Copyright Notice


This article explores the extent to which the criminal law defence of necessity in both its excusatory and justificatory forms can apply appropriately in situations of emergency or natural disaster. Some particular problems are examined, including the possible limitation of the defence to responses to threats to life rather than threats to property and arguments that the defence only applies to persons to whom a defendant owed some duty of protection. The article draws on research into disaster response and recovery to suggest that several formulations of the defence are not well-suited to protect disaster responders, particularly spontaneous volunteer responders, and calls for reform to reconcile more effectively the competing interests and policy imperatives arising in disaster situations.


This article has its genesis in a research project into the legal effects of the Canterbury earthquakes of 2010-2011,1 which revealed an apparent disconnect between the literature on disaster response, management and recovery, which appeared to assume disaster victims and disaster responders would not be liable for offences committed during or in the aftermath of disasters as long as their conduct was directed at protecting themselves or others, and the criminal law literature in common-law systems dealing with the defence(s) of necessity, on which such victims and responders were apparently to rely, which very largely looked at the extremities of the defence and failed to examine how necessity might work in practice in an emergency or disaster context.2

In reality there are two different forms of the defence of necessity. One is justificatory, that is the actor commits no offence because the conduct was ‘the right thing to do’ and therefore not to be considered as involving the commission of an offence. The other is excusatory, often called ‘duress of circumstances’ where although the actor’s conduct is recognised as involving the commission of a criminal offence, the law excuses the conduct because no-one faced with the peril or threat confronting the actor could have been expected to act differently. As Paul Robinson has stressed, in justification defences the focus is on the conduct of the defendant and whether the conduct is justified while in excuse defences the focus is on the defendant and his or her reasons for acting, rather than on what was done.3

This article first canvasses, in Part II, the reasons for correctly classifying and describing the operation of the different forms of necessity so that the conduct of persons invoking the defence – in either or both forms – is fairly and correctly labelled by the courts and by commentators. Doing so will, in turn, send more clear and comprehensible messages to those responding to actual or potential disasters as to the permissible limits of their conduct.

In Part III, the article looks at judicial and legislative formulations of the defence in a range of common-law jurisdictions, identifying those where it exists in both justificatory and excusatory forms and those where only one or other form is recognised, and briefly analysing the key elements of the defences. The discussion then investigates a number of issues which are particularly likely to arise for discussion or decision in the emergency response context, and identifies aspects of the defence which are or may be unduly limiting. These include the application of the defence to acts done by volunteer disaster responders (Part IV) and the related issue of its application where those sought to assisted are “strangers” to the actor (Part V) and some issues of proportionality (Part VI). The article concludes by discussing the relationship between a common law defence of necessity, statutory immunities and the application of a ‘reasonableness’ criterion to persons affected by the impact of a disaster event. The article does not discuss in detail ‘medical necessity’ in the emergency context.4 Where medical resources are strained there will usually be agreement about the need to use a triage process to determining priorities for treatment of victims and therefore over allocation of scant resources even if actual triage practice involves difficult judgments and decisions.

Why The Description of the Defence Matters

In the disaster context, victims or potential victims of a disaster may be able to advance either an excusatory or a justificatory version of the necessity defence. In some jurisdictions both may be recognised and may overlap. However, disaster responders whether part of State-organised agencies or spontaneous volunteers will often be restricted to the justificatory form of the defence because they will have come to the disaster scene after the disaster began, rather than directly experiencing it. Where the disaster event extends over a period of time, the excusatory defence might be available in relation to conduct dealing with unpredictable or unpredicted disaster impacts.

Correct description of the defence of necessity is essential because of the messages sent by its use and acceptance to defendants and, more importantly, to the wider community. Experience in Canterbury and elsewhere shows that after large-scale disasters rescue and recovery tasks are too pressing, and too numerous, to be dealt with solely by officials and official responders. It is therefore not merely socially desirable that volunteers come forward after a disaster, be it a bushfire or an earthquake or a major flood, but essential if responses are to be timely and effective. Folklore about the legal position of both official responders and volunteers, including perceptions of potential legal risks, is likely to significantly affect both whether individuals will act to respond to disasters, and the nature of any response. Research suggests that many official responders to the Canterbury earthquakes were unsure of what their powers and immunities were, for example whether they need be concerned about spray paint damage to buildings which had been marked as searched and therefore clear of survivors.5 Better training and education may improve the position, but a wider social understanding of the nature and availability of a necessity-based defence might allow individuals more opportunity to concentrate on disaster response activities.

