do we have a duty to give aid to our fellow human beings? Lord Atkin’s speech in Donoghue v Stevenson2 gives reason to believe that the answer is affirmative under English law:
The rule that you are to love your neighbour becomes in law, you must not injure your neighbour; (…) Who, then, in law is my neighbour? The answer seems to be – persons who are so closely affected by my act that I ought reasonably to have them in contemplation as being so affected when I am directing mind to the acts or omissions which are called in question.3
However, in Home Office v Dorset Yacht Co. Ltd4 Lord Diplock criticized Lord Atkin’s test in Donoghue. In the view of Lord Diplock omissions give rise to no legal liability:
The very parable of the Good Samaritan which was evoked by Lord Atkin in Donoghue v Stevenson illustrates, in the conduct of the priest and of the Levite who passed by on the other side, an omission which was likely to have as its reasonable and probable consequence damage to the health of the victim of the thieves, but for which the priest and the Levite would have incurred no civil liability in English law.5
It has been confirmed in later judgments that English law does not in principle recognise an affirmative duty to act,6 and thus in general does not require us to give aid to our neigbours in need. In this respect English law differs from most legal systems in Continental Europe. The delictual provisions in the civil codes of for example Germany and France do not distinguish between acts and inaction.
In this thesis the theoretical justifications for distinguishing between acts (or ‘feasance’) and inaction (or ‘nonfeasance’) will be set out in chapter three. Chapter four will proceed with examining the law in two Continental European countries, Germany and France. In chapter five the position in English law will be subjected to a closer analysis. The final aim of this thesis is to investigate if and, if so, to what extent the position of English law on liability for nonfeasance is different from the position of its European counterparts. However, before we can start a more detailed investigation into ‘liability for nonfeasance’, it needs to be decided what definition of ‘nonfeasance’ will be used.
When dealing with liability for nonfeasance, other terms that are generally used to indicate a lack of feasance are ‘inaction’, ‘omission’ or ‘failure to act’. There is a subtle difference between these terms. On the one hand we have ‘nonfeasance’ and ‘inaction’, which are both neutral terms.7 ‘Omission’ and ‘failure to act’ on the other hand, presuppose the existence of a duty, be it legal, moral or otherwise. To say that a person has ‘omitted’ or ‘failed to act’, is to say that he really should have done what he omitted or failed to do.8 In the following the terms ‘omission’ and ‘failure to act’ will therefore only be used in cases in which the existence of a duty to act is undebatable.
So what is nonfeasance? Answers like ‘the opposite of feasance’ do not really help us. In truth, there are many situations in which it is very difficult to draw any logical distinction between feasance and nonfeasance.9 A cardriver who does not brake in time to avoid hitting a pedestrian can be said to have driven negligently. But can it not equally be argued that his negligence did not consist of an act, but rather of a failure to act? Fleming’s answer is negative:
What superficially looks like non-feasance is often, upon correct analysis, a case of misfeasance. A motorist’s failure to brake in time to prevent a collision is not an example of supine inaction: an omission is involved, but it is merely the element that makes his active conduct – driving – negligent.10
In other words: although the failure to brake, in isolation, is nonfeasance, it is part of a larger activity, driving, which has to be qualified as feasance. The problem with this view is that it helps you with answering one question – ‘are we dealing with feasance or nonfeasance?’ – but at the same time gives rise to two new questions:
(i) ‘Is there a larger activity?’, and
(ii) ‘Is the isolated nonfeasance part of this larger activity?’
The answer to the first question is often arbitrary. Fleming does not hesitate to regard ‘driving’ as a larger activity. Yet, when railways fail to provide signals at a level crossing, he regards this as inaction.11 Does this mean that ‘running a railway company’ does not qualify as a larger activity? Similarly, ‘supervising your children’ and ‘babysitting’ are apparently not to be regarded as larger activities, for Fleming treats ‘losing control over children’ as (isolated) inaction.12
If we have identified a larger activity, the second question will prove at least as high a hurdle as the first one. Why is it that we regard a motorist’s failure to brake as part of the larger activity of driving? Let us consider the following example: a motorist drives his car down a country road. Suddenly he sees a severely wounded man ‘stripped of his raiment’, beckoning for help. He decides not to brake. The man continues his journey and a little while later he is confronted with a pedestrian crossing the road. Again he fails to brake, and the pedestrian is killed in the collision.13 Now here we have two instances of failure to brake. Yet, Fleming would probably only regard the second failure to brake as being part of the larger activity ‘driving’. The most likely reason for doing so, is that the traffic rules command the motorist to stop in the second case, but not in the first. In other words: he regards only the second instance of inaction – not braking for a crossing pedestrian – as part of the larger activity – driving – because only in the second instance the motorist has a duty to act. Thus, to Fleming the mere existence of a duty to act can make the difference between inaction and action. For obvious reasons, Fleming’s way of distinguishing between feasance and nonfeasance cannot be adopted in this thesis: by qualifying inaction as part of a larger act, he unwittingly obscures from our view the very duty to act that we are trying to investigate. Instead, we will try to look at nonfeasance itself, isolating it from the acts it may have been accompanied by. Thus, we will regard both ‘not stopping your car to help a man in need’ and ‘not stopping your car to avoid collision’ as instances of nonfeasance. This is of course not to say that similar liability will arise in both cases, because the existence of a duty to act is not at all certain in the first case, whereas it probably is in the second.14
The courts however, do not adopt one uniform definition of nonfeasance. As a result it sometimes proves difficult to establish in individual cases whether the court has made its decision on the presumption that the behaviour complained of is misfeasance or (also) nonfeasance. In those cases close attention should be paid to the court’s reasoning. Most importantly, we must try to establish what particular part of the defendant’s behaviour prompted the court to conclude that he had been negligent. E.g. was it the fact that he hit a cricketball over the fence or that he had failed to build a higher fence? Some of the judgments in the famous cricket-case of Bolton v Stone indeed give reason to believe that this case was concerned with liability for nonfeasance, rather than feasance.15
Thus, although a hard rule that is always able to make the proper distinction between feasance and nonfeasance is impossible to give,16 this thesis will apply the following two guidelines:
I) All action and inaction that is part of the defendant’s behaviour will be regarded in isolation. No inaction is ‘part of an act’ and vice versa.
II) Close attention will be paid to the court’s reasoning; which part of the defendant’s behaviour prompted the court to conclude that he had been negligent?
By not recognizing a general affirmative duty to act, English law does not encourage members of the public to give aid to those in need of help. Quite conversely, English law in fact discourages giving aid, for once someone has started to give aid, he can be held liable for mistakes he makes in doing so. The advice Gregory gave in 1966 – ‘Don’t ever “fease” unless you have to’17 – is still valid under modern English law. Nevertheless, there are apparently enough justifications for the House of Lords to uphold the distinction between feasance and nonfeasance, and to deny the existence of a general duty to act. In the recent case of Stovin v Wise,18 Lord Hoffmann explained:
There are sound reasons why omissions require different treatment from positive conduct. (…) One can put the matter in political, moral or economic terms. In political terms it is less of an invasion of an individual’s freedom for the law to require him to consider the safety of others in his actions than to impose upon him a duty to rescue or protect. A moral version of this point may be called the “Why pick on me?” argument. A duty to prevent harm to others or to render assistance to a person in danger or distress may apply to a large and indeterminate class of people who happen to be able to do something. Why should one be held liable rather than the other? In economic terms, the efficient allocation of resources usually requires an activity should bear its own costs. If it benefits from being able to impose some of its costs on other people (what economists call ‘externalities’) the market is distorted because the activity appears cheaper than it really is. So liability to pay compensation for loss caused by negligent conduct acts as a deterrent against increasing the cost of the activity to the community and reduces externalities. But there is no similar justification for requiring a person who is not doing anything to spend money on behalf of someone else. Except in special cases (such as marine salvage) English law does not reward someone who voluntarily confers a benefit on another. So there must be some special reason why he should have to put his hand in his pocket.
Each of the justifications given by Lord Hoffmann will be examined in the following, together with additional justifications that have been brought forward in academic writings.
In academic debates a mere reference to the sanctity of freedom is often treated as irrefutable and conclusive to the outcome of the debate. We seem satisfied that invasion of an individual’s freedom cannot be tolerated, and often regard an argument to that effect as self-evident and self-justifying, thus rendering additional reasons unnecessary.19 Yet, what an individual’s freedom actually consists of, is a question that has hardly been answered by two ages or two countries alike.20 Hence, Lord Hoffmann is not being very informative when he tells us that omissions require different treatment from positive conduct because ‘it is less of an invasion of an individual’s freedom for the law to require him to consider the safety of others in his actions, than to impose upon him a duty to rescue or protect’.21 At the most we can deduce from this statement that Lord Hoffmann’s understanding of ‘freedom’ includes the right to refrain from acting. Regrettably, Lord Hoffmann fails to give us his definition of ‘freedom’, nor does he tell us why one kind of ‘invasion of an individual’s freedom’, the imposition of liability for wrongful inaction, would be illegitimate where another, the existing liability for wrongful acts, apparently is legitimate.
Freedom or liberty (I shall regard both words as synonymous), is normally said to be the absence of interference by others, or indeed the state, with an individual’s activity. If I am prevented by others from doing what I could otherwise do, I am to that degree unfree.22 This is not to say that interference is necessarily a bad thing. If we want to guarantee everyone a minimum degree of freedom, we will have to restrain everyone from interfering with each other’s freedom. Thus, the freedom of one individual finds its limits where it collides with the freedom of others.23 The difficulty lies in finding the proper standard of interference to apportion a fair share of freedom to each individual. Lawyers and philosophers alike have made numerous attempts to define such a standard of interference. This is not the time, nor the place for a detailed survey of the different standards that have been suggested through the centuries. For present purposes we will have to content ourselves with a summary of the philosophical views that are most often referred to by supporters and adversaries of the rule of no liability for nonfeasance.24
Supporters of the no-liability rule most often refer to the works of Kant for confirmation of their views.25 In his The Metaphysics of Morals, first published in 1797, Kant made a distinction between the doctrine of right on the one hand, and the doctrine of virtue on the other. Regarding the doctrine of right he wrote:
The concept of Right, insofar as it is related to an obligation corresponding to it (i.e., the moral concept of Right), has to do, first, only with the external and indeed practical relation of one person to another, insofar as their actions, as facts, can have (direct or indirect) influence on each other.26 (Footnotes omitted)
Because the doctrine of right only deals with external actions, Kant is said to have been of the view that the law should not require any particular affirmative actions.27 An individual’s choice whether or not to act was an internal affair, which was governed by the doctrine of virtue, rather than of right. Accordingly, a duty to aid others was an ethical, not a legal duty:
(…) duties of benevolence, even though they are external duties (obligations to external actions), are still assigned to ethics because their lawgiving can be only internal.28
Thus, the Kantian idea appears to be, as Howarth puts it, ‘that the state has no business telling people that they are bad people, that they are immoral or wicked’.29 However, Kant himself may not have been as opposed to state interference as is often assumed. The Metaphysics of Morals included a section on the rights of a state, in which Kant wrote that a government was authorised to ‘constrain the wealthy to provide the means of sustenance to those who are unable to provide for even their most necessary natural needs’, for example by imposing taxes to that effect.30 One could argue that the imposition of taxes is different from the imposition of liability in that the former does not involve casting a judgment on an individual’s character; after all, everyone has to pay taxes regardless of one’s (internal) virtue. However, Kant’s suggested solution for the maintaining of abandoned children does appear to involve a straightforward judgment on the taxpayers’ virtue:
(…) the state has a right to charge the people with the duty of not knowingly letting them die, even though they are an unwelcome addition to the resources of the state. Whether this should be done by taxing elderly unmarried people of both sexes generally (by which I mean wealthy unmarried people), since they are in part to blame for there being abandoned children, in order to establish foundling homes, or whether it can be done rightly in another way (it would be hard to find another means for preventing this) is a problem which has not yet been solved (…)31 (Second emphasis added)
Adversaries of the no-liability rule tend to refer to the works of Bentham and Mill.32 Bentham appears to have agreed with Kant that ‘the rules of beneficence’ generally belonged to the ‘jurisdiction of private ethics’.33 However, in cases involving danger to others, he thought that the law ought to interfere:
(…) why should it not be made the duty of every man to save another from mischief, when it can be done without prejudicing himself, as well as to abstain from bringing it on him?34
Unfortunately, Bentham did not proceed to reveal his arguments in favour of imposing such a duty to rescue.35 Mill’s views are more helpful in this respect. In his On Liberty,36 Mill made an attempt to define the proper standard of state interference:
(…) the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number, is self-protection. That the only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right. (…) Over himself, over his own body and mind, the individual is sovereign.37 (Footnotes omitted)
The proper standard according to Mill was what we now know as the ‘harm principle’; the community is only permitted to restrain the behaviour of its members, insofar as this behaviour causes harm to others.38 Consequently we, as individuals, should in principle39 be free to do what we want as long as we do not cause harm to others. From the very moment of its publication, On Liberty has been lauded by some,40 while denounced by others.41 The main objection to the harm principle seems to be its apparent vagueness. If ‘harm to others’ is to be the standard of legitimate interference, what must ‘harm’ be understood to mean? In a chapter devoted to applications of his harm principle, Mill wrote that a man should be free to be drunk if he wants to, provided that he does not harm others in doing so.42 Many of Mill’s contemporaries however, regarded the mere state of drunkenness as in itself harmful to others, because in any case it would set a bad example for the rest of the community.43 Thus interpreted, application of Mill’s harm principle could justify (state) interference with nearly every type of human behaviour. However, such interpretation does not do justice to Mill’s theory. Mill used a common-sense concept of harm, which did not include mere offences,44 nor the kind of negative effects that can be attributed to the free will and consent of the ‘harmed’ person. Thus, one cannot stop a drunkard from being drunk or forbid an atheist preacher from leading his audience astray, because those who may be negatively affected are free to ignore the drunkard and the atheist priest and have themselves to blame if they do not.45
Although Mill’s critics reject the harm principle as the general standard of (state) interference, most of them accept that ‘harm to others’ can be a sufficient reason (though not the only reason) to justify such interference. The reviewer of The Athenaeum predicted this in 1859:
We doubt if anyone, in modern times, will venture to dispute the principle. Some will take it as their guiding rule on principle: others will aver that, whatever other principles they may also allow to act, there is no proper case of application in which this principle does not also apply.46
It has been argued that even Kantian ethics are not altogether inconsistent with the harm principle.47 Thus, the predominant opinion appears to be that if a certain behaviour causes harm (in Mill’s narrow sense of the word) to others, there is a prima facie case for state interference. We must then ask ourselves whether nonfeasance is or can be the kind of behaviour that causes harm to others. This question will be dealt with in more detail in section III:6 of this thesis. For the moment we will content ourselves with Mill’s response:
A person may cause evil to others not only by his actions but by his inaction, and in either case he is justly accountable to them for the injury. The latter case, it is true, requires a much more cautious exercise of compulsion than the former. To make anyone answerable for doing evil to others is the rule; to make him answerable for not preventing evil is, comparatively speaking, the exception. Yet there are many cases clear enough and grave enough to justify that exception.48
Accordingly if we assume, as Mill does, that nonfeasance can cause harm, there appears to be a prima facie case for interference. Nonetheless, there may be good reasons ‘secunda facie‘, ‘at second sight’, to support Lord Hoffmann’s argument that liability for nonfeasance would be too much of an invasion of an individual’s freedom. Regrettably however, Lord Hoffmann does not advance such reasons.
One reason that has been suggested is that we may consider positive duties to be more burdensome than negative duties. This argument is easily met by Atiyah.49 For although positive duties can indeed be quite burdensome, we deny liability even in those cases where there would hardly be any burden. Atiyah gives the example of a person who sees a neighbour’s house on fire. ‘How burdensome would it be to require that person to telephone for the fire brigade?’50 We should also remember that negative duties can sometimes be very burdensome. For although ‘do not act uncarefully’ may seem to be an easy duty to perform, this will become more difficult the more we raise the standard of care. The behaviour of learner-drivers, for example, is measured against the standards of experienced drivers.51 Consequently, the learner driver will either have to drive as carefully as an experienced driver, which to him is nearly impossible, or not drive at all. And if anyone were to suggest that the latter option, ‘not driving at all’, is hardly burdensome, he is mistaken. Where the use of a car may have been a luxury in the past, it is now to many people an indispensable tool of transportation.52
Also, it should be noted that the burdensomeness of a duty is a factor that the courts can take into account at another stage of their judgement, namely when deciding whether the defendant was at fault. If the court finds the performance of a duty too burdensome, it can decide that in the specific case before the court, the defendant has behaved like a reasonable man. Take for example the case of Latimer v AEC Ltd.,53 where the defendant was held not to be at fault for failing to prevent an accident that happened when an employee slipped on the wet factory floor after a flooding. The court was of the opinion that the only possible way of preventing the accident from occurring would have been to close the whole factory altogether until the floor was cleaned. But it regarded this remedy as too burdensome, and concluded that the defendant had behaved like a reasonable man. Hence, the burdensomeness of a duty does not have to prevent the imposition of liability in a whole category of cases, but can be taken into account when deciding in specific cases whether the behaviour complained of was reasonable.54 Thus, if Lord Hoffmann’s ‘freedom-argument’ relies on the presumption that the imposition of liability for nonfeasance would be too burdensome, it is difficult to support his view.
When discussing the notion of freedom, Samuel von Pufendorf – a 17th century writer of the Natural law school – used to emphasise the importance of the individual’s ability to choose. The very reason for holding people responsible for their acts, he argued, is that they voluntarily choose to act the way they do:
And just as the chief reason for considering a man responsible for his own acts is that he undertook them of his own will, so we must especially observe that the freedom of the will is by all means to be asserted, at least in regard to the acts for which a man is commonly held to account before a human court.55
In his speech in Stovin v Wise, Lord Nicholls appears to have adopted a notion similar to Pufendorf’s ‘freedom of the will’ to justify the feasance-nonfeasance distinction:
Liability for omissions gives rise to a problem not present with liability for careless acts. He who wishes to act must act carefully or not at all. (…) With liability for omissions, however, a person is not offered a choice. The law compels him to act when left to himself he might do nothing.56
It must not be doubted that every duty that the law imposes – whether positive or negative – restricts to a certain extent our freedom to choose. Yet, Lord Nicholls argues, positive duties are more restrictive because they deny us the option to avoid being subjected to them. In general, negative duties do offer us this option. If we do not feel competent enough to meet a specific negative duty, we can avoid being subjected to it by not acting at all. Thus, if I feel I cannot live up to the standard of the reasonable driver, I can decide not to drive at all. However, if the law were to require a witness of an accident to act as the reasonable rescuer would do, he would not have an alternative option that would allow him to avoid being subjected to this duty to rescue.
Lord Nicholls’s justification of the feasance-nonfeasance distinction would appear to be a powerful one. There are however some objections to his argument. Firstly, the law already imposes liability in some cases in which we do not have a choice in the sense here discussed. If I blamelessly become the source of a danger, the law imposes on me a duty to prevent this danger from culminating in harm.57 Thus, if through no fault of my own I fall on the road (e.g. I was pushed by a stranger), I have a duty to stand up and clear the road in order to avoid endangering other road-users, even though I had no choice which would have allowed me to avoid being subjected to this duty. Therefore, if the law recognises Lord Nicholls’s version of ‘freedom of the will’, it does not consistently apply it.
Secondly, it is not entirely true to say that we do not have an option that allows us to avoid being subjected to affirmative duties. For in theory, we could lock ourselves in our houses in an attempt to avoid witnessing accidents or other occasions that could require our positive action. True, few people would regard this as a realistic option, but the same can be said in some cases involving negative duties. E.g. ‘not driving at all’ in order to avoid being subjected to the relatively high standard of care required of cardrivers is to many of us not a realistic option either. Hence, the distinction between cases of feasance and of nonfeasance as it is drawn here, appears to be one of degree, not of principle. Imposing positive duties does not leave us without an option to avoid being subjected to these duties, but rather with less of an option than is commonly the case in respect of negative duties.
When caught red-handed with his pockets full of sweets, a child may try to avoid punishment by pointing at his bolder sibling who did the actual stealing. Parents are rarely impressed by this ‘why pick on me?’ argument. After all, one child’s fault is not erased by the fact that another child was equally or even more at fault. Though essentially still the same argument, the ‘why pick on me?’ argument may seem more convincing in complex situations involving nonfeasance. Let us for example recall the (in)famous case of Kitty Genovese who was raped and murdered in broad daylight. Dozens of people watched from their windows and ignored her cries for help. Nobody called the police.58 Why would we allow Kitty’s family to ‘pick’ on one of these people, by bringing a claim in negligence? After all, were the other spectators not equally to blame? And, is it not true that the rapist/murderer was the ‘real’ culprit? As with the sweets-stealing siblings, the answer is fairly straightforward: the question is not whether others were equally or even more culpable, but whether this individual defendant was at fault.59 In this respect there is no reason to differentiate between feasance and nonfeasance.60 Thus, we allow a plaintiff to ‘pick’ on one of a large group of potential defendants because they all individually failed to do what a reasonable man would have done.61
This is not to say that the presence of other people has no bearing whatsoever on the duty of an individual. Depending on the circumstances, a reasonable man may sometimes refrain from acting or limit his actions because there are other people who are in a better position to intervene than he is. The passer-by in the street who witnesses Kitty’s ordeal may be required to run to a phonebox and call the police if he is the only one there. However, another conclusion may be reached if he notices that there are dozens of people witnessing the crime from their windows. Under such circumstances it can be argued that the passer-by may legitimately believe that his intervention is unnecessary, relying on one of the other witnesses to make the phonecall. Thus, the presence of other people – especially if they are in a better position to intervene – can be taken into account when we decide whether the defendant’s behaviour was reasonable. However, if and when we decide that his behaviour was unreasonable, there is no (good) reason to deny a claim against him merely because someone else was equally or even more culpable. Thus, we allow a plaintiff to single out, or ‘pick on’ the defendant of his choice. Any unfairness that might arise from this system, ought to be overcome by the Civil Liability (Contribution) Act 1978, which allows the defendant to bring an action against anyone else ‘liable in respect of the same damage’ to recover a contribution from them.62
In the second half of this century, the law of torts has increasingly been subjected to economic analyses. Terms like ‘wealth maximization’, ‘price deterrence’, ‘internalisation of externalities’ and ‘the cheapest cost avoider’ have become commonplace in textbooks on the law of torts. This is not the right place to examine the various economic theories that have been developed.63 However, the concept of ‘price deterrence’ requires some attention, if only because Lord Hoffmann directly refers to this concept in his speech in Stovin v Wise. Price deterrence has been put forward as an economic justification for the imposition of tort liability on the person who causes loss to others. Essential to the concept of price deterrence is the presumption that if we hold a person liable for the losses his behaviour causes, he will eventually be unable to afford to continue or repeat his harmful behaviour. Take for example two factories that produce a similar type of car. Factory A causes no harm in doing so. Factory B however, uses a production method that has a negative effect on others, say, it causes pollution to the neighbouring land. If the social costs represented by the pollution are not transferred to factory B, there are ‘externalities’. If we ‘internalise’ these externalities to the activity of factory B, the price of the cars of factory B will rise to a level reflecting the ‘true social costs’ of the product. As a consequence, some consumers will buy the cars of factory A instead of B, and eventually factory B will go out of business and the pollution will stop.64 Thus, imposition of tort liability removes externalities and consequently enhances efficiency. As is the case with the general argument of deterrence, the argument of price deterrence is open to considerable criticism.65 Let us however presume that it has some merit in justifying the imposition of liability. Then, why is it that Lord Hoffmann alleges that this theory does not provide a ‘similar justification for requiring a person who is not doing anything to spend money on behalf of someone else’66? If it is true that to accomplish efficiency the externalities caused by actions need to be internalised, then why is this not similarly true for the externalities caused by inaction? Inaction, like action, can cause externalities. Factory C may produce cars without causing any pollution or danger, but if it fails to provide accurate instruction manuals with the cars, this might very well lead to mishandling of the cars and consequently to damage, ‘externalities’. There is no (good) reason why the economic reasons to put factory B out of business do not equally apply to factory C. Apparently, Lord Hoffmann’s assumes that inaction does not cause loss, and therefore does not cause externalities. From the example with the instruction manuals, we see that this point of view is not always easy to maintain. In section III:6 the question whether nonfeasance can cause loss will be subjected to a closer investigation.
