TEACHING CAUSATION IN CRIMINAL LAW: LEARNING

TO THINK LIKE POLICY ANALYSTS

BRENDA MIDSON*

I INTRODUCTION
One of the most persistent ideals in the context of legal education is that of teaching
students to think like lawyers.1 While the precise range of cognitive skills that enable
one to think like a lawyer may be subject to much debate, 2 one of these skills is
undoubtedly the ability to extract legal principles from cases and statutes and apply them
to the facts of a legal problem.3
It has become apparent through teaching causation in criminal law that, while
extracting and applying the law from cases is easy enough when the principles are clear,
students often struggle when relying on cases in which judges employ unexpressed
policy-based reasoning. In the context of causation, James Gobert argues that:
[T]he struggle that courts and commentators have had with causation issues may indicate either
that causation is a much more complex phenomenon than the questions might suggest or that the
questions are the wrong ones to be asking. Obtuse and strained analyses often appear in cases
involving intervening causes in order to allow the courts to impose sanctions on a defendant
thought to be deserving of punishment or to avoid imposing liability on a defendant not thought
to be so deserving.4

Law students must be able to extrapolate outcomes from legal principles and given
facts. When policy grounds for decision-making are not clearly articulated, students
struggle to find the law to apply. As Gobert identified, these difficulties are not often
resolved by a reading of the case law, which illustrates that cases with almost identical
factual matrices can result in different outcomes. This is particularly apparent in homicide
cases, where medical treatment is asserted as breaking the chain of causation. Using some
of these cases as examples, this paper proposes that, in complex cases, the question of
causation cannot be answered simply by applying the legal principles without reference to
a range of policy considerations. To students, these are invisible factors in judicial
decision-making. They account for the variation of outcomes that occur in the application
of causation principles but, because they are not explicitly referred to in case law, it is
almost impossible for students to employ them in problem-solving. The challenge for
legal education is to teach legal reasoning so that students are better able to identify and
apply unarticulated policy reasons.
The question of causation in criminal law provides an instructive example of judges
use of unexpressed policy reasoning and the challenges that this poses for the teaching
and learning of legal reasoning and problem-solving. In demonstrating how the doctrine

of causation can be used to teach reasoning skills to students more effectively, this paper
begins with an analysis of the causation principles that purport to apply in
Commonwealth jurisdictions, followed by an analysis of case law in which the
application of these principles has resulted in disparate outcomes. In turn, this gives rise
to a discussion of the way in which unexpressed policy-based reasoning thwarts the
attempts of students to distil the law from cases. Because the principles that are held to
apply are often expressed in imprecise terms, there is little guidance for students on what
is actually required to break the chain of causation. The role of legal education is to bridge
this gap in legal reasoning and, thus, Part IV below offers suggestions as to how law
curricula could incorporate methods by which students can improve their identification
and use of policy-based arguments. This includes helping students to look beyond explicit
ideas and consider underlying values and policy concerns that impact upon courts
reasoning. It is proposed that effectively teaching legal reasoning involves teaching
enhanced case analysis skills which students can apply within a problem-based learning
method.

II CAUSATION
While there are numerous limitations and exceptions to the principle, it remains
fundamental to Anglo-Saxon criminal law that liability arises out of the proof, beyond
reasonable doubt, of the accuseds commission of the actus reus of an offence, while
concurrently possessing the requisite mens rea.5 But proof of the actus reus and mens rea
is not always sufficient to establish liability. In a number of offences, the prosecution
must also prove that the accuseds act caused a particular result. A clear example is in
homicide cases, where the act of the accused must have caused the death of the victim. In
the majority of homicide cases, establishing causation is uncomplicated because it is not
disputed that, for example, the infliction of grievous bodily injury by the accused caused
the death of the victim. Other cases prove to be more difficult, particularly where there is
an intervening event a novus actus interveniens or where there are multiple causes
of death. In such cases, it may be that the act of the accused is not legally causative of
death, even though a simple application of the but-for test would suggest otherwise.
Whether the death of a victim was caused by an act of the accused is a question of fact
for the jury, but the jury can make this decision only in accordance with the legal
principles explained to them by the judge. 6 In Commonwealth jurisdictions, there have
been a number of approaches taken to determining issues of causation where there is more
than one proximate, or immediate, cause of death. Two of these approaches have been
particularly dominant: the reasonable foreseeability test and the substantial cause test.
The reasonable foreseeability test, which asks whether any intervening event was a
reasonably foreseeable consequence of the accuseds actions, 7 was applied by Brennan
and McHugh JJ in Royall v R.8 Brennan J, relying on the English case of R v Roberts,9
said:
Foresight or reasonable foreseeability marks the limit of the consequences of conduct for which
an accused may be held criminally responsible. 10

In Roberts, the victim jumped from a moving car to escape an assault from the
accused. The accused was charged with assault occasioning actual bodily harm.
Stephenson LJ set out the test for establishing whether the actions of the accused had
caused the victims injuries in the following terms:
Was it a natural result of what the alleged assailant said and did, in the sense that it was
something that could reasonably have been foreseen as the consequence of what he was saying
or doing?11

In applying that test, the English Court of Appeal found that the accused could
reasonably have foreseen that the victim would have jumped from the car, and he had
therefore caused her injuries.
The substantial cause test for causation asks whether the act of the accused is a
substantial and operating cause of death. The primary authority for the substantial cause
test is R v Smith.12 Smith was convicted of murder. One of the grounds upon which he
appealed his conviction was that the jury had been misdirected on causation. Smith had
stabbed a fellow soldier, Creed, with a bayonet, causing one wound in the arm and one in
the back. In respect of the latter wound, the bayonet had pierced the lung and caused a
haemorrhage. Following the stabbing, another soldier attempted to carry Creed to the
medical station, but on the way dropped him twice. At the medical station, staff were
trying to deal with a number of other cases, including two other serious stabbings. They
did not appreciate the seriousness of Creeds injuries. He received some treatment,
including oxygen and artificial respiration, which in the light of the piercing to the lung,
turned out to be thoroughly bad treatment. 13 He died approximately two hours after the
original stabbing. There was evidence that had Creed received immediate and different
treatment he might not have died, and indeed that his chances of surviving were as high as
75 per cent.14 The case was decided on the principle that
if at the time of death the original wound is still an operating cause and a substantial cause, then
the death can properly be said to be the result of the wound, albeit that some other cause of death
is also operating. Only if it can be said that the original wounding is merely the setting in which
another cause operates can it be said that the death does not result from the wound. Putting it
another way, only if the second cause is so overwhelming as to make the original wound merely
part of the history can it be said that the death does not flow from the wound. 15

The Queens Bench Division held that the direction to the jury had been correct, and
that no reasonable jury, properly directed, could come to any conclusion other than that
death resulted from the original wound.
Royall16 is the seminal Australian case on causation. Royall was charged with murder
after his girlfriend, Kelly Healey, fell from the window of a sixth-floor flat. He admitted
assaulting Healey during the course of a violent argument but said she then locked herself
in the bathroom and that she jumped out of the bathroom window when he broke the door
down to check on her. The prosecution relied upon three scenarios in arguing Royall was
guilty of murder: that he pushed Healey; that she fell in the course of avoiding his attack;
or that she jumped because of a fear of life-threatening violence from him. The case was
left to the jury to decide whether Royall had caused Healeys death by any of the three
means alleged by the prosecution. One of the grounds upon which Royalls appeal was

based was that the judge had given erroneous directions on the issue of causation.
In the High Court, both the substantial cause and reasonable foreseeability tests were
discussed in some detail, along with the natural consequence test. It is interesting to note
that Mason CJ cited Roberts17 as authority for this latter test.18 He set out the test in the
following terms:
where the conduct of the accused induces in the victim a well-founded apprehension of physical
harm such as to make it a natural consequence (or reasonable) that the victim would seek to
escape and the victim is injured in the course of escaping, the injury is caused by the accuseds
conduct.19