To simply say a person has a defence of necessity does not distinguish between justification and excuse. While excusatory necessity may well be appropriate for disaster victims or survivors who had no real choice but to offend, disaster responders will more than likely want their conduct to be considered as deliberate, correct and the choice of a lesser evil; and hence, to raise a justification defence. There is a real public interest in promoting an understanding of the difference between the two forms of this defence. Here we may draw upon the arguments for ‘fair labelling’ in the criminal law, although these have more commonly been applied to sentencing rather than to trial and adjudication. Accurate and fair labelling will ‘symbolise the degree of condemnation that should be attributed to the offender and signals to society how that particular offender should be regarded’.6

Chalmers and Leverick7 note that while fair labelling is desirable in the case of successful defences, they argue that the normal practice of an acquittal without reasons prevents the sending of a clear signal to the public. That argument has less force where, as would normally be the case in emergency necessity cases where damage to, or taking of, property was in question, the trial will be by judge alone and a statement of reasons will be available. Any such prosecutions will have a high media profile and therefore signals can be conveyed to the general community which will indicate the manner in which both justification necessity and excuse necessity defences function. Findlay Stark summed up the message which the different findings would send:

If members of society are ready to undertake disaster response activities which would give rise to a justification defence, that is those which do more good than harm, in circumstances where the harm cannot be remedied by state action because of a lack of resources, society can only benefit. Sending signals which encourage such behaviour is therefore highly desirable. If we do not send clear signals, we risk falling into the problem identified by Dan-Cohen;9 that the criminal law may appear to prompt actors to behave in a way which would invite a justification defence while covertly conveying the judges the need to acquit those who should be excused because adherence to the law could not be expected. Clarity in statutory definition of any necessity defences and in judicial articulation of the relevant principles may well mean that such mixed signals can be avoided.

The Defence(s) of Necessity in Different Jurisdictions

The two forms of the necessity defence are conceptually very distinct. The ‘justificatory necessity defence’, or ‘lesser evil defence’ holds that the conduct of the defendant should not be regarded as criminal if the defendant has done ‘the right thing’ and acted in a way which, while it may be in breach of the law, is a proportionate response to the harm which would otherwise occur.10 The ‘lesser evil’ option may be taken if another person’s rights are violated if ‘the rights violated are of a lesser order of importance by comparison with the value being protected’.11 A justificatory defence means that the conduct is lawful, with the consequence that there can be no accessorial liability.12

While technically a defendant has only an evidential onus to adduce evidence raising a defence of necessity, it is convenient to treat the matter as one where the defendant must establish the basis for the defence. In New Zealand, for example, it has been argued13 that a defendant must show a sufficient case in relation to each of James Stephen’s criteria for necessity:14

  1. the act is needed to avoid inevitable and irreparable evil;
  2. no more should be done than is reasonably necessary for the purpose to be achieved; and
  3. the evil inflicted must not be disproportionate to the evil avoided.

The other form of necessity is excusatory necessity, where a defendant faces an imminent or immediate peril and acts in a way which contravenes the law to avoid the foreseen harmful consequence. The alternative description of ‘duress of circumstances’ is tendentious in relation to persons who choose to respond to a disaster rather than being involuntarily affected by it. In excusatory necessity the conduct is unlawful but the actor is excused because the actor could not be expected to withstand the peril and adhere to the law. The New Zealand Court of Appeal accepted in 1991 that this form of defence might apply if:

  1. the actor believed, on reasonable grounds, there was a threat of imminent death or serious injury;
  2. the circumstances were such that the actor had no realistic choice but to break the law;
  3. the breach of the law was proportionate to the peril involved; and
  4. there was a nexus between the threat of imminent peril of death or serious injury and the choice to respond to the threat by breaking the law.15

This simple division is not reflected in the law of different jurisdictions. In some countries, such as New Zealand, both forms exist and in others statutory provisions blur the lines between the different forms. We may however classify the different jurisdictions as follows.

Excusatory Necessity Only

In the landmark case of Perka v The Queen16 the Supreme Court of Canada recognised an excusatory necessity defence at common law. This requires a situation of imminent peril or danger, that the accused had no reasonable legal alternative to the course of action he or she undertook and the harm avoided by breaking the law was proportionate to the harm inflicted. In those circumstances the accused can be treated as acting in a ‘“morally involuntary’ way. That view was later confirmed in R v Latimer.17 Although Wilson J in Perka v The Queen expressly argued that a justificatory necessity defence co-existed with the excusatory defence,18

the dominant discourse is about excusatory necessity.19

In both Queensland and Northern Territory similarly-worded statutory provisions clearly treat emergency response necessity as a form of excuse.20 They provide that a defendant is excused from liability where there is a sudden emergency which caused a pressure to act which an ordinary person in the same circumstances could not have withstood.