A rather different economic argument is put forward by Honoré in a 1991 tribute to Atiyah:67
The most plausible arguments for sanctioning acts more severely than omissions have so far been economic. One can refrain from a great many harmful acts, like poisoning, at the same time, while it is difficult or impossible to combine more than one helpful act, like providing food. The benefit of abstention is also widely distributed, in that if I abstain from doing harm no one is harmed, while the benefit of helpful acts is confined to the few who get the food I can provide.68 (Footnotes omitted)
In other words: the performance of a negative duty involves a smaller effort than the performance of a affirmative duty, because one can ‘refrain from a great many harmful acts (…) at the same time’. Moreover, the performance of a negative duty has a more widespread effect, because ‘if I abstain from doing harm no one is harmed’. However, both submissions are debatable. As for the effort required, for the individual the performance of a negative duty often requires less effort than the performance of an affirmative duty. However, the performance of a negative duty is not always without effort. Sometimes one cannot perform a negative duty other than by taking affirmative action. When confronted with a traffic sign that tells you not to turn left on a T-junction, you will either have to stop your car or turn right. Moreover, for a negative duty to have its required effect, it needs to be performed by all the members of the public who are in the vicinity of the protected interest. The cumulative effort of these members of the public in performing the negative duty can very well exceed the individual effort that is required for the performance of the corresponding affirmative duty. An example can clarify: during a stormy night an unlucky pedestrian is knocked unconscious by a falling branch and falls down in the middle of a street. During the morning rush hour the unconscious man creates a considerable obstacle. To protect the man from being hurt, we can either impose a negative duty not to drive into the man or an affirmative duty to revive the man and/or remove him from the street. From the individual cardriver the negative duty requires less effort. It is easier to manoeuvre his car around the unconscious man than to have to stop his car, get out and help him. But to avoid the man being hurt, it is not just this cardriver who has to guide his car around the man. Every single cardriver who passes through this street will have to do the same. Conversely, it only takes one cardriver’s effort to reach the same result by performing the affirmative duty. The cumulative effort of the cardrivers in performing the negative duty may in this case very well exceed the individual effort in performing the corresponding affirmative duty, reaching the same effect.
The example of the unconscious man also shows that the performance of a negative duty does not always have a more ‘widely distributed’ benefit; it is only the unconscious man who benefits from the cardrivers’ abstention from driving into him. Yet, affirmative action – e.g. taking the man off the street – will not only benefit the unconscious man, but also all the cardrivers who will be passing through the street; they will no longer have to manoeuvre their car around the unconscious man.
Even in cases where the benefit of the performance of a negative duty is more ‘widely distributed’ than the benefit of the performance of an affirmative duty, this does not force us to the conclusion that imposition of the negative duty is necessarily more efficient. For the ‘output’ of a positive act depends on more than just its ‘distribution’. The example given by Honoré provides a good illustration:69 although everybody in my vicinity enjoys the benefit of my abstention to poison them, this is relatively only a very small benefit. For the fact that I abstain from poisoning a hungry man by no means protects him from being poisoned by somebody else. And even if we all abstain from poisoning this hungry man, he may still die of starvation. If I actively feed him, he will survive. Thus, the benefit of a positive act may often be less ‘widely distributed’, but also larger than the benefit of inaction.
The most elaborate attempt to give an economic justification of the general denial of liability in cases of nonfeasance was made by the influential American legal economists Landes and Posner. In 1973, Posner had voiced the opinion that – at least in rescue cases – there are sound economic reasons for imposing affirmative duties.70 In rescue cases, Posner argued, transaction costs are so high that we cannot rely on the free market principle for efficiency. We must therefore create an incentive for potential rescuers to come to the rescue when such intervention is efficient, by introducing liability for failure to rescue. When the costs of rescue are lower than its benefits, potential rescuers will feel compelled to come to the rescue in order to avoid liability. When the rescue costs are higher than the benefits, the ‘reasonable man standard’ allows potential rescuers to abstain from intervention. By 1978 however, Posner’s views had radically changed. After a detailed study into the economic aspects of altruism, carried out with Landes, Posner came to the conclusion that introduction of a positive duty to rescue would not necessarily promote efficiency.71 In individual cases, Posner and Landes argued, the view that the presence of a duty to rescue would promote efficiency, is broadly sound. As a general rule however, liability for failure to rescue would have negative side-effects that might lead to inefficiency. Landes and Posner had the following negative side-effects in mind:
1) Potential rescuers would avoid liability by ‘substituting away’ from activities that give rise to rescue opportunities.72 E.g. good swimmers would avoid crowded beaches.73 Hence, there would be less potential rescuers on the beach, and those remaining would be more likely to be called upon.
2) Even if imposition of a legal duty to rescue would lead to a rise in successful rescues, this would encourage potential victims to disregard their own safety. Moreover, the number of potential victims would increase as hazardous activities would become more attractive.74
3) Under the present Common Law regime, rescuers enjoy public recognition as altruists. Under a regime of liability for failure to rescue, it would be impossible for a rescuer to prove that he was motivated by altruism. This could again discourage potential rescuers from going to places where there is a relatively high likelihood that they will be called upon (e.g. crowded beaches).75
4) The costs of administering a liability rule are high. ‘Legal error costs in particular might be high because of the difficulty, in many settings (e.g., on a crowded beach), of identifying potential rescuers.’76
It must be seriously doubted whether these predicted side-effects are likely to occur. The prediction of the first side-effect overlooks three important points. Firstly, we learn from experience that imposition of liability rarely dissuades people from indulging in activities that may give rise to such liability. Rather, they take more careful precautions or insure themselves against the increased risk of liability.77 Secondly, there are so many activities that may give rise to rescue opportunities, that it is virtually impossible for potential rescuers to substitute away from them. Not only will they have to avoid crowded beaches, but also sporting grounds, industrial sites, busy roads and areas with a relatively high crime rate. Thirdly, Posner and Landes overlook the fact that society does not consist of categories of potential rescuers on the one hand, and potential victims on the other. If I go to the beach, am I a potential victim or a potential rescuer? ‘Both’, is probably the right answer. Most of the people present on a beach are potential rescuers and potential victims at the same time. So even if we accept that the amount of potential rescuers would decrease after the imposition of liability, the amount of potential victims would decrease correspondingly.
The suggested second side-effect must be met with similar scepticism. People do not risk their own safety just because they know that other people will have a legal duty to rescue them if an accident occurs. In fact, there are clear indications that their own safety is one of the few considerations – if not the only consideration – that keep people from acting hazardously.78 Moreover, most legal systems accept that by taking unreasonable risks, one may incur liability in negligence towards the person who comes to the rescue.79 Thus, liability in negligence provides an additional deterrent against hazardous behaviour. The suggestion that more potential victims would indulge in hazardous activities is, again, highly speculative. But even if we accept that this is true, the amount of potential rescuers would rise correspondingly. For every extra potential victim that comes to the beach, is also a potential rescuer.
When Posner and Landes predict the third side-effect, that rescuers would no longer enjoy public recognition for their altruism, we must wonder what kind of liability rule they had in mind. Thus far, not a single proposal for a legal duty to rescue entails a standard of care that is higher than that of the ‘reasonable man’. A rule of liability for nonfeasance would at the most require a person to take affirmative action if it would be unreasonable not to do so. It would not require acts that are beyond the call of moral duty, acts of heroism or acts of sacrifice.80 In other words: it would not require the kind of acts that deserve special praise and public recognition for being truly altruistic. Hence, the public recognition for jumping into a sea with treacherous tidal currents in order to rescue someone, for going into a burning house to save a cat and even for giving large donations to charities will remain unchanged, because these acts are, and would remain, ‘beyond the call of duty’.
As for the fourth and last side-effect, it is conceivable that imposition of liability in cases of nonfeasance would bring with it an increase in administrative costs, especially in cases of multiple potential defendants. The reason for this increase in costs is not, as Posner and Landes seem to suggest, that it would be difficult for the victim to identify a defendant. In cases with multiple potential defendants (e.g. the case of the crowded beach) it is relatively easy for the victim to identify a defendant, since there are so many of them. However, for the defendant bringing an action under the Civil Liability (Contribution) Act 1978, it would indeed be difficult to identify all the other potential defendants, as well as to establish the contribution that each of them would have to pay. Yet, this is not a side-effect that is particular to cases of nonfeasance. The same problem, and similar costs, occur in cases of feasance involving multiple potential defendants.81
We must therefore conclude that the economic arguments put forward by Lord Hoffmann, by Honoré and by Posner and Landes amount to very little. It is here submitted that in general the imposition of affirmative duties would promote efficiency as much82 (or as little) as the existing negative duties do. Thus, economic analysis does not offer a valid argument for a general denial of liability for nonfeasance.
As was indicated in the introduction to this thesis, the law relating to ‘negotiorum gestio‘ falls outside of the scope of this thesis. Hence, Lord Hoffmann’s statement that ‘English law does not reward someone who voluntarily confers a benefit to another’83 will here remain unchallenged. However, even if we assume this statement to be correct, we need not accept it as an argument against the imposition of affirmative duties. After all, the performance of many negative duties is not rewarded by English law either. Moreover, the mere absence of a remedy in one area of the law does not necessarily provide sufficient justification for the absence of a remedy in another area of the law; it may be that both areas of the law need to be reconsidered.
Another argument that has been used to exclude liability for nonfeasance is that it is difficult to see how nonfeasance can be the cause of any loss.84 In the case of positive action the defendant has made the position of the plaintiff worse. Yet, in the case of inaction the defendant has merely failed to confer a benefit to the plaintiff.85
However, there are also instances of positive action in which the defendant merely prevents a benefit from being conferred to the plaintiff.86 In these cases we do not seem to have any problem with regarding the defendant’s behaviour as a ‘cause’ of the plaintiff’s loss. Let us presume that the Priest and the Levite did not walk by, but waited for the Samaritan and convinced him not to help the victim. It is hardly debatable that the actions of the Priest and the Levite would constitute a ‘cause’ of the victim’s loss. Yet is it not true that the Priest and the Levite merely prevented a benefit from coming the victim’s way? The opposite view would be that ‘convincing somebody to abstain from helping a victim’ actually makes the position of the victim worse. But if this is true, why should ‘convincing oneself to abstain from giving aid’ be regarded as any different?87
The problem that has often been felt with regarding nonfeasance as a ‘cause’, originates in our understanding of the word ‘cause’ as implying a positive interference, not a mere inaction. Historically, the notion of ‘movement’ has been central to our concept of causation.88 Honoré explains:
(…) the reason why movements feature so prominently in our assessment of responsibility is that we have a picture of a world as a matrix into which, by our movements and especially our manipulation of objects, we introduce changes. By moving their bodies people change the world, though they may not know or foresee exactly what changes will come about. It is inherent in this world-picture that positive acts count as interventions and attract at least a minimal responsibility for their outcomes. On the other hand, non-movement is prima facie not an intervention in the world, so that the agent is not responsible for its consequences.89 (Footnotes omitted)
These days, the legal concept of causation is no longer completely governed by the notion of ‘movement’. If and when we find that the defendant had an affirmative duty to act, we have, at least in England, no special difficulty with regarding the defendant’s omission as the ‘cause’ of the loss.90 Thus, a wounded man’s refusal to see a doctor must be regarded as a cause of his death.91 Likewise, a borstal officer’s failure to supervise some borstal trainees is treated as a cause of the damage that the escaped trainees inflict on someone’s yacht.92 Honoré continues his explanation:
(…) superimposed on this picture of the world is a refinement of it, which treats disruptions of the normal course of events as similar to interventions which bring about change, though the change is not now a change in the existing state of affairs but rather in the normal or expected sequence of events. The regular course of things may take the form not only of recurrent natural events, but of regular human conduct such as clearing the street when the snow stops falling. If this regular or expected conduct fails to occur, the failure then counts not, as the first world-view would suggest, as a non-intervention but rather, on a wider view of what amounts to change, as a sort of intervention in the world.93
Thus, if the local government suddenly stops cleaning the streets and collecting the garbage, there can be no doubt that the refusal to do so is a legal cause of the resulting chaos. However, the decision of the House of Lords in East Suffolk Rivers Catchment Board v Kent94 has been the source of some confusion. Owing to a very high tide, a breach was made in a sea wall as a consequence of which the respondent’s land was flooded. The appellants exercised their statutory powers to repair the wall but did the work so carelessly that the flooding continued for 178 days. It appeared from the evidence that by the exercise of reasonable skill the breach in the wall might have been repaired in 14 days. If Honoré’s understanding of causation had been applied in this case, the appellants’ failure to repair the wall properly would doubtlessly have been regarded as the cause of the continued flooding. Honoré would have regarded the 164 extra days of flooding as a ‘change in the normal or expected sequence of events’. However, the House of Lords by a majority rejected the respondent’s claim for damage. Parts of the judgments in this case seem to suggest that the reason why the respondent’s claim is rejected, is because his loss was not caused by the appellants’ incompetence. Viscount Simon argued:
In the present case, the damage done by the flooding was not due to the exercise of the appellant’s statutory powers at all. It was due to the forces of nature, which the defendants, albeit unskillfully, were endeavouring to counteract.95
Similarly, Lord Thankerton’s judgment read:
(…) I have felt throughout that the only real question in this appeal relates to causation, and the difficulty arises from the fact that, prior to the intervention of the appellants, a source of damage had been already created by the action of the flood waters (…) The point on which admittedly the success or failure of the appeal depends is whether the failure of the appellants to close the breach sooner by an efficient method can be held as the causa causans of the damage which accrued during these extra days, to the exclusion of the flood breach as a cause of the damage. However, it was in fact still the action of the water, rendered possible by the original breach, which caused the damage during these days, and failure to stop such action of the water cannot alter the fact that it is the water coming through the breach which causes the damage.96
If these judgments were correct, failures to remove an already present source of damage would not constitute a legal cause. A wounded man’s refusal to see a doctor would not be a cause of his death, the local governments failure to clean the streets would not constitute a cause of the resulting mess and a doctor’s failure to diagnose poisoning would not be regarded as a cause of the patient’s death.97 These implications are unacceptable, and the conclusion that East Suffolk Rivers Catchment Board v Kent cannot be supported on the grounds put forward by Viscount Simon and Lord Thankerton is unavoidable. The question in this case was not whether the appellant’s incompetence had caused the damage resulting from the 164 extra days of flooding. The causal connection was undebatable.98 The real question was whether the appellant had an affirmative duty to abate the flooding or merely a negative duty not to make it worse.99 The recent judgment of the Court of Appeal in Capital and Counties v Hampshire CC100 must be understood in the same way. In one of three separate cases a firebrigade arrived at the scene of an explosion and satisfied themselves that all the fires had been extinguished. They left the scene without inspecting the plaintiff’s premises. A fire later broke out, which severely damaged the plaintiff’s premises. The reason why the plaintiff’s claim was rejected is not that the firebrigade’s failure to inspect the plaintiff’s premises was not a cause of the damage. The reason is that a firebrigade, in the opinion of the majority of the Court of Appeal, merely had a duty not to cause additional damage by their action, not an affirmative duty to extinguish the fire.
The confusion between issues of ‘causation’ on the one hand and issues of ‘duty’ on the other is hardly surprising. By incorporating into our ‘picture of the world’, as Honoré calls it, not only the ‘existing state of affairs’ but also ‘the normal or expected sequence of events’,101 we are asking for trouble. For what we ‘expect’ of people is often – but not always102 – the same as their legal duty. Atiyah summarizes the issue most lucidly:
Many people feel instinctively, if irrationally, that one can only treat nonfeasance as a ‘cause’ if one first starts by assuming that there was a duty to act. (…) the way we view the causation issue depends on whether we think that the defendant ought morally to have done something, and whether we think that this moral duty ought to be translated into a legal duty. This is a matter of legal policy and does not depend on any objective distinction between misfeasance and nonfeasance.103 (Footnotes omitted)
In his already quoted tribute to Atiyah, Honoré put forward a powerful defence of the ‘acts-and-omissions doctrine’.104 In general, contends Honoré, omissions are less culpable than acts that bring about the same or similar outcome, other things being equal. Thus, hitting someone is worse than not preventing someone from being hit.105 However, says Honoré, omissions that violate a distinct duty that we owe to other people or associations are usually on a level with positive acts which violate those duties.106 For a nurse it is usually just as culpable to give the wrong medicine, as it is to give no medicine at all.107 And even if there is a distinct duty, it is sometimes worse to act than to omit:
Take the problem of litter in the streets. If a person is employed by the local council to keep the streets clean, it can make no noticeable difference to the appearance of the streets whether he drops a wrapper in the street himself or fails to pick up a wrapper dropped by someone else. The outcome in each case is that there is one more wrapper in the street than there would otherwise have been, and the urban scene is to that extent less tidy. It is nevertheless probably worse for him to drop a wrapper than to fail to pick one up, since he is in that way adding to the very conditions he is employed to remedy.
Honoré’s contentions are hardly debatable. All other things being equal, feasance is often regarded as being more culpable than nonfeasance, not in the least because many acts imply an additional inaction. When we say that ‘dropping a wrapper is worse than not picking someone else’s wrapper up’, what we mean to say is that ‘dropping a wrapper and not picking it up’ is worse than ‘not picking up someone else’s wrapper’. And if we say that – presuming that we have a duty to protect one another – hitting someone is worse than not preventing someone from being hit by a third person, this must be true, because by hitting we break two duties at once: the negative duty not to hit and the affirmative duty to protect.
Yet, accepting Honoré’s contentions, do we also have to accept that, as he puts it, ‘the acts-and-omissions doctrine is broadly sound’? Surely just the fact that nonfeasance is usually less culpable than feasance – all other things being equal – cannot justify a general denial of liability for nonfeasance? The harmful acts of a rookie doctor are usually less culpable than those of an experienced doctor. Likewise, unintentionally harmful acts are usually less culpable than acts committed with the intent to cause harm. Yet neither against the rookie doctor, nor against the unintentional wrongdoer do we deny the victim a claim.
If all other things are not equal, the culpability of nonfeasance can very well exceed the culpability of feasance. If an inaction causes more harm than an action, the inaction is often seen as more culpable. There can be no doubt that leaving a man to drown where one could easily throw him a rope is more culpable than accidentally breaking someone’s tea-pot. In rescue cases it may even happen that, although the act causes more harm than the inaction, the inaction is still regarded as more culpable than the action. Imagine that the Good Samaritan, trying to do his very best to lift the wounded man ‘on his beast’, slipped and dropped the man, causing him to die earlier than he would have done if the Samaritan would have left him alone. The act of the Samaritan has caused more harm than the Priest’s and the Levite’s inaction. However, it is not unlikely that the behaviour of the Samaritan, who after all did his very best to save a fellow human being, will be regarded as the less culpable. Thus, if culpability is to be the test for imposing liability, there is no (good) reason to make a general distinction between cases of feasance and cases of nonfeasance.108
Whether directly or indirectly, many provisions in Western European codifications of private law have been influenced by the corresponding provisions in the laws of the Roman Empire. The Romans, too, were confronted with cases of damage caused by nonfeasance. The relevant question under Roman law was whether the Lex Aquilia, the Roman statute establishing liability for killing slaves and cattle and for causing damage to goods, was applicable to cases of damage (or death) caused by nonfeasance. Ulpian treats a case in which a stoker-slave falls asleep at the furnace, as a consequence of which the house burns down.109 If the slave has lit the fire in the furnace himself, there seems to be no doubt in Ulpian’s mind that liability arises. The lighting of the fire, however inculpable in itself, has, to use English terminology, raised a duty of care to control the fire. More difficult is the case in which one man lights the fire, but another watches over it carelessly. Ulpian gives the following view, as recorded in emperor Justinian’s Corpus Iuris Civilis, D. 126.96.36.199:
(…) but if one man lit the furnace and another watched it negligently, will the one who lit it be liable? For the one who watched it did nothing, while the one who lit it in the proper way was not at fault. What then are we to say? I think that an actio utilis lies against both the man who fell asleep at the furnace and the man who negligently watched it; nor can anyone say that the man who fell asleep was overtaken by a natural human weakness, because his duty was either to put out the fire or to make certain that it should not spread.110
In 19th century Germany, in parts of which the Corpus Iuris Civilis still had force of law, Ulpian’s words were much discussed. Heavily relying on this text, Pernice came to the conclusion that the Romans did not draw a general distinction between feasance and nonfeasance; if all other requirements set in the Lex Aquilia were fulfilled, the careless inactor, like the careless actor, would incur liability:
The only question that is always asked is whether there is a substantial loss on the one hand and careless behaviour as its cause on the other hand. When a stone falls from a cart and damages an animal, this may have occurred because the stones were badly placed, or because the necessary plank was not pushed in its place. One is an action, the other an inaction. This however, does not make a difference; what matters is that it was bad placement, that the necessary plank was not in its place.111
If Pernice was right, the Roman law position on liability for nonfeasance was very much in agreement with the position that the natural law movement adopted. The natural lawyers abandoned the distinction between feasance and nonfeasance for all practical purposes and subjected liability for both feasance and nonfeasance to the same requirements.112 Pernice’s opinion was, however, not the opinion of the majority of writers in the 19th century. Although his important contemporary Dernburg agreed with Pernice,113 the majority followed Windscheid, who thought that the Romans generally denied Aquilian liability in cases of nonfeasance.114 According to Windscheid, there was one exception:
(…) one will only be obliged by inaction, if action had become required due to an earlier or simultaneous act.115 (Footnotes omitted)
Thus, according to Windscheid, one only had a ‘duty to act’ if one’s earlier or simultaneous behaviour had created the need for intervention. Hence the liability of a man who carefully lights a fire, but does not control it properly and the liability of a doctor who performs a skillful operation but neglects the aftercare.116
With distinguished scholars like Pernice, Dernburg and Windscheid having such different views on the subject,117 it is hardly surprising that it proves difficult to establish the true nature of liability for nonfeasance under Roman law. Although one cannot be absolutely certain, it is now generally assumed that the Lex Aquilia initially did not apply to cases of nonfeasance.118 This assumption is based on Gaius’ Institutes, in which it is written that there was only a direct action under the statute when the defendant had done the damage with his own body, corpore suo.119 When the damage was not a direct consequence of the wrongdoer’s physical behaviour, the Lex Aquilia did not apply. Thus, it appears that there was initially no remedy against someone who persuaded a slave to go down a dangerous well where he was killed,120 nor against him who killed a slave by pushing a third party into him.121 If the Lex Aquilia originally required that damage be done corpore suo, this requirement must have blocked liability in cases of nonfeasance. For the very notion of nonfeasance implies the absence of direct physical contact between wrongdoer and victim.122
No matter what the original scope of the Lex Aquilia may have been, over the centuries it came to apply also to cases in which the damage was not done corpore suo. If the direct action on the Lex did not lie, one was granted a decretal action to the same effect;123 an actio in factum or utilis.124 It is likely that along the same lines decretal actions have been developed for certain instances of loss caused by nonfeasance. By the time of Emperor Justinian’s reign (6th century A.D.) we can distinguish at least three types of cases in which liability for nonfeasance could arise:
1) The wrongdoer had created a risk and failed to control it; hence the liability of the doctor who neglected the aftercare,125 of the stoker who fell asleep126 and of the pruner who failed to warn for falling branches.127
2) The wrongdoer had been trusted with the specific task of doing what he failed to do; hence the liability of the man who did not light the fire in the furnace, but was supposed to watch it and failed to do so128 and of the sailors who failed to control their ship in order to prevent a collision.129
3) The wrongdoer was in a special position of control over the person who directly inflicted the harm; hence the liability of the master who knew that his slave intended to wound or kill someone, but failed to prevent it.130
It is worth noting that the Roman term ‘neglegentia‘, from which our ‘negligence’ derives, originally appears to have denoted a (negligent) failure to perform a specific duty; nonfeasance. For active misconduct the term culpa was used.131
Since the introduction of the German Civil Code (BGB) in the year 1900, the Roman texts have lost force of law in Germany. The liability rules for the German equivalent of ‘negligence’ can now be found in § 823 BGB:
§ 823 [Duty to compensate for damage]
(1) A person who, wilfully or negligently, unlawfully injures the life, body, health, freedom, property or other right of another is bound to compensate him for any damage arising therefrom.
(2) The same obligation is placed upon a person who infringes a statute intended for the protection of others. If, according to the provisions of the statute, an infringement of this is possible even without fault, the duty to make compensation arises only in the event of fault.132
Thus, the German legislator did not appear to differentiate between liability for feasance and for nonfeasance.133 However, when interpreting the requirement of ‘unlawfulness’ in section (1) of § 823, the courts have taken a different approach towards actions on the one hand and inaction on the other.
