Causation in law: a matter of common sense not moral philosopy
Summarising Causation for the Jury
This section examines the legal principles governing causation, particularly those cases where there the causal responsibility is contentious because of an intervening or supervening event.
Causation may be specified as a physical element of the offence, for example, in murder the law requires the defendant to cause the death of another person. In many (if not most) criminal cases, causation is not in issue, and the courts simply direct the jury that causation is a question of fact. However, the apparent simplicity of such a direction is misleading. In ordinary speech and common usage the notion of causation is used loosely. In the criminal law, causation has a distinctive legal meaning, and the courts developed several tests to assist the jury deal with causation. The law is a state of uncertainty following the recent decision of the High Court in Royall (1990) 172 CLR 387, BFW at p 551. The High Court delivered separate judgments and several formulations of the test for causation emerged. There are three principal tests: the natural consequences test, the reasonable foreseeability test and operating and substantial cause test. It is important to appreciate that all of these tests are objective, and are not based on what the defendant subjectively intends or foresees. It is most important for you to separate the causation question from the issue of mens rea: the subjective state of mind of the defendant.
The importance of recognising causation as an objective test was emphasised in Hallett [1969] SASR 141, BFW at p 556. The Supreme Court of South Australia held that the defendant's foresight of death or serious injury is irrelevant to the issue of causation. Causation arises as an issue in a trial when some other intervening or supervening event occurs between the defendant's conduct and the death. It is in these case that the courts have attempted to formulate tests to guide the jury.
The early cases developed and applied the operating and substantial cause test. In Hallett [1969] SASR 141 he defendant had attacked the victim (Whiting) on a beach rendering him unconscious. The defendant claimed that he had not drowned Whiting, but left him unconscious in what he thought was a position of apparent safety with his ankles in a few inches of water. Whiting claimed that he did not cause the death. The forensic evidence suggested that the immediate cause of death was drowning in shallow water whilst unconscious. The defendant was convicted of murder.
The question arose on appeal whether the jury could find that the defendant caused the death of Whiting by leaving him on the beach with an incoming tide. The Supreme Court held that it was the defendant's original blow (rendering the victim unconscious) which originated the events which led to drowning. It could not be said that the actions of the sea on the deceased broke the chain of causation. The court approved old cases where it had been held that the ordinary operation of natural causes did not prevent the death from being caused by the defendant. However, they did recognise, obiter, that in some cases the extraordinary operation of natural forces may be regarded as breaking the chain of causation [BFW at p 559]. The Supreme Court in Hallett approved the approach taken to causation in the earlier English decision Smith [1959] QB 35, which is outlined in BFW at 559, see jury direction at 551. The test is whether, at the time of death, the original conduct is still an operating cause and a substantial cause.
Bad medical treatment: Jordan, Smith, Evans and Gardiner
It is important to appreciate that the common law does not require the defendant's conduct to be the sole cause, or even the main cause, of the victim's death. This point is clear from the medical treatment cases. The question has arisen whether abnormal medical treatment can break the chain of causation. One of the first cases is Jordan (1956) 40 Cr.App.R. 152 [BWW at 141]. The defendant was charged and convicted of murder arising from a stabbing. The victim was admitted promptly to hospital and the wound was stitched up. He died a week later of pneumonia. Forensic evidence admitted during the appeal showed that the immediate cause of death related to the administering of an abnormal amount terramycin, to which the victim was known to be allergic. A post-mortem showed that the original wound had "mainly healed".
There is some controversy over what the court decided in this case. The headnote reads "that death resulting from any normal treatment employed to deal with a felonious injury may be regarded as caused by the felonious injury, but the same principle does not apply where the treatment employed is abnormal.". This summary in the headnote is not supported by the judgment itself. The Court of Appeal simply reaffirmed that whether the medical treatment was sufficient to break the chain of causation was a question of fact, and as such a matter for the jury. However, the Court of Appeal was convinced that "whatever direction had been given to the jury, and however correct it had been, the jury would have been precluded from saying that they were satisfied that death was caused by the stab wound".