Justificatory Necessity Only

In the United States of America the dominant view of necessity as a matter of justification or ‘lesser evils’. As summed up by Shaun Martin, the position is:

That stance, Martin argues, reflects a maximisation of net social utility.22

In the United Kingdom the defence is also usually seen as justificatory, and is often described in terms of the test postulated by James Stephen – set out above in the discussion of New Zealand law – and adopted by Brooke LJ in the conjoined twins case, Re A (Children).23

The justificatory version of the statute has been embodied in both Victoria and Western Australia, in similar terms which require that the defendant have acted reasonably in response to an emergency.24

Both Excusatory and Justificatory Necessity

As noted, New Zealand has recognised both forms of the defence. Although the decided cases seem almost exclusively to be concerned with excusatory necessity, it appears the common law in New South Wales, South Australia and Tasmania is more or less the same as in New Zealand.25

In the Commonwealth of Australia (in relation to crimes against the commonwealth) and the Australian Capital Territory the defence has been put into statute. The identically worded provisions do not clearly state a rationale for the defence.26 This may have been a deliberate tactic to avoid disputes as to the basis for the defence. In any case it is likely that both forms of the defence are available.

The Volunteer Problem

A necessity defence will only work properly in the emergency context if it deals both with individuals who are compelled by circumstances to act (for example the disaster victim who must find food for herself or persons for whom she is responsible) and for the person who chooses to be a volunteer in disaster response (for example the person who commences digging into rubble to rescue earthquake victims or who collects material which may be of use to rescuers or to other victims). This has a major implication for designing the defence, because we may want the defence to cover people who do not provide assistance directly to disaster victims but rather to rescuers. Take the following hypothetical:

Should a complaint be made about Petra’s taking of the food and water in these circumstances, would Petra have a necessity defence to a charge of theft? If so, would it be a justificatory or an excusatory defence? It is clear that Petra need not have acted at all. She could have left it to the authorities to provide sustenance for the rescue workers. Further, it is most likely that the rescue effort would have continued without her conduct. However there is no question that the rescue workers would be refreshed, encouraged and ready to continue if supplied with food and water. As such, rescue of persons trapped in the rubble is likely to be swifter, so that the victims will have a greater chance of survival or of less serious complications to their injuries. In that sense, there is a good argument that Petra’s act should be considered justified on a lesser of two evils basis. The invasion of the rights of the shop owner is clear, but it is surely less significant than the potential benefit to the earthquake victims. There is also a good argument that we would wish this kind of behaviour to be replicated in the future. If so, a justification defence will send a better signal to the general public as to how to behave than would an excuse defence.

The ‘Stranger’ Problem

It has been argued, judicially and by academics, that any justificatory necessity defence cannot extend to acts done to assist ‘strangers’, something which would largely exclude volunteer responders from the defence. Ian Dennis puts forward a restrictive view:

Note the qualifications that the ‘emergency’ context is equated with self-preservation and, more remarkably, that an altruistic impulse to save someone applies only where there is a duty of care. Surely altruism is inconsistent with any element of a duty. Where D is under a legal duty to protect V, or more precisely to take reasonable care to ensure V is not harmed, it would be unjust to convict D for conduct in breach of the law if this is the only way in the circumstances that V could be protected from harm. But the real argument here is not a true necessity ‘justification’ argument of prioritising the harms that may flow from the conduct, but rather of prioritising the duties to which D is subject.

A narrow view was also taken by Wilson J in Perka v The Queen, expressly ruling out a justification necessity defence for the rescue of persons to whom the actor owed no duty:

Similarly, if surprisingly, the New South Wales Court of Criminal Appeal has held that excusatory necessity is only available where the actor was responding to threats to the actor ‘or others that he or she was bound to protect’.29 A rather wider view has been expressed in England, where Lord Woolf CJ considered the defence would apply in cases where an actor is responding to threats to persons ‘for whom the situation makes him responsible’, instancing innocent civilians trapped in a building subject to a bomb threat.30

It is evident that with any adoption of the narrow existing duty approach such view would almost invariably rule out any justificatory necessity argument for volunteer disaster responders. Even Lord Woolf’s extension of it would be hard to apply to responders who come onto the scene of a disaster after it has occurred. It seems impractical, as well as profoundly illogical, to limit a justification defence in this way. If the ‘lesser evil’ is the breaking of the law and the ‘greater evil’ is the death of a ‘stranger’ to the rescuer, breach of the law must surely be considered to be justified.


As the preceding discussion makes clear, where justificatory necessity is raised, the criminal law courts must determine whether the defendant’s actions were designed to achieve the lesser of two evils – that is, that the harm caused was proportionate to the benefits intended to be gained.31 While the principle is clear, applying it is not straightforward. In particular, there has been a tendency to posit a simple binary test that avoiding the loss of human life is more important than avoiding harm to property of any kind. A more nuanced approach is needed.