If one of the protected interests of section (1) has been infringed through a positive act, this act is presumed to have been unlawful. The only way for the defendant to rebut this presumption is by invoking a special justification for his act (e.g. self-defence).134
However, if one of the protected interests of section (1) has been infringed by inaction, this inaction is not automatically presumed to have been unlawful. Inaction is only regarded as unlawful if it violated a ‘duty to act’ (Pflicht zum Handeln).135
During the second half of this century the different approach taken by the courts towards cases of feasance and of nonfeasance has been heavily criticized by many influential authors.136 According to these authors one should not presume positive acts to be unlawful, merely on the ground of their harmful result. As with inaction, positive acts should only be regarded as unlawful if they violate a pre-existing duty.137 Thus, according to this view, there should no longer be a difference in approach towards cases of nonfeasance and cases of feasance: both actions and inaction should only be regarded as unlawful if a duty is violated. In theory this development could have serious implications for cases of nonfeasance.138 In practice however, the courts have been – and are likely to stay – much more reluctant to recognise the existence of affirmative duties than of negative duties.139
German law roughly distinguishes between three types of affirmative duties. Firstly, the defendant may, explicitly or implicitly, have undertaken a responsibility for the well-being of others. Thus, contractual parties have duties to act in good faith, regardless of the exact wording of the contract. In the absence of a contract, too, an undertaking may give rise to an affirmative duty. Thus, a neighbour who promises to look after a child or dog, can be held liable for not doing so, even though his promise may not have been contractually binding.140
Because their occupations put them in a special position of responsibility, specific protective duties are imposed on nannies, doctors, the police, sportsteachers etc. Similar duties to ensure the safety of others, are imposed on those who have a family relationship or a relationship of love and affection with the person in need of protection. Thus, although the Levite may not be required to render assistance to everyone, he does have a duty to aid his close relatives, his spouse or his close friends.141
Where the first two types of affirmative duties arise out of the special responsibility the defendant has for the safety of the endangered person, a third type arises when the defendant is in some way connected with the danger itself. Examples are the duty of owners of dangerous premises, substances or animals to prevent others from being harmed. Under this heading one usually also includes the duty arising from a ‘previous act’ (‘aus vorangegangenem Tun‘); those who, no matter how inculpably, create a danger, have a duty to prevent the danger from harming others. Thus, like under Roman law, someone who lights a fire will be liable under German law if he fails to control the fire.142
Most of the above affirmative duties have been expressly recognised by the courts through creation of the so-called ‘Verkehrspflichten‘ and ‘Verkehrssicherungspflichten‘.143 These terms are not easily translated. For present purposes it will suffice to say that Verkehrs(sicherungs)pflichten are legal duties, that require someone who, within the scope of his responsibility, establishes a source of potential danger or allows it to remain, to protect the interests and rights of others against such danger.144 Very soon after the introduction of the BGB, the German courts started recognising the existence of Verkehrs(sicherungs)pflichten.145 The following case was considered by the Reichsgericht in 1903:146 On a cold winter evening the plaintiff had lost his balance when walking on slippery stone steps in his hometown. He brought a claim for damages against the local authorities, arguing that they ought to have gritted the steps. The Reichsgericht found for the plaintiff, establishing that the local authorities had a general duty to take care147 that the steps were safe for the public to walk on. This duty, argued the Reichsgericht, arose from the local authorities’ ‘power of disposition’148 over the steps. Although the local authorities in this case were the legal owners of the steps, it was explicitly stated that this duty of care was not limited to legal owners:
The main focus does not lie on the right of ownership as such; those who possess other rights that entail power of disposition may have a similar duty to take care. The question is, whether the person concerned, by dint of his actual and legal relationship with the property, ought to take any kind of care towards third parties when disposing of this property, respectively when handling or using it in legal relations.149
The amount of Verkehrs(sicherungs)pflichten that is recognised by the courts has increased immensely over the years. In the majority of cases they entail an affirmative duty to act, and arise either out of the special relationship the defendant has with a particular thing or piece of property or out of the defendant’s special position of responsibility.150 Examples are the duty of local authorities to ensure the safety of the roads,151 the duty of hosts to ensure the safety of their guests on their premises,152 the duty of care that road-users owe to other road-users153 and the duty of those whose trade or profession may affect others, to use particular care and skill in exercising their trade or profession.154 What standard of care the Verkehrs(sicherungs)pflichten impose, depends on the particular circumstances of each individual case. The most important factors that are taken into account are the likelihood of a particular risk to materialise, the serious nature of its consequences if the risk does materialise and the difficulty of preventing the risk from materialising.155
Another possible source of affirmative duties are the statutory provisions. If a statutory duty is violated, as a result of which someone is harmed, the victim can bring a claim under section (2) of § 823BGB. This claim will only be successful if the statute was intended to protect others (if the statute is a so-called ‘Schutzgesetz‘). Hence, if someone forgets to light his carlights at night, as required by § 17 of the German Traffic Regulations (Straßenverkehrsordnung), he will be liable to those who suffer harm as a consequence of his omission.156 Of particular interest in this context is § 323c of the German Criminal Code (StGB), a provision originating from the days of the Nazi regime:157
§ 323c [Failure to render assistance to those in need]
Whoever fails to render assistance in case of accident, common danger or emergency, although such assistance was needed and could have been expected from him under the circumstances, especially since he could have rendered it without placing himself in significant danger and without violating any important duties, shall be punished by up to one year’s imprisonment or by fine.158
After the collapse of the Nazi regime, the question arose whether § 323c StGB159 was still in force. The Großer Senat für Strafsachen answered this question in the affirmative, stating that the duty to render assistance in emergencies is a ‘moral obligation that has existed from time immemorial’ and adding that ‘aid to neighbours in need was, in particular, always an imperative command of Christian doctrine’.160
Thus, German law continued to impose a criminal law duty to rescue on those who can do so without the risk of harming themselves. At first sight, this statutory provision would seem to be ‘intended for the protection of others’, in the sense of § 823 (2) BGB. Indeed, it is likely that the German legislator in 1935 was of the opinion that the criminal duty of § 323c StGB would give rise to a private law duty under § 823 (2) BGB.161 However, legal opinion has changed and it is now the predominant view that § 323StGB was not intended to protect individual persons but only society as a whole.162 This legalistic argument does seem peculiar, especially considering the following words by Haager:
(…) It is sufficient, that a statutory provision mainly pursues another goal, but simultaneously seeks to protect certain individual interests. (…) Only statutory provisions that are exclusively intended to protect the order of society, the entire state as such, its external intactness and its internal constitution and administration, fall completely outside of the context of such ‘Schutzgesetz‘, like the criminal provisions on high treason and treason, on resistance against state authority, on breach of the constitution etc.163
It is hard to imagine that § 323c StGB was exclusively intended to protect the ‘entire state as such’ and not (at least) also certain individual interests. It is likely that there are other, more profound reasons that lie behind the denial of a civil duty to rescue under § 823 (2) BGB. In a judgment in 1988, the Oberlandesgericht of Frankfurt used an argument which strongly resembles the ‘Why pick on me?’ argument:
The criminal provision (of § 323c StGB, JSK) is not a ‘Schutzgesetz‘ in the sense of § 823 (2) BGB, because there is no reason to hold the person who omits to render assistance liable in the same way as the person who inflicted the harm.164 (Footnotes omitted)
Even in the absence of a Verkehrspflicht or a Schutzgesetz, liability for nonfeasance may arise if the defendant’s refusal to act was intentional and contra bonos mores.165 § 826BGB reads:
§ 826 [Intentional damage contra bonos mores]
A person who, in a manner contra bonos mores, intentionally causes damage to another, is bound to compensate him for the damage.166
‘Intention’ is taken to exist, if the defendant was aware of the harmful nature of his behaviour. It is not necessary that the motive of his behaviour was to cause harm.167 However, not every intentional refusal to act is regarded to be contra bonos mores. Schäfer, for example, submits that an intentional refusal to render assistance to someone in need is not contra bonos mores, unless such refusal will lead to death, serious bodily harm or a dangerous road accident and unless rendering assistance is easy and can be done without disregarding one’s own important interests.168
There are relatively few cases involving nonfeasance in which successful claims have been brought under § 826 BGB. Contractual parties have been held liable for intentionally failing to disclose relevant information in the course of the negotiations.169 Also, claims have been allowed against persons who noticed that their signature had been forged on an official document, but failed to give a warning to that effect.170
Apart from the general provisions of §§ 823 and 826, the BGB imposes certain specific affirmative duties in the §§ 831 to 838. Thus, if one has failed to exercise the proper supervision or control, one will be liable for damage caused by one’s employees, one’s animals and the buildings or structures on one’s land.
Finally, affirmative duties to act can arise from contractual and pre-contractual relationships. Where English law only imposes contractual duties if a contract has been concluded, German law recognises that contract-type duties can arise from pre-contractual relationships. Thus, a contract-type claim may lie in the event of carelessness of the defendant during the pre-contractual phase (‘culpa in contrahendo‘), even if no contract is ever concluded between the parties.171 Also, German law recognises that contracts may have protective effects towards third parties (Schutzwirkung für Dritte).172 The famous ‘vegetable leaf case’173 provides an example of both doctrines at work at the same time. The plaintiff went with her mother to the defendant’s self-service store. While her mother stood at the till, the plaintiff slipped over a vegetable leaf. She sued the defendant for breach of his duty to provide safe access. The Bundesgerichtshof had no doubt that if the plaintiff’s mother had been injured in the same way as her daughter, the defendant would have been liable for culpa in contrahendo.174 But even the plaintiff, who did not have a pre-contractual relationship with the defendant, was allowed a claim on the basis of the Schutzwirkung of her mother’s pre-contractual relationship with the defendant:
It accords with the long-standing case-law of this Senate in particular that in special circumstances even bystanders who do not themselves participate in a contract are included in the protection afforded by it (…)175
Thus, German law imposes affirmative duties under a variety of headings: § 823BGB in connection with the Verkehrspflichten, § 823 (2)BGB and its Schutzgesetze, § 826BGB, §§ 831 to 838BGB and finally the law of contract. It must however be noted, that, in the absence of express statutory provisions, the courts have thus far only imposed affirmative duties, when the defendant was in some way connected with the endangered person or with the danger itself. If a case like the one of the Good Samaritan were to be considered now, it seems unlikely that the German courts would have held the priest or the Levite liable for the harm that was caused by their inaction.
Until the seventeenth century, the French law of delict had been a mixture of Roman law, canon law and local laws and customs.176 There was no general rule of liability, neither was there a clear distinction between the compensatory and the punitive function of awarding damages. In 1689 Domat published his ‘Les loix civiles dans leur ordre naturel‘, a most influential work which would eventually provide the fundamentals of the modern French law of delictual liability. Central to Domat’s understanding of delictual liability was the notion of faute. Domat distinguished three types of fautes: those that consist of a crime or offence, those that are a breach of agreements and those that are neither, ‘like when one thoughtlessly throws something out of a window and spoils someone’s clothes, when animals that are not properly looked after cause damage, when one causes a fire through carelessness, or when a building under the threat of falling down is not repaired and collapses on another building and causes damage’.177 Delictual liability is concerned with only the latter type of fault. When under a duty to act, Domat continued, a defendant could be held liable for a failure to prevent harm:
Those who, being able to prevent harm from occurring, when a certain duty committed them to prevent this harm, failed to do so, may be held liable depending on the circumstances. Thus, a master who sees and allows the infliction of harm by his servant, is held liable.178
Domat’s ideas were adopted, expanded and improved by Pothier, the most influential French scholar of the eighteenth century. By the end of that century, the French legislator decided to draft a liability rule that was meant to be universally and eternally applicable.179 Domat’s theories were clearly the source of inspiration of the articles 1382 and 1383 of the French Civil Code (C. civ.), which came into force in 1805:180
1382 Any behaviour whatever of man which causes damage to another obliges him by whose fault (‘faute‘, JSK) it occurred to make reparation.181
1383 Each one is liable for the damage which he causes not only by his own behaviour but also by his negligence or imprudence.182
When defining the notion of faute, the French usually distinguish between faute de commission (fault by commission) and faute d’abstention (fault by abstention).183 With regard to the latter, a further distinction is made between abstentions dans l’action (abstentions within an action) and abstentions pures et simples (pure abstentions).184 When we discussed a similar distinction made by Fleming – between ‘real’ nonfeasance and nonfeasance that is part of a larger activity – in the second chapter of this thesis, we concluded that such a distinction is probably not very useful. The same view is adopted by Tunc and the brothers Mazeaud in their influential Traité Théorique et Pratique de la Responsabilité Civile.185 However, the courts have continued to use the distinction.
In cases of abstentions dans l’action, the courts do not seem to have had any problems with imposing liability.186 The so-called ‘Branly-case‘ provides what is probably the most well known example of an abstention dans l’action; Professor Turpain, the author of an article with the title ‘Historique de la T.S.F‘ (‘History of Wireless Telegraphy’) had omitted to mention the name and the works of the plaintiff, Edouard Branly. The French Cour de Cassation decided:
That the court should have investigated whether Turpain, when writing a history of wireless telegraphy in which the name and the works of Branly were willfully omitted, had behaved like a prudent writer or historian, aware and conscious of the duty to be objective that rested upon him.187
Thus, abstentions dans l’action are measured against the same standard as actions are:188 i.e. the French equivalent of the ‘reasonable man standard’.189 It is worth noting that the Cour de Cassation was apparently of the opinion that imposition of liability was not too much of an invasion of Professor Turpain’s freedom of speech and expression.190 On the same principle those who, however blamelessly, create a risk, have an obligation to take reasonable care to prevent this risk from materialising.191 Hence the liability of a railway company for failing to provide the appropriate warning signals at a level-crossing.192
In the area of abstentions pures et simples, too, the Branly-case would play a vital role. Initially, the courts showed a reluctance, similar to that in Germany, to impose liability for ‘pure’ nonfeasance. In a 1924 judgment, the Cour de Cassation held that:
(…) if everyone is liable for his negligence, an omission can only bring about liability if the person who is held liable had an obligation to do that which was omitted.193
At the time, this judgment was interpreted as meaning that liability for nonfeasance would only be imposed if it violated an affirmative duty that had been created by statute or at least by custom.194 Obviously, this requirement restricted the scope of liability for abstentions pures et simples.195 The main reason for the reluctance of the courts is said to have been that one should be allowed at least the freedom not to act. Additionally, it was doubted whether a causal connection could exist between nonfeasance and harm.196 In the 1940’s, the recognition of affirmative duties by the courts was accelerated by a development in the area of criminal law. Unlike the Germans,197 the French regard the commission of any criminal offence which causes harm to others as a ‘faute‘ for the purpose of articles 1382 and 1383 C. civ. No further inquiry into the intended scope of the criminal provision is necessary. This concept is referred to as the ‘unity of criminal and civil faults’.198 Thus, when the legislator introduced a duty to act under criminal law, this was bound to have an immediate effect on the interpretation of ‘faute‘ in the law of negligence.199 The first two paragraphs of article 63 of the French Criminal Code (C. pén.), after it was amended in 1945,200 read:
Article 63 Any person who, by his immediate action and without danger to himself or others, could have prevented either a felonious act or a misdemeanor against the person, willfully fails to do so, shall be punished by jailing for no less than one month nor more than three years and by fine from 12,000 to 500,000 francs, or either punishment, unless more severe punishments are provided by this Code or special law.
Any person who willfully fails to render or to obtain assistance to an endangered person when such was possible without danger to himself or others, shall be subject to like punishments.201
The concept of ‘unity of criminal and civil faults’ meant that the courts could translate the criminal duties imposed by article 63 C. pén. into private law duties. In 1947, the Tribunal Correctionnel d’Aix decided to award 25,000 francs in damages to the plaintiff, who had nearly drowned when he fell through ice into a deep canal. The defendant, who was the plaintiff’s father-in-law, had walked away from the scene after refusing to assist a third person who tried to rescue the plaintiff by handing him an iron bar to which he might cling.202
In a later case, a doctor who was held liable by the Cour d’Appel of Paris for failing to assist a patient in need, argued that article 63 C. pén. was merely intended to protect the public interest, and should not give rise to private law claims. The Cour de Cassation dismissed the submission:
(…) if the provisions of article 63, 2nd paragraph of the Criminal Code concern the general interest of society and repress the disturbance caused by a delictual abstention to law and order, they also aim to protect private interests;203
One noteworthy consequence of the civil liability arising from article 63 C. pén., is that members of the medical profession are only allowed to strike in non-emergency cases.204
It should be mentioned that a new Criminal Code has come into force in 1994. The former article 63 C. pén. is now incorporated in a larger section, which increases the penalties imposed by article 63 C. pén. and introduces a further affirmative duty to assist in the case of a natural disaster, even when there is no immediate danger to any person.205
The decisive breakthrough towards a complete recognition of liability for nonfeasance came with the aforementioned Branly-case. Technically, the Branly-case was only concerned with an abstention dans l’action and not with abstentions pures et simples. The Cour de Cassation however, probably strongly influenced by the works of the brothers Mazeaud,206 did not draw such distinction:
(…) fault within the articles 1382 and 1383 C. civ. can consist of an abstention as well as a positive act; (…) an abstention gives rise to liability of the person responsible – even if there was no malice and intention to harm – when the act that was omitted should have been done either by virtue of a legal, statutory or contractual obligations, or by virtue of the demand for objective information within a profession, particularly when a historian is concerned.207
If we were to interpret this judgment to the letter, it would not cause a big change in the area of liability for nonfeasance.208 In the majority of the subsequent caselaw, however, the courts have proceeded on the presumption that the Branly-case represents a breach with the distinction between liability for feasance and for nonfeasance.209 At present, legal theory appears to have adopted a similar position. Whenever they think it appropriate the courts should be able to impose legal duties, whether affirmative or negative.210
In practice however, the courts appear to adopt a rather cautious approach towards imposition of affirmative duties.211 In particular, the personal characteristics of the defendant are taken into account. Thus, the courts are reluctant to impose affirmative duties on children.212 Also, the courts are not likely to hold a defendant liable, who did not understand that his intervention was necessary or who was physically or mentally unable to render the required assistance.213 If, on the other hand, the nonfeasance was malicious and harm was intended, caselaw and legal theory agree that liability should arise.214
Finally, it should be noted that the Code civil contains express provisions imposing liability for damage caused by the persons for whom one is responsible,215 by the things one has in one’s control,216 by the animals one owns,217 as well as for damage caused by the collapse of one’s buildings.218
Central to the later medieval law of torts were the action of trespass and the action upon the case.219 An action of trespass would almost certainly not have been available in cases of nonfeasance. For by analogy with the Roman requirement that damage be done corpore suo,220 the defendant’s behaviour was only considered a ‘trespass’ if it had directly harmed the plaintiff.221 This requirement of ‘directness’ is illustrated by Fortescue J in Reynolds v Clarke:
If a man throws a log into the highway, and in that act it hits me; I may maintain trespass, because it is an immediate wrong; but if as it lies there I tumble over it, and receive an injury, I must bring an action on the case; because it is only prejudicial in consequence, for which originally I could have no action at all.222
Because in cases of nonfeasance the harm is caused indirectly, the action of trespass would not have lain.
The action upon the case on the other hand, would typically have lain in cases of indirect or ‘consequential’ loss. In 1762, Sir John Comyns, Chief Baron of the Exchequer, included in his Digest a title for the ‘Action upon the Case for Negligence’. ‘Negligence’, rather unlike its wide modern connotations, was understood to mean ‘neglect’ or ‘omission’.223 An action upon the case based on such an omission would have lain for ‘negligence in a man’s trust’ or ‘duty’,224 for negligence in a man’s office,225 for a man’s ‘neglect in doing that, which by law he ought to do’, for his ‘neglect to do that, which he has undertaken’, for negligence in taking care of one’s animals and for negligently keeping a fire.226 As we will see in the remaining sections of this chapter, many similar categories still feature in the modern law.
In the course of the nineteenth century the division between the action of trespass and the action upon the case was overshadowed by the development of a new tort: the tort of negligence. ‘Negligence’ was no longer understood as merely referring to omissions, but came to cover cases of both acts and inaction.227 In Blyth v Birmingham Waterworks Company,228 Baron Alderson defined negligence as follows:
Negligence is the omission to do something which a reasonable man, guided upon those considerations which ordinarily regulate the conduct of human affairs, would do, or doing something which a prudent and reasonable man would not do.229
On the face of it, this definition does not indicate that there was any difference in the law’s approach towards feasance and nonfeasance; apparently both types of behaviour were measured against the standard of the reasonable man. However, as Pollock was careful to point out, Baron Alderson’s definition must have been based on the presumption that the party whose conduct was in question, was already under a ‘duty of taking care’.230 And it was at this ‘duty’-stage of the trial, that the judicial approach in cases of nonfeasance differed from that in cases of feasance. For acts and their results, Pollock explained, the actor was generally speaking held answerable. For mere omission he was not, unless he was under some specific duty.231 An example of the judicial reluctance to impose affirmative duties can be found in Gantret v Egerton. The defendants owned land which was intersected by a canal and cuttings, and bridges ‘leading to certain docks of the defendants’.232 Leon Gantret, who was lawfully passing over and using this land and these bridges, fell into one of the cuttings and drowned. The plaintiff claimed that the defendants had wrongfully and improperly kept and maintained their land, canal, cuttings and bridges, and had allowed them to continue to be in a condition which rendered them unsafe for persons lawfully passing along. Herschell, for the plaintiff, argued that the defendants would have to prove that they did not owe a duty of care towards the plaintiff. At this point, according to the Bar Reports, Willes J interrupted Herschell:
There is no duty to do anything, but there is a duty to abstain from doing anything that would injure.233
In the Law Reports, where this case appears under the name Gautret v Egerton, he is said to have added the following illustration:
No action will lie against a spiteful man who, seeing another running into a position of danger, merely omits the warning.234
Because the plaintiffs declaration did not reveal the breach of a specific affirmative duty, it was held that the plaintiff had no cause of action. As we will see in subsection V:6 (2), below, Gantret/Gautret v Egerton is no longer good law.235
More than a hundred and thirty years have passed since Willes J gave his decision in Gantret/Gautret v Egerton. In that time it has been generally assumed that English law persisted in its reluctance to impose liability for nonfeasance. In the words of Markesinis: ‘it is a well-known fact that the common law, first for historical reasons and later on philosophical grounds, has taken a hostile view towards imposing tortious liability for pure omissions’.236 Nevertheless it proves surprisingly difficult to find binding authority for this no-liability rule.237
True, there have been occasional references to the rule in obiter dicta.238 Yet cases in which it is part of the ratio decidendi are scarce. One of the few exceptions, which for that reason features in almost every textbook as authority on the matter, is said to be Smith v Littlewoods, in which Lord Goff stated:
Why does the law not recognise a general duty of care to prevent others from suffering loss or damage caused by the deliberate wrongdoing of third parties? The fundamental reason is that the common law does not impose liability for what are called pure omissions.239
Although the no-liability rule was here clearly an important argument in Lord Goff’s speech, it must be doubted whether Smith v Littlewoods provides binding authority. Firstly, Lord Goff was the only one of their Lordships to refer to the doctrine of no liability for nonfeasance.240 Secondly, Lord Goff himself went on to express his doubts about the no-liability rule, saying that it ‘may therefore require one day to be reconsidered’.241
Encouraged by Lord Goff’s latter statement Markesinis expressed in a 1989 article the hope and expectation that the no-liability rule would soon belong to the past.242 Many other influential scholars too, have argued in favour of abandoning the rigid distinction between cases of feasance and of nonfeasance.243 However, in its recent decision in Stovin v Wise244 the House of Lords made clear that such development will not take place in the near future. In a powerful speech Lord Hoffmann not only confirmed the principle of no liability for nonfeasance but went on to defend it. The most relevant part of this speech was quoted in section III:1 of this thesis. Lord Goff and Lord Jauncey agreed with Lord Hoffmann. The minority of their Lordships, consisting of Lord Nicholls and Lord Slynn, also adhered to the general no-liability rule. After stating that the distinction between liability for acts and liability for omissions is ‘fundamentally sound’,245 Lord Nicholls said:
The recognised legal position is that the bystander does not owe the drowning child or the heedless pedestrian a duty to take steps to save him. Something more is required than being a bystander. There must be some additional reason why it is fair and reasonable that one person should be regarded as his brother’s keeper and have legal obligations in that regard. When this additional reason exists, there is said to be sufficient proximity. That is the customary label.246
On the facts of the case, Lord Nicholls and Lord Slynn were of the opinion that there were specific circumstances that gave rise to a duty of care.
Thus, Stovin v Wise provided new and ample authority for the no-liability rule. Only if we are under a specific duty to act, will we be held liable for our failure to act. The crucial question becomes therefore: ‘when do such specific duties arise?’ Part of the answer was given by Lord Hoffmann in Stovin v Wise:
There may be a duty to act if one has undertaken to do so or induced a person to rely upon doing so. Or the ownership or occupation of land may give rise to a duty to take positive steps for the benefit of those who come upon the land and sometimes for the benefit of neighbours.247
In the following sections some specific circumstances that may give rise to affirmative duties will be assessed and investigated. It should at the outset be noted that the categories of circumstances that are distinguished in these sections sometimes overlap. As we will see, in many cases it is the combination of circumstances that has given rise to the imposition of an affirmative duty.
When the defendant has created a risk, however blamelessly, and then fails to control it, the courts tend to qualify the case as one of feasance, rather than nonfeasance. In chapter II) we criticised this tendency and discussed an example of a cardriver who failed to brake in time to avoid hitting a pedestrian. For the reasons given there, we decided to treat his failure to brake as an instance of ‘nonfeasance’, rather than, as Fleming argues, ‘the element that makes his active conduct – driving – negligent’.248 Accordingly, we regarded the stoker who fell asleep at the furnace as an example of someone who would incur liability for nonfeasance in Roman law.249
Most English courts however, would regard both the cardriver’s failure to brake and the stoker’s failure to control the fire as active. Thus, by qualifying the defendant’s behaviour as active rather than passive, the courts allow themselves to impose liability without having to address the issue of liability for nonfeasance. This technique is not unlike that of the French courts, when they distinguish ‘omissions dans l’action‘ from the ‘real’ nonfeasance, the ‘omissions pures et simples‘. It would perhaps be more realistic for both the French and the English courts to accept that the cardriver’s and the stoker’s conduct are instances of ‘real’ nonfeasance, albeit instances that fall under an exception to the no-liability rule.250
There is no doubt that under English law, as under French and Roman law, creating a risk gives rise to an affirmative duty to take reasonable steps to prevent the risk from culminating in injury.251 Authority for this proposition can be derived from a host of cases.252 In Musgrove v Pandelis253 the defendant’s servant started the engine of the defendant’s car. For some unexplained reason the carburettor exploded and the petrol caught fire. If the servant had closed the petrol tap immediately the fire would have burnt out harmlessly. Although the servant could not be blamed for the explosion that ignited the petrol, he was held to have come under a duty to take reasonable care in preventing the fire from spreading. The defendant was held vicariously liable.