The majority of the judges determining the appeal in Royall20 favoured the natural
consequences test. New Zealand courts have adopted and applied Smiths21 substantial
cause test,22 as have courts in Australia and Canada. 23 The utility of the Smith24 approach
is that it applies both to establish a causal link and to establish that the link was
maintained in cases where there are multiple causes or intervening causes. 25 Under this
test, the chain of causation is not broken unless the act of the accused is no longer a
substantial and operating cause of death. That is, it is only if the subsequent event is so
overwhelming as to make the initial wound merely part of the history that the chain of
causation will be held to be broken.26
At first glance, the legal principles which determine causation might seem clear. For
the student, the problem usually arises in attempting to employ the principles in problemsolving. Indeed, in Royall, McHugh J noted:
Judicial and academic efforts to achieve a coherent theory of common law causation have not
met with significant success. Perhaps the nature of the subject matter when combined with the
lawyers need to couple issues of factual causation with culpability make achievement of a
coherent theory virtually impossible.27

That academic and judicial attempts to achieve a coherent theory of causation have not
been wholly successful foreshadows the difficulties that students face in attempting to
identify the relevant principles to apply. This point is apposite to the need for changes to
legal education approaches, which is discussed in detail in Part IV. To set the scene for
that discussion, the following section canvasses cases in which medical treatment is
asserted as breaking the chain of causation between the accuseds infliction of injury and
the death of the victim. As will be shown, the resulting decisions cannot be explained
simply by reference to the application of the legal principles.

AR V JORDAN28
The appellant stabbed the victim, Beaumont, in the abdomen. Beaumont died eight
days later. The stab wound had penetrated the intestine in two places but, by the time of
death, both injuries had mainly healed. In the meantime, the medical staff administered an
antibiotic, Terramycin, to Beaumont with a view to preventing infection. Beaumonts
intolerance to the drug was discovered after the initial doses, at which time administration
of the drug was stopped; however, another doctor ordered its resumption the following
day. Evidence of two doctors called by the appellant was to the effect that the treatment of
the patient in this way was palpably wrong, 29 as was the intravenous introduction of

wholly abnormal quantities of liquid,30 which led to pulmonary oedema then bronchopneumonia, from which Beaumont died.
The Court of Criminal Appeal drew a distinction between normal treatment and
palpably wrong treatment, and accepted as correct the position that normal treatment
causing death will not negate causation on the part of the person inflicting the original
injury. From a doctrine of precedent perspective, it should be noted that this case was
decided before Smith,31 and the Court in this instance was not prepared to formulate a test
for establishing causation. Hallett J said:
It is sufficient to point out here that this was not normal treatment. Not only one feature, but two
separate and independent features, of treatment were, in the opinion of the doctors, palpably
wrong and these produced the symptoms discovered at the post-mortem examination which were
the direct and immediate cause of death, namely, the pneumonia resulting from the condition of
oedema which was found.32

On that basis, the Court was of the opinion that, if such evidence had been before the
jury, the jury would have felt unable to be satisfied that the death was caused by the stab
wound. In other words, Jordans act did not cause Beaumonts death.

BR V EVANS AND GARDINER (NO 2)33

In this Australian case, the two accused stabbed a fellow prisoner, Hamilton, in the
stomach. The injury was inflicted in April 1974 and, after a bowel resection operation,
Hamilton resumed normal activities, participating in sports activities at Christmas that
year. On 15 March 1975 Hamilton became unwell, and he died on 23 March. The cause of
death was a stricture in the bowel at the site of the resection operation, which is not
uncommon. It was open to the jury at trial to find that the doctors should have diagnosed
the condition and treated it.
The Full Court of the Supreme Court of Victoria applied Smith,34 and held that the real
issue for the jury was whether the blockage of the bowel was due to the stabbing. The
Court was of the view that there was sufficient medical evidence for the jury to support
such a finding. It noted that there were features of the case that made it unusual
namely, that the stab wound was initially treated immediately and in a skilful way, and
that the wound had healed35 and the victim had recovered. 36 However, it is apparent
that these features of the case were not seen by the Court as being sufficiently unusual,
in the Jordan37 sense of that word. Both Evans and Gardiner were convicted of
manslaughter.

CCHESHIRE38
In early December 1987, the appellant shot the victim in the leg and stomach, causing
serious injuries. The victim was operated on and placed in intensive care. While being
treated in hospital, he developed respiratory problems and a tracheotomy tube was placed
in his windpipe. The victim then developed several infections and it was not until early
February 1988 that his condition began to improve. However, by 8 February, he was again
having difficulty breathing and his condition thereafter deteriorated. He died early on 15

February. A post-mortem examination revealed that the victim had suffered from a rare
complication of the tracheotomy namely, a narrowing of the windpipe to the extent
that it caused asphyxiation. The pathologist who conducted the post-mortem gave
evidence that the immediate cause of death was cardio-respiratory arrest
due to a condition which was produced as a result of treatment to provide an artificial airway in
the treatment of gunshot wounds of the abdomen and leg. 39

The defence called its own medical witness to give evidence that, by 8 February, the
wounds of the thigh and abdomen no longer threatened the life of the deceased and his
chances of survival were good, which would seem consistent with the fact that the victim
had shown some improvement. But, according to the Court, precedent established that the
chain of causation will be broken by medical treatment only in the most extraordinary
and unusual case.40 Ultimately, the Court concluded that, even if more experienced
doctors had detected the complication in sufficient time to prevent death, the complication
was a direct consequence of the appellants acts, which remained a significant cause of his
death:
Even though negligence in the treatment of the victim was the immediate cause of his death, the
jury should not regard it as excluding the responsibility of the accused unless the negligent
treatment was so independent of his acts, and in itself so potent in causing death, that they regard
the contribution made by his acts as insignificant. 41

Arguably, Lord Beldams test that an intervening cause will break the chain of
causation only if it is independent of the acts of the accused and so potent in causing
death is no different from the approach taken in Smith.42 In other words, if the act is
not independent of the accused then he or she is responsible for it, and if it is not potent in
causing death then it will not be so overwhelming as to make the original wound merely
part of the history as Smith43 requires.

DAPPLICATION OF AUTHORITIES
Having canvassed the relevant authorities, it is useful to attempt to apply them to a
hypothetical situation to demonstrate the difficulties that students might encounter.
Imagine a scenario in which A inflicts a relatively minor wound to Bs head. A does not
know that B suffers from haemophilia, a congenital bleeding disorder in which blood does
not clot normally. The blow to the head causes bleeding inside Bs brain, such that he will
eventually die if not treated. A drives B to the hospital. As B steps from As car, B is
struck by another car and is killed instantly.
Imagine the same scenario, in which another person, C, inflicts the same relatively
minor wound to Bs head. But, in this scenario, B is not struck by a car and makes it into
the hospital. He is examined by a doctor who takes Bs full medical history and arranges
for a CT scan which detects the presence of an intracranial haemorrhage. Surgeons
successfully treat B via replacement therapy in which clotting factor is dripped into his
vein. However, several weeks following surgery, an infection develops at the intravenous
site.44 The drug administered to clear the infection is ineffective, but this fact is not
diagnosed by medical staff, who continue to administer it. Bs condition deteriorates and

he dies two months after the day on which the initial wound is inflicted. Evidence shows
that, had alternative treatment been administered, B would likely not have succumbed to
the infection.
It is arguable that, in the first scenario, upon an application of the substantial cause
test, students would find that being struck by a car was so overwhelming as to make As
act of striking B merely part of the historical context. Thus, A would not be held liable for
Bs death, since she did not legally cause it. The question is, would students decide upon
the same result in the second scenario, notwithstanding that Cs act is precisely the same
as As?
The facts of the second scenario are not markedly different from those in Jordan45 or,
indeed, Smith,46 Evans and Gardiner47 and Cheshire.48 But all of the cases since Jordan
assert that the rule in that case is restricted to its own particular facts. As Hallett J
observed in Jordan, the case was exceedingly unusual.49 In R v Blaue,50 the English
Court of Appeal held that, while Jordan51 was rightly decided on its facts, it should be
regarded as a case decided on its own special facts and not as an authority relaxing the
common law approach to causation,52 a view endorsed in Evans and Gardiner.53
Cheshire54 approved of the comments in Smith55 to the effect that Jordan56 was a very
particular case depending upon its exact facts. 57 Perhaps the furthest the courts have been
prepared to go in ignoring the decision in Jordan58 entirely is in R v Malcherek, R v
Steel,59 where Lord Lane CJ said:
In the view of this court, if a choice has to be made between the decision in R v Jordan and that
in R v Smith, which we do not believe it does (R v Jordan being a very exceptional case), then
the decision in R v Smith is to be preferred.60