Jordan has been distinguished in later cases. The next case was Smith [1959] 2 QB 42, refer to BFW at 551, facts outlined at 568. The appeal was heard by the English Courts Martial Appeals Court. A soldier was charged with murder as a result of a stabbing in a barrack-room fight. The victim received medical treatment that was described by the court as "thoroughly bad and might well have affected his chances of recovery." The doctor failed to appreciate that the soldier's lung had been pierced, there were no transfusion facilities and he was dropped twice. Under ideal conditions there was a 75% chance of recovering from the stabbing. The defendant was convicted and appealed. The question was whether abnormal treatment interrupted the chain of causation?
The appeal against the conviction failed since a reasonable jury, properly directed as to the principles of causation, could not fail to conclude that the stab wound was still the operating and substantial cause of death. Lord Parker, giving the judgment of the court, stated that "...only if the second cause [in this case the medical treatment] is so overwhelming as to make the original wound merely part of history can it be said that death does not flow from the wound."
The operating and substantial cause test developed in the English decision Smith, was followed in the Victorian Supreme Court decision Evans & Gardiner (No.2) [1976] VR 523. The victim was stabbed in the stomach and received prompt competent treatment and the wound healed. However, 11 months later the victim died as a result of a secondary condition (a bowel blockage). This secondary condition was a common complication arising from the initial treatment. There was evidence that the failure to diagnose this complaint amounted to improper medical treatment.
The Supreme Court of Victoria rejected the submission that the doctor's failure to diagnose the secondary condition made it unsafe for the issue to go to the jury. It held that the real question for the jury was whether the bowel blockage was due to the stabbing. Since there was plenty of evidence to support this conclusion, the convictions were upheld. This decision certainly is difficult to reconcile with the earlier English decision in Jordan. The Supreme Court in Evans took a strong line, approaching the issue in the following way:
"The failure of the medical practitioners to diagnose correctly the victim's condition, however inept and unskillful, was not the cause of death and the real question for the jury was whether the blockage was due to the stabbing"
As a matter of policy, the courts are extremely reluctant to allow the defence counsel to impeach medical treatment given at the time of the injury. Even where the treatment is thoroughly bad the courts do not regard the causal link as having been broken. Evans can be viewed as the high water mark in medical treatment cases. It goes further than the English authorities, concluding that medical treatment, however inept and unskillful, is not a causally relevant factor.
The conduct of the victim: Blaue and Royall
Another intervening event which may affect the causation issue is the victim's own conduct. Consider the case of a person who sustains a serious injury, and then refuses to accept medical treatment. The person dies. There is evidence that the person could have been saved by the medical treatment. This occurred in the English decision Blaue [1975] 1 WLR 1411. The defendant attacked a young girl, stabbing and piercing her lung. She refused a blood transfusion on religious grounds (she was a Jehovah's Witness). She had been told that she would die without a transfusion. She died and the defendant was charged and convicted of murder. The Court of Appeal held that the stab wound was the cause of death, and the fact that the victim did nothing stop this end coming about did not break the causal connection between the act and the death. The Court of Appeal reviewed the authorities on this subject. In Holland (1841), the victim received a minor wounding (a cut finger), but refused amputation and died of lockjaw. Maule J rejected the submission that the obstinate refusal of the victim to submit to proper medical treatment prevented the wounding being the cause of death. He concluded that whether or not the wound was the cause of death was a simply matter for the jury.
The Court of Appeal declined to find that the victim's refusal to accept treatment broke the chain of causation, for two reasons. First, the court doubted whether it was a proper exercise for the jury to determine whether the refusal by a person of deep religious conviction to consent to a blood transfusion was "unreasonable". Secondly, and more importantly, it has long been the policy of the common 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 [or woman], not just the physical man [or woman]".
Similar problems of establishing causation occur in the the so called "fright and flight" cases. In these cases, the victim dies while trying to escape from the defendant, fearing the imminent infliction of physical injury. This issue was raised in the High Court decision of Royall (1990) 172 CLR 378, refer to Materials at p 6; also BFW 551. The defendant was charged with the murder of a woman who died having fallen from the bathroom window of her sixth floor flat. The Crown alleged that the defendant had brought about the death of the victim in one of three ways. First that he had pushed her out of the window. Secondly that he attacked her, causing her to fall from the window as she avoided his attack. Thirdly, that the victim had a well founded and reasonable apprehension that if she remained in the bathroom she would be subjected to further violence which caused her to jump out of the window to escape the violence.