Although no express reference was made to the doctrine of liability for nonfeasance, we may conclude that creating a risk, no matter how blamelessly, gives rise to a duty to prevent the risk from culminating in harm. On the same principle a man who has made a representation that was initially true, but has become false due to a change in circumstances, is under a duty to disclose this change in circumstances:254 he has created a risk and is under a duty to prevent this risk from culminating in harm. Application of this principle does not, in general, give rise to a duty to safeguard one’s belongings against abuse by third parties. Thus, if I leave my car unattended with the keys still in the ignition and a thief drives off with it and kills a cyclist, I am not liable, even though it may rightly be argued that I increased the risk of such an accident happening.255 Only if there are additional circumstances, such as a special relationship with the plaintiff or with the third party, will I be liable for the deliberate wrongdoing of that third party.256
A slightly different, but obviously related issue is whether someone whose blameless behaviour has not just created a risk, but already caused harm to occur, owes the harmed person a duty to minimize the consequences of this harm. The answer is not all that obvious. When in 1903 a railroad engine, without negligence of its crew, cut of an arm and a leg of a trespasser, the Supreme Court of Kansas refused to impose liability on the railway company for its failure to render aid.257 Twenty-seven years later, the Supreme Court of Alabama came to a similar conclusion in the case of Turbeville v Mobile Light & R. Co.258 The defendant company operated a street railroad. Without negligence on the part of the defendant or its servants, one of the defendant’s street cars ran into the plaintiff. The plaintiff got caught under or against the street car and ‘the safety guard’ attached to this car fell on the plaintiff. The defendant’s servants could easily have reversed the car or lifted the safety guard off the plaintiff, but failed to do so. The plaintiff was suffocated by the weight of the safety guard. Brown J. considered:
(…) we know of no principle of municipal law that imposed on the defendants or its servants a legal duty to relieve him by backing the car off his body or lifting it up so as to relieve his peril. It certainly could not be said that such duty rested on a mere bystander who witnessed his misfortune and who was in no way legally responsible for his predicament. If this be so, what reason can be found in the law to say that the defendant or its servants were under a duty to act, when they were guilty of no wrong or negligence in producing or bringing about the unfortunate situation.259
During the second half of this century American law appears to have given way to the strong criticism260 these and similar decisions gave rise to. It is now generally recognized that a duty to act arises where the defendant’s prior conduct, albeit innocent, has caused an injury.261
Under English law, too, it must be presumed that he who innocently causes an injury owes a duty to minimize the consequences of this injury.262
Thus, both in cases in which the defendant created a risk and in cases in which the defendant blamelessly caused an injury, English law allows an exception to the rule of no liability for nonfeasance. A possible justification for allowing an exception in these cases is that the ‘why pick on me?’-argument is easily met: the defendant is not merely one of a large number of bystanders. Moreover, there is a more obvious causal connection between the defendant’s behaviour and the damage than in most cases involving nonfeasance.
As I am walking over a bridge a woman falls into the water. I hear her scream for help and decide to come to the rescue. I throw her a rope, which she grabs and holds on to.263 In English law I was under no duty to come to the woman’s rescue; I was free to ignore her.264 Now that I have thrown the rope and thus embarked on the rescue, the question arises whether my legal position has changed. Am I still allowed to change my mind about the rescue – and let go of the rope – or have I come under a duty to carry it through? And if I have come under a duty to carry through my rescue attempt, when exactly did this duty arise? Was it when I undertook the rescue or was it when the woman in the water swam towards the rope and grabbed it, thus acting in reliance upon my undertaking?
When one analyses the caselaw it is not always clear what, in the courts’ understanding, amounts to an ‘undertaking’. Not only do the courts use different terms – usually either ‘undertaking’ or ‘assumption of responsibility’265 – but the ways the courts interpret the concept varies considerably. One can roughly distinguish two types of interpretation. Sometimes the courts mean that the defendant has expressly or implicitly made a promise. At other times, the judges have interpreted the term to mean no more than that the defendant has voluntarily acted.266 A good example of the first type of undertaking, ‘the implied promise’, is by Lord Devlin in Hedley Byrne & Co. Ltd v Heller & Partners Ltd:267
(…) the categories of special relationships which may give rise to a duty to take care in word as well as in deed are not limited to contractual relationships or to relationships of fiduciary duty, but include also relationships which in the words of Lord Shaw in Nocton v Lord Ashburton are ‘equivalent to contract’, that is, where there is an assumption of responsibility in circumstances in which, but for the absence of consideration, there would be a contract. Where there is an express undertaking, an express warranty as distinct from a mere representation, there can be little difficulty. The difficulty arises in discerning cases in which the undertaking is to be implied.268 (Footnotes omitted)
Although Hedley Byrne was not strictly speaking a case of nonfeasance, we may safely assume that its principle can equally apply in cases involving nonfeasance. This is confirmed by Lord Goff in Henderson v Merrett Syndicates Ltd:269
I wish to add in parenthesis that (…) an assumption of responsibility by, for example, a professional man may give rise to liability in respect of negligent omissions as much as negligent acts of commission, as for example when a solicitor assumes responsibility for business on behalf of his client and omits to take a certain step, such as the service of a document, which falls within the responsibility so assumed by him.270
A case that can be said to have involved an implied promise is Barnett v Chelsea and Kensington Hospital.271 There, a hospital came under a duty to treat three men suffering from poisoning because, amongst other reasons, they had been allowed to enter the hospital and their complaints were passed on to the medical casualty officer.272
It is more difficult to establish on what basis liability was imposed in White v Jones.273 The defendants – solicitors – had failed to change a will in accordance with the testator’s instructions. As a result the testator’s daughters – plaintiffs – received a considerably smaller sum than their father had intended them to have. In one view, the defendants’ implied promise gave rise to a duty towards the plaintiffs. However, it is difficult to see how the defendants’ contractual promise to the testator could have incorporated an implied promise to the plaintiffs in the absence of direct contact between plaintiffs and defendants. Lord Goff was clearly aware of this difficulty:
(…) there is great difficulty in holding, on ordinary principles, that the solicitor has assumed any responsibility towards an intended beneficiary under a will which he has undertaken to prepare on behalf of his client (…)274
Nevertheless Lord Goff found in favour of allowing the plaintiffs’ claim. Of ‘cardinal importance’ to this decision was a peculiarity which Lord Goff regarded as a ‘lacuna in the law’: if the plaintiffs were not allowed a claim in negligence, there would have been no effective remedy for the defendants’ negligence. The only person who had a valid claim on contract – the (executor of the) estate – had suffered no loss:275
(…) it seems to me that it is open to your Lordships’ House (…) to fashion a remedy to fill a lacuna in the law to solve the problem of transferred loss in the case before them. (…) In my opinion, therefore, your Lordships’ House should in cases such as these extend to the intended beneficiary a remedy under the Hedley Byrne principle by holding that the assumption of responsibility by the solicitor towards his client should be held in law to extend to the intended beneficiary (…)276
Thus, Lord Goff was careful to make clear that his decision in White v Jones was instigated by the exceptional circumstances of the case. Considerably different in this respect was Lord Browne-Wilkinson’s speech in the same case. His Lordship considered that in previous cases special relationships had been found to exist where the defendant had voluntarily assumed to act in the matter by involving himself in the plaintiff’s affairs or by choosing to speak.277 By analogy, he argued, the solicitor came into a special relationship with those intended to benefit under the will when he accepted the instructions to draw the will.278 Rather than exceptional, Lord Brown-Wilkinson appears to have regarded this decision as a straightforward application of an overriding principle:
If the responsibility for the task is assumed by the defendant he thereby creates a special relationship between himself and the plaintiff in relation to which the law (not the defendant) attaches a duty to carry out carefully the task so assumed.279
Whereas Lord Goff used the phrase ‘assumption of responsibility’ in its meaning of ‘implied promise’, Lord Browne-Wilkinson speaks of ‘assuming the task’ which appears to mean ‘choosing to act’.280 On one view an undertaking in this sense does not generally give rise to liability in tort. This view appears to have been supported by Stuart-Smith LJ in Capital and Counties v Hampshire CC,281 where he said:
It is not clear why a rescuer who is not under an obligation to attempt a rescue should assume a duty to be careful in effecting the rescue merely by undertaking the attempt.282
On another view, however, an attempt to rescue can give rise to an affirmative duty of care. In Barrett v Ministry of Defence283 a naval airman died after he became so drunk that he passed out into a coma and suffocated on his own vomit. The Queen’s Regulations for the Royal Navy 1967 stated that all officers owed a duty to discourage drunkenness and in the event of alcohol abuse ‘to take appropriate action to prevent any likely breaches of discipline, possible injury or fatality, including medical assistance if … available’.284 The deceased’s widow sued the Ministry of Defence. Beldam LJ held that the provisions in the Queen’s Regulations did not give rise to a duty towards individual sailors because:
The purpose of Queen’s Regulations and standing orders is to preserve good order and discipline in the service and to ensure that personnel remain fit for duty and while on duty obey commands and off duty do not misbehave bringing the service into disrepute.285
However, after the deceased had collapsed, one of the officers had arranged for him to be put on a stretcher and to be taken to his cabin. On the ground that by doing so the officer had voluntarily undertaken responsibility, Beldam LJ held the Ministry of Defence liable:
Until he collapsed I would hold that the deceased was in law alone responsible for his condition. Thereafter, when the appellant assumed responsibility for him, it accepts that the measures taken fell short of the standard reasonably to be expected.286
Despite the fact that the undertaking was not accompanied by reliance, liability was imposed. According to Stuart-Smith LJ Barrett v Ministry of Defence is an exceptional case. Because the deceased was under command of the officer, he argued, there was already a sufficiently close relationship of proximity to give rise to a duty of care.287 This view may seem convincing, but fails to explain why in Barrett v Ministry of Defence no duty of care arose until the officer undertook the rescue. Before this undertaking, Beldam LJ’s words are clear in this respect, ‘the deceased was in law alone responsible for his condition’.288
In the recent child-care case of Barrett v Enfield London BC,289 too, an undertaking in the sense here discussed appears to have been a relevant factor in establishing the existence of a duty of care. Distinguishing Barrett v Enfield London BC from X v Bedfordshire CC,290 Lord Slynn stressed that in the case before him the child in question had already been taken into local authority care.291
When dealing with reliance, one usually distinguishes detrimental and non-detrimental reliance.292
Let us first consider a case of so-called detrimental reliance, by assuming that the young lady to whom I threw the rope has an alternative option that might save her life. She could swim to a sandbank in the middle of the river, although this sandbank will flood during high tide. Relying on my willingness to carry through my rescue attempt, she chooses for (what she thinks is) the safer option and holds on to the rope I threw her. I let go of the rope and the woman, whose firm grasp of the rope has taken its toll on her physical resources, immediately drowns. It should not be doubted that my undertaking coupled with the woman’s detrimental reliance gives rise to a duty of care.293 More doubtful, however, is the scope of this duty of care.294
In one view my duty is merely to make sure that my undertaking, whether initially made in good faith or not, does not end up putting the woman in a worse position than she was before I threw her the rope. I could perform such a duty by dragging her by the rope to the middle of the bridge, thus enabling her to make it to the sandbank after all. She is then in the same position as she was before I threw her the rope, and even if she drowns as the high tide comes in I am not liable. To use Bagshaw’s terminology: liability arises in respect of ‘the detrimental reliance itself’, not ‘the expectation which prompted the detrimental reliance’.295 If this is the law, the doctrine of ‘detrimental reliance’ is in essence nothing but an analogous application of the rule set out in section V:3 (above); by throwing a rope I increased the risk that the woman would not make it to safety on her own strength. All I would need to do to avoid liability is to neutralise the detriment that followed from the woman’s reliance, thus preventing the additional risk I created from culminating in harm.296 This view appears to be supported by Schroeder JA in a Canadian case called The ‘Ogopogo’ where he refers to the principle that ‘even if a person embarks upon a rescue and does not carry it through, he is not under any liability to the person to whose aid he had come so long as discontinuance of his efforts did not leave the other in a worse condition than when he took charge’.297
On another view, however, an undertaking coupled with detrimental reliance gives rise to a wider duty to perform the task that was undertaken. If this is the law, I am not allowed to put the woman back in the same position as she was before I threw her the rope; I will have to take reasonable care in carrying through my rescue attempt. If I do not, I may be held liable for the woman’s whole loss, including the benefits she expected to receive from my undertaking. This is the view that appears to have been adopted by the Court of Appeal in London Borough of Bromley v Ellis,298 although the case is not entirely clear in this respect. An insurance agent had undertaken to arrange the transfer of an insurance to a Mr Ellis free of charge. In the belief that he was thus insured, Mr Ellis refrained from taking out new insurance. He became involved in a car accident and after he found out that he had not been insured after all, he brought a claim for indemnity against the insurance agent. The Court of Appeal held that even though there was no contractual relationship between Mr Ellis and the insurance agent, the latter’s undertaking gave rise to a duty to arrange the transfer with reasonable care.299 If this duty were merely to take care that his undertaking did not end up putting Mr Ellis in a worse position than he was before the undertaking was made, one would have expected the average charge of an insurance agent to be a reducing factor in the assessment of damages. However, nothing in the case report indicates that such reduction was made. Bagshaw argues that a similar view – albeit with regard to detriment rather than detrimental reliance – is adopted in Capital and Counties v Hampshire CC:
(…) the Court of Appeal held that where the fire brigade had caused additional harm, or increased the risk of harm, it should be treated as having caused the damage represented by the difference between the plaintiff’s actual position after the fire, and the position the plaintiff would have been in if the fire brigade had fought the fire in a reasonably competent manner. This seems incorrect, because it means that the damage the fire brigade is said to have negligently caused (for instance by switching off the sprinkler) is more than the damage which it was its duty to take care not to cause.300 (Footnotes omitted)
It must however be noted that Stuart-Smith LJ’s judgment is not quite as clear in this respect as Bagshaw implies it is. True, Stuart-Smith LJ does initially appear to say that the fire brigade should be liable for the whole damage, taking into account the expectation that resulted from the fire brigade’s undertaking.301 Yet he later qualifies this by saying that the fire brigade is not liable for the damage insofar as it can prove that it would have occurred ‘in any event, even if they had never come upon the scene’.302
The predominant view would appear to be that in general an undertaking coupled with detrimental reliance merely gives rise to a duty to make sure that this undertaking does not end up leaving the victim worse off.303
The second type of reliance, non-detrimental reliance, must be regarded as of little legal importance.304 With a small adjustment our example of the drowning woman can again serve as an illustration; if we assume that the woman would never have been able to swim to safety,305 her holding on to the rope I threw her is not a detrimental act of reliance. In fact it is debatable whether her holding on to the rope should altogether be regarded as an act of reliance. After all, we can hardly speak of ‘reliance’ when the person concerned was not offered a real choice; she could either drown immediately or grab hold of the rope for as long as I let her. The more appropriate term would therefore be ‘dependence’ which, as we learn from the parable of the Good Samaritan, is not regarded as a ground to impose liability. No matter how much the victim depended on the Levite, the latter would not have had a duty to aid him in English law. When we speak of non-detrimental reliance we refer to nothing but an individual’s state of mind which, if not accompanied by some sort of tangible behaviour, should not affect legal relations.306
In recent caselaw attention has been drawn to a third type of reliance, the concept of so-called ‘general reliance’.307 The clearest exposition of this type of reliance is by Mason J in the High Court of Australia, where he stated that:
(…) there will be cases in which the plaintiff’s reasonable reliance will arise out of a general dependence on an authority’s performance of its function with due care, without the need for contributing conduct on the part of a defendant or action to his detriment on the part of a plaintiff. Reliance or dependence in this sense is in general the product of the grant (and exercise) of powers designed to prevent or minimize a risk of personal injury or disability, recognized by the legislature as being of such magnitude or complexity that individuals cannot, or may not, take adequate steps for their own protection. This situation generates on one side (the individual) a general expectation that the power will be exercised and on the other side (the authority) a realization that there is a general reliance or dependence on its exercise of power (…)308
‘General reliance’ appears to provide an argument in favour of the imposition of a duty of care on a class of defendants, rather than on any specific defendant. It is a factor in the decision whether it would be fair, just and reasonable to impose a duty of care. The concept of general reliance refers to the reliance of society as a whole, rather than the reliance of an individual plaintiff. Even if the plaintiff himself did not expect the defendant to act, he can invoke the doctrine of general reliance.309 Although it is not immediately apparent whether general reliance is reliance of the detrimental or the non-detrimental kind, it is likely to possess some detrimental characteristics. For in most (if not all) situations in which society as a whole has relied on the defendant to fulfill a task undertaken, society (or elements within it) would have performed this task if only it (they) had known that the defendant was not going to do so.
To illustrate the types of cases in which this policy argument could play an important role, Mason J mentioned the control of air traffic, the safety inspection of aircraft and the fighting of a fire in a building by a fire authority.310 Although the doctrine of general reliance would typically apply in cases in which the defendant is a public authority, there is no obvious reason why the doctrine could not also apply in cases that do not involve public authorities. Indeed, Lord Browne-Wilkinson’s speech in White v Jones contains a fine example of such application:
Although in any particular case it may not be possible to demonstrate that the intended beneficiary relied upon the solicitor, society as a whole does rely on solicitors to carry out their will making functions carefully.311
It is as yet uncertain if, and if so to what extent, the doctrine of general reliance will be of practical importance in cases before the English courts. In Stovin v Wise Lord Hoffmann stressed that if the doctrine was to be accepted, its application would require ‘some very careful analysis of the role which the expected exercise of the statutory power plays in community behaviour’.312 Addressing one of the examples given by Mason J, Lord Hoffmann indicated that in cases involving the fire brigade this careful analysis might lead the courts to decide that liability ought not to be imposed:313
For example, in one sense it is true that the fire brigade is there to protect people in situations in which they could not be expected to be able to protect themselves. On the other hand, they can and do protect themselves by insurance against the risk of fire.314
Soon after Lord Hoffmann gave his speech in Stovin v Wise, a case involving fire brigades, Capital and Counties v Hampshire CC,315 came before the Court of Appeal. The Court of Appeal was not convinced by arguments based on general reliance or more generally on the fire brigade’s undertaking. It was decided that a fire brigade did not owe a duty to extinguish a fire, regardless of whether or not it had accepted an emergency call. Moreover, even if it did attend the fire its only duty was not to make the situation worse.316 Similar views have since been expressed in a case involving the police317 and one involving the coastguard.318
When it comes to ambulance services, however, the Court of Appeal has recently adopted a rather different position. In Kent v London Ambulance Service an ambulance service which had accepted a 999 call turned up excessively late. Kennedy LJ held that an ambulance service could be under a duty once it had undertaken to come to the rescue by accepting an emergency call.319 It appears to have been an important consideration that the public generally – and detrimentally – relies on an ambulance to turn up once a call has been accepted:
(…) if the ambulance service undertakes to attend the person who has been promised that assistance, and those acting on his or her behalf, normally abandon the search for other possible means of transporting the patient to hospital in a situation in which other means of transport may readily be available.320 (Emphasis added)
All in all, it has become rather difficult to predict whether the courts will impose a duty of care on public authorities, and with that whether the doctrine of general reliance will gain more judicial support in the future.321 As Brooke LJ said in Musgrave v The Chief Constable of Hampshire:
This is an extremely difficult area of the law. There are a number of cases which go either way and in the last resort it is a matter for the judge after establishing the facts to decide, if the necessary proximity exists between the parties, (and? JSK) whether it is just, fair and reasonable to impose the duty complained of.322
In summary, the presence of an undertaking – especially if accompanied by detrimental reliance – may give rise to an affirmative duty and thus to an exception to the rule of no liability for nonfeasance. There are several justifications for allowing such an exception in these cases. Firstly, the ‘why pick on me?’-argument is easily met: by his undertaking the defendant has rendered himself easily identifiable. Secondly, there is a more obvious causal connection between the defendant’s conduct and the damage than in most cases of nonfeasance, especially in those cases involving detrimental reliance. Thirdly, although the imposition of liability may still be regarded as an interference with an individual’s freedom, the defendant has had the choice to avoid this interference by refusing to undertake responsibility.323
On the other hand, the ‘culpability-argument’ seems to oppose at least as strongly against imposing liability as it does in all other cases of nonfeasance. For if I undertake but fail to rescue am I not less culpable than if I do not even try in the first place?
As we have seen in section IV:2 of this thesis, German law is less reluctant to impose liability for nonfeasance in cases where the defendant had special relationship with the plaintiff. Thus, we observed, ‘although the Levite may not be required to render assistance to everyone, he does owe a duty to aid his close relatives, his spouse or his close friends’.324 Whether the same is true for English law is not entirely clear. Cases involving relatives suing each other are rare.325 Most English scholars accept that parenthood may give rise to a duty to protect or help one’s children.326 A similar view appears to have been adopted by the Court of Appeal in Surtees v Royal Borough of Kingston upon Thames in which Stocker LJ considered the duty owed by a foster parent to her foster child:
I have no doubt that stated in general terms the duty imposed on Mrs. Hughes was to take such care as in all the circumstances was reasonable to ensure that the plaintiff was not exposed to unnecessary risk of injury, the standard of care being that of a careful parent in the prevailing circumstances.327
Stocker LJ went on to say that when establishing the scope of this parental duty, the courts should take account of the domestic circumstances of the individual parent.328 The Vice-Chancellor emphasised the point:
We should be slow to characterise as negligent the care which ordinary loving and careful mothers are able to give to individual children, given the rough and tumble of home life.329
Given the scarcity of authority on the matter, we can but speculate about the scope of the parental duty. It probably does not incorporate a duty to insure one’s children against the risk of injury.330
Short of attempting to give a complete list of all the other special relationships that may give rise to a duty to protect, a few examples will suffice for present purposes. A school may be liable for releasing its pupils before closing time331 or more generally for failing to take reasonable care of the safety of the person and property of each pupil.332 An employer must not only secure the safety in the workplace but also look after an employee who is injured or falls ill at work.333 Similarly, prison authorities have a duty to provide medical care to their prisoners.334 Finally, the close relationship between police officers gives rise to ‘a duty to comply with a specific or acknowledged police duty where failure to do so will expose a fellow officer to unnecessary risk of injury’. Thus, the Chief Constable of Northumbria was held vicariously liable when one of his police inspectors failed to render assistance to a female police constable who was attacked by a prisoner.335
There are several possible reasons why the courts allow exceptions to the no-liability rule in cases involving a defendant who was in a special position of responsibility towards the plaintiff. Firstly, owing to his special position the defendant is easily identified, hence easily ‘picked on’. Secondly, in most of the above cases the defendant has entered into a special position of responsibility by his own choice; parents usually choose to have children, teachers choose to become teachers, etc. It can be argued that by making this choice one voluntarily accepts a limitation of one’s liberty.336 Thirdly, most of the above protective duties, though not the parental duty, arise in the course of the defendant’s profession. Through the defendant’s salary, it may be argued, he is indirectly rewarded for the performance of his protective duty.
One can roughly distinguish two types of situations in which the defendant can be said to be in a position of control over a potential source of damage. The defendant may have control over potentially harmful conduct of others or over land.
As a general rule, one is not responsible for injury or loss caused by the conduct of others.337 There are however exceptions to this rule, as we learn from a judgment of Dixon J. in the High Court of Australia:
The general rule is that one man is under no duty of controlling another man to prevent his doing damage to a third. There are, however, special relations which are the source of a duty of this nature. It appears now to be recognized that it is incumbent upon a parent who maintains control over a young child to take reasonable care so to exercise that control as to avoid conduct on his part exposing the person or property of others to unreasonable danger. Parental control, where it exists, must be exercised with due care to prevent the child inflicting intentional damage on others or causing damage by conduct involving unreasonable risk of injury to others.338 (Footnotes omitted)
Not just the parent-child relationship,339 but also most of the other relationships mentioned in the previous section give rise to a duty to control. Thus, in addition to the duty to protect its pupils against harm, a school is under a duty to prevent its pupils from causing harm to others.340 Prison authorities may be liable for the damage caused by escaping prisoners.341 Similarly, a mental hospital may be liable for injury caused by one of its patients while he was on authorised leave.342
The possible justifications for allowing an exception to the no-liability rule in cases in which the defendant is in a position of control over the conduct of others are the same as those mentioned in the previous section: the defendant is easily ‘picked on’, the defendant has usually entered into a position of responsibility by his own choice and in some cases he is rewarded for the performance of his affirmative duty.
A fortiori a duty may be imposed in cases where the defendant is both in a position of responsibility towards the plaintiff and in a position of control over the wrongdoer.343 Thus, an employer is under a duty to prevent his employees from causing injury to their fellow workmen.344 Likewise, prison authorities may be liable for failing to prevent prisoners from injuring each other345 or even from injuring themselves.346 Hospitals owe a similar duty to a patient who is a known suicide risk.347
It is often said that ownership of land is a source of obligations as well as of rights.348 Indeed, there are several areas of the law which impose affirmative duties on landowners.