With respect, analysis of the decision in Jordan61 does not bear out its special
treatment. Following the Smith62 approach, it is arguable that the circumstances of
Jordan63 are simply ones in which the second cause of death was so overwhelming as to
make the original wound merely part of the history. 64 But the same argument could be
made in respect of Evans and Gardiner,65 in that a time delay of almost a year between
infliction of injury and death, and a failure to diagnose and treat a condition that was
eminently treatable, renders the initial wound merely a historical setting for the ultimate
cause of death.
As foreshadowed by the scenarios described above, an interesting question for
students to consider is whether, if the intervening cause had been not medical treatment
but an act of another third party, the outcome would have been the same? Or what of the
case where an injury is inflicted by an accused, and that injury causes the death of the
victim who was already susceptible as a result of an initial injury (inflicted by a third
party) which had substantially healed? The accused would not escape liability because of
the operation of the eggshell skull rule.66 But would the inflictor of the original injury also
be liable for the death? Arguably, outside of medical treatment cases, the answer would be
no, on the basis that the original injury had ceased to be a substantial and operating cause,
and rather is merely the historical setting in which the proximate cause of death took
place.
On the face of it, the distinguishing feature of Jordan is that the treatment given in that

case was characterised as palpably wrong.67 In interpreting the meaning of this phrase in
its context, assistance can be gained from HLA Hart and Tony Honor in their seminal
work, Causation in the Law.68 The authors note that Stephens Digest of the Criminal Law
requires
something more than ordinary negligence in order that one who inflicts a wound may be relieved
of liability for homicide.69

Dicta from the cases discussed thus far provide support for this approach; namely, that
for the chain of causation to be broken in medical treatment cases gross negligence is
required. Lord Beldam in Cheshire stated that
[a]cts or omissions of a doctor treating the victim for injuries he has received at the hands of an
accused may conceivably be so extraordinary as to be capable of being regarded as acts
independent of the conduct of the accused but it is most unlikely that they will be. 70

Again, it is difficult to conceive of a case more extraordinary in this context than

Evans and Gardiner.71 The original bowel resection may not have been independent of
the conduct of the accused72 and, even if the resection surgery or other treatment
immediately surrounding the surgery had been improper, acts of the accuseds would still
have been substantial and operating causes, had the victim died at that time. But, at the
time that he did die, the original wounds had healed. The resection had healed. As noted
by the Full Court itself, the wound healed after treatment and the victim
recovered.73 In other words, the acts of the accused were spent. While the failure to
diagnose and therefore treat may not have constituted gross negligence, the treatment was
at that point independent of the conduct of the accused and thus the chain of causation had
been broken.
Interestingly, in that case the victim, Hamilton, died not long before the expiry of the
year and a day rule which provides that, for liability for homicide to ensue, the death
must occur within a year and a day from the date when the injury causing death was
inflicted. In New Zealand, this rule is codified in s 162 of the Crimes Act 1961 (NZ).
After a year and a day, there is an irrebuttable presumption that the death was attributable
to some other cause.74 Had the victim held on for another three weeks, the accused
would not have been liable for his death. In the writers opinion, that outcome would not
have been any more arbitrary than that which actually arose.
Ian Brudner notes the difficulties inherent in the common law approach to causation. 75
He argues that the applicable principles are couched in language notorious for
imprecision.76 While the cases are quite clear on the principle that the intervening act
must be overwhelming to make the accuseds act pale into insignificance, there is no
guidance for students on what is actually required to do this. The question that needs to be
answered, Brudner says, is precisely what it takes to cancel out the responsibility of
someone whose act otherwise qualifies as a proximate cause of death. 77 The only example
of what might be required is contained in Jordan78 but, as noted, this case is routinely
restricted to its own particular facts.
The medical treatment cases can be compared with the applicable rules in cases where
the conduct of the victim is asserted as breaking the chain of causation. In Blaue,79 the
appellant stabbed the victim, Woodhead, piercing her lung. By the time she arrived at the

hospital she had lost a great deal of blood and was told that a blood transfusion was
necessary. She refused the transfusion on ground that it was contrary to her belief as a
Jehovahs Witness, and she persisted in this refusal despite being advised that she would
die if she did not receive the transfusion. She died the following day. The Crown admitted
at trial that, had the transfusion been administered, Woodhead would not have died. The
defence submitted that Woodheads refusal to have a blood transfusion was unreasonable,
and had broken the chain of causation between the stabbing and her death. The Court
pointed out that
[i]t has long been the policy of the law that those who use violence on other people must take
their victims as they find them. This in our judgment means the whole man, not just the physical
man. It does not lie in the mouth of the assailant to say that his victims religious beliefs which
inhibited him from accepting certain kinds of treatment were unreasonable. The question for
decision is what caused her death. The answer is the stab wound. The fact that the victim refused
to stop this end coming about did not break the causal connection between the act and death. 80

It is implicit, in the decision in Blaue,81 that unreasonable conduct on the part of the
victim is necessary to break the chain of causation in that category of case, and this
proposition is supported by Roberts82 the reasonable foreseeability case referred to
above.83 Hart and Honor also suggest that unreasonable refusal of treatment may negate
causation (while noting that such a proposition is despite Blaue).84 In this regard, cases
in which the conduct of the victim is asserted as breaking the chain of causation are
analogous to medical treatment cases in the degree necessary to break the chain of
causation. That is, gross negligence seems to be necessary to break the causal connection
in medical treatment cases; unreasonable conduct on the part of the victim in the other
category of cases. But, apart from Jordan,85 the outcomes in the cases do not actually
seem to support these principles.86

III POLICY-BASED REASONING

Based upon the foregoing, it seems reasonable to suggest that the decisions in the
cases discussed have nothing to do with law and everything to do with policy. 87 If we use
Jordan88 to illustrate, it is not the application of the legal principles of causation that give
rise to the decision, because in a case in which medical treatment is grossly negligent or
palpably bad, the act of the accused is exactly the same as the act of an accused in a case
where medical treatment is merely unskilful, or even proper. In both cases, the act of the
accused either causes the death or it does not: subsequent gross negligence does not alter
the nature of the original wound. But the line of authorities beginning with Smith89 would
hold one accused liable for the death but not the other. As Hart and Honor point out, the
decision in Blaue90 may be correct but the reasoning is unsatisfactory. 91 Further, in my
view, the reasoning underpinning the decision-making in medical treatment cases is also
unsatisfactory because it is not transparent. The cases use the language of causation to
explain why the accused is still liable, rather than the language of policy. In reference to
the reasoning in Blaue,92 Gardner commented:
In itself such a display of respect for the victim is doubtless commendable, but should it really
dictate the fate of the assailant? The same goes for a detectable striving in the other decisions not

to regard medical mistreatment as a novus actus interveniens.93

Similarly, in Evans and Gardiner94 and Cheshire,95 the reasoning seems to be that, as a
matter of policy, an accused will still be liable for the death of the victim where medical
treatment is the proximate cause of death, whether such treatment is proper or improper,
or even thoroughly bad.96
The extent to which this kind of reasoning is problematic, for students and
practitioners, has been noted by John Farrar, who identifies interests, legal values and
other relevant factors that operate in the context of judicial decision-making:
The main limitations here are that we do not know what influence the various variables have on
the ultimate decision of a court. All that you can learn in our present state of knowledge is that
they do operate and that how they will operate in a particular case is to a degree a matter of
intuition.97