Human Life or Lives vs Human Life or Lives

In common law systems there is a clear consensus that one human life may be sacrificed to save the lives of a greater number. This is particularly so if it can be said that the sacrifice has ‘chosen’ herself or himself as the sacrifice by words or conduct, as in discussions of the Zeebrugge ferry incident where a young man who had for ten minutes blocked others from reaching safety via a rope ladder was brushed off the ladder so that many others could reach safety.32 In such a case, a justificatory necessity defence would be available even to a charge of homicide. Other systems of law do not necessarily adopt the same approach, and in Germany the courts have expressly rejected a ‘proportionality’ test which required balancing the loss of one or more human lives against the loss of an equivalent or greater number of lives.33

However there has been little discussion of the complexities that arise where the question is whether one or more human lives can properly be sacrificed to protect property which may enable a far greater number of lives to be saved in the long run. For example, is it justifiable to prioritise recovering significant quantities of medicines over rescuing a single trapped victim if the medicines will probably, in the long term, preserve the lives of a significant number of people? Or to demolish a building in which several persons are trapped if this is the only way to avoid the spread of a fire which will destroy the only hospital at which disaster victims can be treated, something which would cause a greater number of people to die from inadequate medical treatment over the disaster response period? Bronitt and Stephens have commented on legislative provisions which allow hijacked aircraft to be shot down to avoid damage to critical infrastructure that a ‘decision to pre-emptively terminate innocent lives in order to protect property (albeit high value infrastructure) which is remotely connected to protecting human life or limb will be morally controversial’.34 This is undoubtedly true, but it is suggested that in disaster conditions the case for a utilitarian calculus is stronger, and thus it may be necessary to consider whether, overall, more lives will be saved than will be lost.

Taking Property to Save Lives

On the analysis above, it must generally be permissible for disaster survivors or responders to take or use property to sustain their own lives or to save the lives of others. In such cases, the element of necessity is usually seen as separating ‘self-help’ taking by disaster victims of food or other necessaries from less acceptable forms of looting. Stuart Green argues that that looting spans a range from the deliberate taking of goods for the purposes of sale and monetary gain, which may be seen as a particularly serious form of burglary, or theft, to:

While looting of the first kind is common during civil disturbances such as riots, it is rare in the aftermath of natural disasters.36 The opposite form of looting might well be seen by members of the general public, and most lawyers, as a situation in which a defence of necessity ought to exist. It is not clear that it always does.

At the simplest and most meritorious end of the spectrum, a person should not be held to commit any offence by taking medicines which must be stored at cold temperatures to retain effectiveness from a pharmacy without electricity and storing them in better conditions for later use. A fortiori, a pharmacist should be seen as justified if she takes an unused generator, for which there is no more urgent demand, so as to restore power to the medicine refrigerator.

Proportionality is harder to measure where the property is taken for the takers’ own benefit. By taking food or medicine or shelter materials, the takers are placing themselves in a better state than others who might otherwise have accessed the same goods or medicines. If those other persons were hungrier, or more seriously ill, and thus in greater need of the medications, the taking clearly is to the detriment of those in greater need and so greater harm is caused. True in some cases there may only be a limited number of persons involved and no likely competitors for the resources, but in many disaster events or emergencies this will not be so.

Where similar acts are done to assist others, justificatory necessity should in principle be accepted. If a health professional takes prescription drugs from an abandoned and earthquake damaged pharmacy to treat seriously injured emergency victims, the benefit is more than proportionate to the harm suffered by the pharmacy owner. The same is probably true where the drugs are taken and used to treat victims by a person who had no medical expertise, except there the risk of harm to victims who are given inappropriate medication must also be taken into account. However, it is much harder to determine the proper position of the person, whether disaster victim, official responder or volunteer responder, who has taken the resources for the potential benefit of other victims, as where medicines are taken to be given to a doctor or nurse, but no such person is encountered and the medicines are never used. Here the harm is the loss of their potential use by a qualified person. Is the taking justified? Or is it something to be condoned, and therefore brought within the excusatory necessity defence if criminal liability is not to be imposed?

Analysis is made more difficult where property is taken in quantities seen by the taker as sufficient ‘to last until the emergency is likely to end’. Can the taking of more food or medication than is sufficient for immediate needs be justified? There may be some cases where the end of the disaster is reasonably predictable. Experience may indicate how long a flood may last; equally it may indicate the point at which roads may be reopened after blizzards or avalanches. However in many cases disaster survivors will not know, indeed cannot know, how long it will be before external resources become available so that self-help is superseded. If necessity is seen as an excuse, a court will be likely to accept the disaster victim’s own assessment of whether it was sensible, to use a more neutral word than reasonable, to take the food or medicine rather than wait for supplies to be brought in and become lawfully available. However if the issue is justification, it surely must be required that the survivor has no reasonable belief that too much time will elapse before alternative supplies become available.