Firstly, a claim may arise in negligence against an occupier who fails to neutralise a potential danger that occurs on his land. In Goldman v Hargrave349 a tall redgum tree on the defendant’s land was hit by lightening and caught fire. Instead of extinguishing the fire by putting water on it, the defendant decided to let the tree burn out itself. Five days later a fresh gust of wind revived the fire, which spread to the plaintiff’s land and caused extensive damage. The Judicial Committee of the Privy Council decided in favour of imposing liability. In his speech Lord Wilberforce stated that an occupier’s duty was ‘one of a more positive character than merely to abstain from creating, or adding to, a source of danger or annoyance’.350 The occupier was under a ‘measured duty of care’ to ‘remove or reduce hazards to neighbours’.351 Lord Wilberforce, however, did not rest there and added an important observation with regard to the standard of care required from such occupiers:
(…) the standard ought to be to require of the occupier what is reasonable to expect of him in his individual circumstances. Thus, less must be expected of the infirm than of the able bodied (…)352
Thus, using the words of Atiyah, Lord Wilberforce ‘subjectivised’ the standard of care for cases in which the occupier did not create the original danger.353
The case of Smith v Littlewoods Ltd354 also involved a fire which spread to neighbouring land, but differed from Goldman v Hargrave in that the fire was caused by vandals setting fire to a deserted cinema rather than by a natural event. With respect to the question whether the defendants should have secured their cinema against vandals, the two main speeches, Lord Mackay’s and Lord Goff’s, displayed some considerable differences in approach. Lord Mackay made no mention of the no-liability rule for nonfeasance and dismissed doctrinal objections against liability for the act of another with considerable ease.355 Instead, he focused his attention on the requirement of foreseeability. Because human conduct is particularly unpredictable, he argued, the requirement of foreseeability will not readily be fulfilled:
In summary I conclude, in agreement with both counsel, that what the reasonable man is bound to foresee in a case involving injury or damage by independent human agency, just as in cases where such agency plays no part, is the probable consequences of his own act or omission, but that, in such a case, a clear basis will be required on which to assert that the injury or damage is more than a mere possibility.356
Lord Mackay came to the conclusion that the Court of Appeal had failed to establish that the vandals’ act of arson was not a probable consequence of leaving the cinema unattended and unlocked.
Lord Goff expressly rejected Lord Mackay’s use of the ‘mechanism of foreseeability’.357 His Lordship commenced his speech by stressing the principle that in general one is not liable for the act of another; a principle which he said was founded on the rule of no liability for omission.358 However, in exceptional circumstances a duty could be imposed to guard against the wrongdoing of others. Lord Goff mentioned some specific relationships that could give rise to such a duty before touching on a more general circumstance:
(…) an occupier who negligently causes or permits a source of danger to be created on his land, and can reasonably foresee that third parties may trespass on his land and, interfering with the source of danger, may spark it off, thereby causing damage to the person or property of those in the vicinity, should be held liable to such a person for damage so caused to him.359
Since the cinema could not properly be described as a source of danger in the nature of a fire hazard, no liability ought to arise.360
Lord Goff added that once an occupier knew that a third party had created a fire on his premises, he ought to take reasonable steps to prevent such fire from damaging neighbouring property.361 Again, however, the evidence of the case showed no such knowledge on the part of the defendant.
The law of nuisance, too, may create liability for a failure to act in relations between adjoining landowners. In general a person who unlawfully interferes with another’s use or enjoyment of land is liable in nuisance.362 However, the occupier of the premises from which the nuisance originates may be liable even if the nuisance was not created by himself or by someone for whom he is responsible. It was implied in the Court of Appeal’s decision in St Anne’s Well Brewery Co. v Roberts that an occupier is liable for damages caused to neighbouring land by the collapse of a wall, if he knew or ought to have known that his predecessor’s excavations had rendered the wall unstable.363 Similarly, in Sedleigh-Denfield v O’Callaghan364 an occupier of land was held liable for a nuisance which had been created by a trespasser on his land. The trespasser had placed a pipe in a ditch on the defendant’s land in order to carry off rain water. During a heavy rainstorm the pipe became choked with leaves, so that the water overflowed onto the plaintiff’s land and caused damage. The House of Lords found for the plaintiff because even though the defendant had not created the nuisance, he was aware or ought to have been aware of its existence. Both Viscount Maugham and Lord Wright adopted a statement of law in Salmond’s The Law of Torts:365
When a nuisance has been created by the act of a trespasser, or otherwise without the act, authority, or permission of the occupier, the occupier is not responsible for that nuisance unless, with knowledge or means of knowledge of its existence, he suffers it to continue without taking reasonably prompt and efficient means for its abatement.366
In Leakey v National Trust Megaw LJ quoted the very same passage and applied its principles to a case in which the nuisance originated in a natural occurrence.367 Echoing a statement by Lord Wilberforce in Goldman v Hargrave, the Court of Appeal recognised ‘a general duty on occupiers in relation to hazards occurring on their land, whether natural or man-made’.368 Moreover, Lord Wilberforce’s ‘subjectivised’ standard of care was incorporated into the law of nuisance.369 Hence in nuisance as in negligence, the financial and physical resources available to the particular occupier are taken into account when defining the standard of care.
Thirdly, there are statutory provisions that create affirmative duties. The Occupiers Liability Act 1957 imposes a general duty on occupiers of land to take care to see that visitors and their belongings are reasonably safe.370 The Occupiers Liability Act 1984 creates an additional duty towards persons other than visitors in respect of risk of their suffering injury on the premises.371 Although both acts contain a reference to ‘the circumstances of the case’,372 there is no specific mention of the physical and financial resources of the occupier. Hence it remains unclear whether the standard of care is objective or ‘subjectivised’.373 ‘The foundation of occupier’s liability is occupational control, i.e., control associated with and arising from presence in and use of or activity in the premises’.374 Thus, the dangers against which the occupier is required to guard are not necessarily dangers arising from a defect in the land itself, but can equally arise from the use of the land. In Cunningham & Others v Reading Football Club Ltd375 the defendant club was held liable under the 1957 Act for failing to prevent hooligans from injuring police officers by throwing missiles. These missiles included, besides the ‘usual’ sharpened coins, pieces of concrete that had been broken off the ‘east terrace’. According to Drake J. a ‘reasonably prudent and careful occupier’ would not have left the concrete in the bad state in which it was, especially in a situation in which it was probable that the concrete would be broken up and used as missiles.376 The Highways Act 1980 imposes a comparable duty on the highway authority to maintain the highway.377
The justification for creating an exception to the rule of no liability in cases of occupiers of land is twofold. Firstly, the occupier is in the best position to recognise and remove the risk, hence easily ‘picked on’. Secondly, it has been argued that the ownership of land should entail responsibilities as well as rights.378
In this thesis the legal systems of Germany, France and England have been investigated in an attempt to assess whether, and if so to what extent, they impose liability for nonfeasance. If we allow ourselves to reflect upon the results of this investigation, we will find that there are many similarities in the way the different systems have approached the issue.
In the beginning of this century both German and French courts, like their English counterparts, were reluctant to impose liability for nonfeasance in the absence of a specific duty to act. The courts in all three countries recognised specific duties to act, though only on a case-by-case basis. The second world war appears to have provided the metaphorical junction at which the French system went its own way. In both Germany and France, the legislators introduced a duty to rescue in criminal law. The German courts refused to translate this duty into a duty to rescue under private law and appear to have continued to create affirmative duties on an incremental basis. Conversely, the French courts were quick to recognise a duty to rescue under private law. Eventually, this led to a complete breach with the feasance-nonfeasance distinction after the Branly-case in 1951.
Although German and English law appear to have continued to coincide, in that the distinction between feasance and nonfeasance was upheld and duties to act were created incrementally, we observed some important differences. Probably the most noteworthy difference was that, with an open category of Verkehrspflichten, German courts were less restrictive in recognising new affirmative duties than their English counterparts. Another difference, although of more importance in theory than in practice, was the liability for intentionally inflicted harm (par. 826 BGB), which also applied to cases of nonfeasance.
Returning to the parable of the Good Samaritan, we can conclude that the priest and the Levite would have had nothing to fear from English law. In Germany they would have been liable under criminal law and arguably under par. 826 BGB if it could have been proved that they ‘passed by on the other side’ with the intent to harm. Under French law they would also have incurred criminal liability, while the victim would have had a claim for damages in negligence.
The question we need to ask ourselves, is whether English law should start moving carefully towards a breach with the rigid feasance-nonfeasance distinction. The mere fact that, as we observed in chapter III of this thesis, there are very few (if any) justifications for making the distinction, ought to be a sufficiently strong argument in favour of abandoning it. Our present legal regime aims to compensate loss at the expense of the person whose negligent behaviour caused it. Or, as Lord Hutton put it in the recent case of Darker v Chief Constable of West Midlands, ‘the predominant requirement of public policy is that those who suffer a wrong should have a right to a remedy’.379 If we accept our present regime, the exclusion of one particular type of wrong – omissions – from this regime needs to be justified. Moreover, it has been argued that the law should reinforce the rules of our shared morality wherever it is not oppressive, impracticable or pointless.380 Of course, we need not accept that the moral obligation to aid other people should necessarily be translated into a legal duty. However, we cannot ignore the sense of unease caused by the fact that by upholding the feasance-nonfeasance distinction the law not only fails to encourage altruistic behaviour, but in effect discourages it.
We observed that both German and French law are cautious not to set the standard of care too high in cases involving nonfeasance. In the absence of special circumstances, the German courts are reluctant to impose liability for mere negligent nonfeasance. Yet, in cases in which the defendant intended to inflict harm, liability will more readily arise. The French courts take the personal characteristics of the defendant into account when defining the standard of care in cases of nonfeasance. Thus, we found that the French courts are not likely to hold someone liable who does not understand that his intervention is necessary, nor someone who is physically or mentally unable to render aid. A similar approach could be discerned in some cases before the English courts. In Surtees v Royal Borough of Kingston upon Thames381 it was held that a mother owes her children a duty to protect them, but that the standard of care depends on the domestic circumstances of the individual parent.382 From Goldman v Hargrave we learned that an occupier of land is under a ‘measured duty of care’ to remove or reduce hazards to neighbours. Again, the individual circumstances of the occupier were taken into account when defining the standard of care.383 In contrast, in cases of feasance the financial resources and physical and mental abilities of the defendant are usually ignored.384
The reason behind this difference in approach is a theory which we addressed in section III:2 of this thesis, and which was summarised by Lord Nicholls:
Liability for omissions gives rise to a problem not present with liability for careless acts. He who wishes to act must act carefully or not at all. (…) With liability for omissions, however, a person is not offered a choice. The law compels him to act when left to himself he might do nothing.
This gives rise to a difficulty if positive action requires expenditure. The law requires him to act reasonably. But, as Lord Wilberforce observed in Goldman v Hargrave, what is reasonable to one man may be very unreasonable or ruinous to another. The solution adopted is to have regard to the circumstances of the individual. He must act as would a reasonable person in his position. The standard of reasonableness is to be measured by what may reasonably be expected of the defendant in his individual circumstances. Where action calls for expenditure, the court if necessary will have regard to the financial resources of the defendant.385 (Footnotes omitted)
As we observed in section III:2, this version of the ‘freedom of the will’ does not necessarily provide a good argument in favour of excluding cases of nonfeasance from the imposition of liability. The main objection against the argument was that it draws a distinction of degree, not of principle. Imposing positive duties does not leave us without an option to avoid being subjected to these duties, but rather with less of an option than is commonly the case in respect of negative duties. However, this does not mean that we need completely ignore the point. In most cases involving misfeasance we can be reasonably expected to anticipate being subjected to a negative duty of care. Thus, if we start a building company we know we will be expected to implement some safety measures. Whether or not we can afford these measures is a factor that will be ignored by the courts, because if we do not have the necessary financial resources I should not start a company. However, in some cases of nonfeasance this possibility to anticipate being subjected to a duty of care is absent. For example, if I walk over a bridge and a woman falls into the water, the fact that I cannot swim or that I am deaf and did not hear her screams for help cannot reasonably be held against me. In those cases in which the possibility to anticipate being subjected to an affirmative duty is absent, there is good reason to take the individual circumstances of the defendant into regard.
However, the absence of the possibility to anticipate being subjected to an affirmative duty is not a characteristic of all cases of nonfeasance. In some cases of professional undertaking, for example, the defendant knew the duty of care required of people within his profession, and could have anticipated it. Thus, there is no reason to apply a different standard of care in cases like White v Jones. Indeed, it is debatable whether the very case that Lord Nicholls referred to, Goldman v Hargrave, was a case in which the possibility of anticipation was absent. Salmond appears to have been of the opinion that landowners could anticipate being subjected to a duty of care:
(…) an occupier is liable even for a continuing nuisance which already existed on the premises when he first entered into possession of them; for it is his duty either to refrain from taking possession of such premises or else to abate the nuisance so soon as he becomes the occupier.386
If a landowner does not have the financial resources to neutralise hazards that may occur on his land, should we not expect him to consider selling part of his land or indeed all of it?387 It is here submitted that the application of a subjective standard of care is appropriate only in cases in which the defendant could not reasonably be expected to anticipate being subjected to an affirmative duty.
Let us now return to the example with which we started this thesis. The priest and the Levite could not reasonably be expected to have anticipated their encounter with a wounded man in need of their help. This does not prevent us from imposing a duty on them to render assistance to the wounded man. However, it is here submitted that in establishing whether their behaviour was reasonable, their individual circumstances ought to be taken into account. If, for example, they both had a bad back and no ‘beast’ on which to put the man, they could be forgiven for their refusal to carry the man. Yet this does not mean the law should not oblige them to ‘make the best of a difficult situation’ and attempt to obtain assistance from a third party.388
|AC||Law Reports, Appeal Cases|
|ALJR||Australian Law Journal Reports|
|All ER||All England Law Reports|
|Am.J.Comp.L.||American Journal of Comparative Law|
|art. artt.||article, articles|
|BGH||Entscheidungen des Bundesgerichtshof in Zivilsachen|
|BMLR||Butterworth’s Medico-Legal Reports|
|C. civ.||Code Civil|
|Ch||Law Reports, Chancery Division|
|CILSA||The Comparative and International Law Journal of Southern Africa|
|CLR||Commonwealth Law Reports|
|Coll.||Mosaicarum et Romanarum legum collatio|
|C. pén.||Code pénal|
|D.||Digesta (Justinian’s Digest)|
|D.H.||Receuil hebdomadaire de jurisprudence Dalloz|
|ed., eds||editor, editors|
|e.g.||exempli gratia (for example)|
|Esp.||Espinasse, Nisi Prius Reports|
|ff.||and following pages|
|Fort.||Fortescue’s Reports, King’s Bench|
|G.||Gai Institutiones (Gaius’ Institutes)|
|I.||Iustiniani Institutiones (Justinian’s Institutes)|
|JCP||Juris-classeur périodique (La Semaine Juridique)|
|J.Leg.Stud.||The Journal of Legal Studies|
|JSK||Jeroen Sebastiaan Kortmann|
|KB||Law Reports, King’s Bench Division|
|Ld Ray.||Lord Raymond’s Reports, King’s Bench and Common Pleas|
|Lloyd’s Rep||Lloyd’s Law Reports|
|Lloyd’s Rep Med||Lloyd’s Law Reports Medical|
|LMCLQ||Lloyd’s Maritime and Commercial Law Quarterly|
|LQR||Law Quarterly Review|
|LR||Law Review, Law Reports|
|LR CP||Law Reports, Common Pleas Cases|
|LT||Law Times Reports|
|MLR||Modern Law Review|
|MR||Master of the Rolls|
|NE||North Eastern Reporter|
|NJW||Neue Juristische Wochenschrift|
|NJW-RR||Neue Juristische Wochenschrift, Rechtsprechungs-Report Zivilrecht|
|nr, nrs||number, numbers|
|NZLR||New Zealand Law Reports|
|p., pp.||page, pages|
|Pac.||Pacific Reporter (American Law Reports)|
|PIQR||Personal Injuries and Quantum Reports|
|Plc.||Public limited company|
|QB||Law Reports, Queen’s Bench Division|
|Rev.trim.dr.civ.||Revue trimestrielle de droit civil|
|RGZ||Entscheidungen des Reichsgerichts in Zivilsachen|
|s., ss.||section, sections|
|So.||Southern Reporter (American Law Reports)|
|Str.||Strange’s Reports, King’s Bench|
|TLJ||Torts Law Journal|
|Tort L Rev||The Tort Law Review|
|T.S.F.||Télégraphie sans fil|
|U.Pa.LR||University of Pennsylvania Law Review|
|WLR||Weekly Law Reports|
— C.K. Allen, ‘Legal Duties’ (1931) 40 Yale LJ 331-380, reprinted in C.K. Allen, Legal Duties and Other Essays in Jurisprudence (1931), pp. 156-220
— D.K. Allen, Misrepresentation (1988)
— J. Austin, Lectures on Jurisprudence, Vol. II (1863)
— R. Bagshaw, ‘Proximity’s Siren Song: Negligence Liability of Fire Brigades in England’, (1998) 6 Tort L Rev 109-113
— R. Bagshaw, ‘The duties of care of emergency service providers’  LMCLQ 71-92.
— C. von Bar, Verkehrspflichten: richterliche Gefahrensteuerungsgebote im deutschen Deliktsrecht (1980)
— C. von Bar, The Common European Law of Torts, Vol. 1 (1998)
— K. Barker, ‘Unreliable Assumptions in the Modern Law of Negligence’ (1993) 109 LQR 461-484
— J.L. Barton, ‘The Lex Aquilia and Decretal Actions’ in A. Watson (ed.), Daube Noster (1974), pp. 15-25
— J. Bell, S. Boyron and S. Whittaker, Principles of French Law (1998)
— J. Bentham, An Introduction to the Principles of Morals and Legislation (J.H. Burns and H.L.A. Hart, eds 1970)
— I. Berlin, ‘Two Concepts of Liberty’ (1958), Inaugural Lecture delivered before the University of Oxford, reprinted in I. Berlin, Four Essays on Liberty (1969)
— P.B.H. Birks, ‘Doing and causing to be done’ in A.D.E. Lewis and D.J. Ibbetson, The Roman Law Tradition (1994), pp. 32-54
— K.E. Black, ‘The role of law in suicide prevention, beyond civil commitment. A bystander’s duty to report suicide threats’ (1987) Stanford LR 929
— F.H. Bohlen, ‘The Basis of Affirmative Obligations in the Law of Tort’  44 Am. L. Reg. (N.S.), 209
— F.H. Bohlen, ‘The Moral duty to aid Others as a Basis of Liability’  56 U.Pa.LR 217
— F.H. Bohlen, Studies in the Law of Torts (1926)
— M.R. Brazier and others (eds), Clerk & Lindsell on Torts (17th edn 1997)
— M.R. Brazier and J. Murphy, Street on Torts (10th edn, 1999)
— A. Brinz, Lehrbuch der Pandekten, zweiter Band, erste Abteilung (2nd edn 1879)
— W.W. Buckland, ‘Diligens Paterfamilias’ in E. Albertario, P. Ciapessoni and P. De Fransisci, Studi in Onore di Pietro Bonfante (1930), pp. 85-108
— G. Calabresi, ‘Some Thoughts On Risk Distribution And The Law Of Torts’ (1961) 70 Yale LJ 499.
— P. Cane, ‘Negligent Solicitors and Disappointed Beneficiaries’ (1980) 96 LQR 182-185
— P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999)
— P. Cane, The Anatomy of Tort Law (1997)
— P. Catala and J.A. Weir, ‘Delict and Torts: a Study in Parallel’, (1962-1963) 37 Tulane LR 573-620
— J.H. Crabb, The French Civil Code (rev. edn 1995)
— C.C. van Dam, Aansprakelijkheid voor nalaten, Preadvies voor de Nederlandse Vereniging voor Rechtsvergelijking (1995)
— J.P. Dawson, ‘Negotiorum Gestio: the Altruistic Intermeddler’, (1961) 74 Harvard LR 817-865 and 1073-1129
— N. Dejean de la Bâtie, Aubry et Rau, Droit Civil Français, Vol. VI-2 (8th edn 1989)
— A-F Delvilani, La Responsabilité Civile pour Omission ou Abstention (1978)
— A.T Denning, The Family Story (1981)
— H. Dernburg, Pandekten, zweiter Band (7th edn 1903)
— E. Deutsch, Unerlaubte Handlungen, Schadensersatz und Schmerzensgeld (3rd edn 1995)
— E. Deutsch, Haftungsrecht, Vol. 1 (1976)
— J. Domat, Les loix civiles dans leur ordre naturel (rev. edn 1745)
— W. Dütz, ‘Zur privatrechtlichen Bedeutung unterlassener Hilfeleistung (§ 330c StGB)’, (1970) 23 NJW 1822-1826
— L.H. Eldredge, Modern Tort Problems (1941)
— L. Enneccerus and H.C. Nipperdey, Allgemeiner Teil des Bürgerlichen Rechts, Vol. 2 (15th edn 1960)
— R.A. Epstein, ‘A Theory of Strict Liability’, (1973) 2 J.Leg.Stud. 151
— J. Esser/E. Schmidt, Schuldrecht, Vol. I (6th edn, 1984)
— J. Feinberg, ‘The Moral and Legal Responsibility of the Bad Samaritan’, (1985) 23 Archiv für Rechts- und Sozialphilosophie 216-228
— F.J.M. Feldbrugge, ‘Good and Bad Samaritans’ (1966) 14 Am.J.Comp.L. 630-657.
— D.A. Fischer, ‘Causation in Fact in Omission Cases’  Utah LR 1335
— Fishkin, The limits of obligation (1982)
— J.G. Fleming, The Law of Torts (9th edn 1998)
— T.C. Galligan, ‘Aiding and Altruism: a Mythopsycholegal Analysis’, (1994) 27 Univ. of Michigan Journal of Law Reform 439-520
— J-L. Gazzaniga, Introduction historique au droit des obligations (1992)
— J. Glover, Causing Death and Saving Lives (1977)
— S.L. Goren, The German Civil Code (Rev. edn 1994)
— N. Gray and J. Edelman, ‘Developing the law of omissions: a common law duty to rescue?’ (1998) 6 TLJ 240-259.
— Ch. O. Gregory, ‘Gratuitous Undertakings and the Duty of Care’, 1 De Paul LR 30 (1951)
— H. Grotius, De Iure Belli ac Pacis (1631)
— K. Haager, BGB-RGRK (11th edn 1960)
— J.W. Harris, Legal Philosophies (2nd edn 1997)
— H.L.A. Hart and A.M. Honoré, Causation in the Law (2nd edn 1985)
— J. Hartshorne, ‘The liability in negligence of the fire service – the Court of Appeal decides’, (1997) 13 PN 53-58
— H-J Hoffmann, Die Abstufung der Fahrlässigkeit in der Rechtsgeschichte (1986)
— A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991)
— A.M. Honoré, ‘Law, Morals and Rescue’ in J.M. Ratcliffe (ed.), The Good Samaritan and the Law (1966), pp. 225 ff., reprinted and expanded in A.M. Honoré, Making Law Bind, Essays Legal and Philosophical (1987), pp. 256 ff.
— D. Howarth, Textbook on Tort (1995)
— R. Hübner, Grundzüge des Deutschen Privatrechts (5th edn 1930)
— D.J. Ibbetson, A Historical Introduction to the Law of Obligations (1999)
— M.A. Jones, Textbook on Torts (6th edn 1998)
— I. Kant, The Metaphysics of Morals, translated by M. Gregor (1991)
— W. Page Keeton (ed.), Prosser and Keeton on The Law of Torts (5th edn 1984)
— K.J. Kemp, ‘The duty to rescue – compulsion or laissez faire?’, (1985) XVIII CILSA 163-186
— K.J. Kemp, Delictual Liability for Omissions (1978) (Unpublished thesis, University of Port Elizabeth, South Africa)
— J. Kleinig, ‘Good Samaritanism’, Philosophy & Public Affairs 5 (1976) 328
— J. Kohler, ‘Die Menschenhülfe im Privatrecht’, (1887) 25 Jahrbücher für die Dogmatik des heutigen römischen und deutschen Privatrechts 1-141
— H. Kötz, Deliktsrecht (4th edn 1988)
— E. Krupp, ‘Kameradschaftspflichten und Rechtspflichten’, (1938) 8 DR 465
— W. Kunkel, ‘Diligentia’ (1925) 45 ZSS 266-351
— W.M. Landes and R.A. Posner, ‘Salvors, Finders, Good Samaritans, and Other Rescuers: an Economic Study of Law and Altruism’, (1978) 7 J.Leg.Stud. 83-128
— W.M. Landes and R.A. Posner, ‘Altruism in Law and Economics’, (1978) Am. Econ. Rev., K. Larenz/C-H. Canaris, Lehrbuch des Schuldrechts, Vol. II/2 (13th edn 1994)
— F.H. Lawson, Negligence in the Civil Law (1950)
— F.H. Lawson and B.S. Markesinis, Tortious Liability for Unintentional Harm in the Common Law and the Civil Law, Vols I & II (1982)
— S. Levmore, ‘Waiting for Rescue: An Essay on the Evolution and Incentive Structure of the Law of Affirmative Obligations, (1986) 72 Virginia LR 879-941
— A.M. Linden, ‘Rescuers and good samaritans’ (1971) MLR 241
— N. MacCormick, ‘Voluntary Obligations and Normative Powers I’ (1972) 46 PASS 59-78.