The problem with Royall is that the members of the High Court divided over the test of causation which should be applied. The Court however agreed that an generally an act done by a person in the interests of self-preservation, in the face of violence or threats of violence, which results in the death, does not negative the causal connection between the violence or threats of violence and the death. Mason CJ makes this point in Materials at p 8. The generally endorse the view that causation is a matter of common-sense, refer to Burt CJ in Materials at p 8, BFW 552. This common-sense explanation of of causation was affirmed by majority Mason CJ , Deane and Dawson, Toohey and Gaudron JJ. See also Timbu Kolian (1968) 119 CLR 47 at 69, per Windeyer J.
Mason CJ applied a natural consequence test to the victim reaction, in preference to a reasonable foreseeability test: see Materials p 8. Is intention or foresight relevant to causation? Hallett said no. But Mason CJ recognised that it is impossible to maintain the distinction in some cases:
"in some situations, the accused's state of mind will be relevant to that issue [of causation] as, for example, where there is evidence that the accused intended that injury should result in the same way in which it did and where, in the absence of evidence of intention, the facts would raise a doubt about causation".
What does this mean? Foresight or intention (ie mens rea) will relevant to causation in the rare case. Take the person who has an unreasonable phobia of spiders, and the defendant knows that weakness, and intends to cause injury by praying on this fear. Say the victim is so frightened that he or she jumps out of the window. This reaction may not be a natural consequence of exposure to spiders, but causation is established because of the defendant's intention. Brennan and McHugh JJ, in their separate judgments, agree that in some unusual cases, intention and foresight is relevant to intention.
Mason CJ then considered whether the victim's act of self-preservation be reasonable. Several 19th century English cases required the act of self-preservation to be reasonable: Pitts (1842) C & M 284 and Curley (1909) 2 Cr.App.R 96. Mason CJ followed the formulation adopted in Grimes and Lee (1894) 15 NSWLR 209 that did not require that the steps taken to escape be reasonable.
Brennan J started his judgment with a review of the general principles, refer to Materials at p 12. He formulates the general principle. Brennan's approach differs from Mason's. Brennan says that the victim's act of self-preservation must be reasonable having regard to the nature of the defendant's conduct and the fear that it was likely to have induced. In his words, the victim's attempt at self-preservation must be proportionate to that conduct.
Deane and Dawson JJ applied a version of the natural consequences test: Materials at p 14. They said that the judge should avoid directing the jury in terms of foreseeability. They went on to identify some general principles (F and G).
Two basic propositions emerge from Deane and Dawson JJ judgment. The victim's attempt at self preservation does not break the causal link if (a) the victim's fear or apprehension must be well-founded or well-grounded or reasonable in all the circumstances, and (b) the victim's act of escape or self preservation must be the natural consequence of the defendant's behaviour: Materials at p 15.
Toohey and Gaudron JJ showed greatest fidelity to the earlier approach. They rejected the sophisticated notions of causation used in torts (i.e. reasonable foreseeability). Applied the traditional test for causation based on substantial contribution, which is a variant of the operating and substantial cause test in Hallett. The jury must focus their attention on whether an act of the defendant "substantially contributed" to the death, refer to Materials at p 18. The test is an objective one and does not depend on the defendant's appreciation of the consequences of any act. Refer to Materials at 19 "Consistent with this approach the jury may be told that if the victim's reaction to the act of the defendant ... was quite disproportionate to the act or was unreasonable, the chain of causation was broken."
McHugh J took the most radical approach to causation, refer to Materials at p 24. He recognised the need to clarify and rationalise the relevant principles. He was concerned to achieve a coherent theory of causation in both the criminal and civil law. Therefore, he applied the test of reasonable foreseeability, rather than a "natural consequences" test or the "operating and substantial cause" test. He criticised the operating and substantial cause test. In his view it was nothing more than a disguised but for test, refer to Materials at p 24.
McHugh J doubted whether questions of moral culpability could be divorced from causation, refer to Materials at p 24.. He doubted the approach in Hallett that causation had nothing to do with foreseeability. McHugh then considered whether the act of the victim should not be regarded as a novus actus interveniens merely because it was unreasonable. Did the victim's act of self preservation need to be reasonable. McHugh held that persons subjected to violence or the threat of violence do not always think rationally or act reasonably. The instinct of self preservation often causes them to flee or to take action which, while avoiding the immediate danger, places them in greater peril. Indeed in some cases the suicide of the victim should not be regarded as breaking the causal chain of responsibility. He argued that if unreasonable refusal to take medical treatment Blaue would not operate as a novus actus intervenien, then neither should the unreasonable conduct of a victim in escaping an attack.