Farrar argues that, while legal policy is a fluid concept that is difficult to tie down, it
seems to be relevant in determining the scope of the ratio decidendi of cases and whether
their facts are sufficiently analogous to justify following them.98
Christian Witting, writing about policy in duty of care cases, relies on a definition of
policy-based reasoning as normative. In other words, it is reasoning based upon what the
rights and obligations of individuals ought to be.99 Ellie Margolis notes that policy-based
reasoning involves an assessment of whether a proposed legal rule will advance a
particular social goal.100
There is nothing wrong with policy-based reasoning, per se. Paul Wangerin notes that
good lawyers use it all the time. 101 Furthermore, students are not unused to policy
arguments most curricula expect considerations of policy, or social purpose, within
doctrinal courses.102 But one of the difficulties with policy-based reasoning is the
uncertainty to which it gives rise. Witting argues that, while policy-based reasoning
involves a wider focus, and is more forward-looking, it cannot offer definite guidance for
decision-making since there is opportunity for predictive error.103
W Jonathan Cardi, also in the context of tort law, argues that courts deciding
negligence cases use the language of public policy only reluctantly, instead cloak[ing]
policy-based reasoning in doctrinal-sounding language. 104 The concept of foreseeability
is an example. Cardi argues that judges use of indeterminate terms such as
foreseeability in deciding whether a duty of care exists has a harmful effect on the law. 105
Because the term is so vague, it leads judges to treat like cases differently and different
cases alike.106 The same point can be made in respect of the judicial use of the term
causation. In the medical treatment cases, with the exception of Jordan,107 the use of the
language of causation masks the real reason why liability is said to remain in the hands of
the accused. The real reason is a moral claim that it is repugnant to justice that an accused
should be held not liable for their actions if the intervening cause of death is an attempt to
save the victims life.Indeed, this would seem to be the policy behind s 166 of the Crimes
Act 1961 (NZ). That is, if the treatment is applied in good faith, then a person who inflicts
dangerous injury to another person should not benefit in the sense that their liability might
be mitigated by the fact that some other cause contributed to the harm. This policy is at
least enshrined in statute and students can point to it as the basis for holding that an

accused has legally caused a victims death. The difficulty with the case law is that it is
silent as to this kind of reasoning. Hart and Honor note that there are theorists who
insist that the decisions of courts on the extent of a wrongdoers liability are not and should not
be reached by the application of any general principles but by the exercise of the sense of
judgment, unhampered by legal rules, on the facts of each case Instead it should be
recognized that the judge, though he may weigh an indefinite number of considerations each
with some bearing on the question, decides more or less intuitively what the extent of a
wrongdoers responsibility is to be.108

The case law discussed above suggests that at the heart of decisions about causation
are judgements based on moral or policy considerations. This is acknowledged by
Dressler, who notes:
Proximate causation analysis is less a matter of applying hard and fast rules than it is an effort by
the factfinder to determine, for reasons of social policy or out of a conception of justice, on
whom to impose criminal penalties. Consequently, although courts sometimes act as if there is a
foolproof way of identifying the proximate cause of social harm, it is more accurate to think in
terms of factors relating to causal responsibility that help lawyers predict and effect outcomes. 109

But, as Chan Wing Cheong notes, uncertainty results when decisions about liability
are based upon unarticulated policy choices. 110 Justice is not served by decision-making
that could yield a different result for essentially the same criminal act not for any lack
of causation or mens rea, but because of the moral value attached to the proximate cause
of death. When this kind of decision-making is employed, it is impossible to predict
outcomes with any certainty, and this does not assist students in solving problems. It is
our responsibility as legal educators to bridge this gap and ensure that teaching and
learning approaches support the development of enhanced legal reasoning skills in our
students. As David Nadvorney observes:
We want our students to learn the legal reasoning skills necessary to develop sound legal
argument, yet the message we send them, at least on paper, is that today were studying
homicide, tomorrow theft crimes, next week and for the rest of the semester, doctrine of some
other name. Of course, reasoning skills are taught in law classes. Professors highlight it as they
analyze cases and other materials, develop it when they use hypotheticals, and refine it during
Socratic dialogue But it seems as if they hardly ever, except in legal methods or legal process
courses, explicitly teach it.111

Thus the challenge for legal education is how to explicitly teach legal reasoning when
the legal reasoning in the case law is implicit. David Nadvorney suggests that integrating
the teaching of reasoning skills into substantive courses can be done, and that doing so
will enrich both the teaching and learning of law. 112 David Samuelson also believes that
legal reasoning ought to be taught purposely and forthrightly.113 He says:
How courts solve law problems is not best left to the intuition of beginning students, or to their
memory, or to osmosis. Understanding legal decision-making results from learning how logic
and rhetoric operate in the specialized area of legal thinking and problem-solving. Chiefly, it
results from learning that the law possesses both external and internal logic and then from
learning the dynamics of these breeds of logic. Finally, it results from learning how judges justify
legal outcomes on non-legal grounds.114

IV TEACHING AND LEARNING FROM POLICY-BASED REASONING

It is of little help to students (and lawyers), in a system based very much on the
doctrine of precedent and its reliance on material facts, that courts make decisions on the
basis of moral culpability without reference to the legal principles that have been
expressly held to apply. It is difficult for students to reconcile the outcome that a strict
application of the rationes decidendi yield with the decisions in fact made by judges
which are dictated by broader considerations of justice.115
Therefore, it is essential firstly to draw students attention to the fact that invisible
factors do operate in decision-making; and secondly to encourage students to look
beyond the legal principles or rules in a case, to identify what those invisible factors are,
and how to utilise them in problem-solving.
How is this to be achieved? A number of reasons, outlined below, support the view
that combining a problem-based learning method with enhancing students case analysis
skills is an effective approach and easily achieved within current pedagogies. Case law
judicial application of the law to the facts of real problems is real world problemsolving. The utility of cases as a teaching tool (aside from the fact that they contain the
law) is that they demonstrate just how judges go about the process of applying the law
(and policy) to the facts before them. The ability to utilise facts in problem-solving is a
fundamental skill that must be acquired before students can begin to explore the nuances
of policy-based reasoning. Cohen notes that, in the context of criminal cases, this mastery
includes understanding how critical facts are in criminal cases and how ambiguous they may
well be. It also includes a basic understanding of the processes used in criminal cases to present
facts and to resolve the facts to be applied to the law. 116

At the core of the approaches to teaching and learning discussed in this article is the
idea of students as active learners. As Torrey points out, it is widely acknowledged that
participatory student learning is the most effective pedagogy. 117

APROBLEM-BASED LEARNING
The use of problem-solving (as distinct from problem-based learning) as a method of
teaching and assessment is common in law schools.118 This method invites students to
apply knowledge already gained (through lectures, readings and tutorials) to hypothetical
fact situations.119 But problem-based learning is quite different to problem-solving. It uses
real-world problems to focus learning, and supports the development of abstract thinking
and critical thinking.120 While there are a range of methods that fall within the category of
problem-based learning, the common feature is learning by doing students learn by
being active problem-solvers.121 Problem-based learning enables students to consider
issues in a real-world context rather than single legal category problems, which
encourage the collation and synthesis of information by students and the ability to apply
knowledge gained thereby to the facts of the problem.122 It is important to note that the use
of hypothetical problems in teaching students is not necessarily the same thing as
problem-based learning. A hypothetical problem usually contains only one or two issues,
whereas problems in problem-based learning approaches raise several. These issues must

be marshalled by the students before each can be analysed and solved. 123 As Mryon
Moskovitz notes, clients present lawyers with problems, not hypotheticals. 124 So, for the
purposes of problem-based learning, problems are integrated stories with elements that
must be identified, extracted and organised into a coherent structure. 125 The more that
these stories resemble real life, the more motivated students are to engage with the
material.126