Taking Property to Save Property

Public policy is best served if a justificatory defence is available to any post-disaster conduct which removes or reduces a serious risk of damage to, or destruction of, property while not causing disproportionate harm.37 Thus, a person who takes timber, plastic sheeting or tarpaulins from a disaster-damaged shop and uses them to weatherproof another building in order to protect property which might otherwise be at risk, may be able to claim his or her conduct is justified and therefore not criminal provided no greater harm has been caused. Judging proportionality in such cases is not simple. At first sight we may use a utilitarian calculus, as in Simon Gardner’s example: ‘I break your window (replacement cost, £100) so as to save my chair (replacement cost, £200) from a fire’.38

However, as Gardner accepts, such a calculus omits any consideration of any sentimental value that may attach to particular property. Moreover in a disaster or emergency situation a comparison in solely fiscal terms ignores the different levels of importance attaching to items which are, or are not, important to survival or recovery. Not only may there be no efficient market to provide replacement items, but in a post-disaster situation priorities, and values, may differ sharply from the normal. Let us suppose that the window, replacement cost in normal times $200, is broken to enable the salvage of an artwork valued at $5000 but there are no available replacement windows, so the building cannot be made weatherproof. What is the real value of the damage to the window? It may be a very high cost if as a result the building becomes unsuitable for the storage of perishable foodstuffs, or impossible to heat and therefore uninhabitable. Normal market price is not a suitable criterion, so a court must assess the social benefits that might have flowed from potential alternative uses of the building.

Further, as a matter of practicality, much depends on whether the actor was authorised or instructed by the civil authorities to take or use the relevant property. If so, there may be some possibility of invoking statutory immunities conferred upon disaster response agencies and those working for them or under their auspices. The issue of such immunities requires extended discussion.

Statutory Immunities or Defences and the Fit With Necessity

In many jurisdictions the legislature has provided a degree of immunity or protection to members of emergency response agencies, including police, and to persons working under the direction of such an agency. For example, the Civil Defence Emergency Management Act 2002 (NZ) (CDEMA) provides that no liability will attach to conduct damaging a dangerous structure in the course of demolishing it under the orders of the civil authorities where a state of emergency has been proclaimed.39 Other sections of the Act authorise particular forms of behaviour by a civil defence controller, a police officer or any person acting under the authority of a controller or police officer where a state of emergency is in force. Thus, under s 87 a controller or police officer or a person acting under their authority may enter onto, and if necessary break into, any premises in the area over which the state of emergency extends if the actor believes on reasonable grounds that doing so is necessary to save the life of any person, or to prevent injury to that person or to rescue an injured or endangered person. Another section authorises the removal from any place of a vehicle which is impeding civil defence management and allows the vehicle to be broken into for that purpose.40 There is also a very broad power to requisition virtually any kind of property where this is believed to be necessary by a civil defence controller or a constable.41 Similarly, no liability will attach to any person damaging a dangerous structure in the course of demolishing it under the orders of the civil authorities where a state of emergency has been proclaimed.42 While this kind of statutory protection may resolve some issues, many problems remain.43 Two major ones are of direct concern.

Spontaneous Volunteers

In every jurisdiction so far inspected, any statutory protections do not extend to ‘spontaneous’ volunteers – those with no prior organisational who simply assist with disaster management or recovery because they feel needed – but only to people who are acting under the auspices of the regular disaster management structure. However, in almost all large-scale emergencies there will be spontaneous volunteers and it is clearly valuable to society that their skills and energies are available. If statute does not protect them, it is desirable the common law does.

Secondly, and perhaps less understandably, the statutory immunity only extends to activity which takes place after the declaration of a state of emergency. However it is highly likely that both local emergency service personnel and spontaneous volunteers will be acting well ahead of the declaration of emergency as well as afterward. In the Canterbury earthquakes, the delay was of the order of a few hours; in other cases the delay might well be longer as communications may be disrupted completely. There is no provision for retrospective validation of any acts done prior to the state of emergency being declared. Retrospectivity brings its own problems, but it may be necessary.

The New Zealand statute does not provide expressly for conduct done with a good faith but erroneous belief that what was done was empowered by CDEMA or other statutes and under s 25 of the Crimes Act 1961 a mistake of law will be no defence to any criminal charge. However a mistaken belief of fact (for example, that a building is dangerous and must be demolished) should, on general principles, provide a defence.44

Excusatory Necessity, Reasonableness, Reasonable Firmness and Disaster Psychology

It is frequently stated that an excuse of necessity is not acceptable unless the actor has met a socially acceptable level of steadfastness in adhering to the law. If most people would resist the pressure and adhere to the law, those who do not are in essence blameable and should not be excused. Thus the defendant must have displayed an ordinary person’s level of firmness in the face of the peril. Duff sees this as essentially a statement of a normative standard to which citizens can be expected to adhere:

