R v Hardie

The defendant, Hardie, had been living with a woman in her flat. Their relationship deteriorated and she asked him to leave. In a state of considerable emotional distress at the breakdown of the relationship, Hardie consumed a number of Valium tablets that belonged to his partner. Valium is a benzodiazepine tranquilliser, a prescription-only medicine commonly used to treat anxiety and to produce a calming effect. The defendant took the tablets not as a recreational drug but in an attempt to steady his nerves during an acutely stressful situation.

At some point after ingesting the Valium, Hardie started a fire in the bedroom of the flat. He was subsequently charged with arson, contrary to section 1(1) and section 1(3) of the Criminal Damage Act 1971. The prosecution's case was that he had either intended to endanger life or had been reckless as to whether life would be endangered by the fire. The defendant raised the defence that he had been so affected by the Valium that his judgment and awareness were substantially impaired at the time he set the fire.

At first instance, the trial judge directed the jury in accordance with the conventional rule applicable to voluntary intoxication. The jury were directed that, because the defendant had voluntarily taken Valium, his intoxicated state could not be raised as a defence to the charges. This direction treated Valium in the same manner as alcohol and other intoxicating substances for the purposes of the voluntary intoxication rules. Hardie was convicted and appealed to the Court of Appeal.

The case therefore required the Court of Appeal to consider the nature of the drug that Hardie had consumed, and in particular whether the general rule governing voluntary intoxication by alcohol and dangerous recreational drugs should apply without modification to therapeutic, prescription-grade drugs taken in circumstances of personal distress rather than reckless misuse.

It was not in dispute that arson as charged in this case could, on the particular facts, constitute a crime of basic intent for the relevant purpose. The contested issue was whether the defendant's state of mind at the time of the act, alleged to have been substantially altered by the Valium, could legally be raised before the jury and, if so, on what basis.

The primary issue before the Court of Appeal was whether voluntary intoxication caused by the consumption of a non-dangerous therapeutic drug, such as Valium, should be treated identically to voluntary intoxication caused by alcohol or dangerous recreational drugs, so as to preclude any defence to a crime of basic intent as a matter of law.

A subsidiary but closely related issue was whether, if a distinction is to be drawn between different categories of intoxicating substance, the defendant's own conduct in taking the drug was reckless in a legally relevant sense — that is, whether he had knowingly taken a risk that the drug might produce dangerous, disinhibited or unpredictable behaviour.

The Court was also required to determine the correct jury direction to be given in cases involving therapeutic drug intoxication, and whether the failure to give such a direction at trial rendered Hardie's conviction unsafe.

The Court of Appeal began its analysis by revisiting the foundational principles established in DPP v Majewski [1977] AC 443. In Majewski, the House of Lords confirmed the general rule that voluntary intoxication by alcohol or dangerous drugs cannot negative the mens rea required for crimes of basic intent. The rationale is that the deliberate act of getting drunk or taking dangerous drugs is itself a form of recklessness sufficient to substitute for the mental element of the basic intent offence. Majewski established a clear policy position: a defendant cannot excuse criminal conduct on the basis of a state of intoxication that he has deliberately brought upon himself through the consumption of substances known to produce disinhibition, aggression or unpredictable behaviour.

However, the Court in Hardie observed that the Majewski rule rests upon a particular assumption about the nature of the substance consumed. Alcohol and dangerous drugs are known, by common experience and social understanding, to have the propensity to produce unpredictable, aggressive or antisocial behaviour. It is that very characteristic which justifies the conclusion that the person who voluntarily takes such substances is reckless. The taking of such substances is socially and legally recognised as inherently dangerous in terms of the behavioural consequences it may produce.

The Court drew an important contrast with Valium and comparable benzodiazepine tranquillisers. Such drugs are prescribed by medical practitioners precisely because their expected and normal therapeutic effect is to calm, sedate and pacify the patient. Unlike alcohol or stimulant recreational drugs, Valium does not ordinarily produce aggression, unpredictability or disinhibition. On the contrary, its common pharmacological action is to reduce anxiety and suppress agitation. Accordingly, the Court reasoned, it cannot be assumed that a person who takes Valium is, by that act alone, reckless in a manner analogous to a person who drinks alcohol to excess or takes illegal substances.

The Court considered the earlier decision of R v Bailey [1983] 1 WLR 760, a case concerning a diabetic who had failed to eat after taking insulin, thereby inducing a hypoglycaemic state during which he committed a violent offence. In Bailey, the Court of Appeal had held that a self-induced incapacity arising from the failure to take food after an insulin injection did not automatically attract the Majewski rule. What mattered was whether the defendant had been reckless — that is, whether he knew that his conduct, in failing to eat, was likely to make him behave unpredictably or dangerously. The Bailey approach recognised that not all states of self-induced impairment should be treated alike, and that the concept of recklessness must be applied by reference to the particular facts.

Applying the reasoning in Bailey, the Court in Hardie held that the relevant question is not simply whether the defendant voluntarily consumed a substance, but whether, in doing so, he was reckless. Recklessness in this context requires that the defendant took a risk of which he was aware — specifically, the risk that consuming the drug would render him dangerous, aggressive or unpredictably harmful to others. Where a person takes a therapeutic drug that is not known to produce such effects, he cannot be said to have been reckless merely because he took it without a prescription or in a larger dose than recommended, provided he had no reason to anticipate dangerous consequences.