BCASE SYNTHESIS
Nevertheless students still require case analysis skills. In common law systems, cases
play a significant role in legal problem-solving. As McMunigal points out, in the context
of teaching criminal law,
[c]ases are wonderful teaching tools, allowing students to see the criminal law applied to
concrete and often compelling factual scenarios and to gain insight into the policies behind the
law from the rationales provided for deciding cases. Analyzing cases is a crucial skill for
students to master.127

The case method, as employed in law schools, asks students to examine appellate
decisions to discover legal reasoning and discern the over-arching legal principles. 128
There have been numerous criticisms of the case method, including that it does not allow
for students to think critically about the law, 129 and that it relies overly on a positivistic
view of the nature of law. Lloyd Weinreb notes:
No one would assert anymore that the only question of interest to aspiring lawyers is what the
law is, which question is to be answered exclusively by an examination of cases. The legislative
function is recognized unstintingly. And the relevance of empirical disciplines like psychology
and sociology, as well as historical, philosophic, and economic insights, is not doubted. 130

Notwithstanding, the case method still has a crucial role within legal education. 131 As
Aaronson points out,
the case method provides students with simulated practice in how appellate courts formally
reason, and predicting what courts will do is a core skill central to a lawyers claim to
professional expertise.132

Compared to the United States, the history of legal education in Commonwealth

jurisdictions is relatively short. In part, this is due to the late development of law as an
autonomous discipline in the English university system. 133 Sugarman notes that the
common law formed the nucleus of English legal education and scholarship. 134 According
to Sugarman, the common law frame of mind involves an assumption that:
although law may appear to be irrational, chaotic and particularistic, if one digs deep enough and
knows what one is looking for, then it will soon become evident that the law is an internally
coherent and unified body of rules.135

However, of course, this assumption is not borne out in reality. As Sugarman observes,
principles are inseparable from interpretation and theory, which are shaped by values. He
notes that this is what gives rise to the
schizophrenia of the first-year law student: when is it that s/he is supposed to talk about law;
and when is it that s/he can talk about policy? 136

It is necessary to bear this in mind. However, in avoiding an overly positivistic view

of the law, we must also be mindful of the centrality that case law plays in our legal
system. While in New Zealand, as in many other jurisdictions, the criminal law has been
codified, case law plays a fundamentally important role in interpreting and applying the
statute. Furthermore, in the context of causation, the statute is silent as to when an
accuseds act is held to be legally causative of death (apart from s 166 of the Crimes Act
1961 (NZ) which applies only where the immediate cause of death is medical treatment).
While s 158 of the Crimes Act 1961 (NZ) defines homicide as the killing of a human
being, the cases of R v Storey137 and R v Grant138 imply that to kill means to cause the
death of. Any argument that studying appellate decisions is not essential for an
understanding of the law clearly cannot be sustained. But learning the case law is not
simply a matter of learning the rules from a case and applying that rule to the facts of a
problem. Instead, students must be able to synthesise ideas from groups of cases in order
to determine the law to be applied to any given set of facts. 139 In doing so, students must
be able to look beyond the explicit ideas and consider any unarticulated ideas, such as
underlying policy concerns, that may dictate the courts reasoning. 140 Larry Teply and
Ralph Whitten suggest that reading and analysing cases is still the most important method
for learning legal reasoning.141 They add that cases are useful in demonstrating to students
how courts in some substantive areas articulate rules that do not fully express the policies
that the courts are enforcing:
[T]he cases in these situations allow the students to look beyond the articulation of the rules and
identify the real reasons for the decisions in the area. 142

Teply and Whitten add that these types of cases are instrumental in teaching students
that they cannot simply swallow what the courts say but instead must read critically and
widely in order to predict what an outcome should be on a particular issue. 143 Jane
Gionfriddo argues that the ability to synthesise cases is a complex skill, and students need
to understand what a group of cases explicitly says and what inferences can reasonably be
drawn, in order to come to grips with all important ideas and nuances. 144 However, it
should also be acknowledged that there is not necessarily one coherent set of principles to
be applied and students are often forced to pick and choose between competing
interpretations from the cases.
The process of adopting problem-based learning and the teaching of enhanced case
analysis skills is ongoing at Te Piringa Faculty of Law. My own teaching in criminal law
involves a mixture of methods,145 which build on research and case analysis skills students
learn in a first-year Legal Method course. Within criminal law, content is primarily
delivered via a weekly two-hour lecture to a class of approximately 70150 students in
the third year of their undergraduate degree. Despite the large size of the class, questions
are frequently raised by students during lectures. I also direct questions to the class as a
whole, particularly when recapping particular subject areas or when discussing various
policy approaches to criminal law. In addition to the weekly lecture, students attend six
smaller group tutorials spread evenly across the teaching year. Tutorials are based around
hypothetical fact situations requiring students to apply the relevant law and suggest
potential outcomes. Assessment is also largely based around problem-solving (both in a

test on criminal procedure and in the provision of a legal opinion on an area of the
substantive law), although there is the opportunity to write an essay in the final exam. The
opinion must rely on both statute and case law as authorities. In terms of reading material,
students purchase a course materials book, containing mostly cases, and are advised to
read a number of texts for supplementary reading.
Causation is taught in two different contexts. First, it is covered within the general
principles of mens rea and actus reus, as one of the means by which concurrence of actus
reus and mens rea can be established. But it is also taught within the context of homicide.
There is support for the teaching of general principles within the context of specific
crimes,146 as it is often difficult to consider these concepts as abstractions. There is usually
a homicide question in the final exam.
Currently, students participate in one tutorial (of a total of six per year) in which
causation is in issue. This tutorial topic is on homicide and asks students to consider
causation in the context of (i) an assault victim drowning in his own blood; and (ii) an
assault victim receiving improper treatment causing death. Class and tutorial discussion
focuses on extracting the important ideas from the cases. As Gionfriddo suggests, we
should begin by evaluating what courts have explicitly articulated before proceeding to
work with the ideas inferentially supported by other cases. 147 As mentioned above,
students begin learning case analysis in a first year Legal Method course. This is achieved
by tracking the development of the law of negligence or the postal rule through the line of
precedent cases. Samuelson describes a similar process, by which he uses the
development of the law of negligence to teach legal reasoning. Once students have
considered relevant cases at length and attempted to resolve issues in a hypothetical case,
Samuelson says that students begin to realise that the validity of their answers turns upon
their skill in drawing appropriate analogies and distinctions, and on their sensitivity to
the logical employment of legal principles. 148 Similarly, by the time Te Piringa Faculty of
Law students study Crimes in the third year of their degree, they should have sufficient
case analysis skills to begin synthesising the relevant authorities. In the context of
causation, this entails beginning with the principle enunciated in Smith149 that a wound is
deemed to have caused death if, at the time of death, the wound is an operating and
substantial cause.150 The cases of Evans and Gardiner151 and Cheshire152 supplement the
approach taken in Smith153 by noting that it is only in the most unusual case that the chain
of causation will be taken to be broken. The case synthesis is then enhanced by drawing
an analogy between the explicit policy in Blaue154 that those who use violence on others
take their victims as they find them and the implied policy in Smith,155 Evans and
Gardiner156 and Cheshire,157 to conclude that medical treatment will almost never break
the chain of causation. This is akin to acknowledging the existence of the eggshell skull
rule. Finally, rather than restricting Jordan158 to its own particular facts, students are
encouraged to find that Jordan159 has been overruled by Smith160 and subsequent cases,
despite the reluctance of courts in those cases to hold so expressly. To encourage students
to reason in this way, 10 questions (identified later in this article) 161 are postulated and
discussed in class.
Assessment on causation normally takes the form of an opinion on homicide which