Implicit in this approach is that the law limits the defence to those who deserve sympathy because their reasonable efforts to adhere to the law have been overborne. As Westen and Mangiafico put it: ‘It holds actors to the firmness – the steadfastness to avoid wrongdoing-that the law believes they ought to possess under the circumstances’.46

While disaster responders are likely to have more than average powers of control and commitment or they would not undergo the risks and hardships of emergency response or rescue, disaster victims will include many people who are of less than ordinary firmness and will try to escape the area of a disaster or minimise its impact even if doing so means acting unlawfully. For example, people who have been through one earthquake may well be ready to flee an area of seismic activity quickly, even if this means driving at unsafe speeds or taking somebody else’s transport. Indeed they may be hyper-vigilant and readier to see dangers than would people who have not been through the disaster process. Arguably a truly excuse-based defence would recognise such increased vulnerability.

The Reasonableness Element and the Psychology of Disasters

Research findings suggest that it may be quite unreasonable to expect all disaster victims to behave in a ‘reasonable manner’ or in line with ‘reasonable people’s’ expectations. For the most part major disasters are outside the experience of ordinary people. Those who have experienced disasters before may be readier to respond in a way which a hypothetical reasonable person who had not experienced the same prior stress and possible trial would consider unreasonable. Many of the persons affected by the February 2011 Canterbury earthquake were responding in ways conditioned by several months of prior seismic activity. The same may be expected for persons who have had direct experience of a bushfire or similar emergency.

A full survey of the psychological literature would be out of place here, but some of the research into the effects of the Canterbury earthquakes of 2010-2011 is enlightening. One writer described the cumulative effect of those quakes thus:

A group of researchers state that psychosocial difficulties linked to the aftermath of different natural disasters include:

More specifically, the researchers found that a majority of their research subjects revealed an element of hypervigilance or being constantly on the alert for the next earthquake, something which is ‘likely to manifest in individuals being hypersensitive and alert to the slightest sounds, sensations and movement’.49 It is highly likely that such people will respond to a possible risk in ways which a person without that prior experience may regard as unreasonable. This is not to suggest that the mental effects of a disaster would trigger a defence of insanity or diminished responsibility. The issue is whether an excusatory necessity defence should take into account the psychological impact of prior experience on the defendant. Z offends by driving at an excessive speed away from a disaster scene, or by failing to stop at a bushfire checkpoint, because of Z’s prior experience of similar disasters. Z might justly wish to argue that prior experience caused a level of hypervigilance which resulted in actions which a person without such a background and emotional state would see as reasonable. Is not Z’s behaviour all the more excusable because of the compounding impact of the prior and current events? Yet on the current law a ‘reasonableness’ requirement may well preclude people significantly affected by such prior experience from relying on the defence.50


This article has identified a number of areas in which the defences of necessity, whether excusatory or justificatory, are not easily applied in cases arising from natural disasters or other emergencies. The range of difficulties identified suggests there is an urgent need to revisit the ways in which both the courts and the legislatures have formulated the defences. Confusion as to the underlying rationales for the different forms of the defence has led to the defence being stated in terms which are likely to restrict its application to disaster response conduct, and may discourage volunteer responses to such disasters. Judges might well be assisted by adopting a more structured approach to determining whether a defendant’s response to an urgent peril should be seen as justified and therefore raising a justificatory defence. Thus a judge might ask first whether the defendant was pursuing a legitimate goal (preservation of life or property or minimisation of harm to a person or to property), secondly whether the actions taken had a rational connection to that goal, thirdly whether there were alternative courses of action which would have caused lesser harm to the interest of those affected by the actions, and lastly whether the defendant’s conduct imposed a disproportionate burden on others in that the damage to their interests was disproportionate to the benefits conferred by the defendant on herself or others.51 Legislators considering statutory change might well analyse the issues to be covered in a similar way. However it may occur, reform is necessary – and overdue.