The Court emphasised that Hardie had consumed the Valium in circumstances of emotional crisis, to calm himself down. There was no suggestion that he had taken the drug in order to experience intoxication, to produce disinhibition or to generate any pleasurable altered state. The drug was not a recreational substance; it was a medicine belonging to his partner. His conduct in taking it was not socially analogous to the conduct of a person who deliberately gets drunk or takes recreational drugs knowing the risks of consequent harmful behaviour. The Court therefore held that there was no proper basis for treating his consumption of Valium as an act of recklessness equivalent to the voluntary intoxication condemned in Majewski.

The Court also addressed the trial judge's direction, finding it to have been materially defective. By directing the jury in straightforward Majewski terms — that voluntary intoxication by Valium could not be a defence to a basic intent crime — the trial judge had effectively withdrawn from the jury a question of fact they were entitled to consider. The jury had not been given the opportunity to assess whether, on the particular facts, Hardie's consumption of Valium had been reckless in the legally relevant sense. This was an error of law that rendered the conviction unsafe.

The Court articulated what the correct direction should have been. Where a defendant raises evidence of impairment caused by a therapeutic drug not commonly known to produce dangerous or unpredictable behaviour, the jury should be directed to consider two questions: first, whether the defendant was in fact so affected by the drug as to lack the requisite mental element for the offence; and secondly, whether the defendant was reckless in taking the drug — that is, whether he knew or ought to have appreciated that taking it was likely to render him dangerous or to cause him to act in a harmful manner. Only if the jury find recklessness on the defendant's part does the Majewski route to conviction become available.

The Court took care to confine its reasoning to drugs of the type exemplified by Valium — that is, drugs taken for genuine therapeutic purposes, whose known and normal pharmacological effect is calming rather than disinhibiting. The Court explicitly declined to extend this reasoning to drugs which, whatever their technical medical classification, are known in practice to produce aggressive or unpredictable behaviour in those who take them. The line the Court drew was based not on legal categorisation but on the pharmacological reality of the drug's typical effect on human behaviour.

It is notable that the Court's reasoning implicitly treats the question of recklessness as a matter for the jury, to be assessed on the evidence in each case. This is consistent with the general approach taken in Bailey, where the court recognised that whether a self-induced condition renders a defendant reckless depends on what the defendant knew or ought to have known about the likely effects of his conduct. The Court in Hardie similarly placed emphasis on the defendant's state of knowledge and the reasonable expectations associated with the drug in question.

The Court of Appeal allowed the appeal and quashed the conviction. The trial judge had erred in directing the jury that the Majewski rule applied to Hardie's consumption of Valium as if it were equivalent to intoxication by alcohol or dangerous drugs. The correct position is that intoxication caused by a non-dangerous therapeutic drug does not automatically attract the Majewski rule; rather, the jury must consider whether the defendant was reckless in taking the drug, having regard to the known properties of the substance and the circumstances in which it was consumed.

The Court held that, on the facts as presented, there was no proper basis upon which the jury could have been directed that Hardie was reckless simply by virtue of consuming Valium. He had taken a therapeutic tranquilliser in circumstances of emotional distress, without reason to foresee that it would produce dangerous or harmful behaviour. The defective direction had deprived the jury of a question they were entitled to consider.

The Court ordered a retrial, at which the jury would be directed in accordance with the principles set out in the judgment. At the retrial, the jury would be entitled to consider whether the Valium had impaired Hardie's mental state so as to negative the mens rea of the offence, subject to the question of whether he had been reckless in taking the drug.

R v Hardie establishes a significant and principled exception to the general rule in DPP v Majewski [1977] AC 443. The case recognises that the policy rationale underpinning the Majewski rule — that voluntarily taking a dangerous intoxicant constitutes the recklessness necessary to ground liability for basic intent offences — does not apply uniformly to all forms of self-induced impairment. Where the substance consumed is a therapeutic drug whose normal pharmacological effect is calming rather than disinhibiting, the simple fact of voluntary consumption cannot be treated as substituted recklessness. The case therefore introduces a doctrine of differentiated treatment of intoxicating substances within the criminal law, requiring courts to assess the known properties of the drug in question rather than applying a blanket rule.

The case has particular contemporary relevance in an era of widespread prescription medication. As greater proportions of the population take prescribed drugs for anxiety, depression, sleep disorders and related conditions, the question of how far the criminal law should treat the unintended side effects of such medications as equivalent to voluntary drunkenness has become practically important. Hardie provides the analytical framework for addressing this question: the test is whether the defendant was reckless in taking the drug, assessed by reference to whether he knew or ought to have known that taking it might render him dangerous or harmful to others.

The decision has been integrated into the broader framework governing non-insane automatism, self-induced incapacity and voluntary intoxication. It sits alongside R v Bailey [1983] 1 WLR 760 as authority for the proposition that self-induced impairment does not uniformly preclude reliance on lack of mens rea or automatism, and that the law must attend to the specific circumstances and known risks associated with the defendant's conduct. Both Bailey and Hardie represent a rejection of an overly mechanical application of the Majewski rule, insisting instead on a fact-sensitive inquiry into whether the defendant's conduct in bringing about his impaired state was itself reckless.

Academic commentary has noted that Hardie raises as yet unresolved questions about the precise boundary between drugs