includes hypothetical facts such as those outlined in Part II of this article. In 2010, the fact
scenario concerned a victim who was stabbed in the course of a fight. The causation issue
centred on the medical mismanagement of an outbreak of Methicillin-Resistant
Staphylococcus Aureus (MRSA) which was contracted when the stab wound had almost
healed. To this extent, the facts were analogous to the material facts in Jordan.162 Students
were expected to extract and apply the relevant legal principles, and identify both the
relevant facts and which facts were absent from the hypothetical (for example, the
hypothetical did not state that the infection arose at the site of the wound but left this
intentionally vague). Extra credit was awarded to students who identified the
unarticulated policy from Smith163 and the line of cases that adopted it. This was assessed
by determining the extent to which students had considered and given reasoned responses
to the questions postulated in this paper 164 (bearing in mind that not all questions were
necessarily relevant to the issue).
The 2010 problem is now being developed for wider use in 2011. The factual matrix is
being expanded to give rise to a number of criminal law issues and jurisprudential issues
so that it can form the centrepiece of a problem-based learning approach that applies
outside of the Crimes course.
Obviously, resource issues play a large part in determining the particular approach
taken, but problem-based learning can be used within one course or topic or it can inform
a whole curriculum.165 With regard to the latter approach, Peter Daviss question as to why
we have law schools and not justice schools is an interesting one. 166 He argues that justice
needs to become central to law school curricula and consideration of fairness and justice
issues needs to be explicit.167 Davis uses John Rawls construct of the veil of ignorance to
encourage students to consider what a more just and fair society might look like. 168 As a
teacher of both Jurisprudence and Crimes, and assuming some degree of stability in
workload allocation, it is entirely possible to produce problems for use in both courses.
Issues around distributive and corrective justice, including Rawls theory of justice
(currently taught in Jurisprudence in the second-year program at my institution) are
relevant to questions about culpability (including causation) in Crimes at third year. In
Jurisprudence, students are encouraged to think about how the benefits and burdens of a
society should be distributed and how such a distribution should be maintained. This
involves conversations about how citizens inflict harm on others (by taking away
benefits) and ways in which citizens become liable for those harms. In the Crimes
course, students need to again consider the nature of harm (what harms are criminal rather
than civil) and the basis and extent of liability for criminal harms. Causation, as a basis
for imposing liability on people for harm, is thus an important concept in both
Jurisprudence and Crimes, as well as Torts. The use of a complex factual matrix in
all three courses would assist students in understanding how the principle works within
the particular substantive area of law (crimes or torts) as well as enabling them to
appreciate the policy underlying those principles. Margolis argues that policy is usually
learned in doctrinal law courses, often identified as particular dichotomies such as
security versus freedom of action,169 and says that as a result:
We get the message that policy is an amorphous concept, more useful for understanding a

decision after the fact than for predicting or advocating a particular outcome. 170

Combining a doctrinal approach with jurisprudential analysis will encourage students

to identify the policy issues that arise and, where relevant, how they may lead to injustice
in some cases. Janet Weinstein and Linda Morton, who teach interdisciplinary problemsolving courses at California Western University, note that justice is a goal of many law
students.171 Their aim is therefore to enhance students vision of a just society. 172 They say:
Legal education generally confines itself to a narrow, analytic approach to specific legal issues
an approach certainly essential to the practice of law. On the other hand, with our world
becoming more global and more interdisciplinary, students must also learn methods for, and gain
confidence in, tackling larger societal issues. 173

In many of the causation cases, the actions of the accused can be seen as warranting
punishment regardless of whether or not there was another, more proximate, cause of
death. As Gobert points out in relation to the medical treatment cases, irrespective of the
doctors conduct, the conduct of the defendant was sufficiently blameworthy. 174
Regardless of whether the doctors acts or omissions were grossly negligent or simply
negligent, the accuseds act was morally wrong. Also, one of the factors taken into
account in sentencing is the degree to which an offender poses a risk to society. If we
return to the hypothetical fact situations mentioned earlier, the accused A is equally as
dangerous as the accused C, but will not be liable for Bs death because of the fortuitous
(from As perspective) circumstance of B being struck by the car. Sanford Kadish argues
that reducing punishment for an offender, simply because luckily no harm results, does
not make sense in the context of the purposes of punishment. 175 If one of the aims of the
criminal justice system is to protect society from those who pose a risk, regardless of what
harm actually occurs, why employ a results-based approach? 176 Gobert makes the point
that conduct that causes harm may not always be morally blameworthy, and morally
blameworthy conduct may not necessarily cause harm. He argues that crimes should be
defined in terms of acts and mental state, rather than results. Otherwise, the moral
component of culpability is lost.177 Gobert suggests that
it may well be that the most practical approach to questions of causation lies in a frank
recognition that the true issue is not causation at all but attribution whether, under all the
circumstances, it is fair and just to attribute the harmful result to the defendant. 178

Considerations of justice are not the only relevant factors in teaching problem-solving
from policy-based reasoning. In light of all the foregoing arguments, in seeking to employ
policy-based reasoning in solving problems about causation, students are encouraged to
consider the following, non-exhaustive, list of questions:179
i. Should the degree of the doctors negligence affect the accuseds culpability?
ii. Should the dangerousness of the accuseds conduct play a greater role in determining
liability?
iii. Were the defendants Smith,180 Cheshire181 and Evans and Gardiner182 more dangerous
than the defendant Jordan?183
iv. Were they more morally culpable?
v. Should the victims chance of recovery be a material factor?
vi. Should the length of time between the accuseds act and the victims death be a

material factor (the year and a day rule notwithstanding)?

vii. Should the fact that a wound might have almost healed at the time the proximate
cause arose be a material factor?
viii. What are the facts upon which Jordan184 is distinguishable?
ix. What would the outcomes in Smith,185 Cheshire186 and Evans and Gardiner187 be if the
proximate cause of death was something other than medical treatment; for example,
the act of a third party?
x. What are the policy considerations that underpin apportioning blame for harm?
The hypothetical fact situations outlined in Part II of this article provide useful starting
points for students to consider these questions, but require expansion to encourage
students to think broadly about different bases for liability. In other words, while they are
asked to consider whether the accuseds act was causative of death, they might also
consider any liability arising on the part of medical staff (in tort, for example). In light of
the aims of a problem-based learning approach, the facts of the problem should allow
students to identify where there may be gaps where the law does not expressly provide for
such an answer. Further, students should be encouraged to consider whether any express
legal answer is in fact a just one. Presenting the students with problems that ask them to
think about a number of issues outside the square of the doctrinal subject will enable
them come to grips with the complexity of real-life situations, 188 and the fact that the law
as expressed does not always provide neat answers.
It must be noted that the ideas raised in this paper are a work in progress. While some
of the approaches have been adopted at Te Piringa Faculty of Law, there is scope for
further development of legal method, jurisprudence, torts and criminal law courses to
further accommodate the need for students to learn how to identify and apply the
underlying policy from case law.

V CONCLUSION
An over-arching, pedagogical goal of law school should be facilitating the cognitive and
emotional development of students in ways that provide them with a sufficient foundation to
become lawyers who, in pursuing their profession, are able to analyse problems in their full
context. This includes recognising both patterns and uniqueness in different fact situations and
knowing how to synthesise, prioritise and apply appropriate breadth and depth of knowledge. 189

The homicide cases canvassed in this article demonstrate that, where questions of
causation arise, outcomes depend upon the operation of factors that are usually invisible
to the student. While students are required to extract the rules from cases and to apply
them to hypothetical fact situations in order to predict outcomes, this approach is
problematic where students must rely on cases in which it is difficult to separate out
policy reasons from those based on the application of express legal rules. Such policy
considerations can be found in the line of authorities beginning with Smith,190 to the effect
that medical treatment, regardless of how improper, will not break the chain of causation.
Students must therefore be encouraged to consider the invisible factors that judges
employ in legal reasoning, by developing the skills necessary to synthesise the law and
policy from cases, and by actively solving problems in which these issues arise. This can

occur at all stages of the teaching and learning process, but tutorials in particular are a
useful means by which students can engage in discussions about broader policy
considerations without the fear that they may be getting the answer wrong, which is a risk
in formal assessments. As Tracey Meares, Neal Katyal and Dan Kahan argue, a failure to
engage students in the kinds of questions that are being asked in contemporary
scholarship can result in students being unable to deal with criminal justice policy in
practice.191 This means that students must be taught not only how to extract the law from
cases, but also the policy, based on moral considerations, that underpins the decisions. In
other words, one of the essential skills that falls within the rubric of thinking like a
lawyer is perhaps the ability to think like a policy analyst.
*
1

2
3
4
5
6
7

8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

23
24
25

Senior Lecturer, Te Piringa Faculty of Law, University of Waikato, New Zealand.