McHugh formulated the test of causation, refer to Materials at p 25. Any irrational or unreasonable conduct of the victim will be a variable factor to be weighed according to all the circumstances of the case in determining whether the harm was reasonably foreseeable.
It is necessary to formulate a succinct and comprehensible direction on causation for the jury (and law students!). Several basic principles emerge from Royall. The actus reus of murder and manslaughter is causing death of another person. Royall is a very complex case. Remember that the majority of the High Court endorsed the view that causation is a matter of commonsense for the jury to determine, refer to judicial endorsement of Burt CJ in Campbell, see above. The hard causation case arises where some other event occurs between the act of the defendant and the resulting death, such as bad medical treatment or the victim's own actions. This is where the jury will require careful direction, and a number of tests have been developed to assist the jury. A majority of the High Court favour direction which asks whether the death is a natural consequence of defendant's conduct (Mason CJ, Deane, Dawson JJ). Toohey and Gaudron JJ show greatest fidelity to the operating and substantial cause test in Hallett: the question for the jury is whether the defendant's conduct subtantially contributed to the victim's death. Brennan J and McHugh J, in separate judgments, favoured a reasonable foreseeability test: the question for the jury is whether the death is a reasonably foreseeable consequence of the defendant's conduct.
Causation is determined by an objective test, and the jury must not confuse the causation issue with the subjective mental state of the defendant. However, the distinction between causation and mens rea was blurred by three of the Justices (Mason CJ, Brennan and McHugh). Their statements are strictly obiter. Mason CJ, Brennan and McHugh recognised that there will be cases, admittedly rare cases, where the death of the victim is not a natural consequence or reasonably foreseeably consequence of the defendant's conduct. However, the causation link will be established if the defendant subjectively intended or foresaw death in that way. None of the Justices gave an example of this rare case. My example is the defendant who knew that the victims suffered from an unusual phobia, and literally frightened the victim to death.
The conduct of third parties: Pagett
In Pagett (1983) 76 Cr.App.R (English Court of Appeal), the police were attempting to arrest the appellant for a number of serious offences. In the course of the arrest, it was alleged that the appellant had fired on the police. A police officer, acting in self-defence against the defendant, shot and killed the appellant's girlfriend. The defendant had used her body as a shield. The appellant was charged with her murder.
The question arose whether the appellant's unlawful acts could be regarded as the cause of his girlfiend's death? The acts relied upon as the cause of her death were (a) the appellant firing upon the police (b) using the victim's body as a shield.
The Court of Appeal held that in determining whether a homicide may be attributed in those cases where the immediate cause of death was the acts of another, the ordinary principles of causation apply. The court pointed out that the trial judge had erred in directing the jury that it was a matter of law (and therefore a matter for the judge) to determine whether his acts caused the death. He should have merely directed them as to the relevant principles of law relating to causation, leaving it to the jury to determine whether the causal connection between his acts and the death had been established.
The court reviewed the relevant principles of causation, and in particular those external factors that may operate so as to break the chain of causation (traditionally known as novus actus interveniens). They concluded that:
"...a reasonable act of self-preservation, being of course itself an act caused by the accused's own act, does not operate as a novus actus interveniens."
The court cited Pitts (1842) C& M 284, Curley (1909) 2 Cr.App.R. 96. as authority for this broad proposition. In these cases the act of self preservation involved an act by the victim in attempting to escape the violence of the accused, rather than a third party.
The court in Pagett unusually made reference to the treatise on causation by Professors Hart and Honore, Causation in the Law (1985). After a comprehensive review of the authorities Hart and Honore concluded that the intervention of a third party will have the effect of breaking the chain of causation only where the intervention was "free, deliberate and informed." Non-voluntary conduct is not capable of relieving the accused of his responsibility, as in the case of a reasonable act done in self-defence or in performance of a legal duty.
Arguably this case may stretch the bounds of causation too far, since here the accused is said to be causing the death of the victim by the unintended actions of another. Note how we are applying an objective test of reasonable foreseeability to the consequences of the defendant's actions in determining the causal connection.