  • 1. See Jeremy Finn and Elizabeth Toomey (eds), Legal Consequences of Natural Disasters (Thomson Reuters, 2015).
  • 2. One of the few articles which purports to discuss necessity defences in the context of emergency response is by Michelle Cotton, ‘The Necessity Defense and the Moral Limits of Law’ (2015) 18 New Criminal Law Review: An International and Interdisciplinary Journal 35. Unfortunately the paper discusses only a limited sample of emergency situations.
  • 3. Paul Robinson, Structure and Function in Criminal Law (Clarendon Press, 1997) 70. While this distinction is generally accepted, Robert Walker LJ considered it did not matter if necessity-based defences were regarded as justifications or excuses: Re A (Children) (Conjoined Twins, Surgical Separation) [2000] 4 All ER 971, 1063.
  • 4. The leading judicial application is in the conjoined twins case: Re A (Children) (Conjoined Twins, Surgical Separation) [2000] 4 All ER 961.
  • 5. Steve Glassey, ‘Analysis of Urban Search and Rescue Markings Applied Following the 22 February 2011 Christchurch Earthquake’ (2013) 1 Journal of Search and Rescue 29, 41.
  • 6. James Chalmers and Fiona Leverick, ‘Fair Labelling in Criminal Law’ (2008) 71 Modern Law Review 217, 226.
  • 7. Ibid, 245-246.
  • 8. Findlay Stark, ‘Necessity and Nicklinson’ [2013] Criminal Law Review 949, 951.
  • 9. Meir Dan-Cohen, ‘Decision Rules and Conduct Rules: On Acoustic Separation in Criminal Law’ (1984) 97 Harvard Law Review 625.
  • 10. Proportionality is discussed in Part VI.
  • 11. Andrew Simester and Warren Brookbanks, Principles of Criminal Law (Thomson Reuters, 4th ed, 2012) 453.
  • 12. The case law is extensively recounted in R v Quayle [2005] EWCA Crim 1415, [35]-[53].
  • 13. Simester and Brookbanks, above n 11. A similar view was taken by the High Court of Australia in Taiapa v The Queen (2009) 240 CLR 95.
  • 14. Sir James Fitzjames Stephen, A Digest of the Criminal Law (Crimes and Punishments) (MacMillan, 4th ed, 1887) 24-25, a view adopted in many cases, for example, by Brooke LJ in Re A (Children) (Conjoined Twins, Surgical Separation) [2000] 4 All ER 961, 963.
  • 15. Kapi v Ministry of Transport (1991) 8 CRNZ 49, 57.
  • 16. Perka v The Queen [1984] 2 SCR 232.
  • 17. R v Latimer [2001] 1 SCR 3.
  • 18. Perka v The Queen [1984] 2 SCR 232, 276.
  • 19. Edward M Morgan, ‘The Defence of Necessity: Justification or Excuse?’ (1984) 42(2) University of Toronto Faculty Law Review 165.
  • 20. Criminal Code (Qld) s 25; Criminal Code (NT) s 33.
  • 21. Shaun P Martin, ‘The Radical Necessity Defense’ (2005) 73(4) University of Cincinnati Law Review 1527, 1535-1536. See also Stephen S Schwartz, ‘Is There a Common Law Necessity Defense in Federal Criminal Law?’ (2008) 75(3) University of Chicago Law Review 1259; Michele Cotton, ‘The Necessity Defense and the Moral Limits of Law’ (2015) 18 New Criminal Law Review: An International and Interdisciplinary Journal 35.
  • 22. Martin, ibid.
  • 23. Re A (Children) (Conjoined Twins, Surgical Separation) [2000] 4 All ER 961, 1052.
  • 24. Crimes Act 1958 (Vic) s 322R; Criminal Code (WA) s 25. Section 322R of the Victorian Act as enacted in 2014 replaced the earlier s 9AI but is in essentially the same terms.
  • 25. R v Loughnan [1981] VR 443 (SC); R v Rogers (1996) 86 A Crim R 542 (NSWCA); and Taiapa v The Queen [2009] HCA 53, [36]-[37], respectively. See also Behrooz v Secretary of Department of Immigration and Multicultural and Indigenous Affairs [2004] HCA 36, [15].
  • 26. Criminal Code (Cth) s 10.3; Criminal Code (ACT) s 41.
  • 27. Ian Dennis, ‘On Necessity as a Defence to Crime: Possibilities, Problems and the Limits of Justification and Excuse’ (2009) 3(1) Criminal Law and Philosophy 29, 48. Curiously Dennis does not, in a broad survey of academic views of necessity, refer to an important article by Winnie Chan and Andrew Simester, ‘Duress, Necessity: How Many Defences?’ (2005) 16(1) King’s Law Journal 121.
  • 28. Perka v The Queen [1984] 2 SCR 232, [276].
  • 29. Mattar v The Queen [2012] NSWCCA 98, [7].
  • 30. R v Shayler [2001] 1 WLR 2206 (CA) at [49].
  • 31. This article deals only with the criminal law, but in some cases involving justificatory necessity guidance may be gained from the defence of necessity as it applies in tort law. This permits damage to property where this is ‘reasonably necessary’ to save human life:Esso Petroleum Co Ltd v Southport Corporation [1956] AC 218, or to avoid or minimise damage to property, whether belonging to the defendant or to another: Cope v Sharpe (No 2) [1912] 1 KB 496; The Saltpetre Case (1600) 12 Co Rep 12, 13. This is so even if later events or the conduct of others dispelled the danger, or otherwise prevented, the anticipated harm from occurring: Cope v Sharpe (No 2) [1912] 1 KB 496; New South Wales v McMaster [2015] NSWCA 228. Note that the criminal law justificatory defence has been stated in narrower terms in some jurisdictions and, of course, tort law has no equivalent of excusatory necessity.