— B.S. Markesinis, ‘Negligence, Nuisance and Affirmative Duties of Action’ (1989) 105 LQR 104
— B.S. Markesinis, W. Lorenz and G. Dannemann, The German Law of Obligations, Vol. I (1997)
— B.S. Markesinis, The German Law of Obligations, Vol. II (3rd edn 1997)
— B.S. Markesinis and S.F. Deakin, Tort Law (4th edn 1999)
— Antonius Matthaeus, De Criminibus (1644)
— H. Mazeaud and L. Mazeaud, Traité Théorique et Pratique de la Responsabilité Civile, Vol. 1 (4th edn 1947)
— H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (6th edn 1965)
— J.S. Mill, On Liberty (G. Himmelfarb, ed., 1974)
— M.A. Millner, Negligence in Modern Law (1967)
— L. Mitteis, Römisches Privatrecht bis auf die Zeit Diocletians, Vol. I (1908)
— Th. Mommsen, P. Krueger and A. Watson, The Digest of Justinian, Vol. I (1985)
— W. Münzberg, Verhalten und Erfolg als Grundlagen der Rechtswidrigkeit und Haftung (1966)
— A. Pernice, Labeo – Römisches Privatrecht im ersten Jahrhunderte der Kaiserzeit, zweiter Band, erste Abteilung (2nd edn, 1895)
— A. Pernice, Zur Lehre von den Sachbeschädigungen nach Römischem Rechte (1867)
— F.T. Piggott, Principles of the Law of Torts (1885)
— F. Pollock, The Law Of Torts (1887)
— R.A. Posner, ‘Strict Liability: a Comment’, (1973) 2 J.Leg.Stud. 205-221
— R.A. Posner, Economic Analysis of Law (4th edn 1992)
— W.L. Prosser, The Law of Torts (2nd edn 1955)
— W.L. Prosser, Selected Topics On the Law of Torts (1953)
— S. von Pufendorf, Elementorum Jurisprudentiae Universalis (edn of 1672)
— S. von Pufendorf, De Officio Hominis et Civilis (edition of 1682) Translation: F.G. Moore (1927).
— S. von Pufendorf, De Jure Naturae et Gentium (edn of 1688)
— A. Pyle (ed.), Liberty: Contemporary Responses to John Stuart Mill (1994)
— J.M. Ratcliffe, (ed.), The Good Samaritan and the Law (1966)
— D. Reckseen, ‘The duty to rescue’ (1972) Indiana LJ 321
— D.A.J. Richards, ‘Kantian Ethics and The Harm Principle: a Reply to John Finnis’ (1987) 87 Columbia LR 457-471
— F.M. de Robertis, La Responsabilita’ Contrattuale nel Sistema della Grande Compilazione, Vol. I (1983)
— W.V.H. Rogers, Winfield and Jolowicz on Tort (15th edn, 1998)
— A.W. Rudzinski ‘On rescuers obligation to another; the Ogopogo lands in the Supreme Court of Canada’ 22 UTLJ 98
— J. Salmond, The Law of Torts (5th edn 1920)
— H.C.F. Schoordijk, De verbintenis uit en de verplichting tot altruisme (1987)
— W.A. Seavey, ‘Reliance on Gratuitous Promises or Other Conduct’  64 Harvard LR 913
— M.S. Shapo, The duty to act (1977)
— C.H. Sieburgh, ‘Faute als grond voor aansprakelijkheid’, Groninger Opmerkingen en Mededelingen 1998, pp. 68-90
— J.C. Smith, Liability in Negligence (1984)
— J. Smith and B. Hogan, Criminal Law (8th edn. 1996)
— J.C. Smith and P. Burns, ‘Donoghue v. Stevenson – The Not So Golden Anniversary’ (1983) 46 MLR 147
— M. Spence, Protecting Reliance: The Emergent Doctrine of Equitable Estoppel (1999)
— G. Spencer Bower and A.K. Turner, The Law of Actionable Misrepresentation (3rd edn 1974)
— J. Stapleton, ‘The Normal Expectancies Measure in Tort Damages’ (1997) 113 LQR 257-293
— J. Stapleton, Product Liability (1994)
— J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985)
— J.F. Stephen, Liberty, Equality, Fraternity (1873)
— H. Stoll, Vertrag und Unrecht, Vol. 2 (3rd edn 1943)
— C.L. Ten, Mill On Liberty (1980)
— F. Terré, P. Simler and Y. Lequette, Droit civil; Les Obligations (6th edn 1996)
— J.A.C. Thomas, Textbook of Roman Law (1976)
— P. le Tourneau and L. Cadiet, Droit de la Responsabilité (1996)
— A. Tunc (ed.), International Encyclopedia of Comparative Law, Vol. XI Torts (1983)
— G. Viney, Traité de Droit Civil; Introduction à la Responsabilité (2nd edn 1995)
— G. Viney, Traité de Droit Civil, Vol. IV (1982)
— J. Voet, Commentarius ad Pandectas, Vol. I (6th edn 1731)
— W. Weimar, ‘Besteht eine Schadensersatzpflicht eines Krankenhauses bei Abweisung Schwerverletzter?’ JR 1975, 145
— Weimar, ‘Schadensersatzpflicht bei Schädigungen durch Unterlassen’, (1937) 7 DR 77
— E.J. Weinrib, ‘Law as a Kantian Idea of Reason’ (1987) 87 Columbia LR 472-508
— E.J. Weinrib, ‘Omissions and responsibility’ (1980) 30 Phil.Q. 1
— E.J. Weinrib, ‘The Case for a Duty to Rescue’, (1980) 90 Yale LJ 247-293
— H. Welzel, ‘Zur Dogmatik der echten Unterlassungsdelikte, insbesondere des § 330c StGB’, NJW 1953, 327
— B. Windscheid, Lehrbuch des Pandektenrechts, zweiter Band (9th edn 1906)
— C.A. Wright (1941) 19 Canadian Bar Review 465
— R. Zimmermann, The Law of Obligations (1990)
— Alexandrou v Oxford  4 All ER 328
— Anns v Merton London Borough  AC 728
— Baker v T.E. Hopkins & Son Ltd  3 All ER 225
— Barnes v Hampshire CC  1 WLR 1563
— Barnett v Chelsea and Kensington Hospital  1 QB 428
— Barrett v Enfield London BC  3 All ER 193
— Barrett v Ministry of Defence  3 All ER 87
— Blyth v Birmingham Waterworks Company (1856) 11 Ex. 781
— Bolton v Stone  1 All ER 1078
— British Railways Board v Herrington  AC 877
— Briess v Woolley  AC 333
— Bulman v Furness Rly Company (1875) 32 LT 430
— Capital and Counties v Hampshire CC  2 All ER 865
— Carmarthenshire CC v Lewis  AC 549
— Cherry Ltd v Allied Insurance Brokers Ltd  1 Lloyd’s Rep 274
— Costello v Chief Constable of the Northumbria Police  1 All ER 550
— Cunningham & Others v Reading Football Club Ltd  PIQR P141
— Darker v Chief Constable of West Midlands  3 WLR 747
— Donoghue v Stevenson  AC 562
— East Suffolk Rivers Catchments v Kent  4 All ER 527
— Ellis v Home Office  2 All ER 149
— Fisher v Ruislip-Northwood Urban DC and Middlesex CC  KB 584
— Gantret v Egerton (1867) 16 LT 17 reported under the name Gautret v Egerton in the Law Reports  LR 2 CP 371
— Goldman v Hargrave  2 All ER 989
— Hargrave v Goldman (1963) 110 CLR 40
— Hedley Byrne & Co. Ltd v Heller & Partners Ltd  AC 465
— Henderson v Merrett Syndicates Ltd  2 AC 145
— Hill v Chief Constable of West Yorkshire  AC 53
— Holgate v Lancashire Mental Hospitals Board  4 All ER 19
— Home Office v Dorset Yacht Co. Ltd [ AC 1004
— Hudson v Ridge Manufacturing Co. Ltd  2 QB 348
— Hughes v Lord Advocate  1 All ER 705
— Invercargill City Council v Hamlin  1 All ER 756
— Kasapis v Laimos Bros.  2 Lloyd’s Rep 378
— Kelly v Metropolitan Rly Co.  1 QB 944
— Kent v London Ambulance Service  Lloyd’s Rep Med 58, reported under the name Kent v Griffiths in (1999) 47 BLMR 125
— Kirkham v Chief Constable of Manchester  2 QB 283
— Knight v Home Office  3 All ER 237
— Latimer v AEC Ltd  AC 643
— Leakey v National Trust  QB 485
— Leigh and Sillivan Ltd v Aliakmon Shipping Co. Ltd  2 All ER 145
— London Borough of Bromley v Ellis  1 Lloyd’s Rep 97
— Marc Rich & Co v Bishop Rock Marine Co. Ltd  1 WLR 1071
— Midland Bank Trust Co. Ltd v Hett, Stubs & Kemp  Ch 384
— Morrison v Sheffield Corporation  2 KB 866
— Musgrove v Pandelis  2 KB 43
— Musgrave v The Chief Constable of Hampshire, Court of Appeal 10 February 1999 (unpublished; transcript by Smith Bernal)
— Nettleship v Weston  2 QB 691
— Newton v Edgerley  1 WLR 1031
— The ‘Ogopogo’  1 Lloyd’s Rep 257
— Ogwo v Taylor  3 All ER 961
— Oll v Secretary of State for Transport  3 All ER 897
— Perl Ltd v Camden London Borough Council  QB 342
— Philcox v Civil Aviation Authority, The Times 8 June 1995
— Pyrenees Shire Council v Day (1998) 72 ALJR 152
— Reeves v Commissioner of Police of the Metropolis  2 All ER 381
— Reynolds v Clarke (1725) 1 Str. 634, reported under the names Reynolds v Clerk in 8 Mod. 272 and Reynolds v Clark in Fort. 212
— Ross v Caunters  Ch 297
— Sedleigh-Denfield v O’Callaghan  AC 880
— Selfe v Ilford and District Hospital, The Times 26 November 1970
— Sheppard v Glossop Corporation  3 KB 132
— Smith v Leurs (1945) 70 CLR 256
— Smith v Littlewoods  AC 241
— St Anne’s Well Brewery Co. v Roberts (1929) 140 LT 1
— Stansbie v Troman  2 KB 48
— Stovin v Wise  3 All ER 801
— Sutherland Shire Council v Heyman (1985) 157 CLR 424
— Surtees v Royal Borough of Kingston upon Thames  PIQR P101
— Topp v London County Bus Co  3 All ER 448
— Turbeville v Mobile Light & R. Co. 221 Ala. 91, 127 So. 519 (1930)
— Union Pacific Rly Co. v Cappier 66 Kan. 649, 72 Pac. 281 (1903)
— Van Oppen v Bedford Charity Trustees  1 All ER 273
— Van Oppen v Bedford Charity Trustees  3 All ER 389
— Ward v Tesco Stores Ltd  1 All ER 219
— Weld-Blundell v Stephens  AC 956
— Wheat v E. Lacon & Co. Ltd  AC 552
— White v Jones  3 WLR 730
— White v Jones  2 AC 207
— Wilkinson v Coverdale (1793) 1 Esp. 75
— Wright v Lodge  4 All ER 299
— X v Bedfordshire CC  3 WLR 152
— Yuen Kun Yeu v A-G of Hong Kong  2 All ER 705
— RG 30 October 1902, RGZ 52, 373
— RG 23 February 1903, RGZ 54, 53
— RG 1 June 1937, RGZ 155, 257
— BGH 30 April 1953, BGHZ 9, 373
— BGH 16 December 1953, VersR 1954, 100
— BGH 15 June 1954, BGHZ 14, 83
— BGH 30 December 1954, BGHZ 16, 95
— BGH 25 February 1958, VersR 1958, 308
— BGH 5 November 1962, NJW 1963, 148
— BGH 19 May 1967, VersR 1967, 801
— OLG Köln, 2 November 1971, NJW 1972, 497
— BGH 28 January 1976, BGHZ 66, 51
— BGH 17 May 1982, NJW 1982, 2815
— OLG Frankfurt 27 October 1988, NJW-RR 1989/II, 794
— Cour de Cassation 24 December 1924, D.H. 1925, 120
— Cour de Cassation 8 April 1932, D.H. 1932, 297
— Tribunal Correctionnel d’Aix 27 March 1947,  D. 1947, 304
— Cour de Cassation 27 February 1951, D. 1951, 329
— Cour de Cassation 16 March 1972, D. 1972, 394
— Cour de Cassation 13 December 1972, D. 1973, 493
- Bürgerliches Gesetzbuch
- § 823 BGB
- s. (1)
- s. (2)
- § 826 BGB
- §§ 831-838 BGB
- § 823 BGB
- Civil Liability (Contribution) Act 1978
- s. 1(1)
- Code Civil
- art. 1382 C. civ
- art. 1383 C. civ
- art. 1384 C. civ
- art. 1385 C. civ
- art. 1386 C. civ
- Code pénal
- art. 63 C. pén
- D. 188.8.131.52
- D. 9.2.2.pr
- D. 184.108.40.206
- D. 9.2.8
- D. 220.127.116.11
- D. 18.104.22.168
- D. 22.214.171.124
- D. 126.96.36.199
- D. 9.2.31
- D. 188.8.131.52
- D. 9.2.45.pr
- D. 184.108.40.206
- D. 220.127.116.11
- D. 50.16.226
- Highways Act 1980
- s. 41(1)
- Iustiniani Institutiones
- I. IV.III.16
- Occupiers Liability Act 1957
- s. 1(3)(b)
- s. 2(2)
- Occupiers Liability Act 1984
- s. 1(1)
- s. 1(2)(a)
- s. 1(3)
- s. 1(3)(c)
- § 138 StGB
- § 323c StGB
1 The Holy Bible, Authorized King James Version
2  AC 562.
3 Ibid. at 580.
4  AC 1004.
5 Ibid. at 1060. A similar point was made by Windeyer J in Hargrave v Goldman (1963) 110 CLR 40 at 65-66
6 E.g. Home Office v Dorset Yacht Co. Ltd  AC 1004; Yuen Kun Yue v A-G of Hong Kong  2 All ER 705; Stovin v Wise  3 All ER 801 and Capital and Counties v Hampshire CC  2 All ER 865.
7 Contra: L.B. Curzon, Dictionary of Law (4th edn 1993), p. 262, where it is suggested that the term ‘nonfeasance’, like ‘omission’, presupposes the existence of a duty.
8 J. Austin, Lectures on Jurisprudence, Vol. II (1863), p. 101 (Lecture XX) and A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), pp. 33 and 36-37.
9 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 60.
10 J.G. Fleming, The Law of Torts (9th edn 1998), p. 163.
11 Ibid., p. 167.
12 Ibid., p. 171. Cf. D. Howarth, Textbook on Tort (1995), p. 180.
13 Example taken from H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (6th edn 1965), nr 525.
14 Cf. A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), pp. 43-44; P. Cane, ‘Negligent Solicitors and Disappointed Beneficiaries’ (1980) 96 LQR 182-185 at 184 and H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (6th edn 1965), nr 525.
15 Bolton v Stone  1 All ER 1078 at 1078, per Lord Porter, at 1083, per Lord Normand and at 1087, per Lord Radcliffe.
16 Cf. A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), pp. 39-41.
17 Ch. O. Gregory, ‘The Good Samaritan and the Bad’ in J.M. Ratcliffe (ed.), The Good Samaritan and the Law (1966), pp. 23-41 at p. 28.
18  3 All ER 801 at 819.
19 E. Lucas, ‘Mill on Liberty’ (1869) 13 NS The Dublin Review 62, reprinted in A. Pyle (ed.), Liberty (1994), pp. 255 ff. at p. 255.
20 J.S. Mill, On Liberty (G. Himmelfarb, ed., 1974), p. 64.
21 Cf. J. Kohler, ‘Die Menschenhülfe im Privatrecht’, (1887) 25 Jahrbücher für die Dogmatik des heutigen römischen und deutschen Privatrechts 1-141 at 2.
22 I. Berlin, ‘Two Concepts of Liberty’ (1958), Inaugural Lecture delivered before the University of Oxford, reprinted in I. Berlin, Four Essays on Liberty (1969), pp. 118 ff. at p. 122. I shall continue to adopt a ‘negative’ definition of freedom, as I regard this definition to be sufficiently accurate for the purposes of this thesis. For a description of the difference between the notions of ‘negative’ and ‘positive’ freedom see Berlin, ibid., pp. 122-134.
23 A.T Denning, The Family Story (1981), p. 179.
24 See for a more detailed investigation E.J. Weinrib, ‘The Case for a Duty to Rescue’, (1980) 90 Yale LJ 247-293.
25 Ibid. at 266-267 and N. Gray and J. Edelman, ‘Developing the law of omissions: a common law duty to rescue?’ (1998) 6 TLJ 240-259 at 1-2 with further references.
26 I. Kant, The Metaphysics of Morals, translated by M. Gregor (1991), p. 56.
27 E.J. Weinrib, ‘Law as a Kantian Idea of Reason’ (1987) 87 Columbia LR 472-508 at 489 and idem, ‘The Case for a Duty to Rescue’, (1980) 90 Yale LJ 247-293 at 266.
28 I. Kant, The Metaphysics of Morals, translated by M. Gregor (1991), p. 47.
29 D. Howarth, Textbook on Tort (1995), p. 184.
30 I. Kant, The Metaphysics of Morals, translated by M. Gregor (1991), p. 136.
31 Ibid. at 136-137.
32 E.J. Weinrib, ‘The Case for a Duty to Rescue’, (1980) 90 Yale LJ 247-293 at 266 and C.K. Allen, ‘Legal Duties’ (1931) 40 Yale LJ 331-380 at 368.
33 J. Bentham, An Introduction to the Principles of Morals and Legislation (J.H. Burns and H.L.A. Hart, eds 1970), p. 292. (First published in 1789)
34 Ibid., p. 293.
35 E.J. Weinrib, ‘The Case for a Duty to Rescue’, (1980) 90 Yale LJ 247-293 at 280.
36 J.S. Mill, On Liberty (1859).
37 J.S. Mill, On Liberty (G. Himmelfarb, ed., 1974), pp. 68-69.
38 See for a more detailed account of Mill’s harm principle C.L. Ten, Mill On Liberty (1980), pp. 52-67.
39 There are exceptions: ibid. pp. 40-41.
40 E.g. the review in (1859) 1635 The Athenaeum 281-282, reprinted in A. Pyle (ed.), Liberty (1994), pp. 1 ff. and J. Morley, ‘Mr. Mill’s Doctrine of Liberty’ (1873) 20 The Fortnightly Review 234-256, reprinted in A. Pyle (ed.), Liberty (1994), pp. 271 ff.
41 Most vehemently by E. Lucas in ‘Mill on Liberty’ (1869) 13 NS The Dublin Review 62, reprinted in A. Pyle (ed.), Liberty (1994), pp. 255 ff. and J.F. Stephen in Liberty, Equality, Fraternity (1873).
42 J.S. Mill, On Liberty (G. Himmelfarb, ed., 1974), pp. 167-168.
43 E.g. the review of On Liberty in (1859) 1635 The Athenaeum 282, reprinted in A. Pyle (ed.), Liberty (1994), pp. 1 ff. at pp. 4-5 and R.W. Church, ‘Mill on Liberty’  2 Bentley’s Quarterly Review 434-473, reprinted in A. Pyle (ed.), Liberty (1994), pp. 210 ff. at pp. 246-247. On the harmfulness of setting bad examples in another context see J.F. Stephen in Liberty, Equality, Fraternity (1873), p. 131 and especially the second edition of this work (1874), p. 144, footnote.
44 J. Morley, ‘Mr. Mill’s Doctrine of Liberty’ (1873) 20 The Fortnightly Review 234-256, reprinted in A. Pyle (ed.), Liberty (1994), pp. 271 ff. at pp. 293-294.
45 J.S. Mill, On Liberty (G. Himmelfarb, ed., 1974), pp. 145-149 and A. Pyle, ‘Introduction’ in A. Pyle (ed.), Liberty (1994), p. xvi.
46 Review of On Liberty in (1859) 1635 The Athenaeum, pp. 281-282, reprinted in A. Pyle (ed.), Liberty (1994), pp. 1 ff. at pp. 1-2.
47 D.A.J. Richards, ‘Kantian Ethics and The Harm Principle: a Reply to John Finnis’ (1987) 87 Columbia LR 457-471.
48 J.S. Mill, On Liberty (G. Himmelfarb, ed., 1974), p. 70.
49 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), pp. 62-63.
50 Ibid., p. 63.
51 Nettleship v Weston  2 QB 691.
52 B.S. Markesinis, The German Law of Obligations Volume II (3rd edn 1997), pp. 471-472.
53  AC 643.
54 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 63 and K.J. Kemp, ‘The duty to rescue – compulsion or laissez faire?’, (1985) XVIII CILSA 163-186 at 168.
55 ‘Quemadmodum autem hoc maxime nomine homo actionum suarum autor habetur, quia ipse voluntario eas suscepit: ita circa voluntatem id cum primis observandum, eius spontaneitatem utique asserendam esse saltem circa illas actiones, de quibus ab homine in foro humano ratio solet exigi’. S. von Pufendorf, De Officio Hominis et Civilis (edn of 1682), lib. I, cap. I, § 10. Translation: F.G. Moore (1927). Cf. also idem, Elementorum Jurisprudentiae Universalis (edn of 1672), lib. I, def. XVII, § 7 and more generally idem, De Jure Naturae et Gentium (edn of 1688), lib. I, cap. IV.
56 Stovin v Wise  3 All ER 801 at 809. Cf. K.J. Kemp, Delictual Liability for Omissions (1978) (Unpublished thesis, University of Port Elizabeth, South Africa), p. 386.
57 Cf. below, section V:3.
58 J.M. Ratcliffe in his introduction to J.M. Ratcliffe, (ed.), The Good Samaritan and the Law (1966), pp. IX-X and Ch. O. Gregory, ‘The Good Samaritan and the Bad’, in J.M. Ratcliffe, (ed.), The Good Samaritan and the Law (1966), pp. 23-41 at 33 with further references in footnotes 30 and 31.
59 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 63.
60 E.J. Weinrib, ‘The Case for a Duty to Rescue’, (1980) 90 Yale LJ 247 ff. at 262.
61 Of course the other requirements of liability in negligence would also have to be met. In the case of Kitty Genovese, there are indications that calling the police would not have prevented Kitty’s death, because the police would either have come too late or not at all. Hence, it is doubtful whether the failure to call the police caused Kitty’s death.
62 Civil Liability (Contribution) Act 1978, s. 1(1).
63 See for a comprehensive summary of the various economic theories on the law of torts D. Howarth, Textbook on Tort (1995), pp. 15 ff.
64 J. Stapleton, Product Liability (1994), p. 102. Cf. also A. Pigou, The Economics of Welfare (4th edn 1932), p. 183 and G. Calabresi, ‘Some Thoughts On Risk Distribution And The Law Of Torts’ (1961) 70 Yale LJ 499.
65 J. Stapleton, Product Liability (1994), pp. 108 ff. and pp. 118 ff. and P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), pp. 361 ff.
66 Stovin v Wise  3 All ER 801 at 819, quoted above in the opening of chapter III.
67 A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), pp. 31 ff.
68 Ibid., pp. 31-32.
69 A.M. Honoré, ‘Are Omissions Less Culpable?’, pp. 31 ff. (Quoted above)
70 R.A. Posner, ‘Strict Liability: a Comment’, (1973) 2 J.Leg.Stud. 205-221 at 218-219. Cf. J.G. Fleming, The Law of Torts (9th edn 1998), p. 164, footnote 102.
71 W.M. Landes and R.A. Posner, ‘Salvors, Finders, Good Samaritans, and Other Rescuers: an Economic Study of Law and Altruism’, (1978) 7 J.Leg.Stud. 83-128.
72 Ibid., p. 120.
73 R.A. Posner, Economic Analysis of Law (4th edn 1992), p. 190.
74 W.M. Landes and R.A. Posner, ‘Salvors, Finders, Good Samaritans, and Other Rescuers: an Economic Study of Law and Altruism’, (1978) 7 J.Leg.Stud. 124.
75 Ibid., p. 124. Cf. R.A. Posner, Economic Analysis of Law (4th edn 1992), p. 190.
76 Ibid. p. 124.
77 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 364.
78 Ibid., pp. 364-365.
79 Cf. for English law Baker v T.E. Hopkins & Son Ltd  3 All ER 225 and Ogwo v Taylor  3 All ER 961.
80 E.J. Weinrib, ‘The Case for a Duty to Rescue’, (1980) 90 Yale LJ 247 ff. at 261.
81 Ibid. at 262.
82 Fleming states that by failing to impose liability for nonfeasance in cases of easy rescue, the law ‘clearly abjures the economic standard of efficiency’: J.G. Fleming, The Law of Torts (9th edn 1998), p. 164, footnote 102.
83 Stovin v Wise  3 All ER 801 at 819, quoted above, section III:1.
84 This seems to be the main argument that Epstein relies on, when defending the no liability rule: R.A. Epstein, ‘A Theory of Strict Liability’, (1973) 2 J.Leg.Stud. 151-204, at 201.
85 F.H. Bohlen, ‘The Moral duty to aid Others as a Basis of Liability’  56 U.Pa.LR 217-244 and 316-338 at pp. 220-221; L.H. Eldredge, Modern Tort Problems (1941), p. 13 and J.G. Fleming, The Law of Torts (9th edn 1998), p. 163.