The phrase is of uncertain origin but was given particular prominence by the movie The Paper Chase. A
literature search for the purposes of this article revealed a large number of articles using this phrase in the
context of legal education. See, eg, Kate E Bloch, Cognition and Star Trek: Learning and Legal
Education (2009) 42 John Marshall Law Review 959; Neil P Cohen, Teaching Criminal Law: Curing the
Disconnect (2004) 48 Saint Louis University Law Journal 1195; Larry O Natt Gantt II, Deconstructing
Thinking Like a Lawyer: Analyzing the Cognitive Components of the Analytical Mind (2007) 29 Campbell
Law Review 413; Jane K Gionfriddo, Thinking Like a Lawyer: The Heuristics of Case Synthesis (2007) 40
Texas Tech Law Review 1; Cynthia G Hawkins-Len, The Socratic MethodProblem Method Dichotomy:
The Debate over Teaching Method Continues (1998) Brigham Young University Education and Law
Journal 1; Keith H Hirokawa, Critical Enculturation: Using Problems to Teach Law (2009) 2 Drexel Law
Review 1; Myron Moskovitz, From Case Method to Problem Method: The Evolution of a Teacher (2004)
Saint Louis University Law Journal 1205; Nancy B Rapoport, Is Thinking Like a Lawyer Really What
We Want to Teach? (2002) 1 Journal of the Association of Legal Writing Directors 91; Kurt M Saunders and
Linda Levine, Learning to Think Like a Lawyer (1994) 29 University of San Francisco Law Review 121.
Rapoport, above n 1, 93.
Bloch, above n 1, 963.
James J Gobert, The Fortuity of Consequence (1993) 4 Criminal Law Forum 1, 12.
Woolmington v DPP [1935] AC 46.
R v Cheshire [1991] 3 All ER 670, 674 (Cheshire).
Dennis Klimchuk, Causation, Thin Skulls and Equality (1998) 11 Canadian Journal of Law and
Jurisprudence 115, 117.
(1991) 172 CLR 378 (Royall).
(1971) 56 Cr App R 95 (Roberts).
Royall (1991) 172 CLR 378, 399.
Roberts (1971) 56 Cr App R 95, 102.
[1959] 2 QB 35 (Smith).
Ibid 42.
Ibid.
Ibid 423.
(1991) 172 CLR 378.
(1971) 56 Cr App R 95.
Royall (1991) 172 CLR 378, 389.
Ibid.
(1991) 172 CLR 378.
[1959] 2 QB 35.
See, eg, R v McKinnon [1980] 2 NZLR 31. It is also important to note that, in New Zealand, s 166 of the
Crimes Act 1961 (NZ) also applies in cases where treatment is the immediate cause of death. It states that
[e]very one who causes to another person any bodily injury, in itself of a dangerous nature, from which
death results, kills that person, although the immediate cause of death be treatment, proper or improper,
applied in good faith. The Court in R v Kirikiri [1982] 2 NZLR 648, 651 held that s 166 was declaratory of
the common law, and thus Smith was relevant to the interpretation of the section.
Stanley Yeo, Blamable Causation (2000) 24 Criminal Law Journal 144.
[1959] 2 QB 35.
Yeo, above n 23, 150. Yeo argues that, where there is more than one substantial cause of death, a test of
foreseeability is the most appropriate measure for determining whether a defendant should still be liable. In
other words, did the defendant foresee, or ought he or she to have reasonably foreseen, the intervening
event?

26

27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66

67
68
69
70
71
72
73
74
75

76

For an accuseds act to be legally causative of death, it must also be a dangerous act. In New Zealand, this is
expressly required by s 166 of the Crimes Act 1961. Outside of s 166, some cases and commentators use the
term mortal wound to express the requirement that the original injury must be more than trifling. See, eg,
Paul K Ryu, Causation in Criminal Law (1958) 106 University of Pennsylvania Law Review 773. H L A
Hart and Tony Honor, Causation in the Law (Oxford University Press, 2nd ed, 1985) 2412 identify three
senses of mortal wound as (i) one sufficient to cause the death of a person of average constitution under
normal circumstances; (ii) one highly likely to cause the death of a particular victim, given the victims
constitution and likelihood of medical assistance; and (iii) one that in fact causes death even though it was
not mortal in either of the two preceding senses (for example, a scratch that the victim neglected).
Royall (1991) 172 CLR 378, 448.
R v Jordan [1956] 40 Cr App Rep 152 (Jordan).
Ibid 157.
Ibid.
[1959] 2 QB 35.
Jordan [1956] 40 Cr App Rep 152, 1578.
R v Evans and Gardiner (No 2) [1976] VR 523 (Evans and Gardiner).
[1959] 2 QB 35.
Evans and Gardiner [1976] VR 523, 534.
Ibid.
[1956] 40 Cr App Rep 152.
[1991] 3 All ER 670.
Ibid 672.
Ibid 677.
Ibid.
[1959] 2 QB 35.
Ibid.
I am grateful to the anonymous referee who pointed out a flaw in the original hypothetical fact situation and
suggested an amendment.
[1956] 40 Cr App Rep 152.
[1959] 2 QB 35.
[1976] VR 523.
[1991] 3 All ER 670.
Jordan [1956] 40 Cr App Rep 152, 153.
R v Blaue [1975] 3 All ER 446 (Blaue).
[1956] 40 Cr App Rep 152.
Blaue [1975] 3 All ER 446, 449.
[1976] VR 523.
[1991] 3 All ER 670.
[1959] 2 QB 35.
[1956] 40 Cr App Rep 152.
R v Smith [1959] 2 QB 35, 43.
[1956] 40 Cr App Rep 152.
[1981] 1 WLR 690.
Ibid 696.
[1956] 40 Cr App Rep 152.
[1959] 2 QB 35.
[1956] 40 Cr App Rep 152.
R v Smith [1959] 2 QB 35, 43.
[1976] VR 523.
The eggshell skull or thin skull rule asserts that those who use violence on other people must take their
victims as they find them: Blaue [1975] 3 All ER 446, 450. In other words, a person who inflicts violence
on another is responsible for the consequences, notwithstanding that the victim may have been particularly
susceptible because of his or her thin skull. This principle applies regardless of whether the person who
inflicted the violence was aware of the susceptibility or not.
Jordan [1956] 40 Cr App Rep 152, 157.
Hart and Honor, above n 26.
Hart and Honor, above n 26, 355.
Cheshire [1991] 3 All ER 670, 675.
[1976] VR 523.
The words used by Lord Beldam in Cheshire [1991] 3 All ER 670, 677.
Evans and Gardiner [1976] VR 523, 534.
Ibid 523.
lan Brudner, Owning Outcomes: On Intervening Causes, Thin Skulls, and Fault-Undifferentiated Crimes
(1998) 11 Canadian Journal of Law and Jurisprudence 89.
Ibid 92.