  • 32. JC Smith, Justification and Excuse in the Criminal Law (Stevens, 1989) 77-78.
  • 33. Michael Bohlander, Principles of German Criminal Law (Hart Publishing, 1st ed, 2008) 106-109. For further discussion of this issue, see Michael Bohlander, ‘In Extremis – Hijacked Airplanes, “Collateral Damage” and the Limits of the Criminal Law’ [2006] Criminal Law Review 579, 592; Kai Moller, ‘On Treating Persons as Ends: The German Aviation Security Act, Human Dignity and the German Federal Constitutional Court’ [2006] Public Law 457; Simon Bronitt and Dale Stephens, ‘Flying Under the Radar – The Use of Lethal Force Against Hijacked Aircraft: Recent Australian Developments’ (2007) 2 Oxford University Commonwealth Law Journal 265, 269.
  • 34. Bronitt and Stephens, above n 33, 275.
  • 35. Stuart P Green, ‘Looting, Law, and Lawlessness’ (2007) 81 Tulane Law Review 1129, 1132-1133.
  • 36. EL Quarantelli and Russell R Dynes, ‘When Disaster Strikes (It Isn’t Much Like What You’ve Heard & Read About)’ (1972) 5 Psychology Today 67, 67-68; Caroline C Nobo and Rebecca D Pfeffer, ‘Natural Disasters and Crime: Criminological Lessons from Hurricane Katrina’ in Rob White (ed), Climate Change from a Criminological Perspective (Springer, 2012) 181.
  • 37. The law of tort clearly allows this, see Cope v Sharpe (No 2) [1912] 1 KB 496; The Saltpetre Case (1606) 12 Co Rep 12, 13.
  • 38. Simon Gardner, ‘Direct Action and the Defence of Necessity’ [2005] Criminal Law Review 371, 376.
  • 39. For these powers, see Civil Defence Emergency Management Act 2002 (NZ) (CDEMA) s 85(1).
  • 40. CDEMA s 89.
  • 41. CDEMA s 90.
  • 42. CDEMA s 85(1).
  • 43. For a fuller discussion, see Jeremy Finn and Elizabeth Toomey, ‘Providing for legal issues in disaster management: lessons from New Zealand and the USA’ (Paper Presented at Proceedings of ANZDEM Conference,Gold Coast, 5-7 May 2014) 28, 46.
  • 44. The applicability of a necessity defence where the defendant mistakenly believes a peril exists is too large an issue to explore here. It is notable that in the law of tort a mistaken belief in the existence of the peril is fatal to the justificatory defence: Cope v Sharpe (No 2) [1912] 1 KB 496, 504; New South Wales v McMaster [2015] NSWCA 228, [222].
  • 45. Antony Duff, ‘Choice, Character, and Criminal Liability’ (1993) 12 345, 358.
  • 46. P Westen and J Mangiafico, ‘The Criminal Defense of Duress: A Justification, Not an Excuse - And Why It Matters’ (2003) 6 Buffalo Criminal Law Review 833, 908.
  • 47. Marie Crowe, ‘“Quake Brain”; Coping with the Series of Earthquakes in Christchurch’ (2011) 20 International Journal of Mental Health Nursing 381, 381.
  • 48. Casey Rowney et al, ‘“I Laugh and Say I Have Earthquake Brain!”: Resident Responses to the September 2010 Christchurch Earthquake’ (2014) 43(2) New Zealand Journal of Psychology 4, 4.
  • 49. Ibid, 7-8. Studies after the earthquakes also indicate a likely correlation between severe impact of the disasters on an individual and errors in carrying out set tasks – in essence, that such victims are more likely to process information wrongly or to make slips in performing actions, see William S Helton et al, ‘Natural Disaster Induced Cognitive Disruption: Impacts on Action Slips’ (2011) 20 Consciousness and Cognition 1732. The consequences of such an impact fall outside the current study, but are of obvious relevance when considering liability for negligence.
  • 50. Stanley Yeo has advocated a test of what is reasonable for the particular defendant. See, for example, Stanley Yeo, ‘Compulsion and Necessity in African Criminal Law’ (2009) 53 Journal of African Law 90; Stanley Yeo, ‘Revisiting Necessity’ (2010) 56 Criminal Law Quarterly 13, 30. Such a test would be difficult to apply in practice.
  • 51. This proposal is derived from Kai Möller’s suggested approach to determining proportionality in the area of human rights, see Kai Moller, ‘“Balancing as Reasoning” and the Problems of Legally Unaided Adjudication: A Rejoinder to Francisco Urbina’ (2014) 12(1) International Journal of Constitutional Law 222. I am grateful to the anonymous reviewer for this journal who alerted me to this discussion.

Send mail to Author

Send Cancel