86 J. Stapleton, ‘The Normal Expectancies Measure in Tort Damages’ (1997) 113 LQR 257-293 at 281-282.
87 H.L.A. Hart and A.M. Honoré, Causation in the Law (2nd edn 1985), p. 448.
88 Ibid., p. 29.
89 A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), p. 41.
90 H.L.A. Hart and A.M. Honoré, Causation in the Law (2nd edn 1985), p. 139.
91 Ibid., p. 447.
92 Home Office v Dorset Yacht Co. Ltd  AC 1004.
93 A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), p. 41.
94  4 All ER 527.
95 Ibid. at 531.
96 Ibid. at 539.
97 Cf. Barnett v Chelsea & Kensington Hospital Management Committee  1 All ER 1068.
98 Cf. East Suffolk Rivers Catchment Board v Kent  4 All ER 527, 537 per Lord Atkin, dissenting. Lord Atkin’s speech was referred to with approval by Lord Salmon in Anns v Merton London Borough  AC 728 at 766-767.
99 Ibid. at 543, per Lord Romer and H.L.A. Hart and A.M. Honoré, Causation in the Law (2nd edn 1985), p. 140.
100 Capital and Counties v Hampshire CC  2 All ER 865.
101 A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), p. 41, quoted above.
102 J. Stapleton, Product Liability (1994), p. 123.
103 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), pp. 63-64.
104 A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), pp. 31 ff.
105 Ibid., p. 48.
106 Ibid., p. 33.
107 Ibid., p. 48.
108 See also P. Cane, The Anatomy of Tort Law (1997), pp. 63-64. Contra: K.J. Kemp, Delictual Liability for Omissions (1978) (Unpublished thesis, University of Port Elizabeth, South Africa), p. 615 and idem, ‘The duty to rescue – compulsion or laissez faire?’, (1985) XVIII CILSA 163-186 at 172.
109 D. 18.104.22.168 and Coll. 12.7.7
110 (…) ceterum si alius ignem subiecerit fornaci, alius neglegenter custodierit, an tenebitur qui subiecerit? nam qui custodit, nihil fecit, qui recte ignem subiecit, non peccavit: quid ergo est? puto utilem competere actionem tam in eum qui ad fornacem obdormivit quam in eum qui neglegenter custodit, nec quisquam dixerit in eo qui obdormivit rem eum humanam et naturalem passum, cum deberet vel ignem extinguere vel ita munire, ne evagetur. Translation: F.H. Lawson, Negligence in the Civil Law (1950), pp. 103-105.
111 A. Pernice, Zur Lehre von den Sachbeschädigungen nach Römischem Rechte (1867), pp. 171-172: Es wird stets nur danach gefragt, ob eine substantielle Beschädigung einerseits, eine Unachtsamkeit als deren Ursache anderseits vorliege. Wenn ein Stein vom Wagen fällt und ein Thier beschädigt, so kann das geschehen, weil die Steine schlecht geschichtet sind, oder weil das nothwendige Hinterbrett nicht vorgeschoben ist. Dort Handlung, hier Unterlassung. Darauf kömmt es aber nicht an, sondern auf die schlechte Packung, auf das Fehlen des nothwendigen Brettes. (My own translation)
112 H. Grotius, De Iure Belli ac Pacis (1625), lib. II, cap. XVII, I; C. von Bar, Verkehrspflichten: richterliche Gefahrensteuerungsgebote im deutschen Deliktsrecht (1980), pp. 7-8 and R. Zimmermann, The Law of Obligations (1990), pp. 1032 and 1044.
113 H. Dernburg, Pandekten, Vol. 2 (7th edn 1903), p. 361, footnote 11.
114 Windscheid relied heavily on D. 22.214.171.124 (another text by Ulpian), which seems to indicate that the Lex Aquilia does not apply to cases of nonfeasance; B. Windscheid, Lehrbuch des Pandektenrechts, Vol. 2 (9th edn 1906), p. 974, footnote 9. Cf. for a survey of the various interpretations of D. 126.96.36.199 K.J. Kemp, Delictual Liability for Omissions (1978) (Unpublished thesis, University of Port Elizabeth, South Africa), pp. 75-83.
115 Ibid. p. 974: (…) ein Unterlassung verpflichtet nur insofern, als das Tun durch eine vorhergehende oder begleitende Tätigkeit geboten war. (My own translation)
116 D. 9.2.8.
117 Cf. for a similar dispute Antonius Matthaeus, De Criminibus (1644), lib. 48, tit. 5, cap. 6, num. 5 (p. 560) and J. Voet, Commentarius ad Pandectas, Vol. I (6th edn 1731), p. 537 (ad legem Aquiliam, 3). See for a more detailed investigation; K.J. Kemp, Delictual Liability for Omissions (1978) (Unpublished thesis, University of Port Elizabeth, South Africa), pp. 70-74 and footnotes 44 and 45.
118 See also Pernice, who changed his view on the subject, as far as pre-Justinian law was concerned: A. Pernice, Labeo – Römisches Privatrecht im ersten Jahrhunderte der Kaiserzeit, Vol. 2, Part 1 (2nd edn, 1895), p. 128, footnote 1.
119 G. III, 219. Cf. I. IV, III, 16.
120 G. III, 219.
121 D. 188.8.131.52.
122 J. Limpens, R.M. Kruithof and A. Meinertzhagen-Limpens, ‘Liability for One’s Own Act’, in A. Tunc (ed.), International Encyclopedia of Comparative Law, Vol. XI Torts (1983), ch. 2, p. 36. An additional indication that the Lex Aquilia initially did not apply to cases of nonfeasance lies in the original wording of the Lex; both ‘occidere’ (‘killing’ or ‘slaying’) and ‘urere, frangere, rumpere’ (‘burning, breaking, spoiling’) appear to denote postitive acts rather than inaction. See D. 9.2.2.pr and D. 184.108.40.206 respectively. Cf. P.B.H. Birks, ‘Doing and causing to be done’ in A.D.E. Lewis and D.J. Ibbetson, The Roman Law Tradition (1994), pp. 32-54.
123 With the exception, perhaps, of the so-called lis crescens; J.L. Barton, ‘The Lex Aquilia and Decretal Actions’ in A. Watson (ed.), Daube Noster (1974), pp. 15-25 at p. 23.
124 G. III, 219; I. IV, III, 16 and R. Zimmermann, The Law of Obligations (1990), p. 979.
125 D. 9.2.8.
126 D. 220.127.116.11.
127 D. 9.2.31.
128 D. 18.104.22.168.
129 D. 22.214.171.124. and D. 126.96.36.199.
130 D. 188.8.131.52 and 9.2.45.pr.
131 L. Mitteis, Römisches Privatrecht bis auf die Zeit Diocletians, Vol. I (1908), p. 322. See also W.W. Buckland, ‘Diligens Paterfamilias’ in E. Albertario, P. Ciapessoni and P. De Fransisci, Studi in Onore di Pietro Bonfante (1930), pp. 85-108 at 102, J.A.C. Thomas, Textbook of Roman Law (1976), p. 249 and F.M. de Robertis, La Responsabilita’ Contrattuale nel Sistema della Grande Compilazione, Vol. I (1983), pp. 195, 197 and 199. Contra: H-J Hoffmann, Die Abstufung der Fahrlässigkeit in der Rechtsgeschichte (1986), 6, footnote 25.
132 ‘823 [Schadensersatzpflicht]
(1) Wer vorsätzlich oder fahrlässig das Leben, den Körper, die Gesundheit, die Freiheit, das Eigentum oder ein sonstiges Recht eines anderen widerrechtlich verletzt, ist dem anderen zum Ersatze des daraus entstehenden Schadens verpflichtet.
(2) Die gleiche Verpflichtung trifft denjenigen, welcher gegen ein den Schutz eines anderen bezweckendes Gesetz verstößt. Ist nach dem Inhalte des Gesetzes ein Verstoß gegen dieses auch ohne Verschulden möglich, so tritt die Ersatzpflicht nur im Falle des Verschuldens ein.’ Translation: S.L. Goren, The German Civil Code (rev. edn 1994), p. 153.
133 RG 30 October 1902, RGZ 52, 373 at 376; R. Hübner, Grundzüge des Deutschen Privatrechts (5th edn 1930), p. 609; R. Zimmermann, The Law of Obligations (1990), p. 1045 and C. von Bar, The Common European Law of Torts, Vol. 1 (1998), p. 125.
134 H. Kötz, Deliktsrecht (4th edn 1988), nr 94; K. Larenz/C-H. Canaris, Lehrbuch des Schuldrechts, Vol. 2 (13th edn 1994), p. 364; B.S. Markesinis, The German Law of Obligations, Vol. II (3rd edn 1997), p. 68; J. Limpens, R.M. Kruithof and A. Meinertzhagen-Limpens, ‘Liability for One’s Own Act’, in A. Tunc (ed.), International Encyclopedia of Comparative Law, Vol. XI Torts (1983), ch. 2, p. 15; and F.H. Lawson, Negligence in the Civil Law (1950), p. 30.
135 E. Deutsch, Haftungsrecht, Vol. 1 (1976), p. 127; Idem, Unerlaubte Handlungen, Schadensersatz und Schmerzensgeld (3rd edn 1995), nrs 38 and 86; J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 823, nr 297 and B.S. Markesinis, The German Law of Obligations, Vol. II (3rd edn 1997), pp. 74-75.
136 L. Enneccerus and H.C. Nipperdey, Allgemeiner Teil des Bürgerlichen Rechts, Vol. 2 (15th edn 1960), par. 209; W. Münzberg, Verhalten und Erfolg als Grundlagen der Rechtswidrigkeit und Haftung (1966), pp. 109-112; H. Kötz, Deliktsrecht (4th edn 1988), nrs 98-103 and K. Larenz/C-H. Canaris, Lehrbuch des Schuldrechts, Vol. 2 (13th edn 1994), pp. 364-370 (with further references on p. 365, footnote 37).
137 …or if they were intentional; H. Kötz, Deliktsrecht (4th edn 1988), nr 98. Cf. ibid., nr 103 and K. Larenz/C-H. Canaris, Lehrbuch des Schuldrechts, Vol. 2 (13th edn 1994), p. 365.
138 B.S. Markesinis, The German Law of Obligations, Vol. II (3rd edn 1997), p. 71.
139 R. Zimmermann, The Law of Obligations (1990), p. 1045.
140 E. Deutsch, Haftungsrecht, Vol. 1 (1976), p. 128 and J. Esser/E. Schmidt, Schuldrecht, Vol. I (6th edn, 1984), pp. 354-355.
141 E. Deutsch, Haftungsrecht, Vol. 1 (1976), p. 128 and J. Esser/E. Schmidt, Schuldrecht, Vol. I (6th edn, 1984), p. 355.
142 E. Deutsch, Haftungsrecht, Vol. 1 (1976), pp. 128-129 and J. Esser/E. Schmidt, Schuldrecht, Vol. I (6th edn, 1984), pp. 354-356.
143 On the terminological distinction between Verkehrspflichten and Verkehrssicherungspflichten, see H. Kötz, Deliktsrecht (4th edn 1988), nr 233.
144 J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Book 2, Recht der Schuldverhältnisse (12th edn 1985), § 823, nrs 311-312; B.S. Markesinis, The German Law of Obligations, Vol. II (3rd edn 1997), p. 75; C. von Bar, The Common European Law of Torts, Vol. 1 (1998), nr 104 and idem, Verkehrspflichten: richterliche Gefahrensteuerungsgebote im deutschen Deliktsrecht (1980), p. 113 with further references to the relevant caselaw in footnote 77.
145 RG 30 October 1902, RGZ 52, 373.
146 RG 23 February 1903, RGZ 54, 53.
147 Ibid. at 56; ‘Pflicht zur Anwendung von Sorgfalt’.
148 Ibid. at 56; ‘Verfügungsmacht’.
149 Ibid. at 56: ‘Der Schwerpunkt liegt hierbei nicht in dem Eigentumsrecht als solchem; eine gleiche Pflicht zur Anwendung von Sorgfalt besteht möglicherweise auch für den aus einem anderen Recht als dem Eigentum Verfügungsberechtigten. Die Frage ist, ob der Betreffende vermöge seiner tatsächlichen und rechtlichen Beziehung zu der Sache bei der Verfügung über diese Sache, bezw. bei der Handhabung oder dem Gebrauch derselben im Rechtsverkehr irgend welche Sorgfalt gegenüber Dritten anzuwenden habe.’ (My own translation).
150 E. Deutsch, Haftungsrecht, Vol. 1 (1976), p. 128.
151 H. Kötz, Deliktsrecht (4th edn 1988), nrs 241-242; BGH 30 April 1953, BGHZ 9, 373; BGH 15 June 1954, BGHZ 14, 83 and BGH 30 December 1954, BGHZ 16, 95.
152 BGH 25 February 1958, VersR 1958, 308 and BGH 19 May 1967, VersR 1967, 801.
153 C. von Bar, Verkehrspflichten: richterliche Gefahrensteuerungsgebote im deutschen Deliktsrecht (1980), pp. 56-57 and J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 823, nrs 346-347.
154 BGH 16 December 1953, VersR 1954, 100. See for a more detailed investigation into the caselaw C. von Bar, Verkehrspflichten: richterliche Gefahrensteuerungsgebote im deutschen Deliktsrecht (1980), pp. 43-82; idem, The Common European Law of Torts, Vol. I (1998), nrs 104-105; J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 823, nrs 321-445 and B.S. Markesinis, The German Law of Obligations, Vol. II (3rd edn 1997), pp. 74-95.
155 C. von Bar, Verkehrspflichten: richterliche Gefahrensteuerungsgebote im deutschen Deliktsrecht (1980), p. 114; H. Kötz, Deliktsrecht (4th edn 1988), nr 234 and J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 823, nr 313.
156 E. Deutsch, Haftungsrecht, Vol. 1 (1976), pp. 125-126.
157 Reichsgesetzblatt 1935 (1), 839.
158 ‘323c [Unterlassene Hilfeleistung]
Wer bei Unglücksfällen oder gemeiner Gefahr oder Not nicht Hilfe leistet, obwohl dies erforderlich und ihm den Umständen nach zuzumuten, insbesondere ohne erhebliche eigene Gefahr und ohne Verletzung anderer wichtiger Pflichten möglich ist, wird mit Freiheitsstrafe bis zu einem Jahr oder mit Geldstrafe bestraft. Translation: J.J. Darby, The Penal Code of the Federal Republic of Germany (1987), p. 231.
159 Then § 330c StGB.
160 Großer Senat für Strafsachen 10 March 1954, BGHSt 6, 147 at 151: ‘Es kann nicht bestritten werden, daß gegenseitige Hilfe in Notfällen einem von jeher bestehenden sittlichen Gebot entspricht. Die Hilfe für den notleidenden Nächsten war insbesondere immer ein Hauptgebot der christlichen Lehre.’ Translation: J.P. Dawson, ‘Negotiorum Gestio: the Altruistic Intermeddler’, (1961) 74 Harvard LR 817-865 and 1073-1129 at 1104.
161 Weimar, ‘Schadensersatzpflicht bei Schädigungen durch Unterlassen’, DR 1937, 77; E. Krupp, ‘Kameradschaftspflichten und Rechtspflichten’, DR 1938, 465 at 467 and H. Stoll, Vertrag und Unrecht, Vol. 2 (3rd edn 1943), p. 197. Cf. W. Weimar, ‘Besteht eine Schadensersatzpflicht eines Krankenhauses bei Abweisung Schwerverletzter?’ JR 1975, 145 at 146.
162 H. Welzel, ‘Zur Dogmatik der echten Unterlassungsdelikte, insbesondere des § 330c StGB’, NJW 1953, 327 at 328; J.P. Dawson, ‘Negotiorum Gestio: the Altruistic Intermeddler’, (1961) 74 Harvard LR 817-865 and 1073-1129 at 1107, footnote 80; W. Dütz, ‘Zur privatrechtlichen Bedeutung unterlassener Hilfeleistung (§ 330c StGB)’, NJW 1970, 1822-1826, at 1825; J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 823, nr 591 and OLG Frankfurt 27 October 1988, NJW-RR 1989 (2), 794. A similar approach is taken towards § 138 StGB; J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 823, nr 298. Contra: K. Larenz/C-H. Canaris, Lehrbuch des Schuldrechts, Vol. II/2 (13th edn 1994), p. 441, where it is argued that both § 138 and § 323c StGB are Schutzgesetze.
163 K. Haager, BGB-RGRK, Vol. II, part 2 (11th edn 1960), § 823, Anm. 103, p. 1290: ‘Es genügt, daß eine Vorschrift in der Hauptsache andere Zwecke verfolgt, nebenher aber zum Schutz bestimmter einzelner Interessen bestimmt ist. (…) Nur Gesetze, die ausschließlich die Ordnung der Gesamtheit, das Staatganze als solches, seine äußere Unversehrtheit, seine innere Verfassung und Verwaltung, zu schützen bestimmt sind, wie die Strafgesetze über Hoch- und Landesverrat, über Widerstand gegen die Staatsgewalt, über Verfassungsbruch und andere, fallen vollständig außerhalb des Rahmens eines solchen Schutzgesetzes, auch wenn sie gewissen Personen nützen.’ (My own translation) Cf. J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 823, nrs 580-585.
164 OLG Frankfurt, 27 October 1988, NJW-RR 1989/II, 794 at 795: Die Strafbestimmung ist kein Schutzgesetz i.S. des § 823 II BGB, denn es geht nicht an denjenigen, der Hilfeleistung unterläßt, in gleicher Weise haften zu lassen wie den Schädiger. (My own translation) Cf. J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 823, nr 591 and C. von Bar, The Common European Law of Torts, Vol. 1 (1998), nr 601.
165 RG 1 June 1937, RGZ 155, 257 at 285 and BGH 5 November 1962, NJW 1963, 148 at 149-150.
166 ‘826 [Sittenwidrige vorsätzliche Schädigung]
Wer in einer gegen die gute Sitten verstoßenden Weise einem anderen vorsätzlich Schaden zufügt, ist dem anderen zum Ersatze des Schadens verpflichtet. (My own translation)
167 J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 826, nr 71 and B.S. Markesinis, The German Law of Obligations, Vol. II (3rd edn 1997), pp. 894-895.
168 J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 826, nr 51. Cf. W. Dütz, ‘Zur privatrechtlichen Bedeutung unterlassener Hilfeleistung (§ 330c StGB)’, NJW 1970, 1822-1826 at 1825-1826.
169 OLG Köln, 2 November 1971, NJW 1972, 497; BGH 17 May 1982, NJW 1982, 2815 and J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 826, nrs 52-54, with further references.
170 J. von Staudinger/K. Schäfer, Kommentar zum Bürgerlichen Gesetzbuch mit Einführungsgesetz und Nebengesetzen, Zweites Buch, Recht der Schuldverhältnisse (12th edn 1985), § 826, nrs 55-57, with further references.
171 B.S. Markesinis, W. Lorenz and G. Dannemann, The German Law of Obligations, Vol. I (1997), pp. 64-71 and C. von Bar, The Common European Law of Torts, Vol. 1 (1998), nrs 190 and 471-477.
172 B.S. Markesinis, W. Lorenz and G. Dannemann, The German Law of Obligations, Vol. I (1997), pp. 276-282.
173 BGH 28 January 1976, BGHZ 66, 51.
174 Cf. English law, which would probably allow a claim in negligence, not in contract. E.g. Ward v Tesco Stores Ltd.  1 All ER 219.
175 BGH 28 January 1976, BGHZ 66, 51 at 56: ‘Es entspricht seit langem gefestigter Rechtsprechung insbesondere des erkennenden Senats, daß unter besonderen Voraussetzungen auch außenstehende, am Vertragsschluß selbst nicht beteiligte Dritte in den Schutzbereich eines Vertrages einbezogen sind (…)’. Translation: F.H. Lawson and B.S. Markesinis. Tortious Liability for Unintentional Harm in the Common Law and the Civil Law, Vol. II (1982), 146-151 at 149.
176 J-L Gazzaniga, Introduction historique au droit des obligations (1992), p. 243.
177 J. Domat, Les loix civiles dans leur ordre naturel (rev. edn 1745), p. 179 (Livre II, Titre VIII): ‘ (…) comme si par legereté on jette quelque chose par une fenêtre qui gâte un habit: Si des animaux mal gardez font quelque dommage: Si on cause un incendie par une imprudence: Si un bâtiment qui menace ruine n’étant pas réparé tombe sur un autre, & y fait du dommage.’ (My own translation)
178 Ibid., p. 185 (Livre II, Titre VIII, Section VIII): ‘Ceux qui pouvant empêcher un dommage, que quelque devoir les engageoit de prévenir, y auront manqué, pourront en être tenus selon les circonstances. Ainsi un maître qui voit & souffre le dommage que fait son domestique, pouvant l’empêcher, en est responsable.’ (My own translation)
179 C.H. Sieburgh, ‘Faute als grond voor aansprakelijkheid’,  Groninger Opmerkingen en Mededelingen, pp. 68-90 at p. 71. Cf. also G. Viney, Traité de Droit Civil; Introduction à la responsabilité (2nd edn 1995), nr. 14.
180 J.H. Crabb, The French Civil Code (rev. edn 1995), p. xxiii.
181 ‘1382 Tout fait quelconque de l’homme, qui cause à autrui un dommage, oblige celui par la faute duquel il est arrivé, à le réparer.’ The translation is largely based on J.H. Crabb, The French Civil Code (rev. edn 1995), p. 252. It was necessary to amend this translation in one respect: Crabb’s translation of ‘fait’ was ‘act’. A ‘fait’ however, can refer to an act as well as to inaction. See J. Mihura in his annotation under the judgment of the Cour de Cassation in the Branly-case,  II JCP 6193, with further references.
182 ‘1383 Chacun est responsable du dommage qu’il a causé non seulement par son fait, mais encore par sa négligence ou par son imprudence.’ Translation: see previous footnote.
183 E.g. H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (1965), nrs 524 and 525; G. Viney, Traité de Droit Civil, Vol. IV (1982), nrs 451-452 and F. Terré, P. Simler and Y. Lequette, Droit civil; Les Obligations (6th edn 1996), nr 688.
184 G. Viney, Traité de Droit Civil, Vol. IV (1982), nr 456; F. Terré, P. Simler and Y. Lequette, Droit civil; Les Obligations (6th edn 1996), nrs 688-690 and J. Limpens, R.M. Kruithof and A. Meinertzhagen-Limpens, ‘Liability for One’s Own Act’, in A. Tunc (ed.), International Encyclopedia of Comparative Law, Vol. XI Torts (1983), ch. 2, p. 36 with further references in footnote 332.
185 H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (1965), nr 525.
186 F. Terré, P. Simler and Y. Lequette, Droit civil; Les Obligations (6th edn 1996), nr 689; J. Limpens, R.M. Kruithof and A. Meinertzhagen-Limpens, ‘Liability for One’s Own Act’, in A. Tunc (ed.), International Encyclopedia of Comparative Law, Vol. XI Torts (1983), ch. 2, p. 36.
187 Cour de Cassation 27 February 1951, D. 1951, 329: ‘que la cour devait rechercher si, en écrivant une histoire de la T.S.F. dans laquelle les travaux et le nom d’Edouard Branly étaient volontairement omis, Turpain s’était comporté comme un écrivain ou un historien prudent, avisé et conscient des devoirs d’objectivité qui lui incombaient’. (My own translation)
188 G. Viney, Traité de Droit Civil, Vol. IV (1982), nr 456.
189 E.g. the ‘personne avisée placée dans les mêmes circonstances “externes” que l’auteur du dommage’, H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (1965), nr 540; or the ‘bon père de famille’,‘placé dans les mêmes circonstances de temps, de lieu et d’action’, P. le Tourneau and L. Cadiet, Droit de la Responsabilité (1996), nr 32.
190 G. Viney, Traité de Droit Civil, Vol. IV (1982), nr 456 and F. Terré, P. Simler and Y. Lequette, Droit civil; Les Obligations (6th edn 1996), nr 689.
191 G. Viney, Traité de Droit Civil, Vol. IV (1982), nr 456 and F. Terré, P. Simler and Y. Lequette, Droit civil; Les Obligations (6th edn 1996), nr 689.
192 Cour de Cassation 8 April 1932, D.H. 1932, 297.
193 Cour de Cassation 24 December 1924, D.H. 1925, 120 at 121: (…) si chacun est responsable de sa négligence, l’omission ne peut entraîner une responsabilité qu’autant qu’il y avait, pour celui auquel on l’impute, obligation d’accomplir le fait omis. (My own translation)
194 H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (6th edn, 1965), nr 544.
195 J. Limpens, R.M. Kruithof and A. Meinertzhagen-Limpens, ‘Liability for One’s Own Act’, in A. Tunc (ed.), International Encyclopedia of Comparative Law, Vol. XI Torts (1983), ch. 2, pp. 36-43 at p. 37.
196 P. Catala and J.A. Weir, ‘Delict and Torts: a Study in Parallel’, (1962-1963) 37 Tulane LR 573-620 at 615.
197 Cf. the references in footnote 52 and the corresponding part of the main text.
198 J. Bell, S. Boyron and S. Whittaker, Principles of French Law (1998), p. 361, with further references.
199 A. Tunc, ‘the Volunteer and the Good Samaritan’ in J.M. Ratcliffe, (ed.), The Good Samaritan and the Law (1966), pp. 43-62 at p. 49.