77
78
79
80
81
82
83
84
85
86

87

88
89
90
91
92
93
94
95
96
97
98
99

100

101

102

103
104
105
106
107
108
109
110

111

112
113
114
115
116
117
118

119
120
121
122
123

124
125
126

Ibid.
[1956] 40 Cr App Rep 152.
[1975] 3 All ER 446.
Ibid 450.
[1975] 3 All ER 446.
(1971) 56 Cr App R 95.
See Roberts (1971) 56 Cr App R 95, 102 (Stephenson LJ).
Hart and Honor, above n 26, 361.
[1956] 40 Cr App Rep 152.
The law often distinguishes between acts and omissions and places a higher degree of accountability on
those who act, compared with those who omit to do something. This is another factor that may affect
outcomes in causation cases, but it is premature to conclude that this distinction was critical to the differing
outcomes in Jordan [1956] 40 Cr App Rep 152 as compared to Evans and Gardiner [1976] VR 523 and
Cheshire [1991] 3 All ER 670. In Smith [1959] 2 QB 35, the positive acts of administering oxygen and
artificial respiration did not break the chain of causation either.
Padfield also suggests that cases like these have nothing to do with causation and everything to do with
policy: Nicola Padfield, Clean Water and Muddy Causation: Is Causation a Question of Law or Fact, or Just
a Way of Allocating Blame? [1995] Criminal Law Review 683, 685.
[1956] 40 Cr App Rep 152.
[1959] 2 QB 35.
[1975] 3 All ER 446.
Hart and Honor, above n 26, 361.
[1975] 3 All ER 446.
Simon Gardner, Causation in Homicide (1992) 108 Law Quarterly Review 24, 26.
[1976] VR 523.
[1991] 3 All ER 670.
As was the case in Smith [1959] 2 QB 35.
John H Farrar, Introduction to Legal Method (Sweet & Maxwell, 1977) 157, 158.
Ibid.
Christian Witting, Tort Law, Policy and the High Court of Australia (2007) 31 Melbourne University Law
Review 569, 573.
Ellie Margolis, Beyond Brandeis: Exploring the Uses of Non-Legal Materials in Appellate Briefs (2000) 34
University of San Francisco Law Review 197, 211.
Paul T Wangerin, A Multidisciplinary Analysis of the Structure of Persuasive Arguments (1993) 16
Harvard Journal of Law and Public Policy 195, 218.
Marin Roger Scordato, Reflections on the Nature of Legal Scholarship in the Post-Realist Era (2008) 48
Santa Clara Law Review 353, 365.
Witting, above n 99, 577.
W Jonathan Cardi, Purging Foreseeability (2005) 58 Vanderbilt Law Review 739, 762.
Ibid 740.
Ibid.
[1956] 40 Cr App Rep 152.
Hart and Honor, above n 26, 291.
Joshua Dressler, Cases and Materials on Criminal Law (West Group, 1994) 181.
Chan Wing Cheong, The Requirement of Concurrence of Actus Reus and Mens Rea in Homicide (2000)
Singapore Journal of Legal Studies 75, 88.
David Nadvorney, Teaching Legal Reasoning Skills in Substantive Courses: A Practical View (2002) 5
New York City Law Review 109, 110.
Ibid 113.
David R Samuelson, Introducing Legal Reasoning (1997) 47 Journal of Legal Education 571, 572.
Ibid.
Lloyd L Weinreb, Teaching Criminal Law (2009) 7 Ohio State Journal of Criminal Law 279, 289.
Cohen, above n 1, 1201.
Morrison Torrey, You Call That Education (2004) 19 Wisconsin Womens Law Journal 93, 103.
Jacquelin Mackinnon, Problem Based Learning and New Zealand Legal Education (2006) 3 Web Journal
of Current Legal Issues <http://webjcli.ncl.ac.uk/2006/issue3/mackinnon3.html>.
Ibid.
Ibid.
Hirokawa, above n 1, 2.
Mackinnon, above n 118.
Mryon Moskovitz, Beyond the Case Method: Its Time to Teach with Problems (1992) 42 Journal of Legal
Education 241, 246.
Ibid.
Ibid 256.
Stephen Nathanson, Developing Legal Problem-Solving Skills (1994) 44 Journal of Legal Education 215,

127

128

129

130
131
132

133

134

135
136
137
138
139
140
141

142
143
144
145

146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166

167
168
169

170
171

172
173
174
175

176
177

224.
Kevin C McMunigal, A Statutory Approach to Criminal Law (2004) 48 Saint Louis University Law
Journal 1285, 1286.
Michael J Greenlee, Theory, Practice, Specialization, and Interdisciplinary Perspectives: Pulling It All
Together at the College of Law (2009) 52 Advocate (Idaho) 25, 25.
Anders Walker, The Anti-Case Method: Herbert Wechsler and the Political History of the Criminal Law
Course (2009) 7 Ohio State Journal of Criminal Law 217, 21720.
Weinreb, above n 115, 280.
Saunders and Levine, above n 1, 129.
Mark Neal Aaronson, Thinking Like a Fox: Four Overlapping Domains of Good Lawyering (2002) 9
Clinical Law Review 1, 6.
Geoff McLay, Toward a History of New Zealand Legal Education (1999) 30 Victoria University of
Wellington Law Review 333.
David Sugarman, Legal Theory, the Common Law Mind and the Making of the Textbook Tradition in
William Twining (ed), Legal Theory and Common Law (Blackwell, 1986) 26.
Ibid.
Ibid 27.
[1931] NZLR 417.
[1966] NZLR 968.
Gionfriddo, above n 1, 3.
Ibid 13.
Larry L Teply and Ralph U Whitten, Teaching Civil Procedure Using an Integrated Case-Text-and-Problem
Method (2003) 47 Saint Louis University Law Journal 91, 99.
Ibid.
Ibid 1023.
Gionfriddo, above n 1, 36.
As Mueller notes, teaching methods are linked with a number of variables beyond the teachers control, such
as class size: Gerhard O W Mueller, Teaching International Criminal Law Teaching Techniques (1988)
1 Touro Journal of Transnational Law 177, 1803.
Weinreb, above n 115.
Gionfriddo, above n 1, 10.
Samuelson, above n 113, 594.
[1959] 2 QB 35.
[1959] 2 QB 35, 423.
[1976] VR 523.
[1991] 3 All ER 670.
[1959] 2 QB 35.
[1975] 3 All ER 446.
[1959] 2 QB 35.
[1976] VR 523.
[1991] 3 All ER 670.
[1956] 40 Cr App Rep 152.
Ibid.
[1959] 2 QB 35.
See below n 179 and accompanying text.
[1956] 40 Cr App Rep 152.
[1959] 2 QB 35.
See below n 179 and accompanying text.
Mackinnon, above n 118.
Peter L Davis, Why Not a Justice School? On the Role of Justice in Legal Education and the Construction
of a Pedagogy of Justice (2007) 30 Hamline Law Review 513.
Ibid 534.
Ibid 538.
Ellie Margolis, Closing the Floodgates: Making Persuasive Policy Arguments in Appellate Briefs (2001) 62
Montana Law Review 59, 623.
Ibid.
Janet Weinstein and Linda Morton, Interdisciplinary Problem Solving Courses as a Context for Nurturing
Intrinsic Values (2007) 13 Clinical Law Review 839, 850.
Ibid.
Ibid.
Gobert, above n 4, 17.
Sanford H Kadish, The Criminal Law and the Luck of the Draw (1994) 84 Journal of Criminal Law and
Criminology 679.
Gobert, above n 4, 8.
Ibid 19.

178
179
180
181
182
183
184
185
186
187
188
189
190
191

Ibid 29.
See above n 161 and 164 and accompanying text.
[1959] 2 QB 35.
[1991] 3 All ER 670.
[1976] VR 523.
[1956] 40 Cr App Rep 152.
Ibid.
[1959] 2 QB 35.
[1991] 3 All ER 670.
[1976] VR 523.
Aaronson, above n 132, 8.
Ibid 18.
[1959] 2 QB 35.
Tracey L Meares, Neal Katyal and Dan M Kahan, Updating the Study of Punishment (2004) 56 Stanford
Law Review 1171, 1172.