200 Ordinance of 25 June 1945,  JO 3862.
201 ‘Article 63 Sans préjudice de l’application, le cas échéant, des peines plus fortes prévues par le présent code et les lois spéciales, sera puni d’un emprisonment d’un mois à trois ans et d’une amende de 12.000 à 500.000 francs, ou de l’une de ces deux peines seulement, quiconque, pouvant empêcher par son action immédiate, sans risque pour lui ou pour les tiers, soit un fait qualifié crime, soit un délit contre l’intégrité corporelle de la personne, s’abstient volontairement de le faire.
Sera puni des mêmes peines quiconque s’abstient volontairement de porter à une personne en péril l’assistance que, sans risque pour lui ni pour les tiers, il pouvait lui prêter, soit par son action personnelle soit en provoquant un secours.’ Translation based on G.O.W. Mueller, The French Penal Code (1960), p. 38. Mueller’s translation was of the text of art. 63 C. pén. after it was amended on 13 April 1954,  JO 3580.
202 Tribunal Correctionnel d’Aix 27 March 1947, D. 1947, 304.
203 Cour de Cassation 16 March 1972, D. 1972, 394 at 395: ‘(…) si les dispositions de l’art. 63, al. 2, c.pén. ont pour objet l’intérêt général de la collectivité et répriment le trouble causé à l’ordre public par une abstention délictueuse, elles ont aussi pour but la protection des intérêts privés;’ (My own translation)
204 K.J. Kemp, Delictual Liability for Omissions (1978) (Unpublished thesis, University of Port Elizabeth, South Africa), p. 564.
205 Nouveau Code Pénal, Livre II, Titre Premier, Chapitre III, Section III (artt. 223-5-223-7). Cf. the accompanying notes in the Circulaire of 14 May 1993.
206 H. Mazeaud and L. Mazeaud, ‘Obligations en général et responsabilité civile’, (1943) 41 Rev.trim.dr.civ. 109-121 at 111 and eidem, Traité Théorique et Pratique de la Responsabilité Civile, Vol. 1 (4th edn 1947), nr 525.
207 Cour de Cassation 27 February 1951, D. 1951, 329: ‘(…) la faute prévue par les art. 1382 et 1383 peut consister aussi bien dans une abstention que dans un acte positif; que l’abstention, même non dictée par la malice et l’intention de nuire, engage la responsabilité de son auteur lorsque le fait omis devait être accompli soit en vertu d’une obligation légale, réglementaire ou conventionnelle, soit aussi, dans l’ordre professionnel, s’il s’agit notamment d’un historien, en vertu des exigences d’une information objective;’ (My own translation)
208 H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (6th edn 1965), nr 544.
209 Ibid., nr 544 with further references to the relevant caselaw in footnotes 8, 9, 10 and 11.
210 H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (6th edn 1965), nr 544; G. Viney, Traité de Droit Civil, Vol. IV (1982), nr 453 and J. Limpens, R.M. Kruithof and A. Meinertzhagen-Limpens, ‘Liability for One’s Own Act’, in A. Tunc (ed.), International Encyclopedia of Comparative Law, Vol. XI Torts (1983), ch. 2, pp. 36-43 at 37.
211 F. Terré, P. Simler and Y. Lequette, Droit civil; Les Obligations (6th edn 1996), nr 690.
212 N. Dejean de la Bâtie, Aubry et Rau, Droit Civil Français, Vol. VI-2 (8th edn 1989), nr 38 with further references.
213 N. Dejean de la Bâtie, Aubry et Rau, Droit Civil Français, Vol. VI-2 (8th edn 1989), nr 38 and H. Desbois in his Note under Cour de Cassation, 27 February 1951, D. 1951, 329 at 331.
214 Cour de Cassation 13 December 1972, D. 1973, 493; H. Mazeaud, L. Mazeaud and A. Tunc, Traité Théorique et Pratique de la Responsabilité Civile (6th edn 1965), nr 534; G. Viney, Traité de Droit Civil, Vol. IV (1982), nr 455 and F. Terré, P. Simler and Y. Lequette, Droit civil; Les Obligations (6th edn 1996), nr 690.
215 Art. 1384 C. civ.
217 Art. 1385 C. civ.
218 Art. 1386 C. civ.
219 D.J. Ibbetson, A Historical Introduction to the Law of Obligations (1999), p.155.
220 Cf. above, section IV:1, footnote 10 and the corresponding part of the main text.
221 D.J. Ibbetson, A Historical Introduction to the Law of Obligations (1999), p. 159 and M.R. Brazier in M.R. Brazier and others (eds), Clerk & Lindsell on Torts (17th edn 1997), par. 1-30.
222 (1725) 1 Str. 634 at 636. This case is also reported in 2 Ld Ray. 1399, under the name Reynolds v Clerk and in 8 Mod. 272 under the name Reynolds v Clark in Fort. 212. However, the passage here quoted does not appear in these reports.
223 D.J. Ibbetson, A Historical Introduction to the Law of Obligations (1999), p. 165, footnote 53.
224 E.g. when by the negligence of a servant cattle perish.
225 E.g. when a sheriff allows a prisoner to escape.
226 J. Comyns, A Digest of the Laws of England, Vol. I (1762), pp. 223-228.
227 Cf. for the opposite view J. Austin, Lectures on Jurisprudence, Vol. II (1863), p. 101 (Lecture XX).
228 (1856) 11 Ex. 781.
229 Ibid. at 784.
230 F. Pollock, The Law Of Torts (1887), p. 355.
231 Ibid., p. 352. See also Bulman v Furness Rly Company (1875) 32 LT 430 at 432, where a distinction is made between ‘active’ and ‘passive’ negligence and F.T. Piggott, Principles of the Law of Torts (1885), p. 208, who distinguishes ‘positive’ and ‘negative’ duties.
232 Gantret v Egerton (1867) 16 LT 17 at 17.
233 Ibid. (1867) 16 LT 17 at 18. This interruption does not appear in this form in the Law Reports,  LR 2 CP 371.
234 Gautret v Egerton  LR 2 CP 371 at 375. This quote does not appear in the Bar Reports, (1867) 16 LT 17.
235 D. Howarth, Textbook on Tort (1995), p. 173.
236 B.S. Markesinis, ‘Negligence, Nuisance and Affirmative Duties of Action’ (1989) 105 LQR 104 at 104.
237 This observation was also made by M.A. Millner in his Negligence in Modern Law (1967), p. 30.
238 E.g. Lord Keith’s famous example of a man ‘who sees another about to walk over a cliff with his head in the air, and forbears to shout a warning’ in Yuen Kun Yeu v AG of Hong Kong  2 All ER 705 at 710. (Note that this example is remarkably similar to Willes J’s example of the ‘spiteful man’ who ‘omits the warning’. (As quoted above) and Lord Diplock’s speech in Home Office v Dorset Yacht Co. Ltd  AC 1004, at 1060. In Smith v Littlewoods Lord Goff referred to this speech as authority for the no-liability rule. Yet the fact alone that liability was imposed in Dorset Yacht indicates that it must be seriously doubted whether the no-liability rule was part of the ratio decidendi.
239 Smith v Littlewoods  AC 241 at 271.
240 Although Lord Keith expressed his agreement with Lord Goff’s speech; Smith v Littlewoods  AC 241 at 249. The other Lawlords came to the same conclusion as Lord Goff, but based this conclusion on the fact that, in short, the damage was not reasonably foreseeable. For a more detailed investigation, see B.S. Markesinis, ‘Negligence, Nuisance and Affirmative Duties of Action’ (1989) 105 LQR 104 and below, section V:6 (2).
241 Smith v Littlewoods  AC 241 at 271.
242 B.S. Markesinis, ‘Negligence, Nuisance and Affirmative Duties of Action’ (1989) 105 LQR 104 at 122-123.
243 E.g. A.M. Honoré, ‘Law, Morals and Rescue’ in J.M. Ratcliffe (ed.), The Good Samaritan and the Law (1966), pp. 225 ff., reprinted and expanded in Honoré, Making Law Bind, Essays Legal and Philosophical (1987), pp. 256 ff.; P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), pp. 60 ff.; J.G. Fleming, The Law of Torts (9th edn 1998), pp. 162 ff. and E.J. Weinrib, ‘The Case for a Duty to Rescue’, (1980) 90 Yale LJ 247-293.
244 Stovin v Wise  3 All ER 801.
245 Ibid. at 806.
246 Ibid. at 807.
247 Ibid. at 819.
248 J.G. Fleming, The Law of Torts (9th edn 1998), p. 163.
249 See above, section IV:1.
250 A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), pp. 31 ff. at pp. 43-45.
251 See for French law above, section IV:3, footnote … and for English law W.V.H. Rogers, Winfield and Jolowicz on Torts (15th edn 1998), p. 119 and J.G. Fleming, The Law of Torts (9th edn 1998), pp. 163-164.
252 C.K. Allen, ‘Legal Duties’  40 Yale LJ 331-380, reprinted in C.K. Allen, Legal Duties and Other Essays in Jurisprudence (1931), pp. 156-220 at 206 with further references in footnote 2; M.R. Brazier in M.R. Brazier and others (eds), Clerk & Lindsell on Torts (17th edn 1995), par. 7-28; Kelly v Metropolitan Rly Co.  1 QB 944; Morrison v Sheffield Corporation  2 KB 866; Sheppard v Glossop Corporation  3 KB 132 at 145, per Scrutton LJ; Fisher v Ruislip-Northwood Urban District Council and Middlesex County Council  KB 584; Hughes v Lord Advocate  1 All ER 705; Hill v Chief Constable of West Yorkshire  AC 53 at 61 per Lord Keith and Wright v Lodge  4 All ER 299.
253  2 KB 43.
254 Cherry Ltd v Allied Insurance Brokers Ltd  1 Lloyd’s Rep 274 and D.K. Allen, Misrepresentation (1988), pp. 85-86. A similar duty lies on a man who unwittingly makes a false representation and later realises that he has made a mistake; Briess v Woolley  AC 333 at 359 per Lord Tucker and G. Spencer Bower and A.K. Turner, The Law of Actionable Misrepresentation (3rd edn 1974), pp. 120-121.
255 Topp v London County Bus Co  3 All ER 448. Cf. Perl Ltd. v Camden London Borough Council  QB 342.
256 Stansbie v Troman  2 KB 48 and below, sections V:5 and V:6.
257 Union Pacific Rly Co. v Cappier 66 Kan. 649, 72 Pac. 281 (1903).
258 221 Ala. 91, 127 So. 519 (1930), with further references.
259 Ibid. at 521.
260 E.g. Ch. O. Gregory, ‘The Good Samaritan and the Bad’ in J.M. Ratcliffe, (ed.), The Good Samaritan and the Law (1966), pp. 23-41 at 26-27.
261 W. Page Keeton (ed.), Prosser and Keeton on The Law of Torts (5th edn 1984), p. 377 and F.V. Harper, F. James and O.S. Gray, The Law of Torts (2nd edn 1986), pp. 720-721 (par. 18.6) with further references in footnote 16.
262 A.M. Honoré, ‘Are Omissions Less Culpable?’ in P. Cane and J. Stapleton (eds), Essays for Patrick Atiyah (1991), pp. 31 ff. at pp. 43-45.
263 Example loosely based on J.B. Ames, ‘Law and Morals’ 22 Harvard LR 97-113 (1908), reprinted in J.M. Ratcliffe, (ed.), The Good Samaritan and the Law (1966), pp. 1-21 at p. 19.
264 B.S. Markesinis and S.F. Deakin, Tort Law (4th edn 1999), pp. 138-141 with further references.
265 See for various other terms K. Barker, ‘Unreliable Assumptions in the Modern Law of Negligence’ (1993) 109 LQR 461-484 at 463.
266 Ibid. at 462. Barker distinguishes a third type of undertaking which is based on ‘choice’.
267  AC 465.
268 Ibid. at 528-529.
269  2 AC 145.
270 Ibid. at 181. Cf. Midland Bank Trust Co. Ltd v Hett, Stubs & Kemp  Ch 384 per Oliver J.
271  1 QB 428.
272 Ibid. at 435-436, per Nield J.
273  2 AC 207.
274 Ibid. at 262.
275 Ibid. at 259-260. Cf. Ross v Caunters  Ch 297 at 303, per Megarry V-C and the decision of the Court of Appeal in White v Jones  3 WLR 730, at 749, per Farquharson LJ and at 751, per Steyn LJ.
276 White v Jones  2 AC 207 at 268. Cf. Leigh and Sillivan Ltd v Aliakmon Shipping Co. Ltd  2 All ER 145.
277 Ibid. at 274.
278 Ibid. at 276.
279 Ibid. at 273.
280 Both their Lordships imposed a duty to act in the absence of any reliance by the plaintiffs on the defendants’ undertaking. Cf. White v Jones  2 AC 207 at 272 per Lord Browne-Wilkinson.
281  2 All ER 865.
282 Ibid. at 885. Cf. also Sheppard v Glossop Corporation  3 KB 132 at 145, per Scrutton LJ.
283  3 All ER 87.
284 Ibid. at 87.
285 Ibid. at 94. Note the resemblance between this dictum and the prevailing opinion in Germany on the question whether or not par. 323c StGB is a Schutzgesetz; above section IV:2 with references in footnote … (51?)
286 Ibid. at 96.
287 Capital and Counties v Hampshire CC  2 All ER 865 at 886.
288 Barrett v. Ministry of Defence  3 All ER 87 at 96 (quoted immediately above).
289  3 All ER 193.
290  3 WLR 152.
291 Barrett v Enfield London BC  3 All ER 193 at 207-208. A similar distinction was made by Lord Hutton, ibid. at 227-228.
292 K. Barker, ‘Unreliable Assumptions in the Modern Law of Negligence’ (1993) 109 LQR 461-484 at 475.
293 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 64 and Kent v London Ambulance Service  Lloyd’s Rep Med 58, reported under the name Kent v Griffiths in (1999) 47 BLMR 125, discussed in subsection V:8, below.
294 J.G. Fleming, The Law of Torts (9th edn 1998), pp. 164-168.
295 R. Bagshaw, ‘The duties of care of emergency service providers’  LMCLQ 71-92 at 75.
296 It should be noted that this view does not imply that liability is merely imposed for the act of ‘undertaking’ rather than the inaction of failing to prevent harm from occurring. The undertaking in itself did not amount to negligence. Cf. above, Chapter II.
297 The ‘Ogopogo’  1 Lloyd’s Rep 257 at 263.
298  1 Lloyd’s Rep 97.
299 Ibid. at 99 per Lord Denning M.R. See also Wilkinson v Coverdale (1793) 1 Esp. 75.
300 R. Bagshaw, ‘The duties of care of emergency service providers’  LMCLQ 71-92 at 74-75 and idem, ‘Proximity’s Siren Song: Negligence Liability of Fire Brigades in England’, (1998) 6 Tort L Rev 109-113 at 111-112.
301 Capital and Counties v Hampshire CC  2 All ER 865 at 882: ‘It is not to be supposed that having arrived on the fire ground the fire brigade would simply have sat on their hands’.
302 Ibid. at 882. It remains unclear why the fire brigade should bear the burden of proof in this respect. Cf. J. Hartshorne, ‘The liability in negligence of the fire service – the Court of Appeal decides’, (1997) 13 PN 53-58 at 58.
303 … and, perhaps, not to cause any ‘substitute harm’, as Bagshaw calls it; R. Bagshaw, ‘The duties of care of emergency service providers’  LMCLQ 71-92 at 76. Cf. M. Spence, Protecting Reliance: The Emergent Doctrine of Equitable Estoppel (1999), pp. 6-7.
304 K. Barker, ‘Unreliable Assumptions in the Modern Law of Negligence’ (1993) 109 LQR 461-484 at 475.
305 … and that no other option than the rope would have given her a chance to survive.
306 Cf. N. MacCormick, ‘Voluntary Obligations and Normative Powers I’ (1972) 46 PASS 59-78 at 64.
307 E.g. Stovin v Wise  3 All ER 801 at 828-830 per Lord Hoffmann and Capital and Counties v Hampshire CC  2 All ER 865 at 876-878 per Stuart-Smith LJ.
308 Sutherland Shire Council v Heyman (1985) 157 CLR 424 at 464.
309 Stovin v Wise  3 All ER 801 at 829, per Lord Hoffmann and R. Bagshaw, ‘The duties of care of emergency service providers’  LMCLQ 71-92 at 80.
310 Sutherland Shire Council v Heyman (1985) 157 CLR 424 at 464.
311 White v Jones  2 AC 207 at 276.
312 Stovin v Wise  3 All ER 801 at 829.
313 In Philcox v Civil Aviation Authority, The Times 8 June 1995, it had already been decided that the Aviation Authority, the other example used by Mason J, owed no duty to the owner of an aircraft to take care in the inspection of his aircraft and the certification of its airworthiness.
314 Stovin v Wise  3 All ER 801 at 829.
315  2 All ER 865.
316 Ibid. at 883-886 and 889, per Stuart-Smith LJ.
317 Costello v Chief Constable  1 All ER 550 at 563, per May LJ. Cf. Alexandrou v Oxford  4 All ER 328 at 338, per Glidewell LJ and at 343-344, per Slade LJ. However, in Kent v London Ambulance Service  Lloyd’s Rep Med 58 at 62, reported under the name Kent v Griffiths in (1999) 47 BLMR 125, Kennedy LJ openly questioned the Court of Appeal’s reluctance to recognise a relationship of proximity in Alexandrou v Oxford. ‘Could there not be a special relationship with all 999 callers whose calls were accepted?’, he wondered.
318 Oll v Secretary of State for Transport  3 All ER 897 at 905, per May J.
319 Kent v London Ambulance Service  Lloyd’s Rep Med 58 at 63, reported under the name Kent v Griffiths in (1999) 47 BLMR 125.
320 Ibid. at 63.
321 The High Court of Australia recently rejected the doctrine of general reliance in Pyrenees Shire Council v Day (1998) 72 ALJR 152. In New Zealand the doctrine of general reliance appears to have been accepted by the courts; cf. Invercargill City Council v Hamlin  1 All ER 756 at 761-764, per Lord Lloyd with further references.
322 Musgrave v The Chief Constable of Hampshire, Court of Appeal 10 February 1999 (unpublished; transcript by Smith Bernal).
323 J.C. Smith and P. Burns, ‘Donoghue v. Stevenson – The Not So Golden Anniversary’ (1983) 46 MLR 147 at 156-157.
324 Section IV:2, footnote …(34?) and the corresponding part of the main text and section.
325 W.V.H. Rogers, Winfield and Jolowicz on Tort (15th edn, 1998), p. 121.
326 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 65; B.S. Markesinis and S.F. Deakin,Tort Law (4th edn 1999), pp. 138-139; W.V.H. Rogers, Winfield and Jolowicz on Tort (15th edn, 1998), p. 121 and M.A. Jones, Textbook on Torts (6th edn 1998), p. 54. Cf. also J.G. Fleming, The Law of Torts (9th edn 1998), p. 169 and Marc Rich & Co. v Bishop Rock Marine Co. Ltd  1 WLR 1071 at 1077 per Saville LJ.
327 Surtees v Royal Borough of Kingston upon Thames  PIQR P101 at P111
328 Ibid. at P111-P112.
329 Ibid. at P124.
330 Van Oppen v Bedford Charity Trustees  1 All ER 273 at 291, per Boreham J, as confirmed by the Court of Appeal in Van Oppen v Bedford Charity Trustees  3 All ER 389 at 412, per Balcombe LJ.
331 Barnes v Hampshire CC  1 WLR 1563.
332 Van Oppen v Bedford Charity Trustees  3 All ER 389 at 414-415, per Croom-Johnson LJ.
333 Kasapis v Laimos Bros.  2 Lloyd’s LR 378 at 381, per Salmon J.
334 Knight v Home Office  3 All ER 237 at 243, per Pill J.
335 Costello v Chief Constable of the Northumbria Police  1 All ER 550 at 563-564, per May LJ.
336 J.C. Smith and P. Burns, ‘Donoghue v. Stevenson – The Not So Golden Anniversary’ (1983) 46 MLR 147 at 157.
337 Weld-Blundell v Stephens  AC 956 at 986, per Lord Sumner.
338 Smith v Leurs (1945) 70 CLR 256 at 262. This passage was quoted with apparent approval by both Lord Pearson and Lord Diplock in Home Office v Dorset Yacht Co. Ltd  AC 1004 at 1055 and 1063 respectively.
339 Newton v Edgerley  1 WLR 1031.
340 Carmarthenshire CC v Lewis  AC 549.
341 Home Office v Dorset Yacht Co. Ltd  AC 1004. It should however be noted that liability in this case might have had a different basis. In Hill v Chief Constable of West Yorkshire  AC 53 at 61 Lord Keith suggested that the prison officers had created a ‘potential situation of danger’ by bringing the borstal boys in the locality where the damage was done.
342 Holgate v Lancashire Mental Hospitals Board  4 All ER 19.
343 Home Office v Dorset Yacht Co. Ltd  AC 1004 at 1062, per Lord Diplock and W.V.H. Rogers, Winfield and Jolowicz on Tort (15th edn, 1998), p. 124.
344 Hudson v Ridge Manufacturing Co. Ltd  2 QB 348.
345 Ellis v Home Office  2 All ER 149.
346 Kirkham v Chief Constable of Manchester  2 QB 283 and Reeves v Commissioner of Police of the Metropolis  2 All ER 381.
347 Selfe v Ilford and District Hospital Management Committee, The Times 26 November 1970.
348 E.g. W.V.H. Rogers, Winfield and Jolowicz on Tort (15th edn, 1998), p. 12; P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 68.
349  2 All ER 989.
350 Ibid. at 992.
351 Ibid. at 995.
352 Ibid. at 996.
353 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 69.
354  AC 241.
355 Ibid. at 259-260.
356 Ibid. at 261.
357 Ibid. at 279.
358 Ibid. at 271.
359 Ibid. at 273.
360 Ibid. at 275-276.
361 Ibid. 274, per Lord Goff. Cf. ibid. at 268, per Lord Mackay.
362 W.V.H. Rogers, Winfield and Jolowicz on Tort (15th edn, 1998), p. 494 with further references.
363 St Anne’s Well Brewery Co. v Roberts (1929) 140 LT 1 at 7, per Scrutton LJ.
364  AC 880.
365 Ibid. at 893-894, per Viscount Maugham and at 910, per Lord Wright.
366 J. Salmond, The Law of Torts (5th edn 1920), p. 260.
367 Leakey v National Trust  QB 485 at 516.
368 Ibid. at 516; cf. Goldman v Hargrave  2 All ER 989 at 995.
369 Ibid. at 524-527.
370 Occupiers Liability Act 1957, ss. 2(2) and 1(3)(b).
371 Occupiers Liability Act 1984, ss. 1(1) and 1(3).
372 Occupiers Liability Act 1957, s. 2(2) and Occupiers Liability Act 1984, s. 1(3)(c).
373 The pre-1984 case of British Railways Board v Herrington adopted a subjectivised standard towards trespassers;  AC 877 at 899, per Lord Reid, at 920, per Lord Wilberforce and at 941-942, per Lord Diplock.
374 Wheat v E. Lacon & Co. Ltd  AC 552 at 589, per Lord Pearson and Occupiers Liability Act 1984, s. 1(2)(a).
375  PIQR P141.
376 Ibid. at P152.
377 Highways Act 1980, s. 41(1).
378 P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 68. Cf. J.C. Smith and P. Burns, ‘Donoghue v. Stevenson – The Not So Golden Anniversary’ (1983) 46 MLR 147 at 158.
379 Darker v Chief Constable of West Midlands  3 WLR 747 at 771. See also ibid., at p. 750, per Lord Hope: ‘It is the principle that a wrong ought not to be without a remedy’.
380 A.M. Honoré, ‘Law, Morals and Rescue’ in J.M. Ratcliffe (ed.), The Good Samaritan and the Law (1966), pp. 225 ff., reprinted and expanded in A.M. Honoré, Making Law Bind, Essays Legal and Philosophical (1987), pp. 256 ff. at p. 269.
381  PIQR P101.
382 Ibid. at P111 and P111-P112, per Stocker LJ and at P124, per the Vice-Chancellor.
383 Goldman v Hargrave  2 All ER 989 at 995-996, per Lord Wilberforce.
384 P. Cane, The Anatomy of Tort Law (1997), p. 43.
385 Stovin v Wise  3 All ER 801 at 809.
386 J. Salmond, The Law of Torts (5th edn 1920), p. 259.
387 Cf. P. Cane, Atiyah’s Accidents, Compensation and the Law (6th edn 1999), p. 69 and F.H. Lawson and B.S. Markesinis, Tortious Liability for Unintentional Harm in the Common Law and the Civil Law, Vol. I (1982), p. 77.
388 K.J. Kemp, Delictual Liability for Omissions (1978) (Unpublished thesis, University of Port Elizabeth, South Africa), p. 386.
© 2001 Jeroen Kortmann. This HTML edition © 2001 University of Oxford.
This research has been supported by the network ‘Common Principles of European Private Law’, within the Training and Mobility of Researchers project (TMR) funded by the European Commission (Contract No ERB FMRX-CT97-0118), administered from Münster by Professor Dr Reinhard Schulze. The partner universities are: The University of Oxford (UK), Universidad de Barcelona (Spain), Humboldt-Universität zu Berlin (Germany), Université Jean Moulin Lyon III (France), Westfälische Wilhelms-Universität Münster (Germany), Katholieke Universiteit Nijmegen (Holland) and Università di Torino (Italy).
The contents of this page may be downloaded and printed out in single copies for individual use only. Making multiple copies without permission is prohibited.